Digitized by the Internet Archive

in 2016

https://archive.org/detaiis/bothaiiavoiume707unse

r

South African National

Biodiversity Institute

JSotbalia

A RECORD OF

CONTRIBUTIONS

PROM THE

NATIONAL HERBARIUM

REPUBLIC OF SOUTH AFRICA

PRETORIA

INDEX TO VOL. VII

EDITED BY

R. A. DYER, M.Sc., D.Sc., F.R.S.S.Af.,

CHIEF, DIVISION OF BOTANY, DEPARTMENT OF AGRICULTURAL AND TECHNICAL SERVICES

AND DIRECTOR OF THE BOTANICAL SURVEY OF THE REPUBLIC OF SOUTH AFRICA

THE GOVERNMENT PRINTER, PRETORIA

G.P.-S.2541564— 1962-63— 1,000.

CONTENTS

Part I. Published 1958. Page

1. Leaf Anatomy in the Grass Series

Phragmitiformes (Harz) Avdulov. Fig.

1-8. J. M. J. deWet 1

2. Notes on the Loganiaceae, Salvado-

raceae and Oleaceae in South Africa.

I.C. Verdoorn 11

3. Notes on the Genus Royena in South

Africa. B.de Winter 17

4. Notes and Interesting Records of

African Plants. Various Authors 21

5. The Protea Species of the Summer

Rainfall Area of South Africa. Tab.

1. J. S. Beard 41

6. The Albizia Species of South Africa.

L. E. Codd 67

7. Studies of Wood Rotting Fungi: 1,

Cultural Characteristics of some Com-

mon Species. Tab. 1-5. G. C. A.

van der Westhuizen 83

8. New and Interesting Records of South

African Fungi. Part HI. Fig. 1-4.

P.H.B. Talbot 109

9. The Genera Craterellus, Cymatoderma

(Cladoderris) and Thelephora in South

Africa. Fig. 1-9. P. H. B. Talbot. . 117

10. Studies of Some South African

Resupinate Hymenomycetes. Part II.

Fig. 1-38. Including a*. P. H. B.

Talbot 131

* Species Index of Recorded South

African Thelephoraceae 172

Part II. Published

1 . Notes on Some Species of Cladophora

from South Africa. Tab. 1-3; fig. 1-9.

R. H. Simons 189

2. Notes on Aristothamnion purpuri-

forum (Kutz.) J. Ag. Fig. 1-5. R. H.

Simons 201

3. Virus Diseases of Lupins. Tab. 1-5.

Patricia J. Klesser 207

4. Virus Diseases of Cowpeas. Patricia

J. Klesser 233

5. Virus Diseases of Peas and Sweet Peas.

Tab. 1-8. Patricia J. Klesser 253

6. Virus Diseases of Broad Beans.

Tab. 1-2. Patricia .1. Klesser 283

7. Chromosome Morphology in Knipho-

fia. J. M.J.deWet 295

8. Leaf Anatomy of Six South African

Grass Genera. J. M. J. de Wet 299

9. Leaf Anatomy and Morphology in

South African species of Danthonia.

J. M. J. de Wet 303

10. Culm Anatomy in Relation to Tax-

onomy. .1. M.J.deWet 311

1 1. Notes on the Sapotaccae of Southern

Africa. A. D. J. Meeu.se 317

12. An Enumeration of the Maytenus

Species of Southern Africa. W. Marais 381

Page

13. A New Genus of Gramineae. B. de

Winter 387

14. The Opening of the Pretoria National

Botanic Garden. R. A. Dyer 391

15. Notes on Some South African Species

of the Genus Euclea. B. de Winter. . 403

16. Notes and New Records of African

Flowering Plants. Various Authors 407

Part III. Published 1961.

1 . Homonyms in the Prodromi of Thun-

berg and Burman. J. E. Dandy 427

2. South African Labiatae. L. E. Codd 429

3. South African Species of Satureia.

D. J. B. Killick 435

4. X Ruttyruspolia, A Natural Intergene-

ric Hybrid in Acanthaceae. A. D. J.

Meeuse and J. M. J. de Wet 439

5. Notes and New Records of African

Plants 443

Acanthaceae. A. D. J. Meeuse and

A. A. Obermeyer.

Apocynaceae. L. E. Codd.

Araceae. Cythna Letty.

Ebenaceae. B. de Winter.

Gentianaceae. I. C. Verdoorn and W.

Marais.

Gesneriaceae. R. Story.

Gramineae. B. de Winter.

Hypoxidaceae. D. J. B. Killick.

Liliaceae. A. A. Obermeyer.

Malvaceae. A. D. J. Meeuse.

Plumbaginaceae. R. A. Dyer.

Scrophulariaceae. O. Overkott.

Turneraceae. R. Story.

6. The Virus Diseases of Crotalaria,

Glycine and Medicago Species. Tab.

1-5. Patricia J. Klesser 497

7. The Virus Diseases of Beans. Tab.

1-8. Patricia J. Klesser 521

8. A Contribution to Our Knowledge of

the Fresh-water Algae of the Transvaal

Province. Tab. 1-37. Enumeration of

the Taxa. Vegetative Reproduction

in Desmids. M. I. Claassen 559

9. Review of the Flora Zambesiaca, Vol.

I, Part 1, 1960. L. E. Codd 667

10. Addendum: Cladophora isaacii. R.

H. Simons 668

Part IV. Published 1962.

1 . A Revision of the South African

Species of Anthericum, Chlorophytum

and Trachyandra. A. Amelia Ober-

meyer 669

The Identity of the Fungus Causing

Anthracnose of Olives in South Africa.

G. J. M. A. Gorter 769

Leaf Anatomy in the Grass Series Phragmitiformes

(Harz) Avdulov.

By

J. M. J. de Wet.

The taxonomy of the Gramineae is by no means straightforward and is com-

plicated by many factors. Due largely to a lack of data there is still no general agreement

regarding the subdivision of the family into tribes or even subfamilies. The flowers

are not showy, the component parts are often much reduced and the vegetative parts

have reached a high degree of specialization. A close examination of the glumes,

lemmas, paleae, lodicules, androecium and gynoecium reveals a wide degree of diversity

among these organs. It is, therefore, not surprising that most sj^stems of classification

are based mainly on spikelet morphology. Recently, however, evidence regarding the

relations of the species and genera has become available which differs conspicuously

from most older observations. Leaf anatomy (Duval-Jouve, 1875, Pee-Laby, 1898,

Grob, 1895-1897 and Prat, 1932, 1936) was proved to be useful in grass taxonomy.

Similarly, Avdulov (1931) and Hunter (1934) indicated a close correlation between

cytological data and the physiological observations of Trucel (1858) and Harz (1880-

1882). More recently, Reeder (1946, 1953) combining with his own observations the

data of Burns (1892) and Kennedy (1899) indicated that embryo anatomical observations

may be used taxonomically. Reeder and von Maltzhan (1953) also pointed out that

different types of root hair development were observed which allow for a subdivision

of the Gramineae on the same lines as pointed out by Avdulov (1931) and Prat (1936).

It is generally agreed among modern taxonomists that anatomy, cytology, genetics,

ecology and physiology provide more reliable clues towards the true relationships in

the Gramineae than spikelet morphology alone. Subdivisions based on gross morpho-

logy is essential for the easy and rapid identication of individual plants. Such is the

recent classification of Pilger (1954). The leaf anatomy of the genera included in his

subfamilies Festucoideae Hitchc., Micrairoideae Pilger, Oryzoideae Parodi, Olyroideae

Pilger, Bambusoideae Krause and Anomochloideae Pilger will be discussed. An

attempt will be made to evaluate the available data and supplement them with some

personal observations in determining the phylogenetic position of some genera of

controversial affinity.

The terms Panicoid type, Festucoid type, Eragrostoid type and Oryzoid type of

internal leaf anatomy and epidermal histology were fully discussed by Prat (1936) and

de Wet (1954, 1956a). The reader is referred to the diagrams later in this paper for

more details.

Material and Methods: Specimens investigated were obtained from the

National Herbarium, Pretoria. Slides were prepared according to the procedure out-

lined by Prat (1948). Drawings were made with the aid of a Camera Lucida.

Subfamily Festucoideae.

1. Festuceae - Festucinae: Lintonia and Cynosurus were found to be Festucoid in all

respects. The genera Urochlaena, Lasiochloa and Plagiochloa are characterized by

the Festucoid type of internal leaf anatomy and the Panicoid type of epidermis. They

are further characterized by truncate lodicules and ciliated ligules.

2

The genera Lamai ckia, Briza, Dactylis, Poa, Nephelochloa, Cutandia, Scleropoa,

Vulpia and Festuca are Festucoid in all respects according to Prat (1936). Pmt (1936)

also indicated a Panicoid type of leaf anatomy for Uniola and Distichlis, and a Festucoid

type of chlorophyll distribution with variable epidermal traits in Aeluropus and

Fingerhutia. Moflfet and Hurcombe (1949) demonstrated a Eragrostoid type of leat

anatomy and cytology for Tetrachne.

Festuceae-Triodinae: Plectrachne and Notochloa are unknown cytologically and

anatomically. Brown (1950) refers Astrebla to the Chlorideae because of its small

chromosomes and basic number of n = 10. Triodia is characterized by n = 8 (Brown,

1950) and Burbidge (1946) demonstrated a Festucoid type of leaf anatomy.

Festuceae — Sesleriinae : The leaf anatomy of Entoplocainia and Oi'cuttia is of the

Eragrostoid type. They are further characterized by ciliated ligules, large embryos and

the grains are loosely enclosed by the lemmas and palea. Prat (1936) demonstrates a

Festucoid type of epidermis and internal anatomy for the genera Seslevia and Echinaiia.

Myers (1947) lists large chromosomes with basic numbers n = 7 and n = 9 in these

genera respectively. No data regarding chromosomes, anatomy and lodicules are

available for the remaining genera included in this subtribe.

Festuceae - Beckmanniinae : Includes a single genus Beckniannia usually referred to the

Chlorideae. Von Tieghem (1897), Avdulov (1931), Prat (1936) and Krishnaswamy

(1941) indicated relations with the tribe Agrosteae as recognized by Hubbard (1934).

Reeder (1953) studied the embryo anatomy and indicates relations with the genera

Alopecurus, Phleum and Polypogon.

Festuceae -Melicinae: Only two of the genera included in this subtribe have been

studied. Morphologically Melica and Schizachne differ from the Festuceae in spikelet

morphology. Their lodicules are truncate and vasculated, their stigma brpches are

broad and dichotomous, their caryopsis do not adhere to the palea and their embryos

are relatively large. Both genera are characterized by the Panicoid type of histology.

In the case of Melica, Avdulov (1931) demonstrated large chromosomes in multiples

of n = 9.

Festuceae - Glyceriinae: Pleuropogon and Glyceria resemble the Festuceae in super-

ficial characters but do not seem to be related to members of Festuceae. The chromo-

somes are medium large and Avdulov (1931) demonstrated n = 9 in Glyceria and

n = 8 in Pleuropogon. Leaf anatomy is of the Festucoid type. Poagrostis has been

transferred by de Wet (1956a) to the Danthonieae. The genus Puccinella on the other-

hand is Festucoid in all respects (Prat, 1936). Further-more, in the latter genus the

lodicules have appendages, the embryo is small and the grain is tightly embraced by

the hardened lemma and palea.

Festuceae - Psilurinae: The single genus P.silurus is characterized by a Festucoid type

of leaf anatomy (Prat, 1936) and Avdulov (1931) demonstrated large chromosomes and

a basic number of n =- 7. Hubbard (1934) includes Psilurus together with Pholiurus,

Lepturus and Ischnurus in the tribe Leptureae. Prat (1936) demonstrated that thrs

tribe as recognized by Hubbard (1934) is completely Festucoid in all respects. Pilger

(1954) retains Pholiurus in the subfamily Festucoideae tribe Monermeae. Lepturus

and Ischnurus, however, Pilger (1954) refers to the Chlorideae, which evidently is

incorrect.

Festuceae - Loliinae: Includes a single genus Folium which is Festucoid in all respects.

The cytology was investigated by Faworski (1927) and leaf anatomy by Prat (1936).

Festuceae - Bromineae: A truly Festucoid tribe according to Avdulov (1931), Prat

(1936) and Myers (1947).

Festuceae - Streptogyninae: The single genus Streptogyne is unknown cytologically

and anatomically.

3

Festuceae - Centothecinae: Both Pilger (1954) and Hubbard (1934) regards the Cen-

tothecinae as a subtribe of the Festuceae. It definitely does not belong here. Prat

(1936) illustrates an Oryzoid type of leaf epidermis for Centhotheca, Lophatherum and

Zeugites. On the other hand, Orthoclada was found to have the Panicoid type of

epidermis. Chlorophyll distribution is of the Festucoid type in all these genera.

Avdulov (1931) demonstrated n = 12 in Centotheca, a cytology quite foreign to the

Festuceae. In Centotheca the ligule is scarious, a Festucoid type of character, but

the two lodicules are cuneate and without appendages which again are Panicoid charac-

teristics.

Festuceae - PommercuUinae: The genus PommercuUa has not been studied.

Festuceae - Monanthochloinae: The single genus Monanthochloe is characterized by

the Festucoid type of leaf anatomy but it differs from the Festuceae in spikelet

morphology. The lodicules are truncate and the stigma branches are elongated.

2. Triticeae - Henrardiinae : The genus Henrardia is unknown cytologically and anato-

mically.

Triticeae - Eutriticinae: Prat (1936) and Avdulov (1931) indicated that the genera

included in this tribe are characterized by all the peculiarities typical of the subfamily

Festucoideae.

3. Monermeae: Only Pholiurus of the genera included in this tribe is known cytologically

and anatomically. Prat (1936) indicated that this genus is Festucoid in all respects.

4. Aveneae - Aveninae: This subtribe has been fully discussed by de Wet (1954, 1956a).

Schismus, Afrachneria and Prionanthiwn occupy an artificial position in the Aveneae.

Aveneae - Duthiieinae: The single genus Duthiea has not been studied.

Aveneae - Danthoniinae: This subtribe was discussed by de Wet (1956a). A number

of the genera included in this subtribe are still unknown anatomically and cytologically.

Their phylogenetic position in the Gramineae is still uncertain.

5. Arundineae: A primitive tribe closely allied to the Danthoniinae (de Wet, 1956a).

The morphological characters are peculiar and differ from those of typical members

belonging to the subfamily Festucoideae. The lemmas are primitively awned from the

sinus of the lobes, the ligule is ciliated, lodicules are more or less cuneate and vasculated

and the stigma branches tend to be expanded and flattened in most species. Avdulov

(1931) indicated a basic chromosone number of n = 12. Leaf anatomy. Prat (1936)

and de Wet (1954) is of the Panicoid type.

6. Arundinelleae: The genera Arundinella, Trichopteryx, Loudetia and Tristachya are

Panicoid in respect to leaf anatomy and morphology but resemble the Phragmitiformes

of Avdulov (1931) in cytology (de Wet, 1954). Danthoniopsis on the other hand

according to de Wet (1954) is Panicoid in all respects. The genus Gilgiochloa has not

been investigated.

7. Thysanolaeneae: The single genus Thysanolaena is characterized by the Panicoid

type of epidermis and chlorophyll distribution. It is also Panicoid in respect of the

grain which is free from the lemma and palea, the large embryo and lodicules which

are cuneate. Typical Festucoid characters are the disarticulation of the spikelet above

the glumes and the scarious rim which forms the ligule.

8. Phalarideae: Ehrharta and Microlaena are characterized by the Oryzoid type of

leaf anatomy (Prat, 1936) and cytology (Avdulov, 1931). The remaining genera.

Prat (1936), are typically Festucoid.

4

9. Pappophoreae: In the genera Schmidtia and Enneapogon the siliceous cells are

Panicoid but the bicellular hairs are typically Eragrostoid. Prat (1936) recorded an

Eragrostoid type of epidermis of Cottea and Pappophorum. Chlorophyll distribution

is of the Festucoid type.

Small chromosomes and a basic number of n = 10 were recorded by Covas (1945)

in the genera Cottea and Pappophorum. For the genera Schmidtia and Enneapogon

de Wet (1956b) demonstrated a basic chromosome number of n = 9. In the latter

two genera the stigma branches are stalked, the lodicules are cuneate and the embryo

occupies I or more of the length of the grain.

10. Stipeae: The chromosomes are mostly small but medium large in Piptochaetium

and a few species of Stipa. The basic chromosome numbers are n = 1 1 in Piptochae-

tium and some species of Oryzopsis, while n = 12 in some other species of the latter

genus (Johnson, 1945). In Stipa, Myers (1947), Brown (1949, 1951) and Ono and

Tateoka (1953) demonstrate an aneuploid series, perhaps originally derived from n = 6

or n = 5. Internal leaf anatomy is of the Festucoid type and epidermal traits are

variable (Prat, 1936).

11. Nardeae: Prat (1936) demonstrated a Festucoid type of chlorophyll distribution

and Eragrostoid type of epidermis. Avdulov (1931) observed a basic chromosome

number of n “ 13 in the monotypic genus Nardus stricta.

12. Coleantheae: The single genus Coleanthus is Eragrostoid in respect of leaf anatomy.

Lack of material made it impossible to study spikelet morphology.

13. Lygeae: The monotypic genus Lygaeum sparteum has n = 20 and medium large

chromosomes (Myers, 1947). Leaf anatomy is unknown.

14. Phyllorachieae: The genera included in this tribe were fully discussed by de Wet

(1956b). It was pointed out that this tribe is closely related to the Oryzeae.

15. Parianeae: Contains a single tropical New World genus Pariana which is unknown

anatomically and cytologically.

Subfamily Micraroideae: Tribe Micraireae: Includes a single genus Micraira

which was not available for study.

Subfamily Oryzoideae: Tribe Oryzeae: The genera Oryza, Leersia, Potamophila

and Zizania are all characterized by the Oryzoid type of epidermis and Festucoid type

of chlorophyll distribution. Kuwado (1910) and Ramanujam (1938) demonstrated

a basic chromosome number of n — 12 in Oryza. A similar basic chromosome number

and medium large chromosomes are also present in Leersia (Ramanujam, 1938). The

remaining genera included by Pilger (1954) in this subfamily are unknown anatomically

and cytologically.

Subfamily Olyroideae: Tribe Olyreae: The genera Phams and Leptaspis are

characterized by the Panicoid type of epidermis. These genera are included by Hubbard

(1934) in the tribe Phareae. Of the other genera belonging to the Olyreae, Raddia,

Lithachne and Olyra are characterized by the Oryzoid type of epidermis. Chlorophyll

is distributed according to the Festucoid type in all genera investigated.

Subfamily Bambusoideae: This subfamily has been fully investigated by Prat

( 1936) who indicated a very peculiar type of leaf anatomy. Pilger (1954) also includes

the tribe Streptochaeteae of Hubbard (1934) in this subfamily. The genus Strepto-

chaela, the only member of this tribe, has not been studied anatomically. Stebbins

(Unpubl.) reports n - 1 1 chromosomes in one of the two known species of this genus.

5

Subfamily Anomochloideae : This includes a single genus, Anomochloa. No

data regarding cytology, anatomy or detailed spikelet morphology is available.

One tribe of the subfamily Eragrostoideae the Aristideae as recognized by Pilger

(1954) also needs to be mentioned. Pilger (1954) includes the genera Aristida,

Amphipogon and Diplopogon in this tribe. Only the genus Aristida was available for

study. The basic chromosome number is n = 11 and n 12 in Aristida according

to de Wet (1954, 1956b). A further indication that this genus does not belong in the

Eragrostoideae comes from a study of leaf anatomy. The chlorophyll distribution is

of the Pariicoid type, but epidermal traits are variable. For instance in A. ciliata

it is Eragrostoid, in A. barbicollis, A. bipartita and A. aequiglumus the epidermis is

of the Panicoid type and in species such as A. gracilior, A. obtusa and A. sericans the

epidermis is typically of the Festucoid type.

Subdivision of the Phragmitiformes : The family Gramineae has been sub-

divided at various times into two or more subfamilies or series. As early as 1810,

Brown noted two main groups, the Panicoideae and Pooideae. Pilger (1954) recognizes

nine subfamilies. Combining the evidence presented by Harz (1880-1882), Avdulov

(1939), Prat (1936) and the data presented by numerous taxonomists on spikelet mor-

phology at least four series become evident. The characteristics of these, the Phrag-

mitiformes, Paniciformes, Eragrostiformes and Festuciformes were discussed by de

Wet (1954).

The subfamily Festucoideae as recognized by Pilger (1954) includes a diverse

group of species and genera. In the classification of the Gramineae the true Panicoids,

true Festucoids and the true Eragrostoid-chloridoids constitute no problem. A number

of tribes and genera, however, do not belong to any one of these distinct groups and

these are lumped together into a miscellaneous series the Phragmitiformes (Avdulov,

1931). The more outstanding tribes and genera involved are discussed below. Their

position in grass phytogeny was pointed out by Stebbins (1956).

1. Streptochaeteae: Includes a single tropical genus, Streptochaeta. It has 3 lodi-

cules, 6 stamens, 3 stigmas and the palea is split nearly to the base. Bentham (1881)

refers this genus to the Paniceae and Hackel (1889) to the Oryzeae. Hubbard (1934)

indicates that it is distinct from both these tribes.

2. Bamuseae: The genera are distributed throughout the tropics and subtropics. A

few genera extend into the temperate regions. As pointed out by Hubbard (1934) it

includes genera with the most primitive floral structure, but they are highly specialized

vegetatively. Spikelets are all alike, lemmas awnless or rarely awned from the tip,

lodicules usually 3, and 3, 6 or more stamens are present. Most of the species studied

are high polyploids on the basis of n = 12 (Myers, 1947). Hunter (1934) indicates

n = 9 in a species of Arundinaria. Internal leaf anatomy. Prat (1936) is characteristic.

Epidermal traits are variable, but more or less of the Panicoid type.

3. Phareae: This is a tropical tribe, probably of close common ancestry with the

Bambuseae. As is recognized by Hubbard (1934) the genera Pharus and Leptaspis

are included. The spikelets are 1 -flowered, awnless and unisexual. The lodicules

are mostly 3, stigmas 3 and stamens 6. The chromosomes are small and the basic

number is n = 12 (Valencia ex Stebbins unpublished). The leaf anatomy is Panicoid.

4. Olyreae: This tribe differs from the Phareae in the reduction of the stamens to 3 or

2 and in the Oryzoid type of epidermis. It includes the genera as recognized by Hubbard

(1934).

5. Centotheceae: Anatomical characters are Festucoid in respect to chlorophyll distri-

bution, but the epidermal traits are either Panicoid or Oryzoid. This tribe is recognized

to include the genera Centotheca, Lophatherum, Zeugites, Orthoclada and Magastachya.

Both Hubbard (1934) and Pilger (1954) combine these genera as a subtribe of the

Festuceae.

6

6. Parianeae: Contains a single New world genus which was not studied. Hubbard

(1934) indicates that Pariana resembles the Hordeae in superficial inflorescence charac-

ters. This genus appears to present a highly specialized end line.

7. Anomochloeae: This tribe is recognized in the sense of Hubbard (1934) to include

Anomochloa merantoidea of the Brazilian forest. This genus is incompletely known

and it shows no close affinity with any other genus.

8. Phyllorachieae: Chlorophyll distribution is of the Festucoid type, epidermis of the

Oryzoid type and the chromosomes are small in multiples of n = 12. As was done

by Hubbard (1939) the genera PhyUorachis and Humbertochloa are included. This

tribe resembles the Oryzeae closely not only in anatomy and cytology, but also in spikelet

morphology and starch grains. For this reason it would appear as if they developed

from a common stock and might well be united.

9. Oryzeae: The genera Ehrharta, Microlaena and Tetranhena should be added to

the tribe as recognized by Hubbard (1934). The starch grains are angular, the chromo-

somes small and in multiples of 12 and the leaf anatomy is typical. Ehrharta, Microlaena

and Tetrarrhena are southern in distribution and represent the most primitive members

of the tribe. The tribe as a whole has reached a high degree of spesialization in the

tropics.

10. Anindineae: Is a very large tribe. To the genera included by Pilger (1954), the

tribe Danthonieae as recognized by de Wet (1956a) could be added. In other words,

as recognized by Pilger (1954) the subtribe Danthoniinae of the Aveneae, the genera

Schismus, Afrachneria and Priononthium of the Aveneae subtribe Aveninae and from

the Festuceae subtribe Glyceriinae the genus Poagrostis. Further also the genera

Arundinella, Trichopteryx, Loudetia, and Tristachya from Hubbard’s (1934) tribe the

Arundinelleae. These data were fully discussed by de Wet (1954, 1956a and 1956b).

The leaf anatomy may be either Panicoid or Festucoid, the chromosomes are medium

large and in multiples of 6 and 7 (Avdulov, 1931, Calder, 1937, Stebbins and Love,

1941 and de Wet, 1954). The genera are mostly temperate in distribution with a few

montane species extending to the tropics.

1 1. Stipeae: Is recognized in the sense of Hubbard (1934) to include the genus Aristida,

except that Melica is removed. This tribe may be subdivided into two distinct groups

which appear to have developed independantly. The one group includes Aristida and

the other the remaining genera.

12. Unioleae: Pilger (1954) includes Uniola, DistichJis, Monanthochloe, Vaseyochloa

and Aeluropus in various subtribes of the Festuceae. The genus Jouvea which also

belongs here, Pilger refers to the tribe Jouveneae of the Eragrostoideae. These genera

from a natural group in having truncate lemmas with several parallel veins which often

end in awns. They differ from members of the Festucoideae in leaf anatomy, cytology

and spikelet morphology.

13. Pappophoreae: Besides the four genera placed by Hubbard (1934) in this tribe

Orcuttia and Neostapfia are added. The epidermis is more or less Eragrostoid and the

chromosomes are small in multiples of n = 10.

14. 15 and 16. Lygeae, Nardeae and Micraireae: These tribes consist of monotypic

genera, all of which need further investigation. With the present evidence available

they are difficult to place, but do not seem to be related to each other.

Relic Genera: A number of relic genera still remain to be discussed. These

are Schizaclme, Melica, Pleuropogon, Glyceria, Fingerhutia and Coleanthus. They are

temperate in distribution and have vegetative characters resembling the true members

of the Festuciformes. In the case of Melica, the chromosomes are as large as those of

members belonging to the Festuciformes, but the basic chromosome number appears

to be n = 9 (Avdulov, 1931).

7

That they do not belong in the Festuciformes is quite evident. Their lodicules are

truncate and often vasculated, their stigma branches are broad and dichotomous,

their caryopses do not adhere to the palea and the embryo is relatively large. In

Glyceria and Pleuropogon the chromosomes are medium large with basic numbers 9

and 8 respectively (Avdulov, 1931).

The chlorophyll distribution is of the Festucoid type in Melica and of the Festucoid

type in Glyceria and Pleuropogon. In Melica the epidermis is also of the Panicoid

type and of the Festucoid type in the latter two genera. In Fingerhutia the epidermis is

either of the Festucoid or the Eragrostoid type. In Coeleanthus the leaf anatomy is

typically Eragrostoid.

Morphological, cytological and anatomical data suggest an affinity with the

Pappophoreae. It would appear, however, as if they originated from the primitive

Arundineae stock, as did the Pappophoreae. These genera, although showing super-

ficial resemblances to the Festuceae, are not closely related to them.

Table 1. — A tentative key to the classification of the Phragmitiformes.*

A. Leaves with cross veins; tropical. B. Palea split nearly to the base: Strepto-

chaeteae. BB. Palea bilobed at the apex only. C. Plants arboreal: Bambuseae.

CC. Plants not arboreal. D. Lodicules mostly 3; stigmas 3 or 2; spikelet with one

functional flower. E. Stigmas 3; stamens 6: Phareae. EE. Stigmas 2; stamens

2 or 3: Olyreae. DD. Lodicules 2; stigmas 2; spikelet with more than one functional

flower: Centotheceae. AA. Leaves without cross veins; tropical or temperate.

F. Stamens 6 or more : Parianeae. G. Spikelets often enclosed in a sheath-like spathe.

H. Stigma 1; stamens 4: Anomochloeae . HH. Stigmas 2; stamens 6-4: Phyllo-

rachieae, Oryzeae. FF. Stamens 3. GG. Spikelets not enclosed by a spathe.

I. Stigmas 2. J. Lemmas with a single awn from the sinus between the lobes, awns

reduced in some specialized groups. K. Spikelets with several florets: Arundineae.

KK. Spikelets with a single functional floret ; lemma indurated : Stipeae. JJ. Lemmas

awnless or with several veins excurrent into awns. M. Lemmas awnless, laterally

compressed: Unioleae. MM. Lemmas with several parallel veins, which usually end

in awns; rounded on the back: Pappophoreae. II. Stigma 1. N. Spikelet solitary,

surrounded by a sheath-like spathe: Lygeae. NN. Spikelets in spikes: Nardeae.

SUMMARY.

The leaf anatomy was investigated of a number of genera belonging to the series

Phragmitiformes (Harz) Avdulov. Anatomical characters are found to be variable.

Chlorophyll distribution is of the Panicoid or the Festucoid type. Epidermal traits

may be of the Panicoid, Festucoid, Eragrostoid or Oryzoid type. The subfamilies

Festucoideae, Oryzoideae, Olyroideae, Bambusoideae, Micrairoideae and Anomo-

chloideae as recognized by Pilger (1954) are discussed. A number of genera and even

whole tribes included by Pilger (1954) in the subfamily Festucoideae together with the

other subfamilies are regarded as belonging to the series Phragmitiformes. Anatomical,

cytological and morphological data support such a subdivision. The genera Jouvea

and Aristida in the subfamily Eragrostoideae of Pilger are also transferred to the

Phragmitiformes. The Phragmitiformes are classified and the affinities of the tribes

pointed out. A key to the identification of these tribes is tentatively proposed.

It was pointed out that a miscellaneous group of grasses are lumped together in

the Phragmitiformes. Most of these genera are more or less related to the true panicoids,

true festucoids or the true eragrostoid-chloridoids. Others appear to be relic genera

with no obvious affinities to these distinct groups of grasses.

* A number of relic genera and the tribe Micraireae have been left out of this key.

9

Literature Cited.

Avdulov, N. P., 1931 Karyo-systematische untersuchung der Familie Gramineen.

Bull. Appl. Bot. Suppl. 43: 1-428.

Brown, W. V., 1949 A cytological study of cleistogamous Stipa kiicotricha.

Madrona 10: 97-107.

Brown, W. V., 1950 A cytological study of some Texas Gramineae. Bull. Torrey

Bot. Club 77: 63-76.

Brown, W. V., 1951 Chromosome numbers in some Texas grasses. Bull. Torrey

Bot. Club 78 : 292-299.

Burbidge, N. T., 1946 Morphology and anatomy of the Western Australia species

of Triodia R. Br. ii. Internal anatomy of leaves. Trans.

Roy. Soc. So. Austr. 70: 221-236.

Burns, E., 1892 Der Grassembryo. Flora 76: 1-33.

Calder, J. W., 1937 A cytological study of some New Zealand species and Varieties

of Danthonia. Jour. Linn. Soc. London 51: 1-9.

DE Wet, J. M. J., 1954 The genus Danthonia in grass Phylogeny. Amer. Jour.

Bot. 41: 204-211.

DE Wet, j. M. J., 1956a Leaf anatomy and phylogeny in the tribe Danthonieae.

Amer. Jour. Bot. 43: 175-182.

DE Wet, j. M. J., 1956b Chromosome numbers in Transvaal grasses. Cytologia 21:

1-10.

Duval-Jouve, M. j., 1875 Histotaxie des feuilles de Graminees. Ann. Sci. Nat. Bot.

15: 294-371.

Faworski, N., 1927 Vergleichende Karyologische Untersuchung eigener Arten

\on Loliiim. Planta 3: 282-291.

Grob, a., 1896-1897 Beitrage zur Anatomic der Epidermis der Gramineen Blatter.

Beibl. Bot. Jahrb. 7: 1 (Ex Avdulov, 1931).

Harz., 1880-1882 Beitrage zur Systematik der Gramineen. Linneae 43:

1-30 (Ex Avdulov, 1931).

Hubbard, C. E., 1934 Gramineae. J. Hutchinson: The families of Elowering

plants 2: 199-299.

Hubbard, C. E., 1939 Phyllorachis sagittata Trimen. Hooker’s leones Plantarum

Tab. 3386.

Hubbard, C. E., 1946 Henrardia, a new genus of Gramineae. Blumea Suppl. 3:

10-21.

Hunter, A. W. S., 1934 A Karyo-systematic investigation in the Gramineae. Canad.

Jour. Res. 11: 213-224.

Johnson, B. L., 1945 Cytotaxonomic study in Oryzopsis. Bot. Gaz. 107: 1-32.

Kennedy, P. B., 1898 The structure of the caryopsis of grasses with reference to

their morphology and classification. U.S. Dept. Agr.

Div. Agrost. Bull. 19: 1-44.

LEGEND TO FIGURES.

Figures 1-2. — Semidiagramatical drawings of leaf anatomy types, x 500.

Figure 1 . — Panicoid type.

Figure 2. — Festucoid type.

Figures 3-8. — -Semidiagramatical drawings of epidermal types, x 500.

Figure 3. — Panicoid type; Aristida barbicoUis.

Figure 4.— Panicoid type; Aristida bipartita.

Figure 5. — Eragrostoid type; Aristida ciliata.

Figure 6. — Eragrostoid type; Fingerhutia sesleriaeformis.

Figure 7. — Festucoid type; Aristida gracilior.

Figure 8. — Oryzoid type; Olyra lati folia.

10

Krishnaswamy, N., 1941

Kuwada, Y., 1910

Moffet, a. a. and R. Hurcombe, 1949

Myers, W. M., 1947

Ono, H. and J. Tateoka, 1953

Pee-Laby, E. 1898

PiLGER, R., 1954

Prat, H., 1932

Prat, H., 1936

Prat, H., 1948

Ramanujam, S., 1938

Reeder, J. R., 1946

Reeder, J. R., 1953

Reeder, J. R. and K. von Maltzahn,

1953

Stebbins, G. L. Jr. 1956

Stebbins, G. L. Jr. and R. M. Love,

1941.

Tieghem, P. van 1897

Trucel, M. a., 1858

Untersuchungen zur Cytologic und systematik der Gramineen.

Beih. Bot. Centr. Blatt 60; 1-56.

A cytological study of Oryza saliva L. Bot. Magaz. Tokyo

24: 267-281.

Chromosome numbers of South African grasses. Heredity

- 3: 369-374.

Cytology and genetics of forage grasses. Bot. Rev. 13:

319-421.

Karyo-taxonomy in Poaceae J. Chromosomes and taxonomic

relations in some Japanese grasses. Bot. Magaz. Tokyo

LXVI: 18-27.

fitude anatomique de la Feuille des graminees de la France.

Ann Sci. Nat. Bot. 8: 277-346.

Das System der Gramineae. Bot. Jahrb. 76; 281-384.

L’epiderme des Graminees, etude anatomique et systematique.

Ann. Sci. Nat. Bot. 13: 117-187.

La systematique des Graminees. Ann. Sci. Nat. Bot. 18:

165-258.

General features of the epidermis in Zea Mays. Ann.

Missouri Bot. Gard. 35: 341-351.

Cytogenetical studies in the Oryzeae. 1. Chromosome

studies in the Oryzeae.

Additional evidence of affinities between Eragrostis and

certain Chlorideae. (Abstract) Amer. Jour. Bot. 33:

p. 843.

Affinities of the grass genus Beckmannia Hort. Bull. Torrey

Bot. Club 80: 187-196.

Taconomic significance of root-hair development in the

Graimineae. Proc. Nat. Acad. Sci. Wash. 39: 593-598.

Cytogenetics and evolution of the grass family. Amer. Jour.

Bot. 43: 890-905.

A cytological study of California forage grasses. Amer.

Jour. Bot. 28: 371-382.

Morphologic de I’embryon et de la plantule chez les Graminees

et les Cyperacees. Ann. Sci. Nat. Bot. 8: 254-309.

Des formations vesiculaires dans les celluces vegetales. Ann.

Sci. Nat. Bot. 7, 10: 205-282.

11

Notes on the Loganiaceae, Salvadoraceae and

Oleaceae in South Africa.

By

I. C. Verdoorn.

1. Nomination of neotype for Strychnos decussata (Pappe) Gilg,

When describing Atherstonea decussata, the basionym of the above species,

Pappe wrote at the foot of the botanical description (see Silva Capensis ed. 2: 29,

1862) — “ Grows in the thickets and forests of Uitenhage, Olifants Hoek and elsewhere

in the districts of the Eastern Province”. In a footnote he wrote “ named in honour

of Dr. W. G. Atherstone, of Grahamstown, a gentleman whose merits rendered to

South African Botany and Geology rank high, and to whom I am under great obligations

for valuable information towards this work”. The type or types of this species would

therefore be a specimen or specimens from the mentioned localities collected by Pappe

or annotated by him as being Atherstonea decussata. To date no such specimens have

been found and it is probable that no material was preserved. From the excellent

description there is no mistaking what tree Pappe was describing so there can be no

doubt that it is the same species that Harvey described and figured in his Thesaurus

Capensis 2: 41 (1863) as Strychnos atherstonei with Dr. Pappe’s name as a synonym.

One of the specimens cited by Harvey is “ On the Kowie, Dr. Atherstone ”.

Through the kindness of Prof. D. A. Webb, Trinity College, Dublin, I was able

to examine this specimen. According to the notes and labels on it, the sheet contains

two gatherings, both by Dr. Atherstone from a tree on the Kowie in the Bathurst

district. The flowering material, which is mounted on the lower half of the sheet

(and spare material which is in an attached envelope). Dr. Atherstone states was picked

by him in November 1863 from the same tree from which he had sent a fruiting specimen

earlier in the year. (Note. — This is a year after the publication of Pappe’s species

and therefore cannot be the specimen on which he based Atherstonea decussata.) This

fruiting specimen is mounted on the upper half of the sheet but the fruit is missing.

It is the flowering material that I wish to nominate as the neotype of the species.

There is duplicate material of it in the Kew Herbarium and a scrap in the National

Herbarium, Pretoria. Neotype of Strychnos decussata (Pappe) Gilg.: Cape, Kowie,

Atherstone s.n., flowering material, (TCD, neo.!; K, PRE, iso-neo).

2. The relationship between Strychnos innocua Del., S. dysophylla Benth.

AND S. GERRARDII N.E.BR.

In the Kew Bulletin 1938, page 46 Bullock and Bruce reduced Strychnos dysophylla

Benth., together with a number of other described species and varieties, to synonymy

of S. innocua Del. In 1956, after studying the tropical African specimens involved.

Miss E. A. Bruce decided to exclude S. dysophylla Benth. (and three species which she

regarded as being synonymous with S. dysophylla) from that long list of synonyms and

again treat it as a separate species. Her views were published posthumously in the

Kew Bulletin of that year on page 270 under notes by Bruce & Lewis. The main reason

given is that it is “ undesirable to conceal the variation ” in this group.

12

A study of the South African material both in herbaria and in the wild leads to

the conclusion that the broader view of S. itmocua Del. can be more satisfactorily

applied to the whole complex. If S. dysophylla Benth., with its southern distribution,

is restored to specific rank the following are some of the difficulties that are encountered.

i. Specimens in the southern regions, prepared from different parts of the same

tree or in different seasons, could be identified from the key, some as S. innocua and

some as S. dysophylla.

ii. It is very difficult to decide to which of these two species the taxon which was

described as S. gerrardii N.E.Br., and which occurs in Natal, is the more closely related,

in other words, of which is it a subspecies or variety since it obviously belongs to the

same complex and should not have specific status.

In connection with the first difficulty, S. dysophylla as defined by Bruce and Lewis

differs generally from S. innocua. ssp. innocua, of which, to date, there is no authentic

record from southern Africa, in the following features: the branchlets usually smooth

and dark, either brown or grey, instead of pale and farinose; contracted lateral branch-

lets present; lenticels conspicuous, leaves on the whole smaller, seldom over 7 cm

long as against up to 15 cm long, and drying a dark colour; the inflorescence sessile

instead of shortly pedunculate and the flowers only about 5 mm long instead of about

10 mm long. But there are exceptions and one or more characteristics of the tropical

specimens may be found among specimens which, from the locality, or because of

some other features, belong best to S. dysophylla. These exceptions seem to indicate

that the differences are merely subspecific, that certain features have become accen-

tuated on plants in the one region and another set in the other, so that, on general

appearance, the specimens can, roughly, be separated into groups.

With regard to the second difficulty, S. gerrardii N. E. Brown differs from S.

innocua ssp. innocua principally in the comparatively slender branchlets which are

not fairnose, the conspicuous lenticels and the narrower leaves and it agrees with this

taxon in the normally developed branchlets and the peduncled inflorescences. These

latter features seem more important taxonomically and they are the features in which

it differs from S. dysophylla. Therefore it would seem to be more closely related to

S. innocua. Yet Bruce and Lewis suggest (Kew Bull. 1956: 275) that S. gerrardii

may be synonymous with S. dysophylla spp. engleri (Gilg) Bruce and Lewis, giving it

a closer relationship with S. dysophylla.

These difficulties are overcome if S. dysophylla and Y gerrardii are included in the

compound species, S. innocua, and given subspecific status, the same rank as S. innocua

subsp. burtonii (Bak.) Bruce and Lewis and S. dysophylla subsp. engleri (Gilg) Bruce

and Lewis but under one species. This does not conceal the variation in the complex

species and the objection expressed by Bruce and Lewis will be overcome.

New combinations will be necessary in this arrangement and are made below: —

S. innocua Del. subsp. dysophylla (Benth.) Verdoorn.

S. dysophylla Benth. in Journ. Linn. Soc. 1: 103 (1857).

Type: Delagoa, Forbes s.n.

S. innocua Del. subsp. gerrardii (N.E.Br.) Verdoorn.

S. gerrardii N.E.Br. in Kew Bull. 1896: 162 (1896).

Type: Medley Wood 5624 (K, lecto, PRE fragment lecto!).

I nominate Medley Wood 5624, the sheet in the Kew Herbarium seen by Brown

and of which there is a portion in the National Herbarium, Pretoria, as the lectotype

of Strychnos gerrardii N.E.Br., selected from the following syntypes: Berea, Medley

Wood 5624; Cult. Medley Wood Mil', without locality Gerrard 1421.

13

3. Sinking the genus Chilianthus under Buddleia.

The three genera Nuxia, Chilianthus and Buddleia as defined in the Flora Capensis

( Vol. 4, 1 : 1037) are obviously closely related and Chilianthus is the link between the other

two genera which, by themselves, would be easily distinguishable. Being the inter-

mediate group it is not surprising to find that its members have at different times been

classified under either Nuxia or Buddleia.

Bentham classified all 4 known species of Chilianthus as Nuxia species (see Comp.

Bot. Mag. 2: 60, 1836) and from Burchell’s unpublished notes on his specimens it

can be seen that he identified the two Nuxia species he collected, N. floribunda Bth.

and N. congesta R.Br. ex Fres., as “ Chilianthus ”. When describing the new genus

Chilianthus (see Travels 1: 94 (1822), Burchell contrasted it only with Scoparia and

did not mention Nuxia or Buddleia.

The view that Nuxia and Chilianthus are congeneric can hardly be justified, for

although in general appearance the flowers in both genera are similar, having the corolla

tube short and the stamens exserted, yet in the leaves, general habit and certain other

details they are markedly dissimilar. For instance in Nuxia the anther-cells are con-

fluent, the corolla circumscissile near the base, and there are no stipules present, whereas

in Chilianthus the anther-cells are distinct, the corolla not circumscissile and there are

stipules, or at least interpetiolar ridges, present.

The view, on the other hand, that Chilianthus and Buddleia are conspecific has

proved on investigation, to be feasible.

In 1946, E. P. Phillips reduced the genus Chilianthus to a synonym of Buddleia

(see Journ. of S.A. Bot. 12: 113). When doing so he showed that in the last analysis

the only character by which these two genera could be separated is the exserted stamens

of Chilianthus as against stamens included in Buddleia, and then, he continues, that in

one known Chilianthus, C. corrugatus, “ the anthers are very little exserted ”. He

points out too that in most of the known classifications, such as Genera Plantarum

and Pflanzenfamilien, the close relationship of Chilianthus and Buddleia is emphasized.

On first thoughts one may be inclined to wonder why Chilianthus and Buddleia

should not remain distinct since in South Africa the two groups are easily distinguished,

for all our known species of Buddleia have flowers with a tubular corolla and stamens

included while the Chilianthus species have short campanulate corolla-tubes and at

least the anthers exserted. But a study of the whole genus Buddleia reveals that the

representatives in Africa are not typical of the genus as represented in America and,

as Dr. Phillips also pointed out, differ almost as much from the type species, B. americana,

as they do from the species placed in Chilianthus in S. Africa. From this it is obvious

that the two groups of species in South Africa can at most be treated as two sections

of the genus Buddleia.

4. Reasons for again reducing the genus Lachnopylis to Nuxia.

In the Kew Bulletin, 1930 C. A. Smith resuscitated Lachnopylis Hochst. which

Bentham had reduced to Nuxia Lam. giving his reasons for considering it generically

distinct from this monotypic genus. These reasons have never seemed very convincing.

In several publications on the tropical African flora. Smith was followed, for

example by Robyns in the Flora of the Parc National Albert, Vol. 2: 59 (1947), and

Brenan and Greenway in Check-List of the Forest trees and Shrubs of the British Empire

No. 5, 2: 270 (1949). On the other hand, when dealing with the specimens in the

Mascarene Islands, that is in the region where both genera occur, Nuxia on Mauritius

and Reunion and Lachnopylis on Madagascar and the Comoro Islands, Paul Jovet

retains the name Nuxia for all the species and treats Lachnopylis as a sub-genus of Nuxia

(see for instance Notulae Systematicae 13: 97 (1947).

14

Further, the systematists at present working on the Loganiaceae for the “ Flora

of Trop. East Africa ” have independently come to the decision that Lachnopylis is

not generically distinct from Nuxia and so all the tropical African species are to be

placed in the genus Nuxia in that work.

With these two supporting opinions, and for the sake of uniformity in Africa it

has been decided to re-instate Nuxia in the South African Flora as well.

The taxonomic reasons for favouring this treatment are that the points of similarity

between Nuxia verticillata, the type species of the genus, and the species in the Mas-

carenes and Africa which Smith would refer to Lachnopylis, seems to be of more signi-

ficance as generic characters than are the differences.

The similar features are: ternate leaves; inflorescence of compound cymes; a

tubular calyx with erect lobes and lined with appressed hairs; the corolla circumscissily

deciduous from near the base; and the anther-cells confluent. A combination of some

or all of these features are convincing as generic characters. The most important

differences are that Nuxia verticillata has the corolla-tube longer than the calyx with

the style included in the corolla-tube as against the calyx being longer than the corolla-

tube in all the other species with the style exserted from both calyx and corolla.

These are hardly generic differences especially since the general appearance of

the flower is the same in both groups, the corolla-lobes reflexing in N. verticillata

exactly as in the others, although the calyx does not reach to the point where they

reflex; and if one takes the relationship of the style to the persistent calyx instead of

to the deciduous corolla it is longer than the calyx in both groups. These differences

seem no more than specific but might be regarded as worthy of some higher (sub-

generic) ranking if all the species, including those in the Mascarenes, are studied.

Under the present treatment the following combination is necessary: —

Nuxia glomerulata (C.A.Sm.) Verdoorn for Lachnopylis glomerulata C. A. Smith

in Kew Bull. 1930: 24 (1930).

Although C. A. Smith described 8 new species under Lachnopylis in the above

mentioned publication it has been found, now that the genus is better known in the

wild, that only this one (L. glomerulata) can be maintained as a species. Of the other

seven, one, L. suaveolens C.A.Sm., is a synonym of L. glomerulata and the rest have

been reduced under the very variable and widespread, but characteristic, species Nuxia

congest a Fres.

This wide view of N. congesta conforms to that of the workers on the Flora oi

Tropical East Africa.

5. New status for Salvadora australis Schweickerdt.

Comparison of the specimens of Salvadora australis Schweickerdt (Bothalia 3, 3:

248, 1938) with the description and figure of the Madagascar species S. angustifolia

Turrill (Kew Bull., 1918 and Flora Madagascar, 1946), shows a great similarity especially

in features such as the shape and spread of the leaves and the presence of well developed

glands between the filaments. These are the very features which characterize 5”.

australis and distinguish it from S. persica L. Through the kindness of Dr. Renaud

Paulian I was able to examine two specimens of S. angustifolia from Madagascar which

he sent on loan to the National Herbarium, Pretoria. This confirmed the similarity

between the specimens from the two countries and the only evident difference noted

was the persistent, short, dense pubescence on the South African specimens as compared

with the glabrous or partly scaly specimens from Madagascar.

15

Pubescence in this family is not usually of specific significance and, in most cases,

does not warrant even subspecific distinction, but the nature of the pubescence and

the remoteness of the areas of distribution seem, in the case under consideration, to

justify varietal separation. The following combination is therefore made: —

Salvadora angustifolia Turn’ll var australis (Schweick.) Verdoorn stat. nov. S.

australis Schweick. in Bothalia 3, 3: 248 (1938).

6. Lectotype of Jasminum breviflorum Harv. ex Wright.

In the revision of the Oleaceae, Bothalia 6: 549 (1956), I wrote in the introduction

on page 549, concerning the authorship of certain species, that, in my opinion, Harv ey

should get the credit for them rather than the publishing author. I felt that since Harvey

supplied the descriptions these were parallel cases to those cited under Recommendation

46B of the 1956 Code of Botanical Nomenclature. While I still abide by my decision

in the case of Menodora juncea which was published post humously in Harvey’s Genera

Plantarum ed. 2, 1869, I now feel that in connection with Jasminum breviflorum, J.

Gerrardii and Olea enervis, it is more correct to cite the authority as “ Harv. ex Wright ”

and therefore if abbreviating “ Wright ” should be cited.

My reasons are: (1) That “in Wright in Flora Capensis ” may not be what is

intended by “ in a work by another author (see Rec. 46B) ”. (2) I have no proof that

Wright did not change Harvey’s descriptions to some extent; (3) at least in one case

Wright cites a specimen that Harvey could not have seen; and (4) if Harvey is cited

as author there is uncertainty about the holotypes.

The reviewed treatment, that is citing the author in these ca.es as “ Harv. ex

Wright ”, requires the selection of a type for J. breviflorum since Wright cited 2 speci-

mens. I here select the sheet in the Kew Herbarium of Burke from the Magaliesberg

which has on it a label reading “ seen by Dr. W. H. Harvey ” and written to the left

of this label the word “ type ”.

17

Notes on the Genus Royena in South Africa.

By

B. de Winter.

During a study of the accumulated material of this genus in the National Herbarium,

Pretoria and in the Kew Herbarium, London, it became evident that the treatment and

delimitation of the species in the Flora Capensis 4 (1906) was no longer satisfactory.

The following notes are the result of a study of plants in the field and also of the

ample herbarium material available for study.

The most important characters in distinguishing species are in the fruits and mature

leaves, and fruiting specimens are therefore best suited for identification. In contrast

to those of many other groups, the floral parts of Royena are variable within the species

especially in size, and show hardly any characters of diagnostic value on the specific

level. The distribution of the species was found to be a very useful aid in confirming

the identification of incomplete or very young specimens.

A wider concept of species has simplified identification and what is more important,

has, in my opinion, presented a truer picture of the relationship of the various taxa

to each other.

The discussion of the sex of the flowers of R. glabra in Journ. S. Afr. Bot., April,

1953 by Salter has been corroborated by the study of other species in the field as well

as in the herbarium. All the species have proved to be dioecious, with functionally

male flowers, and female flowers with staminodes borne on different plants. The

female plants, which are in the minority in all the populations studied in the field, and

also in material in herbaria, never have pollen and are therefore truly female. The

functionally male plants are apparently hermaphrodite since they very often have fairly

well developed ovaries which, however, normally do not ripen into fruits and soon

abort. Fruits have been noticed occasionally on male plants and have proved to have

some apparently fertile seeds. It seems best therefore to refer to the species as sub-

dioecious as has been done by other authors.

The treatment of the genus in the Flora Capensis lists 17 species. This number

has now been reduced to 13.

A full account with keys and descriptions will be published in the forthcoming

Flora of Southern Africa.

New combinations and varieties.

Royena cordata E. Mey. ex DC. var. scabrida {Harv. ex Hiern) de Winter comb. nov.

Basionym. R. scabrida Harv. ex Hiern in Trans. Cambr. Phil. Soc. 12, 82 (1872).

Royena lycioides {Desf.) A. DC.

(a) subsp. sericea {Bernh. ex Krauss) de Winter comb. nov.

Basionym. R. sericea Bernh. ex Krauss in Flora 27, 824 (1844).

ib) subsp. guerkei {O. Ktze.) de Winter comb. nov.

Basionym. R. guerkei O. Ktze. Rev. Gen. PI. 3, 2, 196 (1898).

18

(c) subsp. nitens {Harv. ex Hiern) de Winter comb. nov.

Basionym. R. nitens Harv. ex Hiern in Trans. Cambr. Phil. Soc. 12, 87 (1873).

Royena hirsuta L.

{a) var. microphylla {Burch.) de Winter comb. nov.

Basionym. R. microphylla Burch. Trav. 1, 348 (1822).

{h) var. rubriflora de Winter var. nova var. microphyllae et var. hirsutae affinis, a var.

microphylla foliis maioribus diflferens, ab ambabus floribus coccineis, nec roseis

modo var. microphyllae nec cremeis modo var. hirsutae.

Type: Natal. Mont aux Sources, lower slopes of Drakensberg, flowers port-wine

red. Aug., 1930, Hutchinson, Forbes and Verdoorn No. 4 (PRE, holo) = Hutchinson

4483 in K.

Royena villosa L.

var. parvifolia de Winter var. nov., var. villosae affinis foliis minoribus minus 3 -5 cm

longis, fructibus minoribus differt.

Type: Transvaal, Soutpansberg : Elim, plants supported by other trees, Ober-

meyer 713. (Tvl. Mus. No. 28377 in PRE, holo).

The distribution of this variety is limited to the Northern and North-eastern Trans-

vaal.

A NEW SPECIES OF ROYENA.

Royena acocksii, species nova affinis R. ramulosae sed ita differt: folia multo maiora

subtus sparsissime strigosa ramuli subglabri rubri-brunnei, flores semper tetrameri

corolla calyci subaequali vel quam calycem paululum maiore.

Rigidly branched shrubs up to 15 ft. high. Branches straight, spreading at nearly

right angles, bark smooth to finely rugose, ashgrey with blackish markings to grey

with a reddish brown tinge; young branches smooth, deep mahogany red, glabrous

or with a few scattered adpressed hairs. Leaves alternate, very shortly but distinctly

petiolate, 1-0 cm to 2-0 cm long and 0-5 cm to 1-0 cm broad, elliptic to obovate-

elliptic, glossy, coriaceous, finely rugose and glabrous above, more or less smooth and

sparsely strigose below (young leaves densely strigose) ; margins usually slightly undulate,

entire, often involute; nerves inconspicuous except for the midrib and sometimes

faintly marked secondary nerves. Flowers tetramerous, solitary in the axils of the

leaves, functionally male (apparently hermaphrodite) or female with the stamens reduced

to staminodes, pendulous, up to 1 cm long but mostly shorter, peduncles 0-5 cm to

0 • 7 cm long, sparsely strigose ; bracts 2, more or less approximate, up to 0 • 5 cm long,

oblanceolate. Calyx deeply divided, lobes 4, triangular, 0-4-0 -5 cm long and 2-5

mm wide at the base, sparsely strigose outside, glabrous inside, rather thick in texture.

Corolla creamy white, urceolate, four-angled, slightly hairy on the angles, otherwise

glabrous; tube short (^|: length of flower); lobes very broadly ovate with an acute

apex, distinctly imbricate. Stamens 8, 3-0 mm long, filaments very short and broad,

glabrous; anthers lanceolate, 2-5 mm long, hairy on the back, less so on the inside

face, dehiscing by longitudinal slits. Ovary borne on a distinct disc, four-angled,

pyramidal, when more mature very broadly oblong, glabrous, or sparsely hairy on

the angles; style as long as the four branches or slightly shorter. Fruits 1*0-1 -3 cm

long, glabrous, broadly oblong to semi-globose, slightly 4-angled, occasionally dehiscing

along the four sutures which are visible as four fine but distinct lines. Calyx accrescent,

^ the length of the fruits, lobes ovate, apices acute, more or less reflexed. Seeds strongly

laterally compressed, 8 (or fewer), glabrous blackish, shiny.

19

Cape Province. — Little Bushmanland ; (Pella) near dry stream bed, bush 12-15

ft. high, Pearson 3546 (K). Kenhardt: Five miles north of Pofadder in the “ Orange

River Broken Veldt” of rocky hills, shrub up to 10 ft. high alt. 3,300 ft., Acocks 14395

(PRE holo; Isotype in K.); 126 miles from Kenhardt, Pole Evans 2259 (PRE).

South West Africa. — Warmbad district: beside streamlet (dry watercourse!)

on farm Witzand, Galpin 14149 (K, PRE); 20 miles from Warmbad on Goodhouse

road, shrub, 6 ft. high Galpin s.n. (K, PRE).

R. acocksii is a very distinct species with tetramerous flowers and relatively small,

leathery, inconspicuously nerved leaves with undulate margins. It can be distinguished

easily from the tropical spp. of Royena with tetramerous flowers by the much smaller

subglabrous leaves. Two of the South African species can have tetramerous flowers,

one of which, R. glandulosa, shows very little resemblance to R. acocksii. In R. glan-

dulosa the calyx lobes are much accrescent and exceed the fruits in length, and the young

parts are covered with glands. The second species, R. ramulosa, only occasionally

has tetramerous flowers but there is a certain similarity in the leaves and fruits which

suggests an affinity with R. acocksii. Their climatic requirements are also similar,

but R. acocksii, is confined to the arid areas near the Orange River in the northern Cape

Province and north of the river in South West Africa, while R. ramulosa has a wider

distribution.

The Lectotype of R. fallens Thunberg.

R. pallens Thunberg Prodr. 80 (1794).

Syn. R. brachiata E. Mey. ex DC. Prodr. 8, 213 (1844).

In the Flora Capensis Hiern took a very broad view of this species, including as

synonyms, amongst others, R. pubescens, R. lycioides and R. sericea. Of these I regard

the first two as distinct and the latter as a subspecies of R. lycioides. The other synonyms

given by Hiern are referable to one or other of the three species. There are fruit and

leaf characters distinguishing these two species and their distributions are also rairly

well defined. At least part of the confusion which existed in R. pallens can be attributed

to the fact that four different plants are present on the three sheets in Thunberg’s

herbarium. Sheet 1 consists of two twigs of the species referred to by E. Meyer and

published by DC. as R. brachiata. Sheet 2 consists of three twigs of R. pubescens

Wind, and a loose fruit of R. lucida. The short description “ R. foliis oblongis obtusis

glabris margine revolutis ” could equally well be applied to sheet 1 and to the specimens

of R. pubescens on sheet 2. Sheet 3 consists of a rather atypical twig of R. pubescens

and a twig of Rhoicissus cirrhiflorus (L.f.) Gilg. & Ben. and is therefore excluded as

a possible lectotype. Since sheet 1 and 2 are equally eligible for selection as the lecto-

type on other grounds, the lectotype should be selected “so as to preserve current

usage ” App. Id. of the ‘ Rules ’ (1952).

By choosing sheet 1 as the lectotype of R. pallens it is possible to retain the name

R. pubescens Willd. in its currently accepted sense. If on the other hand sheet 2 is

chosen, R. pubescens becomes a synonym of R. pallens and the species on sheet 1 will

have to be referred to as R. brachiata— a. name which has not been in current use since

its inception.

Sheet 1 is therefore chosen as lectotype and the name R. pallens becomes restricted

to a rather localised species so far recorded only from the Knysna, Port Elizabeth,

Uitenhage and Bathurst districts.

i

1

21

New and Interesting Records of African

Flowering Plants.

By

Various Authors.

ASCLEPIADACEAE.

Ceropegia occulta R. A. Dyer sp. nov., affinis C. caffrorum Schl. habitu crassiore,

alabastro obtusiore, corollae lobis latioribus coronae lobis interioribus oblongo-

ellipticis obtusis differt.

Herba perennis simplex vel sparse ramosa. Tuber 1-5-4 cm diam. Rami ad 20

cm longi, procumbentes vel suberecti nonnunquam volubiles, 2 mm crassi. Folia

sessilia vel brevissime petiolata, carnosa, late ovata-vel lineari-lanceolata, 1-2 cm

longa, 3-10 mm lata, acuta. Flores 1-2 extra folii axillam exorti, erecti, pedicillis

1 cm longis. Sepala lineari-lanceolata, 3-3-5 mm longa. Corolla 2-2-8 cm longa,

apice obtusa; tubus 1-5-2 cm longus, basi ovoideo-inflatus, plus minusve 5 mm diam.,

in media parte 2-5-3 mm diam. in faucibus 5-6 mm extra et intus glaber; lobi 6-7

mm longi, apice per 2 mm connati, basi et apice 3 - 5 mm lati, medio 2 - 5 mm lati, intus

breviter pubescentes, carinati. Coronae exterioris lobi poculiformes; 0-5 mm lati

cum lobis interioribus connati; coronae interioris lobi incumbenti-erecti, 2-5 mm

longi, oblongo-elliptici, obtusi, subcarnosi.

Plate I.

Cape Province. — Worcester Division; near Tweefontein, foothills under bushes.

Van Breda 85 (PRE, type); Erens and van Breda 2264 (photograph).

Plants forming one to several subglobes tubers 1 - 5-4 cm in diam. with a few

fibrous roots. Stems 1-4 from each main tuber, up to about 20 cm long, rarely showing

signs of twining, 2 mm thick; lower internodes 2-6 cm long, towards apex of branches

reduced to about 1 cm long or less. Leaves sessile or subsessile very variable in shape,

the broadest towards the base ranging from nearly circular to ovate to linear-lanceolate,

1-2 cm long, 3-10 cm broad, fleshy, grooved down face, convex on lower surface,

acute. Cymes laterally produced at nodes, subsessile, 1-2-flowered; flowers produced

successively, sometimes the second aborted; pedicels up to about 1 cm long, glabrous.

Sepals linear-lanceolate, 3-3 - 5 mm long. Corolla 2-2 - 8 cm long, very slightly curved

or straight with tube 1 - 5-2 cm long, inflated at base to about 5 mm diam. 2 - 5-3 mm

diam. above and spreading at the mouth to 5-6 mm; glabrous inside and outside,

finely longitudinally ribbed within the inflation; lobed portion in bud 5-8 mm long

with a slight constriction about the middle and a flattened united apical portion, apicu-

late; lobes about 3 - 5 mm broad at base and the same towards apex, 2-5 mm at middle,

united for 2 mm at apex forming cap about 4 mm wide, apiculate in middle, each lobe

shortly hairy within the margin and keeled down the inner face with the keel broadened

to 1-5 mm at the base; margins only slightly replicate leaving comparatively small

apertures between the lobes. Corona appearing as one series; outer corona of small

spreading pockets about 0-5 mm wide, confluent with the base of inner lobes; inner

lobes incumbent-erect, narrowly oblong-elliptic, 2-5 mm long, obtuse, somewhat

fleshy. Pollinia subglobose, about 0-25 mm diam. with narrow amber-coloured inner

margin which extends into a minute apical point (a feature not observed previously).

22

It is remarkable that this species of Ceropegia, which grows closest to the first

port of entrance to the Union at Cape Town, should be among the last to be named

specifically. No earlier record of the species seems to exist than a collection by P. A. B.

van Breda shortly before 1950. He found it well hidden — hence the name — under the

protection of karoo shrublets near Tweefontein in the Worcester district. In 1950

he and Mr. J. Erens, then head gardener of the Division of Botany, gathered more

material at the same site for cultivation in Pretoria. This material flowered in Septem-

ber of the same year and although it was photographed, no herbarium specimen appears

to have been preserved at the time.

In June, 1956, Mr. van Breda made further collections on request. A few flowers

were present at the time and the holotype consists of this material, and duplicates were

made when the plants continued to bloom at the Division of Botany, Pretoria, during

February, March and April of 1957. Although the species appears to have a fairly

long flowering period, it grows slowly and is not very floriferous. Under natural

conditions the stems are relatively short and rarely show signs of twining, whereas

under cultivation the stems are more robust and twine when placed on a support.

It seems that C. occulta is nearest in affinity to C. caffrorum Schl. and its allies in

the eastern Cape Province near Uitenhage. It differs from these in the shape of the

corolla lobes and the obtuse cage-like structure they form, and also in the more fleshy

obtuse inner corona lobes. The pollinia are specially interesting. It is readily dis-

tinguished from C. africana R. Br., figured in the Botanical Register 626 (1822), and

since recorded from near Oudtshoorn, by the falcate inner corona lobes of the latter.

R. A. Dyer.

CALLITRICHACEAE.

Callitriche bolusii Schonl. et Pax ex Marl, Flora S. Afr. //, ii; 141 (1925).

The valid publication of this name seems so far to have been overlooked. It was

first rendered as Callitriche bolusiana Schonl. et Pax in a note in Engl. Pflanzenw.

Afr. Ill, ii: 169 (1921), but in Marloth’s Flora under C. bolusii there is a series of line

drawings which make it possible to identify the plant without doubt. Contrary to

Marloth’s observations the plant is restricted to the eastern Cape Province and has

not been found in the Transvaal.

Mayda Henderson.

COMPOSITAE.

Helichrysum albirosulatum Killick, sp. nov., affine H. flanaganii Bolus, sed habitu

robustiore, folds argenteo-albis, floribus majoribus paucioribus differt.

Fruticulus prostratus, ad 6 cm altus, omnino indumento argenteo-albo vestitus.

Folia rosulata, plerumque spathulata interdum obovata, 1-4-2 -2 cm longa, 5-6 mm

lata, apice acuta vel obtusa plus minusve uncinata, basi attenuata. Pedunculi suberecti,

8 cm longi, foliati. Corymbi polycephali, semi-globosi, 1 • 3-1 • 5 cm diam. Capitula

5 mm longa, 2-5-3 mm lata. Involucri bracteae 3-seriatae, erectae, aureo-flavae, ab

exteriore gradatim majores. Flores hermaphroditi, 10-12, aureo-flavi; corolla tubu-

losa, supra ampliata. Achaenia papillosa, breviter pubescentia.

Natal. — Underberg District: rocky slopes of Bamboo Mountain, McLean 697.

Estcourt District: top of Little Berg, 6,000 feet, Cathkin Park, Galpin 11838. Bergville

District: Cave Sandstone outcrops, 5,500 feet, Cathedral Peak Forest Influences

Research Station, Killick 1919 (PRE, type).

23

A prostrate, much-branched shrublet, up to 6 cm high, covered all over with a

silvery-white, somewhat glossy, very closely felted indumentum. Leaves rosulate,

spathulate or sometimes obovate, 1 • 2-2 • 2 cm long, 4-6 mm wide, apex acute or rounded,

somewhat hooked, base attenuate. Peduncles suberect, 6-12 cm long, leafy. Corymbs

many-headed, semi-globose, 1-3-2 cm diam., densely araneous underneath. Capitula

5 mm long, 2-5-3 mm wide. Involucral bracts in 3 rows, erect, golden-yellow at the

tips; outer broadly elliptic, 4-5 mm long, 2 mm wide, subtended by a single large

obovate bract 5 mm long, 2 - 5 mm wide ; median ovate, 4 mm long, 2 mm wide; inner

linear-lanceolate, 3-5 mm long, 1 mm wide, apex reflexed. Florets hermaphrodite,

10-12, golden-yellow; corolla tubular, widening above; pappus of numerous bristles

successively shorter towards base ; anthers with an acute appendage at apex, acuminate

tails at base; style branches truncate. Achenes papillose and shortly pubescent.

Helichrysum albirosulatum seems to be restricted to Cave Sandstone outcrops in

the Drakensberg area. Usually it is found on the “ pavements ” situated along the

edge of the “ Little Berg,” the terrace immediately below the main escarpment. This

species with its numerous silvery-white rosettes forms extensive mats, which are a

conspicuous feature of Cave Sandstone Macchia with Passerina montana. Protea

roupelliae and Erica spp. as dominants. The known distribution of Helichrysum

albirosulatum is between Bamboo Mountain, Underberg, in the south, and Cathedral

Peak in the north. It was first collected by E. E. Galpin in March, 1932.

It differs from its nearest ally, H. flanaganii Bolus, in that the plant is more robust,

the indumentum silvery-white and somewhat glossy instead of dull grey and in having

florets which are larger and fewer per head (10-12 instead of 30 or more).

Helichrysum drakensbergense Killick, sp. nova, distinctissima, nullis e specibus notibus

propinqua.

Herba decumbens, ad 35 cm alta, omnino cano lanata. Folia spathulata, 3-5-4

cm longa, 1 - 2-1 • 8 cm lata, apice subacuta vel rotundata, basi multo attenuata. Pedun-

culi ascendentes, 15-18 cm longi. Corymbi plani, 2-2-4 cm diam., infra dense araneosi.

Capitula congesta, subcampanulata, 6 mm longa, 3-5^ mm lata. Receptaculum

conicum, nudum. Flores hermaphroditi, circiter 27. Involucri bracteae plerumque

3-seriatae, flavae, nitidae. Corolla pallide lutea. Pappi setae numerosae, plumosae.

Achaenia breviter pubescentia.

Natal. — Bergville District; frequent in alpine grassveld below Organ Pipes Pass,

9,200 feet. Cathedral Peak Forest Influences Research Station, Killick 1879 (PRE, type).

A decumbent to suberect herb up to 35 cm high, grey-lanate all over. Leaves

spathulate, 3-5 cm long, 1-2-2 -2 cm wide; apex subacute or round; base much

attenuate appearing almost petiolate, somewhat stem-clasping. Peduncles ascending,

up to 22 cm long, with single clasping lanceolate bract 2-2 cm long. Corymbs flat-

topped, many-headed, 2-3 cm diam., densely araneous underneath. Capitula congested,

subcampanulate, 6 mm long, 3-5^ mm wide at summit. Receptacle conical, nude.

Florets hermaphrodite, about 27. Involucral bracts usually 3-seriate, imbricate, not

radiating, straw-coloured, glossy; outer linear or lanceolate, 4-5 mm long, embedded

in wool; median lanceolate to spathulate, 5 mm long, 1-2-1 - 5 mm wide; inner similar

to median. Corolla pale yellow; tube narrowly cylindrical, 4 mm long, widening

slightly in upper third then abruptly at summit; lobes deltoid, 0-5 mm long, 0-3 mm

wide at base; anthers with appendage at apex and acuminate tails at base; style

branches truncate. Pappus of numerous plumose bristles with the trichomes progres-

sively shorter towards the base. Achenes 5-ribbed, minutely pubescent.

H. drakensbergense is a very distinct species without any obviously close ally,

consequently its position in the genus is difficult to establish. When one uses Harvey’s

key in Flora Capensis it runs to the section Stoechadina, but its relationship to the

constituent species of that section is obscure.

24

It is surprising that this quite conspicuous plant has not been recorded before.

It grows in alpine grassveld below the Organ Pipes Pass next to a footpath which is

frequently used by climbers making for Cleft Peak or Basutoland and is a socially

growing species forming fairly large communities.

Senecio cryptolanatus Killick, sp. nov., affinis 5”. tugelensi Wood et Evans, sed plantis

acaulescentibus differt.

Herba acaulescens, ad 8 cm alta, rhizomate lanato robusto. Folia rosulata,

spathulata, 5-7 cm longa, 1-1-4 cm lata, apice subacuta vel obtusa, plus minusve

carnosa, glabra, nervis immersis, marginibus integris rubris, in axillis lanata. Pedun-

culi simplices, 17 cm longi, bracteis 8. Capitula solitaria, subcampanulata, 1 cm

longa, 1-3-1 -5 cm lata, calyculata. Involucri bracteae circiter 20, lineari-lanceolatae,

7-8 mm longae, 1-5 mm latae. Receptaculum alveolatum. Flores radii circiter 17,

lutei. Flores disci numerosi. Pappi setae numerosae, barbellatae. Achenia glabra.

Cape. — Barkly East District: Doodman’s Krans Mountain, Drakensberg, 8,900

feet, Galpin 6723 (in part).

Natal. — Bergville District: top of Mont aux Sources, 11,000 feet, Evans 743

(in part); mountain top, 11,000 feet, Mont aux Sources, Allsopp 24; very rare on

streambank on summit of Drakensberg between Indumeni Dome and Cleft Peak,

9,700 feet, Killick 1941 (PRE, type).

Note. — Evans 743 and Galpin 6723 in the National Herbarium are mixtures

consisting of Senecio cryptolanatus and S. tugelensis. Mr. W. Marais, our liaison

officer at Kew, reports that Evans 743 in Herb. Kew. has the same mixture, but that

Galpin 6723 is S. tugelensis only. He also reports that Milford 636 and Galpin 6722

are S. cryptolanatus.

An acaulescent herb up to 8 cm high with a thick and woolly rootstock. Leaves

rosulate, spathulate, occasionally oblanceolate, 3-8 cm long, 0-5-1 -7 cm wide, apex

subacute to obtuse, somewhat fleshy, glabrous, veins immersed, margins red, usually

entire but sometimes distantly serrate, axils with long tufts of white wool. Peduncles

simple, rarely divided into two; bracts 5-10, lanceolate with a subrotund base, white

wool in axils. F[eads solitary, subcampanulate, 1 cm long, 1-3-1 -5 cm wide, calycled.

fnvolucral bracts about 20, linear-lanceolate, 7-8 mm long, 1 - 5 mm wide, with acuminate

apices. Receptacle honeycombed. Ray florets about 17, yellow. Disc florets nume-

rous. Pappus of numerous barbellate bristles. Achenes glabrous.

Senecio cryptolanatus is a distinct species with yellowish green, somewhat fleshy,

rosulate leaves. The plants seen by the author had red-margined leaves, a character

which is not always evident in the dried material. All the collectors describe the

flowers as yellow, but in the dried state the rays are almost white on the inner face and

yellow on the outer. The plant flowers during March and April. The epithet crypto-

lanatus refers to the “ hidden ” tufts of wool in the axils of the leaves, bracts and involucel

bracts.

This species grows on the summit of the Drakensberg Mountains between 8,900

and 11,000 feet usually at the edge of small streams or pools. Its known distribution

is from Doodman’s Krans Mountain in the Barkly East District of the Eastern Cape

to Mont aux Sources in Natal — a range of some 200 miles. S. cryptolanatus was

first collected in March, 1898, by Maurice Evans, one of the earliest collectors in the

Drakensberg area.

S. cryptolanatus has no very close ally, but is probably nearest S. tugelensis Wood

and Evans, which differs in being caulescent. It is rather surprising that S. crypto-

lanatus has been collected in mixed gatherings with S. tugelensis', morphologically

the two species are easily distinguishable and in addition they have a different ecology.

S. cryptolanatus, as already pointed out, is a streambank or pool-edge plant, whereas

S. tugelensis grows in alpine grassveld.

25

Senecio praeteritus Killick, sp. nov., affinis S. brevidentato M. D. Henderson, sed foliis

radicalibus cordato ovatis non oblanceolatis differt.

Herba erecta, ad 40 cm alta, simplex, basi fibrosa. Caules glabri vel breviter

pubescentes. Folia glabra; radicalia cordato ovata, laminis 1-2-2 cm diam. petiolis

2-5-3 cm longis, apice subacutis vel rotundatis, marginibus repando dentatis sub-

reflexis; folia caulina inferiora oblanceolata vel spathulata, 2 -5-5 -5 cm longa, 0-8-1 -8

cm lata, marginibus ut in foliis radicalibus; superiora lanceolata, 1-3-2 -7 cm longa,

1^ mm lata, basi auriculata amplexicaulia, obscure et sparse dentata. Pedunculi

laxe corymbosi, raro simplices. Capitula radiata, calyculata, lutea. Receptaculum

alveolatum. Involucri bracteae circiter 20, lineares. Flores radii circiter 10. Flores

disci numerosi. Achaenia juvenia glabra, tereta. Pappi setae numerosae, breviter

barbellatae.

Natal. — Bergville District: locally abundant, mountain side. Old Basuto Pass,

National Park, 6,000 feet, Galpin 9739; locally frequent on streambank. Cathedral

Peak Forest Influences Research Station, 6,800 feet, Killick 1796 (PRE, type).

Erect herb up to 40 cm high with a fibrous base. Stems simple, usually glabrous

but sometimes minutely pubescent. Leaves glabrous, distinctly paler on the lower

surface; radical cordate-ovate with the blade 1-2 cm diam. and petiole 1-5-3 cm long,

apex subacute to round, margins repand dentate occasionally appearing crenate, sub-

reflexed, teeth thickened; lower cauline oblanceolate to spathulate, 2 -5-5 -5 cm long,

0-4-1 -8 cm wide, margins as in the radical leaves; upper cauline lanceolate, 1-3-2 -7

cm long, 1-4 mm wide, auriculate and somewhat amplexicaul at the base, obscurely

and distantly toothed. Peduncles laxly corymbose, rarely simple, with 1 or 2 linear

bracts 2-6 mm long. Heads radiate, subcampanulate, 6-9 mm long, 6-8 mm wide at

the summit, yellow, calycled. Receptacle honeycombed, angles of pits produced into

subulate processes. Involucral bracts about 20, linear, 5-6 mm long, margins mem-

branous, apex black-tipped, glabrous. Ray florets usually 10; tube 4-5 mm long,

limb twice as long. Disc florets numerous, 7 mm long, widening above middle ; lobes

narrowly triangular, 1 mm long, 0-75 mm wide at base; style branches truncate, con-

spicuously hispid at tips; filaments thickened below anthers; anthers with incurved

apical appendage. Achenes terete, glabrous. Pappus of numerous barbellate bristles.

This new species of Senecio was first collected in November, 1928, by E. E. Galpin.

He found it growing at the side of the Old Basuto Pass in the Mont aux Sources area

of the Drakensberg. Galpin describes the plant as “ locally abundant ”. Twenty-four

years elapsed before it was collected again, hence the epithet praeteritus meaning

“ passed-over ”. The author found the plant growing on a streambank in the upper

reaches of the Indumeni Valley on the Little Berg in the Cathedral Peak area. There

it is locally frequent.

The affinity of S. praeteritus seems to be with S. brevidentatus M. D. Henderson,

which differs in having oblanceolate instead of cordate-ovate radical leaves.

A note by Dr. E. P. Phillips on the Galpin specimen states that the plant is “ near

Senecio cordifolius Linn, f.” At first sight this seems possible; S. cordifolius, a S. W.

Cape species, has the same three leaf shapes. However, this is another example of

homoplastic similarity between unrelated species — in this case between two species of

different genera for, as pointed out by Spencer Moore in Journ. Bot. 4f, 406 (1903),

L’ Heritier in Sert. Angl. 25 transferred S. cordifolius to Cineraria renaming it C. mitellae-

folia. In C. mitellaefolia the achenes are compressed as is characteristic of Cineraria,

whereas in Senecio praeteritus the achenes are clearly terete.

D. J. B. Killick.

26

CONVOLVULACEAE.

Ipomoea bisavium A. Meeuse sp. nov., I. heterosepalae Baker affinis sed pedunculis

longioribus praecipue differt.

Perennis, alte scandens. Caules lignosi teretes, ad ca 1 cm diam. corlice atro-

brunneo vel nigricanti rugoso obtecti, juveniles graciles adpresse puberuli. Folia

firmiter herbacea in siccitate chartacea, ovato-cordata vel elongato-cordata, Integra

vel subcrenata, acuminata vel attenuata, mucronata basi late cordata vel subcordata,

3-7 cm longa 1-5-5 cm lata lobis basalibus rotundatis; lamina utrinque praecipue

subtus minute adpresse pubescenti, glabrescenti costa et nervis secundariis gracilibus

subtus pauce prominentibus petiolis gracilibus subteretibus dense adpresse puberulis

1-3 cm longis. Inflorenscentiae axillares, pauciflorae, monochasiales vel interdum

flores solitarii; pedunculus communis satis gracilis, teres, dense adpresse puberulus,

ad 7 cm longus pedunculis secundariis ad 3 cm longis pedicellis gracillibus minute

pubescentibus substriatis 0-5-2 -5 cm longis; bracteolae submembranaceae ovato-

lanceolatae vel oblongae, acutae, sparse puberulae, subciliatae, ca 6 mm longae 2-3

mm latae, deciduae. Sepala erecta, in vivo dilute virides in siccitate olivacea, extus

pubescentia demum glabrescentia, inaequalia; 2 extoriora cordato-triangula lobis

basalibus biauriculatis, subacuminata, subobtusa, margine plus minusve undulata,

ca. 14 mm longa 8-9 mm lata; sepalum tertium semicordatum-subfalcatum, 10-11 mm

longum 4-5 mm latum, 2 interiora minora, vix auriculata, oblonga-lanceolata, subacuta,

9-10 mm longa 2-3 - 5 mm lata. Corolla infundibularis, alba tubo ad basin intus violaceo ;

tubus ca 2 cm longus, limbus 4-6 cm diam., vix lobatus, 5-angulatus, patens cum 5

fasciis mesopetalis extus pilis albidus strigoso-pilosis. Stamina inaequalia, ad basim

breve glanduloso-pilosa granulis pollinis spinulosis. Pistillum glabrum. Capsula

ovoideo-conoidea, glabra, fusca, 10-12 mm longa, ca 8 mm diam. Semina plerumque

4, ca 4 - 5 mm longa, velutina et ad angulos cum linea pilorum longissimorum fulvidorum

ca 7 mm longorum.

Transvaal. — Zoutpansberg district: about 2 m. S. of Wyllie’s Poort, Meeuse

10181 (flow. 2/4/1957, PRE, type); Meeuse 10237 (fruit 11/5/1957), isotypes in BM,

BR, EA, K, L, SRGH.

A tall perennial climber, reaching the tops of the supporting trees. Stems woody,

terete, attaining a thickness of about 1 cm near the base, in the older parts covered with

a dark brown to blackish rough bark, the youngest twigs firmly herbaceous to wiry,

green, adpressed-puberulous. Leaves firmly herbaceous drying chartaceous, ovate-

cordate to elongate-cordate, entire or faintly crenate, acuminate to attenuate and

mucronate at the apex, broadly and shallowly cordate to subtruncate at the base, with

rounded basal lobes, on both surfaces, especially on the lower one, minutely adpressed-

puberulous mainly on the nerves, more or less glabrescent; the midrib and the sub-

palmately arranged main nerves slender, slightly prominent beneath; the blade 3-7 cm

long and 1 - 5-5 cm wide, the petiole slender, subterete, densely adpressed- puberulous,

1-3 cm long. Inflorescence axillary, cymosely few-flowered or occasionally reduced

to a single flower; common peduncle rather slender, terete, densely adpressed-puberu-

lous, up to 7 cm long; secondary peduncles of the monochasia resembling the common

peduncles but slightly thinner, up to 3 cm long; pedicels slender, minutely pubescent,

somewhat striate, 0-5-2 -5 cm long; bracteoles submembranous, ovate-lanceolate or

oblong, acute, sparsely puberulous and subciliate, about 6 mm long and 2-3 mm wide,

early deciduous. Sepals erect, unequal, light green drying olive-green, finely pubescent

outside, glabrescent, not or hardly accrescent in fruit; two outer sepals cordate-triangu-

lar, biauriculate by the basal lobes of the cordate base, subacuminate, subobtuse at the

very apex, more or less undulate along the margin, about 14 mm long and 8-9 mm wide

near the base; third sepal somewhat asymetrical, semi-cordate-subfalcate, 10-11 mm

long and 4-5 mm wide; the innermost two distinctly smaller, not or hardly auriculate

at the base, oblong-lanceolate, subacute, 9-10 mm long and 2-4-5 mm wide. Corolla

27

funnel-shaped, white with the lower portion of the tube purple-mauve inside ; the tube

glabrous, about 2 cm long, the limb spreading; hardly lobed, 5-angled, 4-6 cm in

diam; the midpetaline areas strigose-pilose with white hairs on the outside. Stamens

unequal, white, at the base shortly glandular-pilose; pollen grains spinulose. Pistil

glabrous. Capsule ovoid-conical, glabrous, brown when ripe, 10-12 mm long and

about 8 mm in diam. Seeds normally 4, about 4-5 mm long, shortly velutinous with

greyish or drab hairs and in addition bearing fulvous hairs 7 mm long on the angles.

This plant was discovered after the manuscript of a revision of the South African

Convolvulaceae had gone to press. Specimens were sent to the herbaria in Brussels,

Kew, Nairobi and Salisbury, where Professor W. Robyns, Mr. W. Marais, Dr. B.

Verdcourt and Dr. H. Wild all failed to match it. Mr. Marais, who also tried to match

it in the British Museum (Nat. Hist.) herbarium, and Dr. Verdcourt reported that its

nearest relationship is obviously with Impomoea heterosepala, a Somaliland species.

Thanks are offered to all the botanists who so promptly answered my queries.

As the plant under discussion was collected in a place which was rather inaccessible

until the opening of the new road to Wyllie’s Poort, it is unlikely to be an introduction

or a garden escape. The locality is on a dry rocky hillside northwest of the new road

between the Punch Bowl and Wyllie’s Poort, about two miles from the Poort. This

locality was visited twice, in April, 1957, for flowering material, and again in May

of the same year to collect the fruits. Only a single plant was seen, and the flowering

and fruiting specimens are from the same individual. This species is named in honour

of Dr. Allan V. Bird, M.D. of Johannesburg and his son Peter for their active

assistance in the gathering of the specimens, which, in the dense thorn scrub of the

type locality, was by no means a sinecure.

I. bisavium is closely related to 7. heterosepala Baker, but according to the report

received from Mr. Marais, differs in the following characters: —

7. heterosepala. 7. bisavium.

Flowers Almost invariably solitary Inflorescence usually several-

flowered.

Peduncles 1-1 • 5 cm long up to bracts and Common peduncle 3-7 cm long.

articulation.

Pedicels 1 -5-2 -5 cm long. Up to 3 cm long.

Bracts Small and narrow. Elliptic, foliaceous.

Sepals Ovate-cordate, acute. Cordate, acute.

The seeds and apparently the colour of the corolla of 7. heterosepala are unknown,

but the differences indicated above, apart from the complete geographical separation,

appear to be sufficient to warrant specific distinction.

Viable seeds were obtained which have germinated quite well and it is hoped that

plants can be raised, so that this species can be figured for “ Flowering Plants of Africa ”

in the near future.

A. D. J. Meeuse.

Correction.

“ I wish to correct an erroneous statement which appeared in Dr. Meeuse’s

recent excellent account of the South African Convolvulaceae (Bothalia 6, 4: 1958).

On reading through I noticed on page 753 the siatement that the type of Impomoea

aquatica Forsk. at Copenhagen is not in accordance with the usual conception of that

species. This statement should have applied to Ipomoea biloba Forsk. and a note

concerning this is to be published in the Kew Bulletin. This error is entirely due to

a misleading statement in the postscript of a letter I sent to Dr. Meeuse and I must

apologise to him and to his readers.”

B. Verdcourt.

28

EUPHORBIACEAE.

Euphorbia rowlandii R. A. Dyer, sp. nov., habitu acaule ramis 5-7-angulatis podariis

corneis confluentibus cyma solitaria capsula plus minusve acute triangulata dis-

tinguitur.

Planta succulenta acaulis perennis armata, basi profunde ramosa ad 1-5 m alta.

Rami suberecti simplices vel raro ramulosi, 5-7-angulati in segmentis 7-15 cm longis

basin versus 3-5 cm latis superne angustioribus constricti, podariis corneis 2-aculeatis

confluentibus, aculeis ad 1 cm longis. Cyma solitaria plus minusve 2-4 mm supra

aculeos emittenta, breviter pedunculata, 3-cyatheis. Cyathium primum masculinum,

cyathea lateralia 2, bisexualia; involucrum 4-5-5 mm diametro, glabrum, lobis parvis

subquadratis fimbriatis et glandulis 5 transverse oblongis 2-5-3 mm latis contiguis

integris flavis munitum. Ovarium breviter stipitatum; styli circiter 1-5 mm longi

infra medium connati. Capsula circiter 9 mm lata plus minusve acute triangulata

pedicello circiter 5 mm longo exserta.

Plate II.

Transvaal. — Soutpansberg District; on rocky outcrop eight miles north of

Punda Maria in Kruger National Park, Rowland Jones 48 in National Herbarium,

Pretoria, No. 28636 (PRE, type); 48a in PRE No. 28637; Codd 5370.

A succulent shrub 1-2 m tall and about the same width, with a suppressed main

trunk rising only very shortly above ground and producing many spreading-erect

branches. Branches 5-7-angled, very rarely rebranched, constricted into segments

7-15 cm long; segments 3-5 cm broad near their base and narrowed gradually upwards

to the base of the segment above, v/ith the constriction about 2 cm broad; angles

acute, winglike, with a narrow continuous horny margin, paired spines and rudimentary

leaves. Spines in pairs about 1 cm apart, 5-10 cm long without prickles or with only

rudimentary ones on either side of the leaf-base or scar. Cymes solitary, 2-4 mm

above the spines, shortly pedunculate, consisting of 1 central male cyatheum and 2

lateral bisexual cyathia arranged in a plane parallel to the main axis; peduncle about

2 mm long and about as thick, bibracteate; involucre cup-shaped, glabrous, about

4 - 5-5 mm diam. with 5 glands and 5 small subquadrate, fimbriate lobes ; glands con-

tiguous, transversely oblong, 2-5-3 mm in their greater width, nearly flat on upper

surface, yellow. Ovary on a short gynophore about 1 mm long with a small rim-like

calyx ; styles about 1 - 5 mm long, united below the middle, with free portions about

1 mm long, spreading, bifid at tips; ovule attached within a hood, filling the cell;

capsule purple, more or less acutely 3-lobed, about 9 mm broad, seed globose, about

2-5 mm diam.

The first recorded specimen of the species was collected by Dr. L. E. Codd in

1949 when he visited Lt. Col. Rowland Jones, then in charge of the northern sector of

the Kruger National Park with headquarters at Punda Maria. Lt. Col. Rowland Jones

subsequently forwarded both flowering and fruiting material. It is now nearly 10

years since the plants were recorded by the two collectors as rare on the sand stone

ridges 8-9-2- rniles north-west of Punda Maria and although a constant search has been

maintained no further records have been made.

Several miles distant from the site of E. rowlandii occurs E. confinalis R. A. Dyer

which undoubtedly is a fairly near relative in spite of the fact that it grows into a tree

of about 25 ft. tall, generally has fewer angles to the branches and shows significent

differences in the inflorescence. There is a close superficial resemblance to E. water-

bergensis R. A. Dyer, which however, has more slender, parallel-angled branches and

differs in details of the inflorescence.

29

The combination of characters to be given taxonomic importance are the trunkless

habit, the segmented branches with 5-7 angles, the continuous horny margin to the

angles, the shortly pedunculate solitary cymes with 3 cyathia developed in a plane parallel

to the main axis and the relatively deeply 3-lobed capsule.

R. A. Dyer.

Euphorbia cussonioides Bally, sp. nov., affinis E. obovalifoUae A. Rich., sed foliis rotun-

dioribus, ramulis brevioribus, angustioribus, ad ramorum cacumina congestis,

ovarii denique calyci sine processis distinguenda.

Arbor 20-25 m alta, succulenta, spinosa, trunco erecto, cylindrico, ad 12 m alto,

80 cm diametro, ramis primariis paucis, spiraliter dispositis, ascendentibus, teretibus,

rare ramificis, ramulis numerosis ad cacumina coronatis. Ramuli usque 75 cm longi,

segmentati, inter segmenta constricti, ramifici, 3- raro 4-angulati, angulis valde com-

pressis, 2-3-5 cm latis, margine undulato vel obtuse dentato. Podarii in apicibus

dentium dispositi, cornei, brunneogrisei, breviter triangulati, 5 mm longi, aculeis binis

ad 1 • 5 mm longis, saepe obsoletis instructi. Folia sessilia, in transversum subrotunda,

ad 9-5 mm longa, ad 10 mm lata, succulenta, mox decidua. Cymae 1-4, ex podario

uno, 1-2 mm supra aculeos productae, perbreviter pedunculatae, 3 cythia gereantes.

Pedunculum bibracteatum. Bracteae late deltoideae vel late oblongo-ovatae, mem-

branaceae, margine denticulate, 3 mm latae, 2-3 mm longae. Cyathium primum mas-

culinum, deciduum. Cyathium bisexuale 6-5 mm diametro, in forma poculi, 5 glandulis

flavoviridibus, in transversum reniformibus, 3-4 mm latis, 1-3 mm longis, margine

integro, leviter incrassato, 5 lobis subquadratis, fimbriatis munitum. Capsida breviter

exserta, erecta, glabra, profunde trilobata, 10-12 m longa, 14-15 mm diametro.

Calyx obtuse trilobatus, lobis breviter dentatis munitus. Styli usque basim divisi,

erecti, ad 1 • 2 mm longi, apice bifidi. Semen subglobosum, 4 mm longum, 3 • 5 mm

diametro, griseomaculatum.

Plate III.

Kenya Colony.— Southern Prov.; Ngong Dist., at Ngong, McDonald {Bally

E 42) 8/1 /39 (EA, type) (EA, type; Central Prov.: Fort Hall Distr., Meru-Sagana,

Jackson {Bally E 171); Ndarugu, Bally; Thika, Bally; Gorge of Chania River, Piers.

Embu Distr., Bally, (Phot. CVI, 36).

With a height up to 25 metres this species is one of the tallest and most imposing

of the tree Euphorbias. Its comparatively late discovery in 1939, hardly twelve miles

from Nairobi, is partly due to its habitat, tall mixed Highland Forest in which its

lofty crown merges with the closed canopy, and partly to its general appearance. The

crown of E. cussonioides consists of comparatively few ascending branches, crowned

by clusters of short, thinly winged branchlets which, at a distance, resemble foliage;

its general appearance is not unlike that of Cussonia holstii, which is expressed in the

name of the new species. The type of forest in which E. cussonioides occurs comes under

the category of “ Mixed Highland Forest ” which, in the neighbourhood of Nairobi,

consists mainly of Brachylaena hutchinsii, Croton macrostachys, C. megalocarpus,

Calodendrum capense, Cussonia holstii, Olea chrysophylla, Stryclmos spp. Teclea spp.,

Warburgia ugandensis.

The range of distribution of Euphorbia cussonioides, as it is known at present, is

restricted to the Kenya Highlands at altitudes between 4,500 ft. and 6,000 ft., where it

occurs in a belt extending from the foot of the Ngong Hills just South of Nairobi north-

east wards for about 90 miles as far as Embu, on the southern slopes of Mr. Kenya where

the annual rainfall is between 34 and 60 inches. In spite of its fleshy branches and its

small deciduous leaves this species is not a xerophyte.

The Kikuyu-name is “kithuri”; in Embu the tree is known as “ mramba ”.

3871936-2

30

Fig. I. — Euphorbia cussonioides Bally.

I. a: cyme 1 X 4

b: bisexual involucre 1x4

c: bract 1x4

d: gland 1 X 10

e: lobe 1 X 10

II. spine-shield 1 X 10

III. seed 1 X 10

IV. leaf. 1 X 1

The timber is very soft, white and even-grained and should be well suited for the

manufacture of matches. The copious latex is white. It hardens when drying to a

guttapercha-like consistency.

A tree, 20-25 m high, fleshy, spiny, with an erect, cylindrical bole to 12 m high and

to 80 cm diam. Branches few, spirally disposed, ascending, terete, sparsely branched,

bearing at their tips numerous 3-4-angled, fleshy, green branchlets, branchlets up to

31

75 cm long, segmented, with constrictions between the segments; the segments up to

20 cm long, branching from the constrictions, 3- or rarely 4-angled, with angles much

compressed, 2-3-5 cm wide, their margins undulate or obtusely dentate. Spine-

shields disposed at the apex of the teeth, grey-brown, shortly triangular, 5 mm long,

with a pair of thin spines up to 1-5 mm long, sometimes obsolete. Leaves sesile.

transversely subrotund up to 9-5 mm long, 10 mm wide but mostly smaller, fleshy,

soon deciduous. Cymes produced from the spine-shield, 1 • 2 mm above the spine-pair,

one to four, very shortly pedunculate, bearing one central and two lateral cyathea each.

Peduncle fleshy, with two lateral bracts. Bracts broadly deltoid, or obovate-oblong,

membranous, 2-3 mm long, 3 mm wide, with a denticulate margin; central cyathium

male, soon deciduous, lateral cyathia bisexual, 6-5 mm diam., cupshaped, bearing 5

not contiguous, reniform, greenish-yellow glands with entire, slightly thickened, raised

margin, and with 5 subquadrate, fimbriate lobes. Capsule shortly exserted, erect,

glabrous, deeply 3-lobed, 10-12 mm long, 14-15 mm diam. Calyx obtusely 3-lobed;

its lobes shortly and irregularly dentate. Styles divided to the base, fleshy, erect,

1-2 mm long, with bifid apex. Seeds subglobose 4 mm long, 3-5 mm diam., grey

with darker brown-grey spots.

P. R. O. Bally.

RHAMNACEAE.

Ziziphus rivularis L. E. Codd, sp. nov., Z. pubescenti Oliv. affinis, sed cymis sessilibus

ovariis trilocularibus stylis 3 folds glabrescentibus differt.

Frutex vel arbuscula ad 7 m altus inermis; ramuli cano-tomentosi. Folia petiolata

lanceolata vel ovato-lanceolata serrulata glabrescentia, basi rotundata obliqua, apice

acuta; petiolus 4-10 mm longus; lamina 3 -5-6 -5 cm longa, 1 -2-3-4 cm lata. Stipulae

minutae. Cymae axillares sessiles, floribus 1-5; pedicelli 1-1-5 mm, deinde frugibus

maturatis 2-5-3 mm longi. Calyx 5-lobatus coriaceus, externe pubescens; lobi

trianguli, 1-5 mm longi, apicem versus crassiores. Corolla minuta; petala 5, obovata

unguiculata. Stamina 5, petalis opposita. Orbis carnosus, obscure 5-lobatus. Ovarium

in orbe depressum, cellulis 3; ovula solitaria; styli 3, subulati. F/-Mcto drupa globosus

indehiscens, diametro 6-7 mm.

Transvaal. — Soutpansberg: south end of Wyllie’s Poort, Gerstner 6011; 6037;

Codd 3009; 4823; 8348; Kruger National Park, Baiandbai, Lang in TM 32131;

Shingwedzi River, Lamont 26; Pongola River, van der Schijff3?>l9. Nelspruit: Kruger

National Park, beside stream in Lebombo Mountains, Crocodile Bridge Division,

van der Schijff' 3914 (PRE, type). Barberton: 6 miles south of Komatipoort, on banks

of Komati River, Codd 1111 .

Mocambique. — Goba, Hornby 712; Umbeluzi, Torre 6503.

Shrub or small tree up to 7 m high, unarmed; bark grey, smooth; branchlets and

innovations covered with a short, greyish tomentum. Leaves alternate, petiolate,

lanceolate to ovate-lanceolate, serrulate, apex acute, base rounded, often oblique,

from which three main nerves arise, sparingly pubescent on the nerves below and along

the margin, becoming glabrous with age; petiole 4-10 mm long; blade 3 -5-6 -5 cm

long and 1 - 2-3 - 4 cm broad. Stipules minute. Inflorescence an axillary, sessile cyme,

consisting of 3-5 (rarely 1 or 2) shortly pedicellate flowers; pedicels 1-1-5 mm, elon-

gating to 2 - 5-3 mm in the fruiting stage. Calyx 5-lobed, coriaceous, pubescent without ;

lobes triangular, 1 - 5 mm long with a thickened keel towards the apex within. Corolla

minute; petals 5, obovate, clawed, truncate at the apex, attached between the calyx

lobes. 5, opposite the petals ; filaments 1 mm long; anthers 2-celled, medifixed.

Disc fleshy, obscurely 5-lobed. Ovary superior, embedded in the disc, 3-celled; ovules

solitary, erect; styles 3, free above the disc, subulate; stigma scarcely differentiated.

Fruit globose, drupaceous, indehiscent, usually solitary in the leaf axil, 6-7 mm in

diameter, dark brown when ripe, shortly pedicellate; endocarp and septa relatively

thin, semi-woody; seeds usually 3 (rarely 1 or 2), compressed.

32

Material of this species has been accumulating at the National Herbarium for

some years. Specimens were sent to Kew Herbarium and were reported to be unmatched

in any species represented there. An attempt to name it in Suessenguth’s treatment

of the Rhamnaceae, Pflanzenfam. 20d (1953), was unsuccessful. It is unique among

South African members of the genus in being unarmed and in having a 3-celled ovary

with 3 styles, and a fruit in which the endocarp and septa between the cells do not

develop into a hard, stony centre. Nevertheless, these are not characters which would

exclude it from the genus Ziziphus.

Fig. 2. — Ziziphus rivularis L. E. Codd (van der Schijff 3974). 1, fruiting twig,

natural size; 2, transverse section through fruit, x 3; 3, flowering twig, X 10; 4, flower,

X 10; 5, section through flower, x 10.

33

In vegetative characters it shows a close resemblance to Z. pubescens Oliv., but

there are important floral differences. The main distinguising features are summarised

below: —

Z. pubescens. Z. rivularis.

Leaves Persistently tomentose. Glabrescent.

Cymes Shortly pedunculate, several- Sessile, 1- to 5-flowered.

flowered.

Ovary 2-celled. 3-celled.

Styles 2. 3.

Fruit Ovoid, apiculate. Globose.

Endocarp and

septa Thick and woody. Relatively thin, semi-woody.

Z. rivularis has so far been recorded only from the northern and eastern Transvaal

and from the neighbouring region of Mocambique. Most collectors record that it

is found on stream banks or in water courses, often among rocks, and this ecological

character has suggested the specific name.

It may be noted that Suessenguth, l.c. (1953) cites the genus as Zizyphus Mill.,

Gard. Diet. Abridg. Ed. 4 (1754) and refers to a discussion on the spelling of the generic

name by Loew in Flora der Juden, 3: 139 (1924). This work has not been consulted,

but the main facts appear to be as follows: The species Z. jujuba Mill, was known in

ancient times as “ zizyphon ” (Greek), “zizyphus” (Latin) and “Zizuf” (Arabic).

It was listed by Dodonaeus (1616) as Zizyphus and by Linnaeus, Sp. PL Ed. 1 : (1753),

as Rhamnus zizyphus. However, Tournefort, Inst. Rei Herb. Ed. 3: 627, t. 403 H719),

although he cites Dodonaeus, spells the generic name as Ziziphus and Miller, basing

his genus on Tournefort’s reference, adopted this spelling consistently in his fourth

and subsequent editions. Even if it is argued that this is a philological error, it was

deliberately used by Miller and would appear to be the spelling that must be taken up

according to the rules.

L. E. CODD.

SELAGINACEAE.

Dischisma struthioloides Killick, sp. nov., affine D. squarroso Schlechter, sed folds

angustioribus acutioribus differt.

Fruticulus multo ramosus, erectus, 30-60 cm altus, ramis divaricatis, partim albo-

lanatis, dense foliatis. Folia subreflexa, sessilia, anguste triangula, 3-5 mm longa,

basi 1-1 -5 mm lata, Integra, utrinque glabra, plus minusve glauca. Spicae terminales,

cylindratae, 1-3 cm longae; bracteae folds similes sed basi 2-5 mm latae et dimidio

inferiore cidatae. Calycis segmenta 2, libera, navicularia, 4 mm longa, 0-7 mm lata,

cidata. Corolla alba, tubo gracid, antice ad infra medium fisso, 1 • 1 cm longo, lobis

4, subaequadbus, oblongis, 2-5 mm longis, 1 mm latis, lobo quinto rudimentario,

0-8 mm longo, subulato. Stamina 4, superiora 1 • 5 mm longa, inferiora 0-5 mm longa,

antheris 1-5 mm longis, medifixis. Ovarium 1 mm longum; stylus 3 mm longus,

stigmate simpdei, 2-5 mm longo.

Cape Province. — Namaqualand: 11 miles east by south of Hondekdp Bay,

Strandveld on dunes, c. 500 feet, Acocks 14941 (PRE, type); Hondekdp, Pillans 163.

34

Fig. 3. — Discbisma struthioloides Killick ; A, habit, x I; B, flower x 10- C

pistil, X 20. 5 ,

35

An erect, much-branched shrublet, 30-60 cm high. Branches divaricate, white-

lanate in parts, densely foliate, older portions covered with persistent leaf-bases. Leaves

slightly reflexed, sessile, narrowly triangular, 3-8 mm long, 1-2 mm wide at the base,

glabrous, somewhat glaucous. Spikes terminal, cylindrical, 1-3-5 cm long; bracts

similar to the leaves but 2-5 mm wide, almost ovate and ciliate in the lower half. Calyx

segments 2, free, boat-shaped, acute, 4 mm long, 0-7 mm wide, ciliate. Corolla white;

tube slender, 1 ■ 1 cm long, slit to just below half-way, lobes 4, subequal, oblong, 3 mm

long, 1 mm wide, vestigial fifth lobe at base of slit, 1 mm long, subulate. Stamens

4, didynamous, upper 1 • 5 mm long, lower 0-5 mm long; anthers 1 • 5 mm long, medi-

fixed. Ovary 0-5-1 mm long; style 3 mm long, with few scattered glands; stigma

2-5-3 mm long.

This new species of Dischisma from Namaqualand appears to have been collected

only twice; first by Mr. N. S. Pillans in 1924 and then by Mr. J. P. H. Acocks in 1948.

The specific epithet struthioloides points to the superficial similarity of this plant to

species of Struthiola in Thymelaeaceae. So marked is this similarity that the Acocks

specimen was originally misidentified as a Struthiola. The nearest ally of D. struthio-

loides is D. squarrosum Schlechter which, however, has broader and blunter leaves.

D. J. B. Killick.

37

Plate I. — Ceropegia occulta R. A. Dyer.

38

Plate II. — Euphorbia rowlandii R. A. Dyer.

T

39

Plate III. — Euphorbia cussonioides Bally.

41

The Protea Species of the Summer Rainfall Area of

South Africa.*

By

J. S. Beard.

Introductory.

The genus Protea, R. Br. (nom. cons.), is distributed throughout Africa south of

the Sahara with two main centres of distribution, the Cape fynbos and the Central

African Brachystegia belt which stretches from Angola through the Rhodesias into

Tanganyika. The former contains about 70 species, and the latter, as at present known,

about 50. Four species only which are not represented within the tropical Brachystegia

belt occur to the north of it, bringing the total for tropicals up to 54. For the summer-

rainfall area of South Africa, which we may designate the sub-tropical belt linking the

tropical and Cape centres of distribution, this paper will deal with 12 endemic species,

bringing the total for the genus to 136. These are, therefore, equally divided between

the Cape and the rest of Africa.

The summer rainfall area of South Africa is, for Proteas, essentially a transitional

region and is thus in many ways of special interest. It is much poorer in species than

the main centres to north and south which it links and the individual plants are less

conspicuous in the landscape. Of the 15 species concerned, one {P. lacticolor) is essen-

tially a Cape fynbos species which extends its range into the southern part of our area,

and two (P. gaguedi and P. hirta) are tropical species of very wide distribution which

come down into Natal. The remaining 12 species are endemic to the summer-rainfall

area. None of them enters Cape vegetation to the southward or tropical vegetation

to the north. Two of them, P. subvestita and P. rouppelliae, are of Cape affinity,

belonging to sections of the genus (Exsertae, Ligulatae) otherwise only represented

among Cape fynbos species. Nine others belong to the typically tropical sections

Lasiocephalae and Leiocephalae, while three belong to a new section which it is proposed

to recognize, the Patentiflorae, which also contains a number of tropical species.

Taxonomy.

There have been only two complete treatments of the genus Protea in South Africa

since R. Brown’s work in 1810: Meisner in de Candolle’s Prodromus, 1856, and

Phillips and Stapf in the Flora Capensis Vol. V of 1913. The latter is therefore the

foundation on which this paper rests. The taxonomy of Protea is in general very

difficult and confused due to the horticultural popularity of the genus in Europe in the

late eighteenth century and the consequent large amount of literature from that period.

Fortunately for the present study, the summer-rainfall area Proteas were not discovered

early enough to share in that confusion, and it is only one solitary intruder from the

Cape flora, P. lacticolor Salisb., which is likely to present difficulty. It may be remarked

that whoever sets out to revise the Cape Proteas will have to undertake some very

laborious research into the eighteenth century work.

* (Preliminary account presented to the Annual Congress of the S.A. Association for the Advance-

ment of Science, Nelspruit, 1956.)

42

Phillips and Stapf divided the South African species of Protea into 12 sections,

arranged in two major groups; the first of these contained eight of the sections and

consisted of arborescent plants, while the remaining four sections were placed in a

group of suflfrutescent plants having underground rootstocks. In the writer’s opinion

this major grouping is artificial and separates sections that should properly be united.

We find the majority of summer rainfall species fall into the sections Ligulatae, Exsertae,

Lasiocephalae and Leiocephalae, a treatment which it is proposed to follow. However,

there are a number of dwarf species which would have to be placed among the Micro-

geantheae if Phillips and Stapf were followed literally : instead, they have been placed

among the Leiocephalae where they clearly belong naturally. Furthermore, there

are several others which may assume either a dwarf or arborescent habit according to

growing conditions and degree of veld-burning.

Two of the newly discovered species described here, P. comptonii and P. rubropilosa,

while closely related to one another, do not fit readily into any of the sections of Phillips

and Stapf: their heads are much larger than in the Lasiocephalae and Leiocephalae,

yet they are very distinctly dilferent from any of the large-headed Cape sections. A

curious feature, taken as diagnostic, is that the heads do not close again after anthesis

while the fruit is maturing but remain wide open or everted, and the section is thus

named Patentiflorae. Both heads and leaves are larger than in the Lasiocephalae and

Leiocephalae, whose heads are uniformly closed after anthesis. The previously known

species P. curvata is placed in this section, which will include also a number of the tropical

species, notably P. angolensis Welw., P. chionantha Engl, and Gilg, P. bella Hauman

and P. madiensis Oliv. The South African species are peculiar in that the growth-

habit and bark of the trees and the leaves in the case of P. rubropilosa and P. comptonii

are more typical of Faurea than of Protea, in particular resembling Faurea galpinii.

The treatment in this paper has perforce been based upon specimens in the South

African herbaria, supported by a large amount of field study. The types of those

species of more antique description are mostly in Europe, but they were obtained on

loan for study for which thanks are due to the Directors of the herbaria at Kew, Berlin-

Dahlem, Brussels and Leiden. Typification has been undertaken as prescribed in the

international code. None of the authors of the species in question indicated holotypes.

In cases where no duplicate specimens of the type have been traced the unicum has been

designated the holotype. In other cases a lectotype has been selected, naming as far

as possible a specimen which is at Kew or some other European herbarium.

Distribution and Ecology.

In mapping the distribution of our species, use may be conveniently made of the

vegetation map of the Union by Acocks, “ Veld Types of S. Africa ”, which accompanies

Botanical Survey Memoir No. 28.

The true Cape fynbos which is the real home of the Cape species of Protea is confined

to the winter rainfall area west of Uniondale. The constant rainfall zone contains a

fynbos mapped as “ False Macchia ” by Acocks which covers widely the various mountain

ranges in the Uniondale, Humansdorp and Port Elizabeth Divisions and is prolonged

in a narrow outlying strip along the Zuurbergen, around Grahamstown and at Riebeeck

East. This False Macchia is much poorer in Protea species than the true Macchia

further west and appears to contain only 19 of them. Fifteen of these do not range

east of Uitenhage, the outlying area up to Grahamstown containing four species only.

North and east of this, across the Keiskama river and along the scarp front of the

Winterberg and Amatola mountains, summer rainfall vegetation begins in the form of

various types of grassveld which continue north right to the Limpopo. Here one can

no longer expect to see Proteas more or less anywhere as is the case in the Cape fynbos.

Habitats are now much more restricted and the Proteas occur in localised patches.

Most of the summer rainfall area is covered with grassveld, which is a fire climax:

much of this again is potentially forest, though whether it has ever actually in whole

43

or part been, forest in recent geological times is another matter. The Proteas do seem

to indicate that forest has at least been rather more extensive than it is today. They

do not themselves occur, of course, in forest but they are found naturally in conjunction

with it. In many places today where there are forests on mountain sides one can study

this conjunction. At the foot of the slopes the forest is on deep soil and is very luxuriant ;

higher up the soil becomes more and more shallow and rocky, and the forest is reduced

in stature. Eventually, if the mountain is capped by extremely steep and rugged slopes,

rock slabs or krantzes, the forest gives way abruptly to an open Protea woodland.

There is little doubt from the relics that occur in protected situations that this open

Protea woodland as we now see it is a remnant, decimated by fire, of a Macchia floris-

tically and physiognomically similar to the Cape fynbos. Acocks several times gives

lists of associated Macchia species: various Ericaceae and Widdringtonia spp. were

undoubtedly prominent but have been largely eliminated by fire. The Proteas are

relatively fire-resistant and have survived. None the less there is a limit to their tolerance

and all too many of the Protea colonies that one sees today are patently in a state of

recession. The forest that they once bordered has long since disappeared, their

associated shrubs have gone and now they themselves are threatened with extinction

as the constant fires repeatedly eliminate their seedlings.

This ecological pattern is typical of the country along and to the east of the eastern

escarpment, where we may only look for Proteas in the high rainfall areas and on sites

too adverse for forest. In the central and western Transvaal the pattern changes into

something much more suggestive of the tropics. Here, certain Protea species inhabit

Acocks’ Sour Bushveld and Bankenveld: in the former they are general and in the

latter occur in the woodland which covers the rocky knolls, in both cases helping to

make up a type of savanna woodland of distinctly tropical appearance. As before,

however, we still find the Proteas occupying poor, rocky ground in the areas of highest

rainfall.

As mentioned above, Acocks’ Veld Types map has been drawn upon as an ecological

basis for mapping Protea distribution, since the Protea species can be simply correlated

with a small and defined number of veld types. The distribution of Proteas is not

continuous, in fact it is highly erratic, and to map separately every single isolated

locality of occurrence would be a formidable and rather unnecessary task. If, however,

it is known in which veld types they are to be found, a map of the distribution of those

veld types will divide the country within which locally Proteas may be found from

that where they are absent. A “ veld type ” is not a plant community, it is really a

landscape unit containing a mosaic of plant communities. Appropriate veld types

contain Protea woodland as one of the components of the mosaic. Proteas occur in

montane macchia in the following veld types: —

Acocks’ Type 1, Coastal Forest and Scrub. 3, Pondoland Coastal Plateau Sourveld.

44, Highland Sourveld, with its unnumbered subdivision to the south, the Dohne

Sourveld, and its separately numbered northern subdivision: 8, North-east Highland

Sourveld. 45, Ngongoni veld of the Natal Mistbelt. 63, Piet Retief Sourveld (in part).

Proteas occur in savanna woodland in the following: 20, Sour Bushveld. 61,

Bankenveld.

A small portion only of the Piet Retief Sourveld is affected, accompanying the

escarpment of the Karroo rocks from Wakkerstroom north to the Swaziland border.

It is probable that this strip should properly be regarded as a narrow extension of the

Highland Sourveld.

With the exception of Protea gaguedi all our species occur within the boundaries

of these veld types as mapped, unless here and there some mountain top carries a patch

of the veld type too small to be separately mapped. P. gaguedi is typical rather of

the ecotone between highland sourveld and bushveld and is thus frequently found

slightly outside the borders of the above-mentioned veld types.

44

Key to Sections.

1. Bracts exceeding or equalling the perianth 2.

Bracts very much shorter than the perianth 4.

2. Inner bracts spathulate, 2 mm wide in lower part, broadening to 10 mm wide at tip. . .1. Ligulatae.

Inner Bracts oblong 3.

3. Inner bracts pubescent, curved outwards away from the flowers: heads campanulate. . .II. Exsertae.

Inner bracts glabrous, curved inwards to enfold the flowers: heads globose or tur-

binate III. Leiocephalae.

4. Heads conical to globose, closing while the seed is ripening; perianth tube densely

shaggy-hairy IV. Lasiocephalae.

Heads everted and open while the seed is ripening; perianth tube glabrous or at most

lightly pubescent V. Patentiflorae.

Alternative Key to Sections.

Based on ratio total length of perianth to length of lip.

1 . Ratio 2 ■ 5 or less : Lasiocephalae.

2. Ratio 2-5 to 3-0: Ligulatae or Patentiflorae.

Length of teeth 12 mm: Ligulatae.

Length of teeth 3-4 mm: Patentiflorae.

3. Ratio 3-0 to 3-5: Leiocephalae.

4. Ratio over 4 : Exsertae.

Key to Species.

Section I. — Ligulatae. Heads large, 8-12 cm long, conical, inner bracts red, spathulate,

silky-tomentose, exceeding the flowers.

. Only species 1. P. rouppeUiae.

Section II. — Exsertae. Heads medium-sized, 5-7-5 cm long, campanulate, inner bracts

white or pink, silky-pubescent, recurved at tip, exceeding the perianths but not the

styles.

Leaves hairy in youth, distinctly veined: Heads 6-7-5 cm long, perianth lip

14 mm 2. P. lacticohr.

Leaves woolly in youth, indistinctly veined: Heads 5 cm long, perianth lip 10 mm

3. P. subvestita.

Section III. — Leiocephalae. Heads small to medium-sized, 4-8 cm long, globose or

turbinate, bracts greenish tinged with red, glabrous, slightly incurved, more or less

equalling the flowers.

1. Trees or divaricate shrubs: stems compound, flowering branchlets 5-15 mm

thick 2.

Low shrubs or smalt creeping plants: stems simple, flowering branchlets 3-5

mm thick 6.

2. Perianth 4 cm long, tube glabrous within 4. P. caffra.

Perianth 4-5-5 cm long, tube fulvously hairy within 3.

3. Leaves falcate and more than 15 cm long 6a. P. rhodantha var. falcata.

Leaves straight and under 15 cm long 4.

4. Most leaves less than 2-5 cm broad 5. P. multibracteata.

Most leaves more than 2-5 cm broad 5.

5. Leaves thin, venation prominent: heads 7-8 cm long 6. P. rhodanl/ia var. rhodantha.

Leaves thick, venation indistinct: heads 5-6 cm long 7. P. dracomontana.

6. Leaves less than 15 mm broad 7.

Leaves more than 15 mm broad 8.

7. Stems creeping: leaves all turned to one side of stem and very narrow, 7-10

mm broad 8. P. parvida.

Stems erect: phyllotaxis normal, most leaves 10 mm or more broad 9. P. simplex.

8. Leaves less than 25 mm broad 9.

Leaves more than 25 mm broad 10.

9. Heads 6 cm long, bracts 5 cm, perianth 4-5 cm, lip 15 mm 10. P. transvaalemis.

Heads 5 cm long, bracts 3-5 cm, perianth 3-5 cm, lip 10 mm 9. P. simplex.

10. Venation prominent, leaves thin and turning black when dry: perianth lip

15 mm 10. P. transvaalensis.

Venation indistinct, leaves thick and turning grey when dry: perianth lip

12 mm 7. P. dracomontana.

45

Section IV, — Lasiocephalae. Heads small to medium-sized, 4-8 cm long (rarely to 10 cm),

conical or globose, bracts greenish, hairy, with silvery or russet pubescence, very much

shorter than the flowers.

1. Adult leaves glabrous; bracts covered with dense silvery pubescence; heads

always solitary, terminal 11. P. gaguedi.

Adult leaves hairy; bracts with thin silvery to russet pubescence; heads often

lateral and/or clustered 2.

2. Leaves densely clothed with fine silky hairs and with longer russet hairs on the

lower half of the midrib and base 12a. P. hirta subsp. hirta.

Leaves apparently glabrescent but with closely adpressed silky hairs, especially

on the base of the midrib 12b. P. hirta svibsp. glabrescens.

Section V. — Patentiflorae. Heads medium to large, 6-10 cm long, everted, bracts red

or green, glabrous or hairy, very much shorter than the flowers.

1. Bracts densely rusty-pilose 13. P. rubropilosa.

Bracts glabrous or virtually so 2.

2. Bracts green, flowers white; leaves 3-6 cm wide 14. P. comptonii.

Bracts red, flowers pink; leaves 1-1 -5 cm wide 15. P. curvata.

1. P. rouppelliae Meisn. in DC. Prodr. 14: Til (1856); Phillips in F.C. 5: 573 (1913);

Bews in FI. Natal & Zululd.: 82 (1921); Burtt-Davy in FI. PI. Afr., t. 133 (1924),

FI. Transv.: 211 (1926).

Type: Magaliesberg, Zeyher 1457 (K, holo.!).

Scolymocephalus lanuginosus O. Ktze in Rev. Gen PI. 3: 279 (1891).

Protea lanuginosa K. Schum. in Just. Jahresb. 26, 1: 364 (1900).

P. transvaaliensis Gdgr. & Schinz in Bull. Soc. Bot. de France 50, 4: t. 13 (1913) non

Phillips. Type: Magaliesberg, Laidley 380 (L, iso.!)

P. rudatisii Engl, in Pflanzenwelt Afr. 3, 1 : 66 (1915) nomen.

A small, gnarled tree 3-5 m in height and usually assuming a neat domed form

with many spreading branches from a short trunk up to 20 cm thick. Bark black,

rough, deeply fissured. Flowering branchlets 8-12 mm thick, hairy when young,

rapidly becoming glabrous and developing a rough black bark. Leaves sessile or so

narrowly attenuate at the base as to appear subpetiolate, closely ranked into terminal

rosettes densely enveloping the flower heads, dark green, from linear-lanceolate to

obovate or even obovate-spathulate, 1-5-4 -5 cm wide by 6-16 cm long (average about

2-5 X 12 cm), 5 mm broad at the base, apex acute, base narrowed; blade leathery,

often somewhat stiffened, glabrous and shiny on the upper surface after shedding an

initial woolly covering, midrib prominent both sides, secondary venation ditto, reticulate,

without a marginal vein. Heads solitary, terminal, 8-12 cm long and as much in

diameter, conical, shortly pedunculate, opening to 90°. Receptacle convex, 15 mm

wide. Bracts about 10-seriate, silky-tomentose and ciliate, the outermost brownish,

short, ovate, obtuse, often recurved to revolute: middle bracts silvery-pink, obovate,

up to 1-5 cm broad; the innermost 8-10 cm long, exceeding the flowers, pink to rose,

spathulate, only 2 mm wide in their lower part, widening to 10 mm at the top. Perianth

1 cm long of which 1 -5 cm for the base, 3 cm for the tube and 2-5 cm for the lip and

teeth; base brown, glabrous, dilated, 3-keeled and 7-nerved; tube white, hairy, rather

slender; lip densely hairy, from white below to deep red above; teeth covered with

pink to crimson hairs, lateral teeth 12 mm, median tooth 8 mm long. Fertile stamens

3; filaments 1 mm long, flattened; anthers linear, 3 mm long; apical glands 0-5 mm

long, oblong, acute; barren stamen acute, eglandular. Style white, 6 cm long, curved,

somewhat flattened, keeled below on the convex side, usually more or less hairy: stigma

red, 9 mm long, slender, apiculate, kneed and bent at the junction with the style. Ovary

4 mm long. Fruit 15 mm long clothed with reddish-brown hairs.

Flowering mainly in autumn.

TABLE OF FLORAL AND VEGETATIVE MEASUREMENTS (MILLIMETRES).

46

47

This is one of the most decorative and horticulturally noteworthy of the Proteas

under discussion, having large and showy heads of a predominantly red colour. It

is also the most widespread in distribution, ranging from Port St. John’s to the Zout-

pansberg, from sea level to 6-7,000 feet, and found in practically all the mentioned

veld types. In the coast belt it ranges only from Port St. John’s to Durban, accompany-

ing the outcrop of the Table Mountain sandstone. Inland, this geological preference

is abandoned and it ranges almost throughout, being absent from the Dohne Sourveld

and the western end of the Transvaal Bankenveld.

There is considerable variation in size and shape of leaves throughout the extended

range of this species and also, though to a lesser degree, in the size of the flower heads.

As however, these variations are not consistent for given localities, it is considered

inadvisable to attempt to recognize varieties. The specimens studied show plants

from the Coast and Midlands of Natal frequently to have the largest and softest leaves

whereas those from the Central Transvaal and the Waterberg are of the opposite extreme

with short, stubby leaves, very stiff. North-eastern Transvaal and upper Natal speci-

mens tend to be intermediate, with specially narrow-leaved types from East Griqualand.

The size of flower-heads is reduced in unfavourable situations and vice-versa. Trans-

vaal specimens appear to have the bracts more noticeably silver-tomentose without

than Natal specimens, while along the eastern Transvaal escarpment heads are markedly

green coloured, red being limited to the lip, stigma and tip of the bracts.

Selected Citations.

Cape. — Matatiele: Baw 4975. Maclear: Galpin 6822. Port St. John’s: Sini

2547.

Orange Free State. — Harrismith: Witzieshoek, Flanagan 1849.

Basutoland. — Mafube: Dieterlen 946.

Natal. — Nkandhla: Qudeni, Gerstner 1862. Pietermaritzburg: Table Mt.,

McLean 145. Inanda: Groenberg, Medley Wood 7918.

Swaziland.— Dalriach : Bolus 12266. Mbabane: Burtt-Davy 2789.

Transvaal. — Pietersburg: Blouberg, Codd and Dyer 8980; 8984; Strydpoort-

hQxgQn, Maguire 9\2. Rustenburg: Magaliesberg, Zpy/ier 1457; LmW/cy 380. Krugers-

dorp: Mogg 20259; 23204. Belfast: Galpin 13091. Lydenburg: Sekukuniland,

Barnard 311. Soutpansberg: Galpin 14950. Pietersburg: Houtbosch, Bolus 10951.

Pilgrim’s Rest: Graskop, Galpin 14486. Barberton: Burtt-Davy 353.

2. P. lacticolor Salisb. in Parad. Lond. t. 27 (1806); Phillips in F.C. 5: 578 (1913).

Protea orientalis Sim in For. Flop. Cap. 296: t. 128 (1907).

Type: Pirie, Sim 1478 (BOL, CTM, iso.!).

A small spreading tree up to 4 m in height generally with a neat, domed form.

Trunk up to 2() cm in diameter with grey smooth bark, horizontally ringed. Flowering

branchlets 5-8 mm thick, somewhat woolly-hairy in youth, later glabrous with smooth

pale brown bark. Leaves sessile, densely crowded into rosettes at the end of the

branchlets, dark green, oblong, 1 ■ 5-3 cm wide by 5-10 cm long (average about 2-5x8

cm), apex obtuse or subacute, narrowed at the base; blade rigid, in youth covered

somewhat with hairs which later disappear or persist only on the midrib or base, minutely

punctate, distinctly veined: midrib yellow, slightly prominent both sides. Axillary

buds conspicuous. Heads solitary, terminal, fairly constant in size, 6-7-5 cm long and

as much in diameter when open, sessile, campanulate, expanding only to a narrow

angle, enclosed by a few bracteolar foliage leaves. Receptacle flat to slightly convex.

48

12-15 mm wide. Bracts 6-8 seriate, the outer pinkish or greenish-brown, acute, silky-

pubescent, ciliate and with an apical tuft of hairs : the inner either white or pink, 1 cm

wide at broadest, tips obtuse, white-ciliate, recurved outwards, 6-7 cm long, exceeding

the perianths but not the styles. Perianth very slender and thin in the upper half which

is eventually spirally coiled up and withdrawn from the styles, reddish-hairy except

at the base, 6 cm long of which 18 mm for the base, 28 mm for the tube and 14 mm for

the lip; lip glabrous, 3-toothed; lateral teeth 1-5-2 mm long, red or white hairy;

stamens all fertile, subsessile; anthers linear, 9 mm long, apical glands ^ mm long,

ovate, obtuse, swollen on the inner face. Style white, glabrous, 6-5 cm long, straight,

tapering above, compressed below, subterete above, grooved on one side; stigma red

or white, very slender, 9 mm long passing with an abrupt thickening into the style.

Ovary 3 mm long. Fruit curved, 5 mm long clothed with long brown hairs.

Flowers in late summer (February to April).

This species is essentially an inhabitant of the Cape fynbos from Worcester to

Uitenhage, entering the summer rainfall area in the Amatola and associated mountains.

Its taxonomic identity has been accepted without critical study, following Phillips

in Flora Capensis. The original description was in 1806 and the name (“ milk-

coloured ”) refers to the white colour of the heads. Actually pink-headed forms seem

to be at least equally common. Perhaps for this reason Meisner changed the name

to P. latericolor (“ brick-coloured ”).

Selected Citations.

Cape. — Stockenstroom : Katberg, Dyer 369. Keiskamahoek : Hogsback, Beard

759. Stutterheim: Mountains, Flanagan 1703. King William’s Town: Pirie, Sim

1478.

Phillips and Stapf in FI. Cap. recognise a variety orientalis of P. lacticolor, based

upon a reduction of P. orientalis Sim of which the type is Sim 1478. After examining

this specimen I am unable to establish any points of difference from the type of P.

lacticolor and accordingly cannot recognise the variety.

3. P. subvestita N.E. Br. in Kew Bull. 1901: 132; Phillips in F.C. 5: 578 (1913);

Bews in FI. Natal & Zululd. (1921).

Type: Faku’s Territory (Pondoland), Sutherland s.n. (K, holo.).

The description given above for P. lacticolor serves for P. subvestita likewise, with

the following amendments: —

Branchlets distinctly woolly-hairy in youth. Leaves covered in youth with a

dense woolly layer which is later dehiscent, leaving the blade glabrous or almost so:

leaves rather more fleshy, so that venation is indistinct except sometimes on the upper

surface and the blade can no longer be seen to be minutely punctate. Pleads and

flowers shorter. Pleads 5 cm long, perianth rather more thickly hairy, teeth of the

lip white-woolly. Perianth 5 cm long of which 15 mm for the base, 25 mm for the tube

and 10 mm for the lip. Anthers 5 mm long. Style 5 cm, stigma 5 mm.

P. subvestita R. Br. is so close to P. lacticolor that there was some temptation to

reduce it in rank as a geographical subspecies. The distinguishing differences are

merely such as one would expect to be associated with life at higher altitude in a more

rigorous climate. P. subvestita is common on high mountains throughout the High-

land Sourveld proper, seldom much below the 6,000 feet contour and favouring the

steepest and most rocky places. None of the species of this section appear to have

much fire resistence and are confined to sites where fires are less fierce.

49

Selected Citations.

Cape. — Xalanga: Cala Mission, Pegler 1650. Maclear: Tent Kop, Galpin 6824.

Umtata: Baziya Mountain, Baur 624.

Basutoland. — Near Qacha’s Nek: Dieterlen 993.

Natal.— Underberg : McClean 678. Estcourt: Cathkin Park, Galpin 11748.

Bergville: National Park, Pardoe, s.n. Klip River: Nr. van Reenen, Medley Wood

5631. Nkandhla: Qudeni, Gerstner 3945.

Note. — Protea mundii Klotzsch in Otto & Dietr. Gartenzeit. 1838: 113.

This is a winter rainfall species to which two collections from the summer rainfall

area were wrongly allocated by Phillips in FI. Cap. : Baur 624, the Baziya Mts., Umtata,

and Flanagan 1703, mountains near Stutterheim. I have examined both above-

numbered specimens in CTM and the former is P. subvestita, the latter P. lacticolor.

It is noteworthy that Flanagan 1703 is also cited in FI. Cap., with a slightly different

locality, under P. lacticolor. P. mundii has thus never been collected east of van

Staaden’s Berg. It is very readily distinguished by its capitate stigmas.

4. P. caffra Meisn. in DC. Prodr. 14: 237 (1856); Phillips in F.C. 5: 585 (1913);

FI. PI. Afr. t. 22, (1921), as P. abyssinica, in error; Burtt-Davy in FI. Transv. 211

(1926).

Type: Magaliesberg, Zeyher 1458 (P. lecto. !, CTM, iso.!).

A small gnarled tree up to 5 m in height, usually assuming a fairly neat domed

form with many spreading branches from a short trunk up to 20 cm thick. Bark

black, rough, deeply fissured. Flowering branchlets 10 mm or more thick (drying to

less), grey-brown, glabrous. Leaves sessile, closely ranked on the flowering branchlets,

pale green drying to grey-green, lanceolate or oblong-lanceolate, 1 • 5-2 • 5 cm wide by

8-12 cm long (average about 2 x 10 cm), 5 mm broad at the base, apex subacute or

obtuse, narrowed towards the base; blade leathery, stiff, glabrous, with a thin car-

tilaginous margin, midrib yellow, prominent both sides, secondary venation very

indistinct even in dried specimens. Heads solitary, terminal, 5-7 cm long and as much

in diameter, broadly globose, pedunculate with a rather solid woody stipes 1 • 5 cm

long, opening to 90°. Receptacle slightly convex, 25 mm wide. Bracts 5-seriate above

the stipes, at first very finely silky-tomentose, soon glabrous; those on the stipes scaly,

greenish or brown, outer bracts of the head greenish developing brown tips, inner

pink to red, obtuse, up to 5 cm long and 2 cm broad, slightly shorter than the flowers.

The bracts are somewhat woody in this species. Perianth pinkish-white, glabrous

except for fugacious hairs on the lip spirally coiled up and withdrawn in mature heads

4 cm long of which 12 mm for the base, 16 mm for the tube, and 12 mm for the lip;

lateral teeth 1 mm long, median tooth under 1-5 mm; teeth tufted with fugacious

tawny-white hairs. Stamens all fertile; filaments 0-5 mm long, flattened, concave;

anthers linear, 9 mm long. Style white, 4 cm long, curved, tapering, flattened, keeled on

both sides, glabrous; stigma 8 mm long, filiform, passing almost imperceptibly into the

style. Ovary 6 mm long. Fruit 12 mm long, thickened towards the top and bent

over, clothed with long brown hairs.

Flowers in summer (November to February).

Protea caffra Meisn. is characteristic of rocky ridges in the Transvaal Sour Bushveld

and Bankenveld, to which areas it is confined except for outliers in the Lulu Mts. of

Sekukuniland, and the mountains of the Leribe district of Basutoland. The specimens

from these outlying areas have been very critically examined and are definitely P. caffra.

In all cases this species inhabits drier, colder conditions than its relatives.

50

Selected Citations.

Basutoland. — Leribe: Dieterlen 319*.

Transvaal. — Waterberg: Naboomspruit, Galpin 11665. Rustenburg: Magalies-

berg, Zeyher 1458. Krugersdorp : Mogg 20258. Heidelberg: Leendertz 7682.

Pietersburg: Strydpoortberge, Welcome Mine, Beard 859. Lydenburg: Lulu Mts.,

Mogg 16885.

5. P. multibracteata P/n7//p^ in Kew Bull. 1910; 230; Phillips in F.C. 5: 586 (1913);

Bews in FI. Natal & Zululd. 82 (1921).

Type: British Kaffraria, Cooper 86 (K, lecto. !).

Protea pegleri Phillips in Kew Bull. 1910: 230. P. natalensis Phillips l.c. : 231. P.

6own7 Phillips l.c. : 232. P. Phillips l.c. ; 232, in part from syntype; Gwen-

kala River, Flanagan 804 (CTM).

An irregular gnarled tree 1-5 m in height. Trunk up to 15 cm thick, bark black,

thick and fissured. Flowering branchlets 5-10 mm thick, glabrous, at first greenish

or reddish. Leaves sessile, often fairly closely ranked at the ends of the branchlets,

light green, from narrowly oblong to linear, 8-25 mm wide by 8-15 cm long (average

about 12 cm long, width highly variable), sometimes somewhat falcate, apex subacute

to subobtuse, narrowed at the base, sometimes so much as to appear sub-petiolate ;

blade soft, quite glabrous, midrib red, prominent both sides, secondary venation

inconspicuous in fresh leaves, moderately conspicuous, finely reticulate and prominent

both sides in dried specimens, uniting near the margin but with no true marginal vein.

Heads terminal, solitary (or in some Eastern Cape specimens, clustered), 4-5-6 cm

long and 5-8 cm in diameter, turbinate, generally peduncled with a small scaly stripes

some 5 mm long, opening to 160°. Receptacle slightly convex, 15-20 mm wide.

Bracts about 7-seriate above the stipes, the outer green, at first with a glaucous bloom,

the inner greenish to red, colour variable; maximum width 1-5 cm (or 2 cm in-some

very large heads), 4 cm long, equalling or slightly shorter than the styles, oblong, obtuse.

Perianth white, fulvously hairy within the tube, the lip edged with fugacious pink or

brownish hairs, either densely or tipped only; perianth otherwise glabrous, 4-5 cm

long of which 12 mm for the base, 18 mm for the slender tube and 15 mm for the lip;

lateral teeth 1 -5 mm long, median 1 mm; stamens all fertile, subsessile, anthers linear,

8 mm long; the perianth spirally coiled up and withdrawn in mature flowers. Style

white, 5-6 cm long, slightly curved or wavy, somewhat compressed, glabrous; stigma

usually pink, 7-9 mm long, filiform, obtuse, passing almost imperceptibly into the

style. Ovary 3 mm long. Fruit 12 mm long clothed with golden-brown hairs.

Flowers in summer (November to February).

P. nmltibracteata Phillips ranges throughout Natal and the eastern Cape in all the

mentioned veld types, from the steamy flats of the Zululand coast to the Drakensberg

at 7,000 feet, and south to the Amatola Mts. This is a moderately severe habitat,

with some extremes of heat and cold, less cold but much hotter than that of P. caffra.

It is everywhere accompanied by the dwarf P. simplex Phillips.

The width of the leaves varies considerably in this species, being typically very

narrow at the southern end of the range and growing steadily broader towards the

north. The distribution is not in contact at any point with either P. caffra or

P. rhodantha.

* One of the two sheets of this number in the Natal Herbarium is P. dracomontana Beard.

51

Selected Citations.

Cape. — Kentani: Pegler 1872. Komgha: Gwenkala River, Flanagan 804. Port

St. John’s: Hutchinson 1828. British Kalfraria: Cooper 86. Stockenstroom: Kat-

berg, Dyer 370. Matatiele: Baur 4839.

Orange Free State. — Drakensberg, Orange Free State: Cooper 951.

Natal. — Umzinto: Umtwalumi, Beard 745. Pinetown: Springfield, Medley

Wood 12822. Eshowe: Gerstner 1863. Stanger: Tugela Beach, Johnson 390; 620.

Richmond: Beardl^6. Bergville: National Park, and McC/con 141 ; Cathedral,

Killick 1061. Nkandhla: Qudeni, Bayer 791. Vryheid: Hlobane, Johnstone 436.

Ubombo: Lebombo Mts., Sargent s.n.

Swaziland. — Lebombo Mts. S. of Stegi: Verdoorn 1679.

Transvaal. — Amersfoort: Wakkerstroom, Beard 747.

Phillips in Flora Capensis sustained P. multibracteata, reducing to synonymy

P. pegleri, P. natalensis and P. baurii which had been simultaneously published in 1910.

P. flanaganii is now also reduced as it has not been possible to establish a separate

identity for this species in the field (see further discussion under P. simplex).

6. P. rhodantha Hook, f., var. rhodantha, in Bot. Mag. t. 7331 (1893); Phillips in F.C.

5: 586 (1913); Burtt-Davy in FI. Transv. 211 (1926).

Type: Cultivated at Kew, Horn s.n. (K, holo.!).

Protea bolusii Phillips in Kew Bull. 1910: 231.

Type: Forbes Reef, Bolus 12265 (BOL, iso.!).

P. stipitata Phillips l.c. 1934: 104.

Type: Haenertsburg, Murray s.n. (PRE, iso.!).

Tree up to 15 feet in height, commonly rather less, branching low down and of

rambling habit, trunk up to 50 cm thick, bark black, fissured, warty. Flowering

branchlets 10-15 mm thick, glabrous, with smooth bark, at first green or pinkish, soon

pale brown. Leaves sessile, crowded towards the end of the branchlets, green, oblong

to oblanceolate, often slightly falcate or oblique, 2-5-5 cm wide by 10-18 cm long

(average about 3-5 x 14 cm), apex subacute, narrowed at the base; blade leathery,

quite glabrous, with thin, transparent, cartilaginous margin, midrib red or yellow,

prominent both sides, secondary venation reticulate, prominent both sides in dried

specimens, with no true marginal vein. Heads terminal, solitary, or, rarely, clustered,

7-8 cm long and 12 cm in diameter, turbinate, pedunculate with a well-marked scaly

stipes 1 cm long, opening to 180° at anthesis. Receptacle slightly convex, 25 mm wide.

Bracts about 6-seriate above the stipes, the outer ovate, subacuminate, green or more

or less pink, at first silky-pubescent below, ciliolate; inner oblong, deep rose-colour,

glabrous, up to 2 cm wide and 5 cm long, equalling or slightly shorter than the flowers.

Perianth white suffused with pink, glabrous except within the tube where it is fulvously

hairy and at first on the lip, 5 cm long of which 15 mm for the base, 20 mm for the tube

and 15 mm for the lip; tube expanded below, slender above, spirally coiled up in old

flowers; lip clothed at first with orange hairs, soon glabrous; teeth subequal, recurved,

1 mm long. Anthers linear, 10 mm long. Style pink or white, 5-5 cm long, distinctly

swollen just above the ovary, stout when fresh, when dry becoming compressed, grooved

or quadrangular, curved, glabrous; stigma 10 mm long, filiform, slightly wavy at the

Junction with the style. Ovary 3 mm long. Fruit 12 mm long clothed with golden-

brown hairs.

Flowers in summer (December to March).

52

P. rhodantha Hook. f. is characteristic of the north-east Highland Sourveld and

thus of the eastern Transvaal and Swaziland. The habitat is wetter and milder than

that of P. multibracteata and more genial than that of P. caffra. At the southern end

of its range in the Barberton mountains and Swaziland this species is represented by

its variety falcata in which the leaves become much longer, narrower and curved.

Selected Citations.

Swaziland. — Forbes Reef: Bolus 12265. Mbabane: Compton 26343.

Transvaal. — Soutpansberg: Between Rumble’s farm and Lejuma Peak, Reynolds

4109; Entabeni, 872. Pietersburg: Haenertsburg, Mw/roy s.n. ; Wolkberg,

Beard 861. Pilgrim’s Rest: Beard 871; Horn s.n.; Mauchsberg, Smuts and Gillett

2261. Barberton: Meeuse 10107.

This species has been most difficult to elucidate. Protea rhodantha (the “ red-

flowered ”) was figured in the Botanical Magazine in 1893 with a description by Sir

Joseph Hooker from a plant raised at Kew, the seed having been presented by Mr.

W. J. Horn in 1886, collected at Pilgrim’s Rest in the Transvaal. Unfortunately it

is somewhat likely that a plant raised in a Kew greenhouse would dilfer from its wild

prototype, particularly in point of size: furthermore the type material that has been

preserved is inadequate, consisting only of a single leaf 14-5 x 3 cm in size and a

small head, detached and bisected, 5 cm long and 5 cm in diameter. It is hardly sur-

prising that the species has been imperfectly understood.

In 1906 a collection of Dr. Bolus, No. 12265, was compared by him at Kew with

the above type material of P. rhodantha and determined as that species. In Kew Bull.

1910 Phillips erected a new species, P. bolusii, upon this specimen but in FI. Cap. (1913)

reduced it to synonymy under P. rhodantha. The Bolus collection and the type are the

only specimens cited in FI. Cap. under P. rhodantha. In recent years specimens of

a taxon found in the Natal Drakensberg have been referred to P. rhodantha on the

authority of Kew, but no plants occurring in the Pilgrim’s Rest district or elsewhere

on the Transvaal Drakensberg have, so far as I am aware, been attributed to that

species. In 1934 Phillips described a new species, P. stipitata, from Haenertsburg,

“ aff. P. rhodanthae Hook. f. sed foliis et capitulis majoribus, involucri bracteis nume-

rosioribus inferne dense et molliter tomentosis differt ”.

The essential step towards an understanding of P. rhodantha was clearly an inves-

tigation as to what Proteas of the section Leiocephalae could be found in the Pilgrim’s

Rest district. This was carried out by Dr. A. D. Meeuse and myself in February,

1957, after first visiting at Haenertsburg the type localities of P. stipitata and P. trans-

vaalensis and collecting them in flower. P. stipitata was found to be very common

near Pilgrim’s Rest, to the west on the road to Ohrigstad, to the east at Kowyn’s Pass

below Graskop and to the south all along the eastern slopes of the Mauchsberg and

Mt. Anderson. At the very summit of Mt. Anderson some dwarf plants were found,

sterile but apparently P. transvaalensis, and at the summit of the Long Tom Pass others,

also sterile but demonstrably P. parvula.

The latter is distinctly different from P. rhodantha in its creeping habit and exceed-

ingly narrow leaves. P. transvaalensis has about the same sized heads and leaves

as the type of P. rhodantha but its leaves are distinctive, turning black when dried and

of different shape: the head also is of different shape, long in proportion to breadth.

It must be admitted that P. stipitata comes closest to P. rhodantha. The leaves are

a very good match and the floral characters agree if we allow for reduction in size due

to stunting in the Kew cultivated specimen. The common and conspicuous distribution

of this form near Pilgrim’s Rest makes it very likely that its seeds would have been

collected by Mr. Horn. It is accordingly concluded that P. stipitata falls into synonymy

under P. rhodantha.

53

The Natal Drakensberg specimens belong to a distinct species that is not found

in the Pilgrim’s Rest district and thus they cannot be P. rhodantha. They are placed

under P. dracomontana.

6a. P. rhodantha Hook. /., var. falcata Beard, var. nov.

Type; Barberton, Beard 810 (PRE, holo.).

A varietate rhodantha folds subter attenuatissimis, saepe quasi petiolatis, demissis,

falcatis tamen subter inflexis, longioribus differt.

At the southern end of the range of this species, in the Barberton mountains, one

finds mainly this varietal form which differs from typical P. rhodantha only in its very

long, drooping and falcate leaves, as follows: —

Leaves frequently so long attenuate at the base as to appear petiolate, grey-green,

drooping, very variable as to size and shape, linear-oblanceolate to oblanceolate,

almost always falcate or at least curved in the lower half, 1 • 5-3 cm wide by 1 5-25 cm

long (average about 2 x 20 cm), apex obtuse, base cuneate; blade glabrous, with a

membranous margin, midrib sub-prominent, secondary venation indistinct.

Swaziland. — Ngwenya Mts. : Compton 835 in herb. J. S. Beard. Mbabane:

Compton 23825.

Transvaal. — Barberton: Makonjwa Mts., Beard 810; Meeuse 10117; Pott

5676; Clarke s.n.

7. P. dracomontana Beard, sp. nov.

Type; Cathedral Peak, Killick 1337 (PRE, holo.!, NH, iso.!).

Frutex divaricata 1-1 ■ 5 m alta, vel saepe nana ramulis simplicibus ex radice crasso

editis. Folia elliptica, 3 cm lata, 10 cm longa, lamina crassa, rigida, glabra, margine

translucente. Capitula terminalia, solitaria vel fasciculata, 5-6 cm longa. Involucri

bracteae 5-seriatae, rubrae vel rubrescente-virides, ciliatae, 4-5 cm longae. Calyx

punicea, tubo intus fulvo-piloso, limbo capillis fugacibus cristato. Stylus 4 cm longus,

stigma 8 mm.

Potentially a divaricate shrub up to 1 metre tall, rarely 1 - 5 m, but most commonly

adopting a dwarf form with numerous simple stems 30-50 cm long arising from a

thick rootstock and repeatedly killed back by fire or frost. Shoots woody, 5 mm

thick, bark reddish, glabrous. Leaves bluish-green, sessile, evenly spaced along the

shoots, elliplic, 2- 5-3 -5 cm wide by 8-12 cm long (average about 3 X 10 cm), apex

obtuse, base cuneate; blade thick, stiff, quite glabrous, midrib red, prominent both

sides in the lower | of the leaf, secondary venation inconspicuous, the main veins

slightly prominent below; margin translucent. Heads temiinal, solitary or clustered,

5-6 cm long by 7 cm in diameter, oblong to turbinate, pedunculate with a small scaly

stipes 0-5 cm long, opening to 180° at anthesis. Receptacle convex, 2 cm broad.

Bracts about 5-seriate above the stipes, red or green edged with red, minutely ciliate,

otherwise glabrous, up to 2 cm broad and 4-5 cm long, equalling or slightly exceeding

the flowers. Perianth pink, fulvously pilose within the tube and on the lip (the latter

hairs fugacious), otherwise glabrous, 4 cm long of which 12 mm for the base, 16 for

the tube and 12 for the lip: teeth subequal, 2 mm long. Anthers linear, 6 mm long.

Style white, flattened and grooved, 4 cm long, glabrous: stigma slender, 8 mm long,

slightly wavy at the junction with the style. Ovary 2-3 mm long. Fruit 12 mm long

clothed with golden-brown hairs.

Plate I.

Flowers in summer.

54

Confined to the Natal Drakensberg from Underberg round to Leribe in Basutoland,

at a high level between 6,000 and 8,000 feet elevation, on the basalt.

Basutoland. — Leribe: Dieterlen 319 (in part).

Natal. — Underberg: Himeville, Bews 36. Estcourt: Giant’s Castle, Bruyns-

Haylett 52; Highmoor Forest Reserve, v. Rensburg 4. Bergville: Cathedral Peak,

Killick 163S-, 1337, 729; Esterhuysen s.n.; ~Ndedema, Esterhuy sen 11367 ; Mweni,

Esterhuysen 14673; National Park, Hutchinson, Forbes and Verdoorn 89; Pardoe

s.n.; Edwards 505; 508.

The various collections of this species which exist in S. African herbaria have

previously been referred to P. rhodantha Hook. f. On acquaintance with the population

in the field, it is found to be a distinct taxon differing completely in its broad, fleshy

leaves from any of the most nearly related species. As it does not occur in the Pilgrim’s

Rest district it cannot be P. rhodantha and is accordingly named as new.

8. P. parvula Beard, sp. nov.

Type: Dullstroom, Galpin 13149 (PRE, holo.!).

Suffrutex caulibus subterraneis, ramulis prostratis glabris. Folia sessilia, erecta,

anguste lineare-oblanceolata, 8 mm lata, 8 cm longa, saepe aliquantum falcata, apice

acuta, ad basin versus attenuatissima. Capitulum 4 cm longum, stipitatum. Involucri

bracteae 5-seriatae, puniceae, glabrae, 3 cm longae. Calyx alba, glabra capillis paucis,

3 • 2 cm longa, limbus dentatus 1 cm longus, dentes 1 mm, cristati. Stylus 3 cm longus,

glaber, compressus; stigma 6 mm filiforme, apice obtusum.

A small woody plant with branching underground stems 1-2 cm thick. Aerial

flowering branchlets prostrate, 3 mm thick, glabrous with pale greenish to reddish

bark. Leaves sessile, closely ranked, all turned to an erect position on the same side

of the stem, light green, narrowly linear-oblanceolate, 7-10 mm wide at the broadest

point by 6-12 cm long (average about 0-8 x 8-0 cm), often somewhat falcate, apex acute

to sub-rotundate, very long attenuate at the base: blade coriaceous, glabrous, with

a thin membranous margin, midrib yellow, prominent below, sub-prominent above;

secondary veins fairly distinct, arching and prolonged close to the margin so as to give

the impression of a continuous marginal vein. Heads solitary, terminal, 4 cm long and

as much in diameter when open, globose, pedunculate with a slender scaly stipes 7 mm

long, opening to 160° at anthesis. Receptacle convex, 15 mm wide. Bracts 5-seriate

above the stipes, pale pink, glabrous, up to 15 mm broad and 3 cm long, equalling

the flowers, obtuse, shortly ciliate. Perianth creamy-white flushed with pink, glabrous

except for a variable pilosity along the inner edges of the tube and the tufted teeth;

3-2 cm long of which 10 mm for the base, 12 mm for the tube and 10 mm for the lip:

lateral teeth 1 mm long with a tuft of hairs. Anthers linear, 6 mm long. Style white,

3 cm long, glabrous, compressed and grooved, bulbously swollen above the ovary;

stigma 6 mm, slender, obtuse, passing almost imperceptibly into the style. Ovary

2 mm long. Fruit clothed with long brown hairs.

Flowers in midsummer, (December to January).

This species is among the smallest of the creeping Proteas. It is localised on

certain mountain tops in the Eastern Transvaal at 5,500 to 7,000 feet elevation, in very

short grassland.

Transvaal. — Belfast: Suikerboskop, Galpin 13149; Bruce 492. Pilgrim’s Rest:

Mariepskop, Hebronberg, Beard 868. Lydenburg: Steenkampsberg, Codd 1715;

Acocks 12922; Mount Anderson, Smuts and Gillett 2380; Makobulaan, Beard 874.

Nelspruit: Kaapsche Hoop, on top of the mountain, Phillips 3464; v. d. Merwe 1538.

55

9. P. simplex Phillips in Kew Bull. 1910: 230; Phillips in F.C. 5: 588 (1913); Bews

in FI. Natal & Zululd. 82 (1921); Burtt-Davy in FI. Transv. 211 (1926).

Type: Maclear, Galpin 6823 (K, lecto. PRE, BOL, GRA, iso.!).

Protea doddii Phillips in Kew Bull. 1911: 82.

Type: East London, Galpin 7936.

P.flanaganii Phillips l.c. 1910: 232 in part, from syntype: Kentani, Pegler 274 (GRA,

PRE, BOL, CTM).

A dwarf shrub with a thick underground rootstock from which arise numerous

simple, ephemeral erect stems, 5 mm thick and up to 50 cm long, glabrous with pink

to green bark. Leaves sessile, deep green, arranged all along the shoots, from narrowly

oblong to linear, very variable as to size, (especially in width), from 8-25 mm wide by

8-12 cm long (average about 16 mm x 10 cm), apex subacute to sub-obtuse, base

cuneate: blade leathery, glabrous, midrib red or yellowish, prominent both sides,

secondary venation prominent. Heads solitary, terminal, 5 cm long and as much in

diameter, turbinate, generally peduncled with a small scaly stipes some 5 mm long,

opening to 160° at anthesis. Receptacle convex, 15 mm broad. Bracts about 5-seriate

above the stipes, the outer green, the inner greenish to red, 3-5 cm long and up to 12

mm broad, equalling or slightly shorter than the styles. Perianth white, fulvously

hairy within the tube, the lip edged and tufted with fugacious pink or brownish hairs;

perianth otherwise glabrous, 3-5 cm long of which 10 mm for the base, 15 mm for the

tube and 10 mm for the lip; teeth up to 2 mm long with fugacious tuft of orange hairs.

Anthers linear, 6 mm long. Style white, 4 cm long, somewhat compressed, glabrous;

stigma usually pink, 5 mm long, filiform, obtuse, passing almost imperceptibly into the

style. Ovary 2 mm long. Fruit 12 mm long clothed with golden-brown hairs.

Flowers in summer.

Found in the Dohne Sourveld, Flighland Sourveld and Coast Belt, accompanying

P. multibracteata, also in the N.E. Highland Sourveld in Swaziland.

Selected Citations.

Cape. — East London: Dodd in Herb. Galpin 7936. Keiskamahoek: Story

3836. Kentani: Pegler 274; 1871. Lusikisiki: Galpin 10999. Xalanga: Cala,

Pegler 1651. Maclear: Galpin 6823.

Natal. — Inanda: Field’s Hill, McrZ/cy (Foot/ 12316. Eshowe: Lnnv? 350. Under-

berg: McClean 725. Bergville: Champagne Castle, Bayer 1440. Lion’s River:

Howick, s.n. Nkandhla: Qudeni, Gcwmcr 639. Pietermaritzburg: McClean

153. Umvoti: Greytown, Galpin 14836.

Swaziland. — Mbabane: Bolus 12264; Burtt-Davy 2896. Forbes Reef, Burtt-

Davy 2767.

After field study and consideration of the herbarium material 1 have come to the

conclusion that there is no valid reason for the separation of P. doddii which is accor-

dingly reduced. There has further been a strong feeling towards reducing both these

species under P. multibracteata. The following long-hand note by E. E. Galpin was

found on an isotype specimen of P. doddii in the Albany Museum: “ I am very doubtful

as to whether P. multibracteata, P.flanaganii, P. simplex and P. doddii are not all varietal

forms of one species. The length of the scaly stipes supporting the flower head and

causing it to be either pedunculate or sub-sessile, from examination of specimens of

P. multibracteata from the same spot appears to me to be a matter of luxuriance ”.

Galpin had a profound knowledge of plants in the field. Support has now been lent

56

to his view by the reduction of P. flanaganii partly under P. multibracteata and of P.

doddii under P. simplex, leaving us with two species, one arborescent and one dwarf,

which clearly stand in a special relation to each other.

The ranges of these two species are almost identical and they almost invariably

accompany one another in the same locality, the main exceptions being in the Natal

Drakensberg where simplex forms are rare, and in Swaziland where P. multibracteata

is absent. A few of the herbarium specimens seem to be intermediate and intermediate

forms have been found in the wild though unfortunately always sterile. There is a

sharp distinction between the species in stature and habit: otherwise the only difference

in vegetative and floral characters is in size, all parts being proportionately smaller in

P. simplex. It is suspected that P. multibracteata carries a gene for miniature habit,

which may be selected by veld burning, leading to the establishment of dwarf populations

which are genetically incapable of assuming an arborescent habit even if protected

against veld fires. This theory is being tested experimentally. In the meantime,

pending clarification of the relationship between them the two species are conserved

separately.

10. P. transvaalensis Phillips in Kew Bull. 1911: 84; Phillips in F.C. 5: 587 (1913);

Burtt-Davy in FI. Transv. 211 (1926).

Type: Pietersburg, Burtt-Davy 5179 (K, holo.!).

..A small dwarf shrub with numerous ephemeral simple stems 20-30 cm long arising

from an underground rootstock. Stems 4 mm thick, glabrous, pale green or pink.

Leaves spaced along the stems, sessile, bright green, black when dried, linear-elliptic

to oblong, 1-5-3 -5 cm wide by 7-12 cm long (average about 2x8 cm), apex obtuse,

base cuneate : blade erect and somewhat stiffened, glabrous, midrib yellow or red-veined,

prominent both sides, secondary venation ditto, margin translucent. Heads solitary,

terminal, 6 cm long and as much in diameter, narrowly cylindrical-turbinate, pedun-

culate with a small stipes 0-5 cm long, or virtually sessile opening to 130° at anthesis.

Receptacle convex, 15 mm wide. Bracts only 4-seriate, pale green, glabrous, the inner-

most commonly pink-tipped and with a fugacious russet pubescence in the upper part,

up to 2 cm broad and 5 cm long, acute, equalling the flowers. Perianth white, fulvously

hairy within the tube, the lip clothed at first with fugacious tawny hairs, otherwise

glabrous; 4-5 cm long of which 15 mm for the base, 15 mm for the broad tube and 15

mm for the lip; teeth indistinct, laterals 1 mm long. Anthers linear, 10 mm long.

Style white, 5 cm long, glabrous, terete: stigma pinkish, 8 mm long, passing almost

imperceptibly into the style. Ovary 3 mm long. Fruit 10 mm long, clothed with

golden-brown hairs.

Flowers in summer.

This is a small low-growing species found in mountain grasslands of the Haenerts-

burg area.

There is some confusion as to the type locality. The sheet bearing the type at

Kew (Burtt-Davy 5179) is marked “ Goedgeluk, Zouipansberg ” and is so cited in

F. Cap. However, in FI. Transv. 1: 211 Burtt-Davy cites his own number 5179 as

“Pietersburg, Houtboschberg ” while his register says “hillsides above Diepkloof”.

We have not been able to trace any such locality at Goedgeluk, but the farm Diepkloof

is a short distance to the north of the Houtboschberg. The type therefore evidently

came from somewhere on the Woodbush ridge.

Transvaal. — Pietersburg District, Haenertsburg area: Houtboschberg, Burtt-

Davy 5179; Suikerbosrand, Thompson s.n. ; Beard 858; Iron Crown, Beard 860;

Wolkberg, Beard 862; Meeuse 9861.

57

11. P. gaguedi Gmel. in Ed. 13, Linn. Syst. Nat. 2,1: 225 (1791).

Protea abvssirdca Willd. in Species Plantarum 1: 52? (1798). R. Brown in Trans.

Linn. Soc. 10: 85 (1810); Richard in Tent. FI. Abyss. 2: 232 (1851); Meisner

in DC. Prodr. 14: 237 (1856); Phillips in F.C. 5: 581 (1913); Baker and Wright

in F.T.A. 6, 1: 199 (1913); Bews in FI. Natal & Zululd. 82 (1921); Burtt-Davy

in FI. Transv. 211 (1926).

Type: none appears to have been preserved.

P. trigona Phillips in Kew Bull. 1910, 230.

Type: Pretoria, Leendertz 679 (PRE, iso.!).

Gaguedi (nom. vern.) Bruce, Abyss. 5: 52 (1790).

Small gnarled tree not exceeding (in South Africa) 4 m in height, generally 2 m,

less in adverse situations. Trunk 10 cm in diameter with pale brown, flaky, bark.

Flowering branchlets 8-10 mm thick, drying to 5 mm, pubescent when young, soon

glabrous and developing a thin dehiscent and papery bark cracking to show a reddish

mealy surface beneath. Leaves sessile, closely ranked, light green, linear-lanceolate

or -oblanceolate, 1 • 5-3 cm wide by 9-18 cm long (average about 2x12 cm), 4-8 mm

broad at the base, sometimes slightly falcate, apex obtuse, base cuneate, blade leathery,

glabrous except in extreme youth when it may be densely pubescent, midrib yellow,

prominent both sides, secondary venation finely reticulate, prominent both sides in

dried specimens, no distinct marginal vein. Heads solitary, terminal, varying con-

siderably in size from 4-10 cm in length and as much in diameter, broadly globose,

generally pedunculate with a well marked scaly stipes 1 • 5 cm long, though some heads

are virtually sessile: opening to 160° at anthesis. Receptacle slightly convex, 15 mm

wide. Bracts about 6-seriate above the stipes, pale green with dense silvery pubescence,

sometimes rose tinted above, 2 cm wide at their broadest, up to 4 cm long, much shorter

than the flowers, obtuse with short white ciliate fringe. Perianth densely hairy, the

base and tube whitish or fulvous, lip pure white or rose tinted, 5 cm long of which 1

cm for the base, 2 cm each for the tube and lip; tube glabrous within, lip covered with

shaggy hairs except on the back which is glabrous or almost so, or with a line of hairs

along the median keel; teeth not readily distinguishable, about 1 mm long. Anthers

linear, 10 mm long. Style white, about 6 cm long, glabrous, flattened or triangular

below, terete above: stigma white, slender, 6-10 mm long, passing with a slight bend,

into the style. Ovary 3 mm long. Fruit 1 cm long clothed with pale golden-brown

hairs.

Flowering mainly in winter.

P. gaguedi Gmel. ranges right through tropical Africa as far as Abyssinia, where

it was first found and described. “ Gaguedi ” is a vernacular name in those parts.

In the Union, it follows the eastern escarpment, with interior outliers only on the

Magaliesberg, right down from the Zoutpansberg into Zululand where a collection of

the writer’s establishes its extreme southerly limit at the gorge of the Insuzi river in

the Nkandhla district, just north of the Tugela. Sometimes this species is found within

the North East Highland Sourveld, often just outside it, under rather drier and hotter

conditions merging to Bushveld. It seems to demand less rainfall than others and is

generally found on the rockiest of ground.

Selected Citations.

Natal. — Utrecht: Thode A385. Ngotshe: Louwsburg, Dyer 5029. Nkandhla:

Insuzi River, Beard 799.

Swaziland. — Mbabane: Miller 5142.

58

Transvaal. — Soutpansberg: Makonde, van Warmelo 5117/9. Pietersburg:

Wolkberg, Gerstner 5607. Pilgrim’s Rest: Galpin 14312. Barberton: Thorncroft

3061. Pretoria: Leendertz 679. Rustenburg: Pegler 941.

This species is the subject of the only name change effected in this paper, being

the one hitherto known in South Africa as P. abyssinica Willd. The evidence support-

ing this change is as follows. A plant in the vicinity of Lamalmon, Abyssinia, was

described in semi-technical terms under its native name Gaguedi by Bruce in his “ Travels

to Discover the Source of the Nile ” 1790, with two illustrations. This plant was named

scientifically as Protea gaguedi by Gmelin in his edition (No. 13) of Linnaens’ Syst.

Nat. in 1791 and as Protea abysinica by Willdenow in Species Plantarum, 1798. As

both authors made it clear that they were describing Bruce’s plant, there is no doubt

that P. gaguedi is the valid name. We next have to decide whether our South African

plants are conspecific with Bruce’s, which might be a thorny problem since the eighteenth

century descriptions are of the briefest and vaguest and Bruce’s illustrations are stylised

and might be almost any Protea. Fortunately, only one species of Protea has ever

been collected in Abyssinia and we can thus be tolerably certain that it is P. gaguedi.

South African specimens believed to be P. gaguedi have been compared at Kew with

Abyssinian material by Mr. D. J. B. Killick at the writer’s request and pronounced

identical. P. trigona Phillips is now reduced to synonymy under P. gaguedi after exami-

nation of an isotype in the Transvaal Museum collection.

12. P. hirta Klotzsch in Flora, 76 (1845) and Beitrag FI. Kap.- und Natal, 140 (1846),

aggregate species.

Shrub or small tree, up to 3 m in height, rarely in dwarf form with underground

rootstock emitting numerous stems up to 70 cm tall, almost invariably simple. Steins

pale green to brown or silvery when young, densely hirsute, at length brown and glabrous,

up to 10 mm thick. Leaves sessile, numerous, russet-green, oblong to oblanceolate,

2-3-5 cm wide by 6-12 cm long (average about 3 X 10 cm), apex rounded or acute,

base broad, rounded and even slightly auriculate: blade leathery, densely clothed with

fine silky white hairs, and with longer russet hairs on the lower half of the midrib and

base : midrib prominent both sides, secondary veins not numerous, irregular, prominent

both sides, uniting to form an irregular marginal vein. Heads terminal or lateral,

solitary or clustered, 4-6 cm long and 5-7 cm in diameter when fully open, conical,

generally pedunculate with a small scaly stipes 1 cm long: opening to 180° at anthesis.

Receptacle slightly convex, 12 mm wide. Bracts 4-8 seriate above the stipes, russet-

green to yellow, often pink in bud, densely silky-tomentose, 12 mm wide at their broadest,

up to 3 cm long and equalling or only slightly shorter than the flowers, obtuse, shortly

ciliate at the tip. Perianth densely whitish-hairy, 3-5 cm long of which 8-10 mm for

the base, 10-20 mm for the tube and 12-20 mm for the lip; tube glabrous within, lip

covered with shaggy hairs except on the back which is glabrous or almost so, or with

a few hairs along the median keel; lateral teeth 2-3 mm long, intermediate much

shorter, tufted with russet hairs. Anthers linear, 6 mm long, connective shortly produced

into a capitate gland. Style white, 3-5 cm long, flattened, curved; stigma 6-8 mm

long, filiform, passing with an S-bend into the style. Ovary 4 mm long. Fruit 6 mm

long, clothed with pale, straw-coloured hairs.

Flowering in mid-summer to autumn.

P. hirta, aptly “ the hairy Protea ” was the earliest of the summer rainfall Proteas

to be discovered, collected in Natal by Krauss at the early date of 1838. The type

locality, ad radicem montium Tafelberge prope fluv. Umlaas, Natal, alt. 800-1,500 /?.

{bunt. Sandstein), is thought to be somewhere near Mariannhill. For some cime there

was doubt as to the identity of Transvaal plants which do not conform entirely to the

Natal type. This difficulty has been resolved by the foundation of a new subspecies,

glabre.scens. In the course of fieldwork in Rhodesia and East Africa it became clear

59

that P. hirta also extended there, although it had hitherto been classified under a number

of different names. After examination of the types of these species they can now be

reduced to synonymy under the aggregate species though more field work will be required

before they can be divided into subspecies. These are:

Southern Rhodesia.— P. swynnertonii S. Moore in J. Linn. Soc. Bot. 40: 184

(1911). The type (Swynnerton 1411) can no longer be found at the British Museum.

From description, however, and from collecting in the type locality, there is no hesitation

in reducing the species.

Angola. — P. welwitschii Engl., Hochgebirgsfl. Trop. Afr. 196 (1892). The syn-

types, (Welwitsch 1600 and 1602 B) show somewhat narrow leaves and if this proves

typical of Angola specimens, recognition of a subspecies will be appropriate.

Belgian Congo (Katanga). — P. obtusifolia de Wild, in Ann. Soc. Sc. Brux. 40,

2: 84 (1921) non Buck ex Meisn. in DC. Prodr. (1856). This was reduced to synonymy

under P. goetzeana Engl, by Hauman in Bull. Jard. Bot. Brux. 17 (1946). In any case

the name is illegitimate, being preoccupied.

Type: Homble 1253 (BR).

Belgian Congo. — P. congensis Engl. Jahrb. 33: 129 (1904).

Type: Descamps s.n. (B).

Uganda. — P. melUodora Dale in Indigenous Trees of the Uganda Protectorate,

non Engler & Gilg. Greenway & Eggeling 7069 (EA).

Tanganyika. — P. goetzeana Engl. Jahrb. 30: 298 (1902).

Type: Goetze 1367 (B).

P. eickii Engl. Jahrb. 33: 130 (1904).

Type: Eick 39 (B).

It is probable that we must also include:

Tanganyika. — P. uhehensis Engl. Jahrb. 27: 380 (1901).

Type: Goetze 720 (B). This specimen has an abnormally large head.

I have no doubt that all these forms represent a single, widespread, variable species.

They are united by such characters as the unusually pale, straw-coloured seed hairs

(which unfortunately darken with age so that this is unreliable in the herbarium) and

the triangular appendage to the anthers.

(a) subsp. hirta.

P. hirta Klotzsch in Flora, 67 (1845) and Beitrag FI. Kap.- und Natali, 140 (1846);

Meisn. in DC. Prodr. 14: 236 (1856); Medley Wood in PI. Natal 3: t. 218 (1902);

Phillips in F.C. 5: 582 (1913) in part; Bews in FI. Natal & Zululd. 82 (1921).

Type: Umlaas Riv., Krauss 202 (B, holo.!).

Growing as it does at the extreme southerly limit of the range of the species, this

subspecies exhibits reduction in size and is commonly found as a low shrub less than

1 m high, often with numerous ephemeral stems arising from a rootstock. Stems

only 5 mm thick. Leaves as for aggregate species. Heads generally solitary and

terminal only, 4-5 cm long, bracts 4-5-seriate. Perianth 3 cm long of which 8 cm for

the base, 10 mnt for the tube and 12 mm for the lip. Style 3 • 5 cm long, stigma 6 mm.

A rare plant, found in the Natal Coast belt from Umzinto to Nkandhla on outcrops

of Table Mountain Sandstone, and still more rarely on the eastern Transvaal escarpment.

Natal. — Pietermaritzburg: Table Mt., McClean 165; Sim 19402. Umvoti:

Seven Oaks, Beard 121 •, 772; 883. Umzinto: Dumisa, Fairfield, Bayer 1418. Pine-

town: Blackhill, Hesom s.n. Near Umlaas River: Krauss 202. Spring Grange, Platt

s.n. Springfield: Moonsamy s.n. Kloof: Dohse Landed and Shepherd 39. Inanda:

Medley Wood 577. Nkandhla: Wylie in herb. Medley Wood 8757.

Transvaal. — Lydenburg: Beard 875.

60

(b) subsp. glabrescens Beard, subsp. nov.

P. hirta sensu Phillips in F.C. 5: 582 (1913) in part; sensu Burtt-Davy in FI.

Transv. 211 (1926).

Type: Pretoria, Meeuse 9052 (PRE, holo. !).

Frutex ramosus a subspecie hirta foliis minoribus demum manifesto glabrescentibus,

capitulis majoribus saepe lateralibus et fasciculatis differt.

This subspecies is characterised as follows: —

A divaricate shrub generally about 1 m heigh, sometimes rather less, sometimes

up to 3 m. Branchlets 5-10 mm thick. Leaves oblong to oblanceolate, very variable

as to size from 12-25 mm in width and 6-12 cm long, (average about 2x8 cm), apex

obtuse, base attenuate; blade coriaceous, densely pubescent in youth, at length

apparently glabrescent but with closely adpressed fine silky hairs, especially on the base

of the midrib. Heads terminal or lateral, very commonly in groups of two to four or

more, 4-6 cm long. Bracts 5-8-seriate, often pink in bud, russet to yellow with fine

silvery pubescence, much shorter than the flowers. Perianth 4-5 cm long of which

10-12 for the base and 15-20 each for the tube and lip: lateral teeth 3 mm, median

2 mm. Style 5 cm long, stigma 8 mm.

Flowers in summer.

Locally frequent in the Bankenveld and Sour Bushveld of the Transvaal on sandy

flats.

Selected Citations.

Transvaal. — Middelburg: Smith 3470. Belfast: Machadodorp, Marais 11.

Waterberg: Warmbaths, Marloth 3807; Geelhoutkop, Breyer 30236. Potgieters-

rust: Galpin 8834. Pretoria: Hammanskraal, Codd 7045; Nature Reserve, Meeuse

9052. Rustenburg: Dyer and Verdoorn 3925. Johannesburg: Burtt-Davy 4002.

Ventersdorp: Mabaalstad, Loww 1900.

13. P. rubropilosa Beard, sp. nov.

Type: Pietersburg Dist., Wolkberg, S. Thompson 828 in herb. J. S. Beard (PRE,

holo. !).

Arbor usque 8 m alta cortice nigro, ramulis glabris, aereis, 1 cm crassis. Folia

sessilia, obovata, 5 cm lata, 15 cm longa, apice rotundata, inferne angustata; lamina

glaberrima, nervis utrinque prominentibus. Capitulum 8 cm longum, sive sessile.

Involucri bracteae 7 seriatae supra stipitem, subacutae, extrinsecus dense rubro-pilosae

ac margine rubro-ciliatae, usque 5 cm longae quam floribus multum breviores. Calyx

6 cm longa, inferne glabra, limbo dense albotomentoso, 2 cm longo; limbi dentes

3 mm rubro-cristati. Stylus puniceus, 7 cm longus, glaber; stigma 12 mm, filifoime.

Gnarled tree with spreading branches up to 8 m in height and 1 m in diameter

(more commonly 5 m heigh and 25 cm in diameter), bark black, flaky. Flowering

branchlets 1 cm thick, glabrous with shiny copper-coloured bark. Leaves sessile,

confined to previous season’s wood, dark green, obovate 5 cm wide by 15 long, apex

rounded, slightly narrowed towards the base: blade leathery, quite glabrous with a

transparent margin; midrib red, prominent both sides, secondary veins irregularly

reticulate, prominent both sides (with no marginal vein) in sicco but indistinct in fresh

leaves. Heads solitary, terminal, 8-10 cm long and as much in diameter, globose-

everted, virtually sessile, opening to 90° at anthesis and thereafter remaining open

until the seed is shed. Receptacle convex, 25 mm wide. Bracts about 7-seriate,

rosy-pink, densely rusty-pilose without and ciliate with tawny hairs 1-2 mm long, the

pilosity giving the bracts a bronze metallic sheen; bracts up to 2 cm broad and 4 cm

long, much shorter than the flowers, acute to rounded. Perianth white with rosy-red

61

veins on the tube and lip, 6 cm long of which 1 • 5 cm for the base, 2 • 5 for the tube and

2 cm for the lip; base and tube glabrous, lip densely white-tomentose on the sides and

keel and recurved at the top ; lateral teeth 3 mm, median 1 mm tufted with rusty hairs.

Anthers linear, 12 mm long, with an apical gland. Style pinkish, 7 cm long, glabrous,

terete; stigma filiform, 12 mm long, passing abruptly into the style. Ovary 3 mm

long, fruit 1 cm clothed with reddish hairs.

Flowers in spring, (August to October).

This species is named from the unusually thickly pilose nature of the bracts. It

is found along the north-eastern Transvaal escarpment, above 6,000 ft., from Lyden-

burg to the Wolkberg.

Transvaal. — Lydenburg, Keet 1123. Mariepskop, Beard 865. The Downs,

Rogers 21964. Wolkberg, L. C. Thompson s.n. ; Sheila Thompson 828 in herb. J. S.

Beard; Schutte s.n. Drakensberg without precise locality, Kilpir N. B. Card. 1000/31.

14. P. comptonii Beard, sp. nov.

Type: Barberton, Compton 19781 (KIRST, Holo.!).

Arbor ramis glabris 2 cm crassis. Folia sessilia, aliquot magna, oblanceolata,

4-5 cm lata, 18 cm longa, glabra margine rubro, nervulis in sicco utrinque prominen-

tibus. Capitulum grande, 10 cm longum, sessile. Involucri bracteae 6-seriatae,

glaberrimae, subvirides, 5 cm longae, quam floribus breviores. Calyx alba, glabra

praeter lateribus dentibusque limbi dense villosis, 8 cm longa, limbus dentatus 3 cm,

dentibus lateralibus 4 mm longis rubido-cristatis. Stylus albus, 8 cm longus, glaber.

Stigma album, filiforme, 12 mm longum.

Gnarled tree up to 5 m high with a short trunk up to 20 cm diameter. Bark

very thick and corky and longitudinally fissured. Flowering branchlets unusually

stout, 1 • 5-2 cm thick, glabrous with glossy pale brown bark. Leaves sessile, crowded

at the end of the branchlets, dark green and fleshy, oblong to oblanceolate, unusually

large for this genus, 3-6 cm wide by 12-24 cm long (average about 4-5 x 18 cm), apex

obtuse, often faintly retuse at the tip, base broadly attenuate, lamina 2 cm broad just

above junction with stem; blade stiff, fleshy, quite glabrous, midrib red below, yellow-

green above, prominent both sides, secondary venation inconspicuous in the fresh

state, becoming prominent when dry on both sides; margin red, translucent. Heads

solitary, terminal, 8-10 cm long by 10-15 cm in diameter, virtually sessile, everted,

opening to 180° at anthesis and thereafter remaining wide open until the seeds are

shed. Receptacle conical, 2-5 cm wide. Bracts about 6-seriate, pale green, quite

glabrous, with a shiny waxy covering, 2 • 5 cm wide at their broadest, up to 5 cm long,

very much shorter than the flowers, obtuse, very shortly white-ciliate. Perianth white,

glabrous except on the sides, keel and teeth of the lip which are densely villous, 8 cm

long of which 2 cm for the base, 3 for the tube and 3 for the lip which is 5 mm broad;

lateral teeth 4 mm long, median tooth 1 • 5 mm, conspicuously tipped with reddish

hairs. Anthers linear, 1 5 mm long, apical gland 1 mm. Style white, 8 cm long, glabrous,

terete or very slightly flattened, tapering; stigma white, slender, 12 mm long, passing

with a slight bend into the style. Ovary 3 mm long, fruit 10 mm, clothed with whitish

hairs.

Flowering in autumn and winter, (May to July).

This species bears the name of Professor R. H. Compton, its first collector. It

is known only from the mountains above Barberton in the eastern Transvaal, along the

road from Barberton to Havelock Mine, altitude 4-5,000 ft., in grass on rocky mountain

slopes, associating with P. rouppelliae Meisn. and P. rhodantha Hook, f., var. falcata

Beard.

Transvaal. — Barberton: Compton \91%\\ Schelpe A\01', 794 and 809

in herb. J. S. Beard.

62

15. P. curvata N.E. Br. in Kew Bull. 1901, 131; Phillips in F.C. 5: 580 (1913); Burt-

Davy in FI. Transv. 211 (1926).

Type: Barberton, Galpin 973 (K, Lecto., NH, GRA, CTM, BOL, iso.!).

A slender tree up to 6 m in height, with long ascending branches. Flowering

branchlets 1 cm thick, bark smooth, grey-pink and shortly tomentose in youth, longitu-

dinally wrinkled. Stem up to 15 cm diameter, bark black, scaly and fissured. Leaves

sessile or so long attenuated at the base as to appear petiolate, closely ranked, light

green, mostly somewhat falcate, linear-oblanceolate, 10-15 mm wide by 10-20 cm

long (average about 1-3 X 15 cm), apex obtuse, long attenuated to 2-3 mm wide at

the base: blade leathery, glabrous; midrib prominent both sides, secondary venation

indistinct, irregular; blade with translucent cartilaginous margin. Heads solitary,

terminal, 6-7 cm long and as much in diameter, globosely everted, pedunculate with

a short scaly stipes 5 mm long, opening to 180° at maturity and thereafter remaining

fully opened till the seed is shed. Receptacle conical, 2 cm wide. Bracts about 4-seriate

above the stipes, grey-tomentose on the lower half, glabrous above, deep red in colour,

15 mm wide at their broadest, up to 4 cm long, very much shorter than the flowers,

obtuse to sub-acute, minutely ciliate. Perianth rose-pink, pubescent except on the

lower part of the base, and the upper part of the back of the lip, 5 cm long of which

1 cm for the base, 2 cm each for the tube and lip; tube pubescent within; lateral teeth

3 mm long, tomentose except at the glabrous tips, median tooth 2 mm filiform. Anthers

linear, 15 mm long, with an apical gland. Style pink, about 6 cm long, flattened below,

trigonous above, glabrous: stigma red, slender, 10 mm long, obtuse, passing with a

slight bend into the style. Ovary 2 mm long. Fruit 1 cm long, and 5 mm thick at

top, clothed with long red to yellow-brown hairs.

Flowers in winter, (June to July).

This extraordinary tree, so peculiar in appearance and habitat, confined as it is

to a single koppie in the bushveld below Barberton, was first collected by Galpin in

1895, his specimen forming the basis of N. E. Brown’s description in 1901. It was

collected again by Thorncroft in 1919. The locality given in Galpin’s register is vague,

“ Kaap River Valley, 2,500-3,000 ft.”. However the labels of all the specimens in

the South African herbaria, though of the same number and written up in Galpin’s

handwriting, bear in many cases different data as to locality. The one in the Natal

Herbarium has “ Caledonian ” which is a siding five miles north of Barberton and

with this direction Dr. P. D. Hamilton was enabled to rediscover the type locality

in 1956. The precise spot is a koppie formed by an outcrop of talcose schist, rising

from the north bank of the Suidkaap River slightly downstream from Caledonian.

As far as is known the species is confined to this single locality, in a bushveld association

with Acacia spp., Pavetta edentula, Combretum transvaalense, C. apiculatum and Bolusan-

thus speciosus.

Transvaal. — Barberton: Galpin 973; Thorncroft 1053; Hamilton 832; 884 in

herb. J. S. Beard.

Reputed Hybrid Forms.

P. hirta X P. simplex. A number of plants exist in a mixed population of the parental

species near the Catholic Mission, Canema, Seven Oaks, Natal: Dyer 4144, Beard

773; 774.

P. gaguedi X P. rhodantha. Along road from Barberton to Havelock Mine six miles

from the Barberton-Kaapmuiden main road, a single plant: Meeuse 10108 (1957).

Both parental species present — Meeuse 10107, P. rhodantha; 10109, P. gaguedi.

63

Doubtful Records.

P. tenax, R. Br. Rogers 28291 (PRE), Culvers, Weenen, Natal. As this species has

only otherwise been collected in the False Macchia of the Eastern Province, it

is supposed that an error in labelling the specimen may have occurred.

P. grandiflora, Thunb. Rogers 22298 (PRE), Rustenburg. The locality is thought to

be incorrect as in the case of the above.

Acknowledgments.

The writer is indebted to the Chief and staff of the Division of Botany, and to

Prof. A. W. Bayer and the staff of the Botany Department, University of Natal, for

advice, encouragement and assistance; and to the Division of Botany for assistance

in the matter of field work.

65

Plate I. — Protea chacomoutana Beard

67

The Albizia Species of South Africa.

By

L. E. Codd.

Most of the South African Albizia species were dealt with by Burtt Davy in his

Flowering Plants and Ferns of Transvaal and Swaziland, Vol. 2, 1932, but several

name changes have rendered it desirable to review them again. The opportunity is

also taken to include those species not listed by Burtt Davy. The territory covered

by the present review is made up of the Union of South Africa, South West Africa,

Swaziland and Basutoland, but excludes Bechuanaland, and is therefore the area to

be dealt with by the proposed Flora of Southern Africa. Eleven species are recorded

as being indigenous in this area, while two exotic species are considered to be sufficiently

naturalised to be included in this treatment. One species found in South African

gardens, A. Julibrissin Durazz., is not included because it shows no signs of becoming

naturalised. This is not due to lack of opportunity because Sim, in manuscript notes,

relates that 100 lbs. of seed of this species were sent to Pondoland and were sown in

error for wattle seed, but there is no evidence that any plants have become established.

In addition to the material in the National Herbarium, Pretoria, specimens have

been seen from the following Herbaria, whose co-operation is gratefully acknowledged:

Bolus, Kirstenbosch, South African Museum, Albany Museum, Natal (Durban),

Natal University (Pietermaritzburg) and the Forestry Department, Pretoria. When

citing specimens, the abbreviation indicating the Herbarium is given only in cases

where the gathering is not represented in the National Herbarium.

Useful assistance and advice concerning certain species with a largely tropical

distribution have been freely given by Mr. J. P. M. Brenan of Kew Herbarium, while

Mr. D. J. B. Killick, our liaison officer at Kew, has helped by consulting botanical

material and literature not available in South Africa. I am also grateful to the Directors

of the Zurich University Herbarium and of the East African Herbarium, Nairobi, for

the loan of type material.

Durazzini based the genus Albizia on A. julibrissin Durazz., naming it in honour

of “ II sig. Cavalier Filippo degl’ Albizzi ” since, he claimed, a member of this illustrious

family first introduced the species into Tuscany from Constantinople in 1749.

Apparently Durazzini considered that two “ z’s ” in the middle of a latinised generic

name would not be appropriate because he deliberately spelt the genus with one “ z ”.

The name was amended to “ Albizzia ” when it was taken up by Boivin (1834) and

Bentham (1844), and this spelling has been in general use, though Mueller (1872)

referred to the original spelling. Since Little (1949) drew attention to his examination

of the rare publication containing Durazzini’s description, several authors have reverted

to the original spelling, so it is unlikely that favourable consideration would now be

given to a proposal to conserve the philologically preferable spelling “ Albizzia ”.

In Albizia species the stipules are herbaceous and are shed very early, unlike the

African species of the closely related genus Acacia, where the stipules usually become

hard and thorny. The main diagnostic character between the two genera, however,

is to be found in the stamens which, in Albizia, are usually longer and are united at

the base into a tube. In Acacia, the stamens are free or very shortly connate at the

base into a shallow cup or irregular perigynous disc.

3871936-3

68

Useful timber is derived from certain species of Albizia, for example, A. adianthi-

folia, A. versicolor and A. suluensis, though pieces of timber size are rare in South

Africa. A. suluensis and A. tanganyicensis contain a substance which causes irritation

to the nose and throat when the wood is worked. The bark of some species is known

to have anthelmintic properties and to contain saponins, and such species are used

medicinally by various native races. According to Watt and Brandwijk, the anthel-

mintic action is not due to the saponins, /but to other unidentified substances.

ALBIZIA'

Durazz. in Magazz. Toscan. 3, 4: 11 (1772);. Little, Amer. Midland Nat. 30: 510

(1945); Lawrence, Gentes Herb. 8: 44 (1949); Brenan, Kew Bull. 189 (1955);

’ as Albizzia, Benth. in Hook. Lond. Journ. Bot. 3; 84 (1844); Harv. in FI. Cap.

* 2: 284 (1861-62); Oliv. in FI. Trop. Afr. 2: .355 (1871); Bak. f., Legi Trbp.

Afr.' 855 (1930); Burtt Davy, Fl. Tvl. 2: 347 (1932). • ^

'Zygia P. Br. sensu E: Mey. Comm. PI. Afr. Austr. 165 (1836); Benth. in Hook. Loncf.

Journ. Bot. 3: 92 (1844),; Harv. in Fl. Cap. 2 : 284 (1861-62).

Trees or shrubs, unarmed. Leaves bipinnate usually with a gland on the petiole

and often elsewhere on the pinnae; stipules herbaceous, usually early deciduous-.

Inflorescence globose or spicate, pedunculate, arranged in terminal panicles or racemes,

or 1-several axillary. Flowers usually 5-merous, hermaphrodite or, rarely, some

flowers in the inflorescence unisexual. Calyx campanulate to tubular, 5-toothed;

teeth shorter than the tube. Corolla funnel shaped, deeply 5-lobed; lobes shorter

than the tube. Stamens numerous, much exserted, connate in a tube at the base; tube

included or exserted; anthers peltate on the filaments. Ovary oblong to lanceolate,

subsessile or shortly stipitate; style filiform, usually longer than the stamens; stigma

scarcely dilferentiated. Pod oblong to linear, flat, few to several seeded, papyraceous

to subcoriaceous, sometimes dehiscing by the separation of the valves. Seed flat,

ovate to orbicular, compressed; funicle filiform.

Key to Species.

Staminal tube included within the corolla, or scarcely exserted:

Leaflets not exceeding 4 mm in width:

Inflorescence spicate 13. A. lophantha.

Inflorescence globose:

Mature leaves and twigs glabrous; calyx and corolla with a

few whitish hairs I. A. brevifolia.

■ Mature leaves and twigs pubescent; calyx and corolla fulvo-

tomentose:

Pinnae 15-35 pairs; leaflets 25-42 pairs 2. A. amara spp-

sericocephala.

. ' ' Pinnae 8-15 pairs; leaflets 16-24 pairs 3. A. harveyi.

Pinnae 2-7 pairs; leaflets 6-14 pairs 4. A.forbesii.

Leaflets exceeding 4 mm in width :

Mature leaflets persistently tomentose 5. A. versicolor.

Mature leaflets glabrous or with scattered hairs:

Leaflets 7-18 pairs or, if less, exceeding 2-4 cm in length;

branchlets not spine-tipped:

Leaflet margin not crisped; stamens usually exceeding

2 cm in length:

Corolla lobes fulvo-tomentose; pedicels less

than 2 mm in length:

Leaflets markedly discolorous; bark

grey, not peeling

in papery flakes

6. A. antmesiana.

69

Leaflets not markedly discolorous;

bark pale brown

to whitish, peeling

in papery flakes

7. A. tanganyicens.s.

Corolla lobes glabrous or with a fringe of

whitish hairs; pedicels

usually longer than 2

mm 12. A. lebbeck.

Leaflet margin crisped; stamens less than 2 cm in

length 8. A. sidiwnsis.

Leaflets 2-6 pairs less than 2-5 cm in length; branchlets often

spine-tipped 9. A. anibelmiiitica.

Staminal tube well exserted beyond the corolla:

Leaflets 3-5 pairs 10. A. evamii.

Leaflets 6-12 pairs II. A. aclianthifolia.

INDIGENOUS SPECIES.

1. A. brevifolia Schinz in Bull. Herb. Boiss. ser. 2, 2: 945 (1902); Bak. f.. Leg. Trop.

Afr. 864 (1930).

Type: Mocambique, Zambesi area, Boruma, Menyharth 944 (Z, holo.!).

A. rogersii Burtt Davy, Flora Tvl. 2: xviii and 348 (1932); Codd, Trees and Shrubs,

K.N.P. 56 (1951).

Type: Transvaal, Soutpansberg District, Messina, Moss and Rogers 66 (PRE, iso.).

A. parvifolia Burtt Davy, l.c. xvii and 348 (1932).

Syntypes: Transvaal, Soutpansberg District, near Messina, Rogers 19247a (PRE,

iso.) and 22118.

Shrub or tree 4-15 m high, usually branching near ground level, bark grey; young

parts pubescent, becoming glabrous with age. Leaves 8-10 cm long and 4-6 cm wide,

sparingly pubescent when young, soon becoming glabrous; petiole 1-2 cm long;

pinnae 6-10 pairs; leaflets 10-20 pairs, linear oblong, 3-5 mm long and 1 mm broad,

oblique at the base, apex obtuse to rounded, not falcate, midrib nearly central. Inflo-

rescence of 8-15 flowers; buds pubescent; peduncle 1 • 5-3 cm long, sparingly strigose;

bracts and bracteoles linear, early deciduous, 4 mm and 1 mm long respectively. Flowers

creamy white; pedicels 1 mm long. Calyx 1-1-5 mm long, 5-toothed, pubescent at

first, becoming glabrous or retaining a fringe of hairs. Corolla 4 ■ 5-5 mm long, 5-lobed;

tube glabrous, 2-5-3 mm long; lobes deltoid, 1-5-2 mm long with a tuft of hairs at

the apex of each lobe. Stamens 1 - 5-1 - 8 cm long, united at the base for 3 mm. Ovary

glabrous, 1 - 5 mm long, shortly stipitate; style filiform exceeding the stamens in length.

Pod several seeded, linear to linear-oblong, 12-23 cm long and 2-4-4 cm broad, flat,

thin textured, often with transverse striations; margin slightly thickened, straight or

indented between the seeds; position of seeds apparent as raised bumps on the valve.

Recorded from Mocambique, Southern Rhodesia, Bechuanaland and the Transvaal.

Transvaal. — Soutpansberg: Messina, Moss and Rogers 66; Pole Evans 2527;

Gerstner 5449; near Messina, Rogers 19247a; Dongola area, Codd 4459; Zoutpan,

Bremekamp and Schweickerdt 2>\0\ Verdoorn O.S.V. 170; Masekwapoort,

8372; 4 miles N.W. of Wyllie’s Poort, Codd 2999; Mpefu Location, Smuts 2053;

Vanetzi Poort, Gerstner 6055; 6231; Kruger Nat. Park, near Punda Maria, Lamont

5; Lebombo Hills, Lamont 20. Pilgrims Rest: Kruger Nat. Park, The Gorge, Codd

6188; van der Schijjf 2323. Waterberg: 11 miles S. of Ellisras, Codd 8493; 8831.

Brits: Farm Welgevonden, Mogg 14641 ; 7 miles N. of Assen, Codd 6566. Rustenburg:

Onkuil, van Nouhuys 61.

70

The holotype of A. brevifolia was kindly sent to us on loan by the Director of the

Botanical Museum, Zurich University, and was compared with an isosyntype of A.

parvifolia present in the National Herbarium, Pretoria. Both are flowering specimens

with immature leaves and are similar in all essentials. When taken into consideration

with a good range of modern material, the conclusion is unavoidable that only one

species is involved. The type of A. rogersii is a fruiting twig with mature leaves.

Observations made in the type area of A. rogersii and A. parvifolia, that is, in the northern

Transvaal between Messina and Dongola, combined with tlae examination of flowering

and fruiting specimens collected from individual trees, leave no doubt that the two are

conspecific. The leaves of the type material of A. parvifolia are so immature that the

leaflets can scarcely be distinguished; thus the number of leaflets and their dimensions

given by Burtt Davy cannot be regarded as representative.

The type locality of A. brevifolia is Boruma, on the lower reaches of the Zambesi

River in Mocambique, and intermediate localities in Southern Rhodesia form a link

with the records from the Transvaal. In the last named region the species is almost

invariably associated with hot, dry, rocky hillsides, on geological formations such as

sandstone, quartzite or granite. Although often found as a small tree branching from

near ground level, it is said to grow to 40 feet tall with a spreading, umbrageous crown.

Information on herbarium specimens indicates that the wood is hard and durable and

is used for axe and pick handles. According to Lamont (formerly a Ranger in the

northern part of ihe Kruger National Park), the bark, being termite resistant, is stripped

and used for tying poles of huts. Notes on characters distinguishing A. brevifolia

from A. harveyi will be found in the discussion on the latter species.

2. A. amara (Roxb.) Boiv. subsp. sericocephala (Benth.) Brenan in Kew Bull. 1955:

190.

Syntypes: Ethiopia, Kotschv 294; Sennaar, Kotschv 244; Abyssinia, Schimper

818, 883.

A. sericocephala Benth. in Hook. Lond. Journ. Bot. 3; 91 (1844); Milne-Redhead in

Kew Bull. 1934: 301; Brenan and Greenw., Ch. List. For. Trees and Shrubs.

Tanganyika, 2: 341 (1949).

Acacia sericocephala Fenzl in Flora, 312 (1844), nomen nudum.

Inga sericocephala A. Rich., FI. Abyss. 1: 236 (1847).

Albizzia struthiophylla Milne-Redhead in Kew Bull. 1933: 144.

Type: Northern Rhodesia, Mazabuka, Milne-Redhead 1207 (PRE, iso.).

Tree 4-12 m tall; branchlets persistently pubescent. Leaves 10-18 cm long and

3-5-5 cm wide, fulvo-tomentose when young, rusty to grey hispid when mature; petiole

0-8-1 -5 cm long; pinnae 15-35 pairs; leaflets 25-42 pairs, linear oblong, 3-3-5 mm

long and 0-5-0-75 mm broad, oblique at the base, apex obtuse to rounded, not falcate;

midrib nearly central. Inflorescence of 12-18 flowers; bracts and bracteoles minute,

linear, early deciduous; peduncle 1-5-2 cm long; peduncle and buds fulvo-tomentose.

Flowers whitish, subsessile. Calyx 1-5-2 mm long, 5-toothed, golden tomentose;

teeth rounded. Corolla 4-5-5 mm long, 5-lobed, golden tomentose, especially on the

lobes; lobes ovate deltoid, 2-2-5 mm long. Stamens 1 -2-1-4 cm long, united at the

base for 2-2-5 mm. Ovary glabrous, 2-2-5 mm long, shortly stipitate; style filiform,

slightly exceeding the stamens in length. Pod several seeded, linear to linear oblong,

12-28 cm long and 2-8-4 cm wide, flat, thin textured, with occasional transverse stria-

tions; margin slightly thickened, straight or occasionally indented between the seeds;

position of seeds apparent as raised bumps on the valves.

71

Recorded from Anglo-Egyptian Sudan and Eritrea, through tropical East Africa

and the Rhodesias to Bechuanaland and north-eastern Transvaal.

Transvaal. — Soutpansberg: Nuanetsi Drift, s.n.; Sibasa area, Wannelo

5115/24; 24 miles N.E. of Sibasa, Code! 6901; 17 miles E. of Sibasa, Codd and Dyer

4494; Shingwedzi River, Forest Officer, Pietersburg, s.n. ; Minga, near Punda Maria,

Gerstner 6219; Kruger National Park, near Punda Maria, van der Schijff'9S6; between

Shangoni and Punda Maria, van der Schijff 2894.

Baker (1930) places A. sericocepfia/a Benth. as a synonym of A. aniara (Roxb.)

Boiv. It was upheld again as a distinct species by Milne-Redhead (1934), but Brenan

(1955) takes an intermediate course and gives it the status of a subspecies of A. ainara.

According to the latter author, specimens referrable to the subsp. ainara are also to

be found in East Tropical Africa, though this subspecies has its main distribution in

Asia. A. struthiophylla has been regarded for some time as being hardly separable

from A. sericocephala, a view confirmed by Brenan, l.c.

Within our region, A. ainara subsp. sericocephala has been found only in the north

eastern corner of the Transvaal, on sandy flats, mainly in the area between Sibasa and

Punda Maria. This area receives a slightly higher rainfall than the typical lowveld to

the south, north and east, where the somewhat similar species A. harveyi occurs. It

may be readily distinguished from A. harveyi by its longer leaves, more numerous

pinnae and pinnules, and by the apex of the leaflet not being distinctly acute and curved.

In stature it resembles A. harveyi, being a small tree usually with a single main trunk

and a relatively narrow crown.

3. A. harveyi Fourn. in Bull. Soc. Bot. Fr. 12: 399 (1865); Bak. f. Leg. Trop. Afr.

865 (1930); Burtt Davy, l.c. 348 (1932); Brenan and Greenw., l.c. 341 (1949);

Codd, l.c. 56 (1951).

Type: Bechuanaland, banks of the Chobe, Lake Ngami and adjoining forests.

J. McCabe (K, holo.).

A. pallida Harv. in El. Cap. 2: 284 (1861/62), non Fourn.

Type: As above.

A. hypoleuca Oliv. in El. Trop. Afr. 2: 356 (1871).

Type: As above.

Tree 5-10 m tall, usually with a distinct main trunk; bark grey; branchlets pubes-

cent. 7-12 cm long and 4-5- 5 cm wide; young parts fulvous; rhachis persis-

tently strigose pubescent; petiole 1-2 cm long; pinnae 8-18 pairs; leaflets 12-24 pairs,

discolorous, pubescent when young, becoming glabrescent with age, linear oblong,

3-6 mm long and 1-1-5 mm broad, oblique at the base, apex acute, falcate; midrib

distinctly nearer the upper margin. Inflorescence of 10-15 flowers; buds fulvo-tomen-

tose ; peduncle 1 • 5-3 cm long, hispid ; bracts and bracteoles linear, 3 mm and 1 • 5

mm long respectively, hispid, persisting to the flowering stage. Flowers creamy white,

subsessile. Calyx 2-2-5 mm long, 5-lobed, fulvous; lobes deltoid, 0-5 mm long.

Corolla 5-5-5 mm long, 5-lobed, pubescent; tube 3 mm long; lobes ovate, 2-2-5 mm

long. Stamens 1-3-1 -6 cm long, united at the base for 2-5 mm. Ovary puberulous,

2 mm long, very shortly stipitate; style filiform, exceeding the stamens in length. Pod

several seeded, linear to linear-oblong, 6-12 cm long and 1-8-2- 8 cm wide, flat, thin

textured, sometimes with transverse striations; margin slightly thickened, straight or

slightly indented between the seeds; position of seeds apparent as raised bumps on the

valve.

72

Distributed from Tanganyika through Nyasaland and the Rhodesias to the Caprivi

Strip, Bechuanaland, Mocambique, Transvaal and into Swaziland.

South West Africa. — Caprivi Strip: Katima Mulilo, West 3249.

Transvaal. — Soutpansberg : Smuts s.n. ; Messina, Pole Evans 1440; Dongola

Reserve, Krematatfontein, Pole Evans 3526; 54 miles N. of Alldays, Codd 4053; 4460;

near Wyllie’s Poort, Pole Evans 2047 ; Kruger National Park, Punda Maria, Lang in

TM 32251; Cor/r/ 4223; Mabaza, Lm/g in TM 32248. Pietersburg: Blaauwberg, on

flats at base of mountain, Leeman 51. Waterberg: 20 miles E. of Ellisras, Acocks

8820. Letaba: Kruger National Park, Letaba area, Lang in TM 30357; 30862;

van der Schijff 623. Nelspruit: Kruger National Park, near Skukuza Camp, Letty

57; 124 miles N.E. of Malelane, Codd 5218; Malelane, Thorncroft 3111. Barberton:

near Hectorspruit, Bwtt Daw 8004; Komatipoort, Schlechter 11875 (GRA); Rogers

22246.

Swaziland. — Near Makombo, Miller S/T5.

Vegetative specimens of A. harveyi are sometimes confused in the herbarium with

A. brevifolia but the two may be distinguished by the shape of the apex of the leaflet

which, in A. harveyi, is acute and distinctly falcate, while in A. brevifolia it is obtuse

or rounded and almost straight. In ecology the two are distinct as A. harveyi is found

mainly on the flat, dry bush country of the Limpopo valley and the eastern Transvaal

Lowveld, where it is a common constituent of the vegetation of the Kruger National

Park.

4. A. forbesii Benth. in Hook. Lond. Journ. Bot. 3: 92 (1844); Harv. in El. Cap. 2:

284 (1861-62); Benth. in Trans. Linn. Soc. 30: 568 (1875); Burtt Davy, l.c. 348

(1932); Codd, l.c. 54 (1951).

Type: Mocambique, Delagoa Bay, Forbes s.n. (K, holo.).

Tree 8-20 m tall, often branching from near the base with spreading, drooping

branches, making an extensive crown; branchlets pubescent. Leaves 6-10 cm long

and 5-6 cm wide, frequently folded along the midrib, hispid; pinnae 2-7 pairs; leaflets

6-14 pairs, oblong, 5-6 mm long and 2-3 mm broad, oblique at the base, apex obtuse

to acute, oblique to slightly falcate, often mucronate; midrib distinctly nearer one

margin; margin of leaflet inrolled. Inflorescence of 10-15 flowers; buds fulvo-

tomentose; peduncle 2-3 cm long, hispid; bracts and bracteoles linear, 3 mm and 1-5

mm respectively, early deciduous. Flowers creamy white, sessile. Calyx 3-3-5 mm

long, 5-toothed, fulvous. Corolla 6-7 mm long, 5-lobed, fulvous; tube 4-4-5 mm

long, subcylindrical, widening towards the throat; lobes ovate, 2-2-5 mm long.

Stamens I - 3-1 - 5 cm long, united at the base for 2-5-3 mm. Ovary glabrous, 2-3 mm

long, shortly stipitate; style filiform exceeding the stamens in length. Pod several

seeded, oblong to linear-oblong, 8-15 cm long and 3-5-5 cm broad, flat, fibrous, with

marked transverse striations; margin thickened, straight; position of seeds not apparent.

Recorded from the eastern Transvaal lowveld, Mocambique and northern Zululand.

Transvaal. — Soutpansberg: Kruger National Park, near Punda Maria, Rowland

Jones 29. Nelspruit: Kruger National Park, Lower Sabie Road, Codd 5497; 5704;

van der Schijff 3265; 3412; near Malelane, van der Schijff \4n. Barberton: Komati-

poort, Pole Evans HI 6861.

Natal. — Zululand, without locality, Gerstner 4943. Ingwavuma: Mangusi

Forest, Gerstner 4076 (NH); Maputo Forest, Gerstner 4115. Ubombo: Schreiber’s

farm, Gerstner 5562; Mkuzi Game Reserve, Ward 26AI.

This is a distinctive species with a limited distribution. Although not generally

common, it is sometimes gregarious, occurring in local groups, especially on or near

river banks. Linder favourable conditions it grows to a tall tree with a large, spreading,

twiggy crown, usually with somewhat drooping branches.

73

5. A. versicolor Welw. ex Oliv. in FI. Trop. Afr. 2: 359 (1871); Bak. f., l.c. 863 (1930);

Burtt Davy, l.c. 348 (1932); Brenan and Greenw., l.c. 343 (1949); Codd, l.c.

57 (1951).

Syntypes: Angola, Golungo Alto and Zenza do Golungo, Welwitsch, and Nyasa-

land, Maravi Country west of Lake Nyasa, Kirk (all at BM).

Var. mossambicensis Schinz in Bull. Herb. Boiss. ser. 2, 2: 946 (1902); Bak. f.,

l.c. 863 (1930).

Type: Mocambique, Zambesi area, Boruma, Menyharth 77b (Z, holo.).

A. mossambicensis Sim, For. FI. P.E.A. 59, t. 60 (1909).

Type: Mocambique, Umbelusi, Sim 6392 (NLF, hold.!).

Tree 6-18 m tall with a single main trunk and a large, round or spreading crown;

bark grey; branchlets persistently tomentose. "Leaves 18-30 cm long and 12-24 cm

broad; rusty pubescent; petiole 3-6 cm long; pinnae. 1-3 pairs; leaflets 3-5 pairs,

oblong to obovate-oblong or broadly elliptical, 3-5 • 5 cm long and 1 • 5-3 cm broad,

oblique at the base, apex rounded to truncate, often ‘mucronulate. Inflorescence

18-25 flowered; peduncle 4-5-5 cm long, fulvo-tomentose. Flowers creamy white,

sessile. Calyx 5 -5-6 -5 mm long, 5-toothed, fulvo-tomentose; teeth deltoid, 1-5 mm

long. Corolla 7-8 mm long, 5-lobed, tomentose; tube 3-5-4 mm long; lobes ovate,

3-5-4 mm long. Stamens 2 -8-3 -5 cm long, united at the base for 3 mm. Ovary

puberulous, 3 mm long, subsessile; style filiform, slightly exceeding the stamens in

length. Pod several seeded, linear to linear-oblong, 8-22 cm long and 3-5 - 5 cm wide,

flat, thin textured with some transverse striations; margin thickened, straight or slightly

indented between seeds; position of seeds apparent as raised bumps on the valves.

Recorded from Uganda southwards to Angola, South West Africa, Bechuana-

land, the Rhodesias, Mocambique, Transvaal, Swaziland and Zululand.

South West Africa. — Ovamboland: Oshikongo, Loeb 20; 431; Rodin 2609

(BOL). Okavango: Runtu, Maguire 1595; Popa Falls, Maguire 1665. Caprivi

Strip: Curson 940.

Transvaal. — Soutpansberg: Messina, Pole Evans 2040; Soutpansberg Range,

Smuts s.n. ; Soutpan, Galpin 15128; Elim, Obermeyer 564; Makonde, Westphal in

TM 29123; 4 miles S.E. of Sibasa, Codd and Dyer 4491 ; Kruger National Park, Punda

Maria, Lang in TM 32301; Codd and Dyer 4615; Rowland Jones 14. Pietersburg:

Smuts Letaba: Thabina, Davy 2893 ; Swierstra 2\%%. Lydenburg: Buffels-

vlei, Codd 6674. Pilgrims Rest: Oranje 4453; Acorn Hoek, Keet 1488; Bushbuck-

ridge, Pritchard 28; Acocks 12894; Lothian, Forest Dept. 6836. Nelspruit: Legat

in Hb. Burtt Davy 2830; Kruger National Park, near Pretorius Kop, Codd 5694;

Rowland Jones s.n.; van der Schijff 151; 431; 727. Barberton: C. A. Smith 7072;

Malelane, Pole Evans H 15771; Queens River Valley, Galpin 651; Sheba Valley,

Thorncroft 3019.

Swaziland. — Ingwavuma River, West 2105.

Natal. — Zululand, without locality: Wylie 8805; Boocock 4; Gerstner 5086.

Ubombo: Otobotini, McClean in NH 18473 (NH). Hlabisa: Gerstner 2944 (NH);

Dukuduku Reserve, Stephens 97 (NH); Hluhluwe Game Reserve, Ward 1861 (NH);

van Zinderen Bakker 111 (NH); Bayer in NU 17113 (NU). Lower Umfolosi: Near

Ngogodo Siding, Acocks 12976; Umhlatuzi Valley, Lawn 2118a (NH); Entonjaneni:

Near Melmoth Road, Lawn 1392 (NH). Eshowe: Between Eshowe and Nkandhla,

Pole Evans 3622. Mtunzini: Gingindhlovu, Wylie in NH 9243 (NH); 4 miles N.

of Gingindhlovu, Lawn 1864 (NH); Amatikulu, Wood 7589.

74

The type of var. mossanibicensis Schinz has not been seen. The slight deviations

in corolla length and pubescence do not appear to justify its separation even as a variety,

nor does an examination of a good range of modern material indicate that this variable

species should be split up into groups of minor rank. The type of A. mossambicensis

Sim is cited as No. 6392. The holotype is in the Natal University Herbarium and was

sent on loan by Professor A. W. Bayer. It is actually annotated by Sim as A. umbelusiana

(sic), but obviously agrees with his description and plate of A. mossambicensis and not

with his published A. umbalusiana (see notes under A. anthelmintica and A. evansii).

In all respects the type of A. mossambicensis falls within the range of variation accepted

for A. versicolor.

A. versicolor grows to an attractive large tree with a round crown of graceful foliage.

A useful timber is obtained from the tree and is known commercially in South Africa

as Umvangaas. This is an unfortunate choice as it is the name used in the Swazi and

Shangaan languages for the Kiaat {Pterocarpus angolensis) which produces a very similar

timber. The Swazi name for A. versicolor is siVangatsane, or “ Little Kiaat”. Sim

unaccountably states that its timber is useless. Brenan and Greenway, in addition to

recording that the timber is used, state that the natives make a soapy substance from

the roots and that the roots are employed as an anthelmintic and also as a purgative

and enema.

6. A. antunesiana Harms in Engl. Bot. Jahrb. 30: 75 (1901); Bak. f., l.c. 861 (1930);

Brenan and Greenw., l.c. 342 (1949).

Syntypes: Angola, Huilla, Antunes 330, and Nyasaland, Unyika, Goetze 1372.

Tree 3-12 m tall; bark grey, rough; young parts pubescent, becoming glabrous

with age. Leaves 12-25 cm long, and 10-20 cm wide, sparingly pubescent when young,

soon becoming glabrous; petiole 4-6 cm long; pinnae 1-3 pairs; leaflets 5-9 pairs,

ovate-oblong to trapeziform, 2-5-6 cm long and 0-8-2 -8 cm broad, more or less

coriaceous, discolorous, oblique at the base, apex rounded. Inflorescence of 15-20

flowers, the central flower usually larger than the rest; buds rusty pubescent; peduncle

4-5 cm long, rusty pubescent; bracts and bracteoles small, linear, early deciduous.

Flowers creamy white; pedicel up to 1 mm long; central flower larger and purely

male. Calyx 4-5-5 mm long, 5-toothed, narrowing to the base, rusty pubescent.

Corolla 7-8 mm long, 5-lobed, rusty pubescent, paler than the calyx; tube 3-5-4 mm

long; lobes 3-5-4 mm long, ovate oblong, margin inrolled. Stamens 2 -4-2 -8 cm

long, united at the base for 4-5 mm. Ovary glabrous, 2 mm long, shortly stipitate;

style filiform, shorter than the stamens. Pod several seeded, linear oblong, 11-16 cm

long and 2 - 5-4 cm broad, flat, thintextured ; margin slightly thickened, scarcely indented

between the seeds; position of seeds apparent as bumps on the valve.

Recorded from the Belgian Congo, Tanganyika, the Rhodesias, Mocambique,

Angola and northern South West Africa.

South West Africa. — Okavango: de Winter and Wiss 4394. Caprivi Strip:

opposite Andara, de Winter and Marais 4816.

A. antunesiana is closely related to A. coriaria Welw. In the absence of authentic

material of either species in South African herbaria, I accept the opinion of Mr. D. J. B.

Killick, who considers that the two are distinct. He states that A. antunesiana has

longer leaflets which are more oblique at the base, while the leaf rhachis is always

glabrous. A. coriaria sometimes has completely glabrous leaflets, but the rhachis

always shows some pubescence.

75

7. A. tanganyicensis Bak.f. in Journ. Bot. 67: 199 (1929); Leg. Trop. Afr. 862 (1930);

Brenan and Greenw., l.c. 342 (1949).

Type: Tanganyika Territory, Kondowe District, Simbo Hills, B. D. Burtt 716

(K, holo., EA, iso.!).

A. rhodesica Burt Davy, FI. Tvl. 2: xviii and 348 (1932); Codd, l.c. 56 (1951).

Syntypes: Southern Rhodesia, Matopos, Galpin 7082 (PRE, isosyn.) and Victoria

Falls, Allen 174; Rogers 5319.

A. lebbek var. australis Burtt Davy in Burtt Davy and Hoyle, Ch. List Trees and Shrubs

Nyasaland Prot. 53 (1936), nomen nudum.

Tree 3-8 m tall, sparingly branched; branches ascending; bark smooth, white

to light brown, peeling in broad, papery strips; twigs sparingly pubescent, usually

becoming glabrous or nearly so. Leaves 25-40 cm long and 20-30 cm broad, sparingly

pubescent, especially when young; petiole 5-10 cm long; pinnae 2-5 pairs; leaflets

5-12 pairs, oblong, broadly elliptical to ovate oblong, 2-4 cm long and 1-2-2 cm broad,

oblique at the base, apex rounded to obtuse. Inflorescence of 20-25 flowers ; peduncle

3-4 cm long; peduncle and buds rusty hispid; bracts and bracteoles small, linear,

early deciduous. Flowers creamy white, subsessile. Calyx 4 -5-5 -5 mm long, 5-lobed,

fulvous; lobes ovate-deltoid, 1-5-2 mm long. Corolla 7-9 mm long, 5-lobed; tube

usually pubescent, 4-5 mm long; lobes ovate to ovate-lanceolate, 3-3-5 mm long,

fulvo-tomentose, especially towards the apex. Stamens 2-5-3 cm long, united at the

base for about 4 mm. Ovary glabrous, 3 mm long, shortly stipitate; style filiform,

exceeding the stamens in length. Pod several seeded, flat, semi-woody, smooth, linear

to linear-oblong, 15-25 cm long and 3-5 cm broad, light brown; margin scarcely

thickened; position of seeds usually not visible through valves.

Recorded from Tanganyika, Nyasaland, Rhodesia and the Transvaal.

Transvaal. — Soutpansberg: Menne s.n.; Kruger National Park, near Punda

Maria, Lamont s.n.; Codd and Dyer 4549. Waterberg: Near Nylstroom, Pole Evans

H19679; Codd 5601; Repton 3472; Prosser 1722; 19 miles E. of Vaalwater, Codd

986; Leopards Kloof, Hartebeestelaagte, Galpin 13391; Rooiberg, Pole Evans s.n.;

hills near Warmbaths, Burtt Davy 2183; Bolus 11868; Gerstner 5278; Story 1522.

The probability that A. rhodesica would prove to be a synonym of A. tanganyi-

censis was recognised by Mr. B. de Winter while at Kew Herbarium as our liaison officer.

This view was supported after seeing an isotype of A. tanganyicensis kindly sent to us

on loan from the East African Herbarium, Nairobi, and comparing it with material

in the National Herbarium.

The name A. lebbek var. australis is listed by Burtt Davy and Hoyle without

description or citation of specimens. A clue to its identity is provided by the fact

that several sheets in the National Herbarium, Pretoria, are annotated by Burtt Davy

with this name, including Galpin 7082, one of the syntypes of his A. rhodesica.

There has been some confusion in the past between A. tanganyicensis Bak. f. and

A. lebbeck (L.) Benth., but the two are clearly distinct. Vegetative specimens of the

two are very similar, though A. tanganyicensis has more closely placed and more

numerous leaflets (5-12 pairs), as against 3-9 pairs of leaflets for A. lebbeck. In A.

tanganyicensis the individual flowers are subsessile, the calyx is relatively long (4 - 5-5 - 5

mm) and is 5-lobed, with a marked golden brown tomentum, especially towards the

apex, as well as on the lobes of the corolla. In A. lebbeck the flowers have pedicels

2-3-5 mm long, the calyx is 4-4-5 mm long, is shortly 5-toothed, with a short, pale

brown tomentum, while the corolla is almost glabrous with a few whitish hairs on the

lobes. There are other differences, including a marked distinction in ecology and

habit. A. tanganyicensis is a small deciduous tree inhabiting rock formations with

76

an acid reaction, such as granite or quartzite. It is sometimes sparingly branched

near the base, with few, ascending or crooked branches, forming a sparse canopy.

The most striking feature, however, is the bark, which is thin and light brown in colour,

peeling off in broad, papery flakes, leaving a white smooth surface. It has no timber

value and farmers in the Transvaal have remarked that, when the wood is worked, the

dust is a strong irritant to the nose and throat, so that it is locally known as “ sneeze-

wood

8. A. suluensis Gersiner in Journ. S. Afr. Bot. 13; 62 (1947).

Lectotype: Natal, Melmoth District, 2 miles west of Dundulu Store, Gerstiier

4337 (PRE).

Tree 5-15 m tall with a round or spreading crown; bark grey, fissured; young

twigs usually pubescent, soon becoming glabrous. Leaves 13-20 cm long and equally

broad; petiole 2-5 cm long; pinnae 2-4 pairs; leaflets 5-9 pairs, glabrous to sparingly

pubescent, especially on the petiolule, dark green above, underside paler, broadly

elliptical to oblong or obovate, 2-2-5 cm long and 1-1 • 5 cm broad, oblique at the base,

apex rounded to truncate, mucronate; margin crisped. Inflorescence of 20-25 flowers;

peduncle 3-4 cm long; peduncle and buds fulvo-tomentose. Flowers whitish, sub-

sessile. Co/yv 5-toothed, 2- 5-3- 5 mm long, fulvo-tomentose; teeth rounded. Corolla

5-6 mm long, 5-lobed, fulvo-tomentose; tube 3-3-5 mm long; lobes lanceolate, 3

mm long. Stamens 1 - 3-1 - 6 cm long, united at the base for 5-6 mm. Ovary glabrous,

2-3 mm long, shortly stipitate; style filiform, equal to the stamens in length. Pod

several seeded, linear to linear-oblong, 8-16 cm long and 1 -4-2-6 cm broad, flat, light

brown, thin textured with occasional transverse striations; margin slightly thickened,

straight or indented between the seeds; position of seeds apparent as raised bumps

on the valves.

Distribution confined to northern Zululand.

Natal.— Hlabisa: Gerstner 1714 (NBG); 1730 (NBG); 4601 (NH); 6440;

6440a (NH); 6440b (NH); Nhlwati, F. fiaye’r 64 ; Gerstner 13>0; Dukumbane, Gerstner

714; Gwegwede River, Gerstner 6261 (BOL); Mtadhlwana Hill, F. Bayer 14/46;

4 miles N. of Hlabisa, Codd 9611; Hluhluwe Game Reserve, Ward 2829 (NH).

Melmoth: 2 miles W. of Dundulu Store, Gerstner 4337, (PRE, lecto.).

In choosing one of the National Herbarium specimens as the lectotype, I was

influenced by the fact that the preliminary examination leading to the recognition of

this as an undescribed species was carried out by Miss I. C. Verdoorn, who had before

her a good range of material collected by the Rev. Fr. Gerstner and Mr. F. Bayer,

including specimens in better condition than those cited with the published description.

In the selection of a lectotype, however, one is limited to the material cited by the author.

Of the syntypes listed by Gerstner, two are represented in the National Herbarium,

namely, Gerstner 4337 (flowering) and Bayer 64 (fruiting). Although he cites his No.

4601 as being present in PRE, this does not appear to be the case. Because flowers and

leaflets are the most diagnostic characters of the species, Gerstner 4337 is chosen as

the lectotype. The leaf illustrated with Gerstner’s type description is obviously from

his No. 4601, which is an immature specimen with leaflets mainly acute. This is not

representative of mature specimens, in which the leaflets are rounded to truncate at

the apex, with the margins distinctly crisped. The drawing of the flower also conveys

the wrong impression through showing the calyx distinctly lobed when in fact it is

very shortly toothed.

A. suluensis is a graceful tree with a round crown, bearing some resemblance to

A. lebbeck but can readily be distinguished from that species by the crisped margin of

the leaflet, by the golden tomentum of the calyx and corolla and by several other features.

It appears to be endemic in northern Zululand, having been collected only from the

Hlabisa and Melmoth Districts.

77

Gerstner records that the natives pound the bark with water, producing a foaming

mixture which is used as a powerful enema. The timber is said to be hard and durable

with an attractive grain, suitable for furniture. It is also stated that the sawdust pro-

duced while working the timber is somewhat irritant and causes sneezing.

9. A. anthelmintica (A. Rich.) A. Brongn. in Bull. Soc. Bot. Fr. 7: 902 (1860); Oliv.

in FI. Trop. Afr. 2: 357 (1871); Benth. in Trans. Linn. Soc. 30: 564 (1875);

Marloth, FI. S. Afr. 11, 1: t. 29 (1925); Bak. f., l.c. 859 (1930); Brenan and

Greenw., l.c. 341 (1949); Codd, l.c. 53 (1951).

Type: Abyssinia, near Add’erbati, Quartin Dillon.

Var. australis Bak. f., l.c. 859 (1930).

Type: South West Africa, Okahandja, Dinter 269 (K, holo., PRE, iso.).

Var. pubescens Burtt Davy, l.c. xvii (1932).

Syntypes: Transvaal, Soutpansberg District, Waterpoort, Rogers 19347 (PRE,

isosyn.) and Rogers 21504 (PRE, isosyn.).

Besenna anthelmintica A. Rich., El. Abyss. 1: 253 (1847).

Acacia marlothii Engl, in Engl. Bot. Jahrb. 10: 19 (1889).

Type: South West Africa, near Otjimbingwe, Marloth 1317 (B, holo., destroyed;

PRE, iso.).

Alhizzia umbalusiana Sim, Eor. El. P.E.A. 59, t. 55a (1909).

Type: Mocambique, Umbeluzi, Sim 6200 (apparently no material extant).

Shrub or small tree 2-10 m tall; branchlets twiggy, usually with lenticular bark,

often forming abbreviated, spine-tipped shoots; young twigs usually sparingly pubescent,

soon becoming glabrous. Leaves 6-8 cm long and 5-6 cm broad; rhachis glabrous

to sparingly hispidulous; petiole 0- 5-2 cm long; pinnae 2-4 pairs; leaflets 2-5 pairs,

the uppermost usually the largest, glabrous above, glabrous or sparingly pubescent on

the nerves below, obovate, broadly elliptical to subrotundate or ovate, 0-8-2 -5 cm

long and 0-5-1 -8 cm broad, oblique at the base, apex obtuse, mucronulate. Inflores-

cence of 15-25 flowers; peduncle 1-4-2 cm long, often sparingly pubescent; bracts

and bracteoles small, early deciduous. Flowers white to cream; pedicels 1-2-5 mm

long. Calyx 3-4-5 mm long, 5-toothed, deeply cleft on one side, usually glabrous

except for occasionally a tuft of white hairs at the apex of each tooth. Corolla 6-9

mm long, 5-lobed, glabrous except for a tufc of white hairs at the apex of each lobe;

tube 3-5 mm long; lobes ovate-lanceolate, 3-4 mm long. Stamens 2 -3-2 -8 cm long,

united at the base for 3-4 mm. Ovary glabrous, about 3 mm long, shortly stipitate;

style filiform, exceeding the stamens in length. Pod several seeded, linear to linear-

oblong, 8-12 cm long and 2-2-5 cm broad, flat, membranous, straw coloured to light

brown with occasional transverse striations; margin slightly thickened, straight or

indented between the seeds ; position of seeds apparent as raised bumps on the valves.

Recorded from Abyssinia, through tropical East Africa and the Rhodesias to

Angola, South West Africa, Bechuanaland, Mocambique, Transvaal and Natal.

South West Africa. — Ovamboland: near Oshikanga, Rodin 2673. Okavango:

Runtu, Maguire 1615 (NBG). Grootfontein: Watt and Brandwijk 1443; Tsumeb,

Basson 15; Schaap River Hills, Keet 1504; 20 miles S. of Osiri, Liebenberg 4682.

Otjiwarongo: Waterberg, Liebenberg 4774. Okahandja: Dinter 269 (PRE, SAM,

GRA); Bradfleld 452. Gobabis: Liebenberg 4628. Windhoek: Dinter s.n.; Keet

1674; Codd 5793; Otjisewa, Wiss and Kinges 739. Rehoboth: Buellspoort, Strey

2187. Maltahoehe: Kleinfontein, Marloth 5052. Keetmanshoop : Rogers s.n.

(GRA); Gellap Ost, Acocks 15608.

78

Transvaal. — Soutpansberg: Dongola area, Pole Evans 3534; Codd 4331; near

Messina, Rogers 19347; Pole Evans HI 8897; Gerstner 5453; Waterpoort, Rogers

21504; Gerstner 5718; Masekwapoort, Gerstner 5953; Sibasa area, Gerstner 6213;

Kruger National Park, Makuleka, Lang in TM 32263; Pafuri, van der Schijff 642.

Pietersburg: Smuts s.n. ; Leipzig Mission, Blaauwberg, Leipolt 3; near Malipsdrift,

Erens s.n.; Repton 699. Waterberg: 30 miles N.W. of Vaalwater, Smuts 361 . Rusten-

burg: near Matlabas, Acocks 8802. Lydenburg: Barnard 49a; 49b; 4 miles N.

of Buffelsvlei, Codd 6675. Pilgrims Rest: Kruger National Park: Sand River, van

der Schijff 115. Nelspruit: Kruger National Park, near Skukuza, Lang in TM 30866;

Petty 67; Codd 5701; van der Schijff 141: 783. Barberton: near Louws Creek,

Acocks 12879; Codd 1040.

Natal. — Zululand, without locality, Genyt/zer 4944; 5088. Ingwavuma: Gerstner

3767 (NH); near Pongola, IKes'? 2093 (NH); Otobotini, Ger^m^r 3431 (NH). Ubombo:

Mkuzi, Galpin 13691. Hlabisa: Hluhluwe Game Reserve, Ward 1440 (NU). Lower

Umfoiosi: near Umhlatuzi River, Gerstner 6441 (NH).

In South Africa and in drier parts of the tropics, A. anthelmintica is a small tree or

shrub up to about 15 feet tall. According to the Flora of Tropical Africa, it may

reach the stature of a large tree from the trunks of which native canoes are made, but

this statement requires confirmation. The southern form has been separated varietally

by both Baker fil. and Burtt Davy. There seems little justification for this course,

a view also expressed by Mr. Brenan and Mr. Killick after the examination of a wide

range of specimens, including type material. Specimens matching the southern form

can be found from north tropical Africa, while pubescence is not restricted to the

South African specimens. Dr. P. J. Greenway of Nairobi informs me that he knows

the species in the field almost throughout its whole range from the Kruger National

Park to the Kenya — Ethiopian boundary and can see no evidence to support the sepa-

ration of a variety.

The holotype of Acacia marlothii Engl., Marloth 1317, presumed to have been in

Berlin-Dahlem Herbarium, appears to have been destroyed, but a specimen of this

number is in the National Herbarium, Pretoria. It is a sterile twig but is unmistakably

conspecific with Albizia anthelmintica.

There is some evidence that Sim based his A. umbalu.siana on a mixture. The

flowering twig with spine-tipped branchlets illustrated on Plate 55A of his Forest

Flora of Portuguese East Africa would pass as A. anthelmintica, though the leaf and

pod show a closer resemblance to A. evansii. No material of the type, Sim 6200,

has been traced. The greater part of Sim’s phanerogamic collection is in the Natal

University Herbarium, but No. 6200 is not represented. Two specimens have been

annotated by him with the name A. umbelusiana (sic). One of these is No. 6392, the

type of A. mossambicensis Sim (see note under A. versicolor): the other, No. 23118,

a fruiting specimen, is identifiable as A. evansii Burtt Davy, a species with very different

flowers from those described and illustrated as A. umbalusiana. It seems probable

that the leaf and pod of Plate 55A were drawn from this specimen, in which case the

flowering twig with spine-tipped branchlets (excluding the leaf) no doubt corresponds

to the missing type. No. 6200. It may be noted that Sim also lists A. anthelmintica,

but gives the description in inverted commas after Dr. Meller, indicating that he was

not familiar with the species.

The anthelmintic properties of the roots of this species are recognised by a number

of native tribes. Watt and Brandwijk report the presence of sapogenins in the roots

but state that these substances are not responsible for the anthelmintic action. Tests

with pollen carried out by the Division of Veterinary Services, Onderstepoort, showed

that it acts as an irritant to the eyes.

79

10. A. evansii Burtt Davy, FI. Tvl. 2: xvii and 349 (1932). Codd, l.c. 54 (1951).

Type: Transvaal, Nelspruil District, Sabie Game Reserve, Pole Evans HI 6921

(K, holo.).

Shrub or small tree 4-8 m tall, usually branching freely from the base with many

ascending branches; branchlets pubescent, not forming abbreviated spine-tipped

shoots. Leaves 5-8 cm long and 4-6 cm broad; petiole 0-8-2 cm long; petiole and

rhachis hispidulous; pinnae 2-4 pairs; leaflets 2-5 pairs, obovate to trapezifomi-

elliptical, 1-2 cm long and 0-5-1 cm broad, upper surface dark green with a few scattered

hairs, lower surface paler, hispidulous; base cuneate, usually oblique, apex obtuse,

often mucronulate. Inflorescence 6-15 flowered; peduncle 0-8-1 -5 cm long, his-

pidulous; bracts and bracteoles minute, early deciduous. Flowers whitish, tinged

with red; pedicels 1-5 mm long. Calyx 1-1-5 mm long, cup-shaped, obscurely

5-toothed, puberulous. Corolla 6-7 mm long, 5-lobed, glabrous or sparsely pubescent

on the lobes; tube subcylindrical, widening slightly towards the throat, 3-5-4 mm

long; lobes 2-5-3 mm long, ovate-lanceolate, cucullate. Stamens 1-6-1 -8 cm long,

united in a narrow tube for 1 -2-1 -3 cm. Ovary glabrous, 2-5 mm long, shortly stipi-

tate; style filiform, exceeding the anthers by 3-5 mm. Pod several seeded, linear-

oblong, 6-13 cm long and 1-4-2 cm broad, flat, semi-woody, smooth, yellow brown;

margin very slightly thickened, usually straight; position of seeds not apparent through

the valves.

Recorded from the north-eastern and eastern Transvaal, southern Mocambique

and northern Zululand.

Transvaal. — Soutpansberg: Kruger National Park, 2 miles S. of Punda Maria,

Codd 5990. Pilgrims Rest : 15 miles S. of Satara, Con'r/ 4296; 20 miles N.E. of Skukuza,

Codd 5592. Nelspruit: Kruger National Park, 164 miles E. of Skukuza, Codd 5706;

54 miles S.W. of Lower Sabie Camp, Codd 5708.

Natal. — Ingwavuma: Gerstner 3734; Ndumu Game Reserve, Gerstner 3440;

3993; Ward 2051.

Attention has been drawn (page 78) to the possibility that the description and

illustration of A. umbalusiana Sim are drawn from the two species A. anthelmintica

and A. evansii and that possibly Sim 23118 is the basis of the leaf and pod of his Plate

55A. Sim also records the native name Nala for his species and this is the Shangaan

name for A. evansii encountered in the Kruger National Park. However, the descrip-

tion of the flowers of A. umbalusiana could not apply to A. evansii.

There is a superficial resemblance between the leaves of A. anthelmintica and of

A. evansii, but the leaflets of the latter are hispidulous, especially on the under surface,

while those of A. anthelmintica are glabrous or with a few scattered hairs on the nerves,

and only very rarely hispidulous in some South West African specimens. A. evansii

is closely related to A. petersiana Bolle but the leaflets of the latter species are more

numerous and glabrous, while the corolla and stamens are longer. The habit of A.

petersiana is not known, but A. evansii has a very characteristic appearance, unlike

any other South African species. It branches freely from near ground level, forming

numerous ascending branches 12-25 feet tall (see Trees and Shrubs of the Kruger

National Park, Pig. 47). Although never very common, it is sometimes locally grega-

rious on brakish, sandy flats as, for example, near the Lion Pan in the Kruger National

Park.

11. A. adianthifolia (Schumach.) W. F. Wight in U.S. Dept. Agr. Bull. 137: 12 (1909);

Brenan in Kew Bull. 1952: 520.

Type: Gold Coast, Bliguisso, Thonning s.n. (Hb. Haun., Univ. Bot. Mus., Copen-

hagen, holo.).

Mimosa adianthifolia Schumach. in Schumach. et Thonn., Beskrif. Guin. PI. 322 (1827).

80

Zygia fastigiata E. Mey. Comm. PI. Afr. Austr. 165 (1836); Benth. in Hook. Lond.

Journ. Bot. 3: 93 (1844); Harv. in FI. Cap. 2: 285 (1861-62).

Syntypes; Natal, between Umzimkulu and Umkomaas, Drege, and Port Natal,

Drege.

A. fastigiata (E. Mey.) Oliv. in El. Trop. Afr. 2: 361 (1871); Benth. in Trans. Linn.

Soc. 30: 570 (1875), pro parte; Wood and Evans, Natal PI. 1: 24, t. 27 (1898);

Sim, For. FI. Cape Col. 213, t. 62 (1907); Sim, For. El. P.E.A. 59, t. 58 (1909);

Marloth, El. S. Afr. II, 1: t. 30 (1925).

A. gummifera non (Gmel.) C.A.Sm., C.A.Sm. in Kew Bull. 218 (1930), pro parte;

Burtt Davy, l.c. 2: 349 (1932); Henkel, Woody PI. Natal and Zul. 236 (1934);

Codd, l.c. 55 (1951).

Tree 7-15 m tall with a flat, spreading crown; bark grey; young twigs rusty pube-

scent. Leaver 12-22 cm and 8-12 cm broad; petiole 3-4 cm long; petiole and rhachis

rusty pubescent; pinnae 4-7 pairs; leaflets 6-12 pairs, trapeziform, 1-1-6 cm long

and 5-8 mm broad, sparingly hispidulous and dark green above, hispidulous and

paler below; midrib diagonal. Inflorescence 15-25 flowered; peduncles 2-5-6 cm

long, fulvous; bracts and bracteoles minute, early deciduous. Flowers whitish, shortly

pedicellate, of two kinds; outer flowers hermaphrodite, with a few larger male flowers

at the centre. Calyx 4 -5-5 -5 mm long, 5-toothed, fulvo-tomentose. Corolla 8-10

mm long, 5-lobed, grey pubescent; tube subcylindrical, widening towards the mouth,

6-7 mm long; lobes lanceolate — oblong, 2-3 mm long, apex cucullate. Stamens

2-5-3-2 cm long, united into a narrow tube for 1 -8-2-6 cm. Ovary glabrous, 2-5

mm long, subsessile; style filiform, exceeding the stamens. Pod several seeded, linear

to linear oblong, 11-16 cm long and 2 -3-2 -6 cm broad, flat, thin textured, pubescent,

rugose, pale brown; margin thickened, straight or wavy; position of seeds apparent

as bumps on the valves.

Distributed from Senegal to Abyssinia and, southwards, through east Tropical

Africa to Rhodesia, Mocambique, Transvaal, Natal and to Port St. Johns in the Cape

Province.

Transvaal. — Soutpansberg : Tshakoma, Obenneyer 975; Shewass, Legat in

TM 6259; Makonde, Westphal 5; Curson and Irvine 78; Pepeti Falls, Carson and

Irvine 97; Kruger National Park, near Punda Maria, Lamont 43; Codd 6529; van

der Schijjf?>lU.

Natal. — Without locality, Mrs. Saunders s.n. (BOL); Gerrard and McKen 397

(NH). Zululand: Bazwana, Gerstner 3725 (NH). Ingwavuma: Mangazi Forest,

Maputa, .Sover 753. Hlabisa: St. Lucia Bay, Po/c 3646; near Hluhluwe Game

Reserve, Ward 2627. Mtunzini: Ngoye Forest, Schmidt in For. Dept. Herb. 1445.

Eshowe: Forest adjoining Eshowe, Codd 1864. Mapumulo: Umhlali, Kept on 1838.

Inanda: near Umzinyati Falls, McClean and Ogilvie in NH 28868. Durban: Medley

Wood 6135. Pinetown: Amanzimtoti, Gerstner in NH 22080 (NH). Umzinto:

Dumisa, Rudatis 681; Umdoni Park, Smuts 2340. Port Shepstone: 2\ miles W. of

Southbroom, Marais 1147.

Cape Province. — Lusikisiki: miles N. of Embotyi, Codd 9740.

The difficult problem of species limits and synonymy in the A. gummifera — A.

adianthifolia—A. zygia complex has been dealt with by Brenan in Kew Bull. 1952:

520, and his conclusions are followed here.

A. adianthifolia is a conspicuous tree of the coastal forests of Natal, with its

spreading, flat crown of graceful foliage. Sim, in his Forest Flora of the Cape of Good

Hope, states that the wood is fine grained, rather soft and susceptible to borer.

81

EXOTIC SPECIES.

12. A. lebbeck (L.) Benth. in Journ. Bot. Lond. 3: 87 (1844) (as “ A. lebbek ”); Brenan

and Greenw., l.c. 342 (1949).

Type: Linnaean Herbarium, No. 1228/16.

Mimosa lebbeck L., Sp. PI. ed. 1 : 516 (1753).

M. lebbek Forsk., FI. Aegypt. Arab. 177 (1775).

Acacia lebbeck (L.) Willd., Sp. PI. 4: 1066 (1806); DC., Prodr. 2: 466 (1825).

Tree 8-15 m tall with a large round crown; bark grey, fissured; branchlets glabrous

to tomentose. Leaves 15-35 cm long and 12-30 cm broad; petiole 5-8 cm long;

petiole and rhachis pubescent to glabrescent; pinnae 2-4 pairs; leaflets 5-9 pairs,

sparingly hispidulous to glabrous, paler below, oblong to broadly elliptical, 2 -5-4 -4

cm long and 1-2-4 cm broad, oblique at the base, apex rounded, often emarginate;

margin not crisped. Inflorescences 1-3 in axils of terminal leaves, each of 20-25 flowers;

peduncle 4-6 cm long; peduncle and buds brownish tomentose. Flowers white;

pedicels 2-2-5 mm long. Calyx subcampanulate, 4-4-5 mm long, 5-toothed, grey-

brown tomentose. Corolla 6-8 mm long, 5-lobed, almost completely glabrous below

with patches of whitish tomentum towards the apex; tube subcylindrical, 4-5-5 mm

long; lobes ovate-lanceolate 2-3-5 mm long. Stamens 2-2-5 cm long, united at the

base for 5-5 - 5 mm. Ovary glabrous, 2 mm long; style filiform, exceeding the stamens.

Pod several seeded, linear to linear-oblong, 12-20 cm long and 3 -5-4 -5 cm broad,

flat, thin textured, light to dark brown; margin thickened, usually straight; position

of seeds visible as raised bumps in the valve.

A native of tropical Asia and, probably, tropical Africa, widely cultivated in the

tropics. Cultivated and semi-naturalised along the north coast of Natal. According

to Forskahl, the specific name is derived from the Arabian name Labach for the tree.

Transvaal. — Pretoria: Meyer s.n.

Natal. — Mtunzini: Lawn 2118; Gingindhlovu, Lawn 1863 (NH); valley near

Gingindhlovu, Lawn (NH). Inanda: 2 miles S. of Verulam, Codd 9653.

A. lebbeck is growing in the Botanic Gardens, Durban, and the specimens collected

from the coastal areas of Natal and Zululand are from trees that are very likely cul-

tivated or semi-naturalised. The species is included in this treatment because it is

sometimes confused in the herbarium with A. tanganyicensis and A. suluensis. For

the distinguishing characters, see the discussions under those species.

It is a graceful tree of easy culture in a frost-free climate, forming a fairly straight

trunk and a round crown of attractive, dark green foliage. Brenan and Greenway

record that it produces a useful timber and that the leaves are browsed by stock.

13. A. lophantha {Willd.) Benth. in Hook. Lond. Journ. Bot. 3: 86 (1844); FI. Austral.

2: 421 (1864).

Type: It is not known if a specimen is preserved on which Ventenat, Jard. Cels,

t. 20, based his description and figure; if not, the latter will suffice as the type.

Acacia lophantha Willd. Sp. PI. 4: 1070 (1806); DC., Prodr. 2: 457 (1825).

Mimosa distachya Vent. Jard. Cels. t. 20 (1800 or 1801), non Cav. Ic. 3: 48, t. 295

(1794 or 1795).

Type: As above.

Albizzia distachya (Vent.) MacBride in Contrib. Gray Herb. n.s. 59: 3 (1919); Salter

in Adamson and Salter, FI. Cape Penins. 452 (1950).

82

Tree 4-7 m tall; branchlets usually velvety pubescent. Leaves 20-25 cm long and

11-18 cm broad; petiole 3-7 cm long; petiole and rhachis pubescent; pinnae 7-12

pairs; leaflets 20-35 pairs, linear oblong, 7-5-11 mm long and 2-2-5 mm broad,

sparingly pubescent to glabrous, oblique at the base, apex obtuse to rounded, mucronu-

late; midrib nearer to upper margin. Inflorescence spicale, 4-8 cm long, axillary,

1-3 per axil ; peduncle 0 - 8-1 - 5 cm long, fulvo-tomentose. Flowers creamy to yellowish ;

pedicels 1-2-5 mm long. Calyx campanulate, 2-2-5 mm long, 5-toothed, pubescent.

Corolla 5 -5-6 -5 mm long, 5-lobed, appressed pubescent; tube subcampanulate, 4-5

mm long; lobes ovate, cucullate, 1-5 mm long. Stamens 1-3-1 -5 cm long, united at

the base for 1 - 5-2 - 5 mm. Ovary glabrous, 2 - 5 mm long, very shortly stipitate ; style

filiform, exceeding the stamens by 6-7 mm. Pod several seeded, linear to linear

oblong, 6-11 cm long and 1-4-1 -6 cm broad, flat, semi-woody, light to dark brown;

margin slightly thickened, straight or nearly so; position of seeds clearly evident as

raised bumps on the valve; seeds, when mature, loose and rattling in the pod.

Native of the south-western coastal region of Western Australia and naturalised

in South Africa, especially along the coast from Humansdorp to the Peninsula.

Natal.— South Coast, Wood 10588 (NH).

Cape Province. — Humansdorp: Rogers 3021; Fourcade 3934 (BOL). Knysna:

District Forest Officer in PRE 8707. Riversdale: Corente River Farm, Muir in Hb.

Galpin 5092. Caledon: Kogel Bay, Parker 4206 (BOL). Cape Peninsula: Simons-

town, Watt and Brandwijk 1682; Rondebosch, Gerstner 6135; Edinburg Estate, Salter

7356 (BOL).

It is not known when this species was introduced into South Africa. The earliest

herbarium specimen seen is from the Riversdale District, dated December, 1908, when

it was already recorded as “ common on river banks ”. Collectors describe it as almost

a ruderal weed, being common and gregarious along river banks, forest margins and

in wooded ravines.

Index. Page.

Acacia marlothii Engl. = Albizia anthelmintica 78

Albizia adianthifolia (Schumach.) W. F. Wight 79

A. amara subsp. sericocephala (Benth.) Brenan 70

A. anthelmintica (A. Rich.) A. Brongn 77

A. anthelmintica var. australis Bak. f. = A. anthelmintica 77

A. anthelmintica var. piibescens Burtt Davy = A. anthelmintica 77

A. antunesiana Harms 74

A. brevifolia Schinz 69

A. distachya (Vent.) MacBride = A. lophantha 81

A. evansii Burtt Davy 79

A. fastigiata (E. Mey.) Oliv. = A. adianthifolia 79

A. forbesii Benth 72

A. gummifera C.A.Sm. non (Gmel.) C.A.Sm. = A. adianthifolia 79

A. harveyi Fourn 71

A. hypoleuca Oliv. = A. harveyi 71

A. lebbeck (L.) Benth 81

A. lebbeck var. australis Burtt Davy = A. tanganyicensis 81

A. lophantha (Willd.) Benth 81

A. mossambicensis Sim -- A. versicolor 73

A. pallida Harv. = A. harveyi 71

A. parvifolia Burtt Davy = A. brevifolia 69

A. rhodesica Burtt Davy = A. tanganyicensis 75

A. rogersii Burtt Davy = A. brevifolia 69

A. sericocephala Benth. = A. amara ssp. sericocephala 70

A. struthiophylla Milne-Redhead = A. amara ssp. sericocephala 70

A. suluensis Gerstner 76

A. tanganyicensis Bak. f. 75

A. umbalusiana Sim = A. anthelmintica (for the greater part) 77

A. versicolor Welw. ex Oliv 73

A. versicolor var. mossambicensis Schinz = A. versicolor 73

Zygia P.Br. sensu E. Mey. = Albizia 68

Z. fastigiata E. Mey. = Albizia adianthifolia 79

83

Studies of Wood-Rotting Fungi :

1. Cultural Characteristics of some Common

Species.

By

G. C. A. van der Westhuizen.

It is often essential to identify the organisms causing decay to timber. Since a

sporophore is seldom present, diagnostic features other than those important ones

usually provided by the sporophore must be sought. It has long been known that

wood-rotting fungi grow readily on artificial media, but only comparatively recently

have cultural characters been used as in aid in the identification of non-fruiting mycelium.

A difficulty peculiar to the identification of these fungi in culture is that the morphological

characters shown are applicable directly to the species only and cannot be used in the

identification of other species in the same genus.

Among the first workers in this field were Long and Harsch (12) and Fritz (11),

who established criteria for distinguishing wood-rotting fungi in pure culture. This

work was followed by studies of fungi causing decay of a specific host, or the species

of a single genus in artificial culture. Bavendamm (1) and Davidson, Campbell and

Blaisdell (9) showed that the behaviour of decay fungi on media containing gallic

acid and tannic acid is a valuable aid in the identification of the fungus. But the

different workers followed different methods so that it was still difficult or impossible

to identify a fungus from culture unless many different methods were followed.

Recently, Nobles (16) made an exhaustive study of the various methods and incor-

porated many of these into a standard procedure. This procedure has been followed,

with some modification, by da Costa, Matters and Tamblyn (8). The macroscopic

features of the colony are described in terms defined by Long and Harsch (12). Colours

are recorded according to Ridgway’s Colour Standards or the Munsell Book of Colour.

The behaviour of the fungus on 0-5 per cent gallic acid and tannic acid in malt agar,

and the growth rate and appearance on 1 • 5 per cent malt agar are noted. These

features, as well as the nature of the host plant, the presence of fructifications and the

microscopic characters, are expressed by a system of numerals termed the key pattern.

Each digit refers to a specific character. Digits are arranged in eleven columns in

ascending numerical order. For workers to whom Nobles’ (16) paper is not readily

available, the list indicating the meaning of each digit is appended below.

In the present paper, the cultural characters of a number of fungi are described.

Some of them have been described before by other workers, but are included here to

emphasize the similarities or differences between specimens from South Africa and

elsewhere. Others are South African species and their cultural characteristics are

described here for the first time.

Methods and Materials.

The methods followed were those described by Nobles (16) and da Costa et al.

(8) but with small modifications. The fungi were grown on 1 • 5 per cent Difco malt

extract agar in three inch Petri dishes, each plate containing 30 ml. of medium. Inocula

were cut out with a sterile cork borer of 10 mm diameter from the youngest part of an

84

actively growing colony and placed mycelium downwards near one side of the plate.

Six plates of each fungus were incubated at 26° C., one plate being removed and placed

in diffuse daylight at room temperature at weekly intervals. All cultures were examined

at weekly intervals for changes of colour and other characteristics. All colours are

described according to Ridgway’s terminology (19).

The oxidase tests were carried out on media containing 1 • 5 per cent Difco malt

extract agar and 0-5 per cent gallic acid and tannic acid, respectively, as described by

Davidson, Campbell and Blaisdell (9). Incubation was at 26° C. for seven days or

fourteen days for the slower growing species.

List indicating the meaning attached to each digit as used in the key pattern.

[Nobles; (16) p. 291.]

First column: Host.

1. Occuring on broad-leaved trees.

2. Occuring on coniferous trees.

Second column: Colour of mycelial mat.

1. Mat remaining white, pale yellow or pale pink for six weeks.

2. Mat yellow or brown, at least when mature.

Third column: Reaction on media containing gallic acid and tannic acid.

1. Diffusion zone present.

2. Diffusion zone lacking.

Fourth column: Septation of phyphae.

1. Clamp connections present on all parts of mat but may be lacking on fibre

hyphae.

2. Simple septa on all hyphae.

3. Hyphae of advancing zone with simple septa, those of older part of mat with

clamp connections.

4. Multiple clamp connections present, at least in the advancing zone.

Fifth column: Special structures.

0. Contorted incrusted hyphae.

1. Cystidia or gloeocystidia.

2. Setae or setal hyphae.

3. Bulbils.

4. Rigid hyphae with right-angled branches.

5. Cuticular cells, forming a pseudo-paranchymatous layer.

6. Hyphae with numerous interlocking projections.

7. Swellings on hyphae.

8. Lactiferous cells.

9. No special structures.

Sixth column: Chlamydospores.

1. Chlamydospores present.

2. Chlamydospores absent.

o

X

1

2

85

Key to the fungi described in this paper.

86

Seventh column: Conidia.

1. Conidia present.

2. Conidia absent.

Eighth column: Oidia.

1. Oidia present.

2. Oidia absent.

Ninth column: Rate of Growth.

1. Rapid growth, plates covered in one to two weeks.

2. Moderately rapid, plates covered in three to four weeks.

3. Slow, plates covered in five to six weeks.

4. Very slow, plates not covered in six weeks.

Tetith column: Fruiting.

1. Fruiting before the end of six weeks.

2. No fruiting.

Eleventh column: Effect on Agar.

1. Reverse brown, at least in part, before the end of six weeks.

2. Reverse unchanged, or not darker than honey yellow in six weeks.

3. Reverse bleached, at least in part, before the end of six weeks.

1. Amauroderma rude (Berk.) G. H. Gunn. (PI. 1; Fig. 1, PI. 4, Fig. 1.)

Key pattern: 1 2 1 1 7 2 2 2 12 3.

Growth characters: Growth is rapid, the colony reaching a radius of 40 mm in

seven days and covering the plate in two weeks. The advancing zone is even, trans-

parent, appressed for about 0-5 mm, then slightly raised. The mat is at first white,

cottony, but after two weeks, thin strands of mycelium appear and the mycelium darkens

to “ pale brownish drab ” in some places. The colour darkens gradually as the mat

becomes more felty or sub-felty in places and dark “ hairs ” start growing out onto the

side of the dish. After six weeks, the mat is mostly “ buffy brown ” and sub-felty to

crustose in places with patches of “ avellaneous ”, “ olive brown ”, “ wood brown ”

and “ natal brown ”. The mat is tough but separates readily from the agar and a

faint, fragrant odour is given off. Reverse, bleached to “ ochraceous buff ” but later

mottled with lines of ” buckthorn brown ” and “ warm blackish brown ” in the agar.

On gallic and tannic acid media there is good growth, the colonies reaching diameters

up to 45 mm with strong diffusion zones up to 50 mm diameter after seven days.

Hyphal characters.

Advancing zone. — Hyphae hyaline, thin-walled, branched, septate with simple

clamps of the “ eyelet ” type, 3-6 p. wide.

Aerial mycelium. — (a) Hyphae as in the advancing zone, [b) Fibre hyphae nume-

rous, hyaline, without lumen and walls thick and refractive, 2-3 ii wide, (c) Hyphae

with thick, refractive walls and narrow lumina which widen towards the tip and with

deeply staining contents, 4-6 /x wide, {d) Wide hyphae, 4-6 p wide, with slightly

thickened, hyaline walls, clamp connections at the septa and wide lumina with deeply

staining contents, with narrow thick-walled fibre hyphae, 2-3 /x, branching from them.

(e) Narrow fibrelike hyphae, 2-3 /x, with thickened walls, at first hyaline, later brown,

with inflated terminal or intercalary parts up to 20 /x wide, fonning a tough, crustose

skin on the mat.

87

Submerged mycelium. — Hyphae as in the advancing zone.

A. rude is very common in wattle plantations in South Africa and has been reported

as the cause of decay of stumps and roots of Acacia mollissima. It has also been found

on unspecified dead wood (10). With its unique key pattern and combination of

readily recognizable microscopic features, the identification of this fungus from culture

should be relatively easy.

2. Coniophora arida (Fries) Karst. (PI. 1, Fig. 2; PI. 4, Fig. 2.).

Key pattern: ( 1, 2) 1 2 4 9 2 2 2 1 12.

Growth character.s. — Growth is fast, the colony reaching a radius of 38 mm after

seven days and covering the plate in two weeks. The advancing zone is bayed and

hyphae are raised right to the limit of growth. The mycelium is at first raised, silky,

but becomes appressed after two weeks and thin rhizomorphs begin to appear and

radiate from the inoculum over the surface of the agar. After 3-4 weeks the surface

appears uneven or lacunose with hyphae thin and silky and the agar surface almost

liquified. No further changes take place. The colour of the mat is at first hyaline,

later turning pale greyish. The mat remains soft and adherent to the agar which

is slightly bleached or remains unchanged. A faint odour is emitted.

On gallic and tannic acid media, no diffusion zones are found, while the colony

reaches diameters of 60 mm on the former and 15 mm on the latter medium after seven

days incubation.

Hyphal characters.

Advancing zone. — Hyphae smooth, branching, nodose-septate with multiple clamps

arranged in a whorl, often with a whorl of branches from the clamps. Hyphae are

usually 3-6 p wide.

Aerial mycelium. — (a) Hyphae as in advancing zone, 2-5-10 p. (b) Hyphae with

simple clamps, 3-6 g. (c) Wide hyphae with simple septa are present together with

the ordinary hyphae in the rhizomorphs on the surface.

Submerged hyphae. — {a) Hyphae as in the advancing zone, (b) Knobbly, tortuous,

branching hyphae with simple septa 2-6 p. wide.

Comparison with Nobles’ key shows that C. puteana, to which C. arida is closely

related, has an identical key pattern. There is indeed a very close resemblance between

the cultures of these two species and only a few small characters separate them. In

C. arida cultures, the advancing zone is bayed and the mat remains uncoloured while

the mat of C. puteana may be coloured and the margin even. These differences are,

however, so small as to suggest that the two fungi are probably different forms of the

same species.

C. arida has been found to cause a brown rot of coniferous trees but also occurs

saprophytically on deciduous trees such as Acacia mollissima.

3. Ganoderma colossum (Fr.) Bres. (PI. 1, Fig. 3; PI. 4, Fig. 3.).

Key pattern: (1, 2) 2 2 1 9 2 2 2 2 2 2.

Growth Characters. — Growth is moderately fast, the colony reaching a radius of

65 mm after 14 days. The advancing zone is even, slightly raised and is preceded by

a zone of turbid agar about 8 mm wide. At first, the mat is silky but after two weeks,

patches or tufts of cottony mycelium start to form on the sides of the dish. These

later become compacted into woolly or felty pads up to 5 mm wide around the edge

of the mat. The young mat is white and remains so for longer than six weeks, but

the dense mycelium on the sides of the dish turns to “ cream color ” or “ buff yellow ”

after three to four weeks and darkens to “ antimony yellow ” and “ cinnamon buff”.

The mat remains soft and adherent to the agar, while the reverse remains unchanged.

On gallic and tannic acid media there is good growth, the colonies reaching dia-

meters of about 35 mm after seven days but no diffusion zones are formed. On tannic

acid medium, a clear zone is formed slightly larger in diameter than the colony.

Hyphal characters.

Advancing zone. — Hyphae thin-walled, branched, nodose-septate with simple

clamps of the “ eyelet ” type, often with branches arising opposite clamps, 3-6 /x wide.

Aerial mycelium. — (a) Hyphae as in the advancing zone, (b) Thin-walled hyphae

without clamps and few simple septa, 1 • 5-2 /x wide. (c) Hyphae as in advancing

zone but with walls light brown, (d) Hyphae with deeply staining contents and simple

septa, irregularly widened and distended, 5-8 y. diameter.

Submerged mycelium. — {a) Hyphae as in the advancing zone, {b) Much branched

hyphae with simple septa.

G. colossum has been isolated from Pinus hondurensis and Callitris robusta in Natal

on which it causes an extensive collar rot. It also attacks eucalypts (14).

There are no marked morphological features by which G. colossum can be identified

in culture. The restricted host range, the absence of a diffusion zone on the oxidase

media together with the colour of the mycelial mat, which so closely resembles the

colour of the large fructifications, and the microscopic appearance of the hyphae,

should allow the easy identification of a culture of this fungus.

4. Lentinus sajor-caju Fr. (PI. 1, Fig. 4; PI. 4, Fig. 4.).

Key pattern: 12 119 12 2 12 1.

Growth characters. — Growth is rapid, the colony reaching a radius of 55 mm in

seven days and the plate being covered in ten days. The advancing zone is even to

slightly bayed, appressed, silky for 2-3 mm behind the tips and then raised. The mat

is at first cottony to woolly, turning felty to sub-felty with nodulose or granular bands

over the surface. As the culture ages, some areas become sub-felty to crustose. At

first, the mat is white, but, after three weeks, patches of “ light quaker drab ” appear,

but some darken to “ quaker drab ” or ” mouse grey ” so that the culture assumes a

grey, marbled appearance. The agar is bleached but then darkens to “ warm buff”

with dark brown wavy lines giving a mottled appearance.

After two weeks growth, the edge of the mat turns upward and inward and small

hyaline papillae appear, growing out from the mat and appressing their ends against

the glass side of the dish. A sweet, fragrant odour is emitted by the fungus.

On gallic and tannic acid media, there are strong diffusion zones of 25-40 mm on

the two media respectively, after seven days, while the colonies reach diameters of 15

mm and 35 mm.

Hyphal characters.

Advancing zone. — Hyphae thin-walled, branching, nodose-septate with simple

clamps of “ eyelet ” type, 3-5 /x wide.

89

Aerial mycelium. — (a) Hyphae as in advancing zone, {b) Fibre hyphae numerous,

thick-walled, branched or unbranched, 1-5-2 -5 y.. (c) Chlamydospores hyaline,

elliptical, oval or short, cylindrical with rounded ends, 6-8 x 9-12 /x. (d) Hyphae with

thin walls and bladderlike swellings and projections which later turn brown to form a

crustose layer.

Submerged mycelium. — Hyphae as in the advancing zone but with occasional

chlamydospores.

The most striking feature of this culture is the appearance of the papillae which

seem to push the edge of the colony away from the sides of the Petri dish. This feature,

together with the strong fragrant odour given off by the mycelium and the unique key

pattern, should establish the identity of the culture with certainty.

L. sajor-caju causes a white rot of felled deciduous trees.

5. Lenzites palisoti Fr. (PI. 1, Fig. 5; PI. 4, Fig. 5.).

Key pattern: (1, 2) 1 1 1 9 2 2 1 2 2 3.

Growth characters. — Growth is moderately fast, the plate being covered in three

weeks. The advancing zone is even, thin and closely appressed to the medium. The

mat is at first thin, with a chamois texture and later slightly farinaceous with numerous

tiny, clear droplets of liquid on it. Later it thickens, becoming compacted felty and

slightly furrowed in the thicker parts. Thin, tough, smooth pads may develop over

the surface. After six weeks the entire colony is farinaceous. The mat remains white

throughout or a “ cream color ” on some of the thickened pads. The agar is bleached

and the reverse is yellowish.

The mat is tough, adherent to the agar and emits a strong, sweetly fragrant, slightly

musty odour.

On the oxidase test media, there are strong diffusion zones with growth up to 32

mm diameter on gallic acid and 10 mm diameter on tannic acid in seven days.

Hyphal characters.

Advancing zone. — Hyphae thin-walled, simple or branched nodose-septate with

simple clamps often branching from the clamps, 1 • 5-4 g. diameter.

Aerial mycelium. — {a) Hyphae as in advancing zone, {b) Fibre hyphae numerous,

thick-walled, with narrow lumina which widen characteristically at the tips, 3-4 g.

diameter. Walls sometimes dilated into a vesicle as well, (c) Oidia elongate, cylin-

drical, with rounded ends or barrel-shaped to nearly globose, 4-9 x 2-4 n.

Submerged mycelium. — Hyphae as in advancing zone but more tortuous and with

more clamps, 2-4 /x diameter.

The strong, fragrant odour emitted by this fungus and the pure white mat with

its dewy appearance, should give an indication of its identity which may be confirmed

by the presence of the characteristic fibre hyphae and the unique key pattern. The

digit denoting swellings on hyphae is not included in the key pattern because these

swellings are only seen very occasionally.

The key pattern denotes the presence of clamps on the hyphae but this fungus

has a tendency to revert to the haploid condition after repeated subculture. This is

accompanied by loss of vigour as demonstrated by a retardation in the growth rate,

an increased production of oidia and growth with a very strongly indented margin.

As the characters otherwise remain unchanged and new isolates are invariably diploid,

the key pattern is presented as applicable to new isolates.

L. palisoti is a very common cause of white rot on timber in use but occurs on a

number of living trees as well (10).

90

6. Polyporus arcularius Batsch ex Fries (PI. 1, Fig. 6; PI. 4.).

Key pattern: 1 2 1 1 6 2 2 2 (1, 2) 3.

Growth characters. — Growth is rapid, the colony reaching a radius of 43 mm in

seven days and covering the plate after 14 days. The advancing zone is even, thin,

appressed for 2-3 mm, then raised. The aerial mycelium is at first white, cottony to

woolly, soon becoming more compact and granulose to lacunose felty, with crustose

or skinlike and brittle areas. The white mat soon shows irregular patches of “ russet ”

to “ cinnamon brown ” with a thin brown zone line appearing where the mat meets the

agar at the sides of the dish. With progressing age, the culture assumes a mottled

appearance with white areas mingled with “ russet ”, “ mars brown ”, “ Prouts brown ”

and “ mummy brown ” patches and lines. The crustose, dark-coloured patches often

have raised edges where they meet the white areas. The mat is tough after six weeks

and a faint, yeasty odour is noticeable. Stalks of pilei appear after 2-3 weeks on some

cultures and elongate into numerous contortions but bear normal pilei at their ends.

The reverse of the colony may be bleached or turn “ honey yellow ”. Lines of “ antique

brown ” to “ Prout’s brown ” appear in most cultures in the bleached agar.

On gallic and tannic acid media, strong diffusion zones of 55 mm diameter appear

under colonies which reach diameters of 30 mm and 50 mm on the two media, respec-

tively.

Hyphal characters.

Advancing zone. — Hyphae thin, branching, nodose-septate with simple clamps,

2-4 fji wide.

Aerial mycelium. — {a) Hyphae as in the advancing zone, {b) Fibre hyphae nume-

rous, walls thick, refractive, branched, 1 • 5-2-5 g. (c) Hyphae from dark skin-like areas,

nodose-septate, thick-walled, often without lumina, buffy brown in KOH with numerous

short side branches or knoblike projections, interlocked to form a tough, coherent mat

which defies separation into single elements. Fruit body: Basidia hyaline, 14-16 x 4-5

P with four sterigmata. Basidiospores: smooth, oblong-elliptical, hyaline, 6 x 2-5 /x.

Submerged mycelium. — Hyphae as in the advancing zone.

The above description corresponds closely with Nobles’ description (16) except

for colour, which in my cultures was always browner. No chlamydospores were

observed in these cultures. The formation of normal fructifications with their charac-

teristic large pores took place very readily on some of the cultures.

P. arcularius is found fairly frequently on decayed wood of broad-leaved trees in

contact with the soil but does not seem to be important as a wood destroyer in South

Africa.

7. Polyporus sanguineus Linn, ex Fries (PI. 2, Fig. 1; PI. 4, Fig. 7.).

Key pattern: (1, 2) 2 1 1 9 1 2 2 1 (1, 2) (2, 3).

Growth characters. — Growth is fast, the colony reaching a radius of 50 mm in

seven days and covering the plate in twelve days. The advancing zone is even, appressed

for 1-2 mm then raised, hyaline. The texture is at first cottony to woolly, lacunose

in the denser parts and later granular, turning to velvety or chamois near the inoculum.

From the inoculum, broad, shallow furrows radiate out to some distance behind the

advancing edge. At about four weeks incubation, the dense parts of the colony become

compacted into tough, felty to velvety pads, often raised above the level of the surround-

ing mat. On some of these, pores are developed which shed white spore deposits if

the plate is inverted.

91

The mat is at first white but after one to three weeks, tinges of “ pale salmon color ”

appear which later deepen to “ salmon color ”, “ apricot orange ”, ” carnelian red ”

or “ flame scarlet ” in the older parts with occasional “ cinnamon rufous ” near the

edge of the newest growth. The reverse is unchanged in some cultures and bleached

in others. A faint, pleasant odour is present.

On gallic and tannic acid media, there are strong diffusion zones of up to 65 mm

diameter, while colony diameters vary from 35-55 mm on gallic acid and 0-55 mm on

tannic acid media.

Hyphal characters.

Advancing zone. — Hyphae thin-walled, smooth, branched, nodose-septate with

simple clamps of the “ eyelet ” type, 2-5 p. wide.

Aerial mycelium. — (a) Hyphae as in the advancing zone, (b) Fibre hyphae with

thick, refractive walls and no lumina, branching, very numerous, 2-A p. wide,

(c) Chlamydospores smooth, hyaline, terminal or intercalary, rare in some cultures,

more numerous in others, with walls slightly thickened, 4-9 X 6-14 p.

Fruit body. — (a) Basidia clavate, 4 • 5-6 p wide with four spores. (6) Basidiospores

hyaline, short, cylindrical, 4-5-6 x 2-5-3 /x.

Submerged mycelium. — (a) Nodose-septate hyphae, (b) fibre hyphae and (c) chlamy-

dospores as above.

The cultures described here are very similar to those described by Nobles (16)

and Matters et al. (15). Like those described by the latter authors, the South African

fungi tend to bleach the medium but differ from them by having higher growth rates.

Bose (2) noticed that cultures of P. cinnabarinus differed from those of P. sanguineus,

two species considered to be identical by Wakefield (2) and Lloyd (13), in that the hyphae

of the latter develop a pigment on their walls so that the whole mat looks orange,

while the hyphae of the former remain white and the characteristic orange-red colours

only appear when basidia initials are formed in the cultures. Colour is consequently

confined to poroid areas of cultures of P. cinnabarinus. It has been noticed tha.

amongst the cultures described here, there are some which tend to form colour at a

later stage of development and in localized patches only. In the others, colour is

formed at an early stage and is more or less of an even hue over the whole surface of

the culture. The sporophores from which these cultures were made were, however,

identified as those of P. sanguineus. This phenomenon is considered to be caused by

differences between different strains of the same organism.

P. sanguineus can be readily recognized in culture by the bright orange-red colours

formed, without recourse to microscopic examination.

This fungus is widely distributed throughout South Africa and has been found on

living trees (Doidge 10) and on dead wood in which it causes a white rot.

8. Polystictus hirsutus Wulf. ex Fries (PI. 2, Fig. 2; PI. 4, Fig. 8.).

Key pattern; (1, 2) 1 1 1 9 (1, 2) 2 2 1 (1, 2) 3.

Growth characters. — Growth is rapid, a radius of 55-60 mm being reached in one

week, while the plates are covered in 9-10 days. The advancing zone is even, appressed

or slightly raised. The mat is at first cottony but soon becomes woolly and compacted

to felty or sub-felty in some places round the inoculum, but finally tough, dense, felty

and slightly lacunose, with mounds of compact waxy mycelium over the surface.

92

Mycelium frequently grows out between the dish and cover to form poroid fructifica-

tions. The predominating colour is white but patches of “ cream color ” occur near

the sides of the dish and on the poroid mounds. The reverse is rapidly bleached and

cleared.

Growth on gallic acid medium does not occur or produces a mere trace of mycelium

with a strong diffusion zone. On tannic acid, the colony may reach 40 mm diameter

in seven days, while a strong diffusion zone, slightly wider, is formed. Odour is lacking

or may be faint, sweetish.

Hyphal characters.

Advancing zone. — Hyphae hyaline, branched, nodose-septate with simple clamps,

3-5 p. wide.

Aerial mycelium:— {a) Hyphae as in the advancing zone. ^ {b) Fibre hyphae very

numerous, with walls thick and refractive, lumina narrow or lacking, aseptate, branched,

1-5-3 /X.

Fructication. — Basidia 4-5-6 /x diameter, having four spores. Basidiospores

hyaline, even, cylindrical, 5-7 x 2-5 ju,.

Submerged mycelium. — (a) Hyphae as in advancing zone, {b) Chlamydospores

observed in one isolate, terminal and intercalary, thin-walled, smooth, hyaline,

10-15 X 4-5-7/x.

The key pattern for P. hirsutus is identical with those of other fungi from which

it can, however, be distinguished on the macroscopic appearance in culture. Nobles

records, however, that the cultures of Polyporus pubescens and P. zonatus are closely

similar to those of P. hirsutus. Although neither of these two species has been grown

in culture by me yet, the former may not be a serious consideration in this respect as

it is not at all common in this country. P. zonatus is more common and great care

should be exercised in identification of cultures when there is the risk of confusing these

two closely related species.

P. hirsutus causes a white rot of coniferous and hardwoods and has been reported

on broad-leaved trees.

9. Poria vaillantii (D.C.) Fr. (PI. 2, Fig. 3; PI. 5, Fig. 1.).

Key pattern: 1 1 2 1 9 2 2 2 (1, 2) (1, 2) 2.

Growth characters. — Growth is moderately fast to fast, the plate being covered in

twelve to twenty days. The advancing zone is even, appressed for 0 - 5 or 1 mm behind

the extreme tips, than raised in a dense cottony or woolly mound but sinking again

towards the inoculum. After three weeks, dense strands of hyphae form and radiate

from the inoculum to the newer growth where their ends fan out into cottony plumes,

which cling to the sides of the dish. With advancing age, the mycelium, clinging to

the glass sides, is torn and becomes rather ragged through shrinkage of the agar. After

two to four weeks, wide, shallow pores are formed on the dense mycelium of some

cultures. Creamy spore deposits are formed if the plates are inverted.

The colour remains pure white throughout, except in the older parts of the fruc-

tification, which turn a pale cream colour. The aerial mycelium is easily stripped

from a thin layer of mycelium which adheres to the agar. The reverse remains

unchanged.

The fungus is negative for oxidase and no diffusion zones are formed. Growth

on gallic acid may reach a diameter of 60 mm in one week, but no growth takes place on

tannic acid.

93

Hypbal characters.

Advancing zone. — Hyphae thin-walled, branched, nodose-septate with simple

clamps, often branching opposite a clamp, 2-5-6 fi diameter.

Aerial mycelium. — [a) Hyphae as in the advancing zone, (b) Fibre hyphae without

lumina, refractive, unbranched, 2 -5-3 -5 /^. (c) Wide hyphae, irregularly swollen,

6-10 fj. in diameter.

Fruit body. — (a) Hyphae as above, {b) Basidia: clavate with four sterigmata

6-8 X 16-20 /X. (c) Basidiospores hyaline, oblong-ellipsoid, slightly flattened on one

side, 4-5-6 x 3-5 /x.

Submerged mycelium. — (a) Hyphae as in the advancing zone, (b) Crystals nume-

rous.

The cultures described here agree very closely with the descriptions of Nobles (16)

and Matters et al. (15). No chlamydospores were seen, though they were noted by

Cartwright and Findlay (3) but fibre hyphae were numerous in both the fructification

and the hyphal strands. This fungus may readily be recognised by the thin mat around

the inoculum with its radiating strands of mycelium, which spread out into fan-shaped

areas near the younger parts of the colony. In test tubes, these fan-shaped structures

form a plug of hyphae near the edge of the slope, while the hyphal strands lead up to

it over a thin mycelial mat.

10. Schizophyllum commune Fries. (PI. 2, Fig. 4; PI. 5, Fig. 2.).

Key pattern: 1 1 (1, 2) 1 9 1 2 2 2 (1, 2) (2, 3).

Growth characters. — Growth is moderately fast, the colony reaching a radius of

22-48 mm in seven days, while the plate is overgrown after 3-4 weeks. The advancing

zone is bayed or almost even, thin, appressed or raised. The mat is at first cottony

but soon denser patches of felty mycelium are formed. As growth proceeds, more or

less fan-shaped areas of dense, felty mycelium may develop with shallow furrows

radiating from the inoculum. From these compact, felty patches of mycelium, the

normal fructifications usually develop. At this stage, the mat is usually tough and

free from the agar. The colour remains white throughout, but fructifications may be

“pale buff”. There is a slight musty odour. The reverse remains unchanged or may

bleach slowly.

On the oxidase reaction media, the colonies reach diameters of 35-65 mm on gallic

acid and 35-50 mm on tannic acid, while a strong diffusion zone is produced on tannic

acid medium only.

Hyphal characters.

Advancing zone. — Hyphae are thin branching, smooth, nodose-septate with simple

clamps, 2-5-5 fi wide.

Aerial mycelium. — (a) Hyphae as in the advancing zone but more frequently septate,

(b) Fibre hyphae unbranched, with thick, refractive walls and narrow lumina, 1 - 5-3 /x.

(c) Thin-walled hyphae with short, narrow branches, (d) Chlamydospores: inter-

calary and terminal, hyaline, thick walled, sub-globose, oval or pyriform or elliptical,

often divided by a single septum, 4-20 X 4-8 /x.

Fruit body. — Basidia: obovate or obclavate, often swollen at the top, with four

sterigmata, 15-20 x 3-4 /x. Basidiospores: hyaline, cylindrical, rounded at the ends,

obliquely apiculate, 2-3 x 5-7 /x.

Submerged mycelium. — {a) Hyphae as in the advancing zone, {b) Chlamydospores

as above.

94

Schizophyllum commune is a very common saprophyte of timber (10) but has also

been reported as a parasite of living trees (18). In culture, it is readily recognizable

by the production of mature sporophores within a few weeks. Some cultures do not

produce fructihcations, but the presence of the thin-walled hyphae with the minute

side branches, and the chlamydospores, which are usually present in large numbers,

provide useful features for identification. The absence of a diffusion zone on gallic

acid medium, in combination with these morphological characteristics, makes S.

commune one of the easiest basidiomycetes to recognise in pure culture.

11. Stereum hirsutum (Willd.) Pers. (PI. 2, Fig. 5; PI. 5, Fig. 3.).

Key pattern: (1, 2) 1 1 4 9 2 2 2 1 2 (2, 3).

Growth characters. — Growth is rapid, the plate being covered in less than two

weeks. The advancing zone is thin, even, appressed, but merges into the loose, cottony

mycelium a few millimeters behind it. At two weeks, the culture has a thin, even

cottony texture with little balls of dense compacted mycelium scattered through it.

Gradually, the mycelium becomes denser and becomes compacted in some areas into

smooth, felty or chamois areas.

The colour is at first hyaline, gradually turning creamy white to “ cream color ”

with patches of “ pinkish buff”, “ light buff” to “ ochraceous tawny ” on thickened

parts at the rim of the dish. On some of the cultures are thickened pads exuding drops

of dark amber liquid.

The reverse of the colony turns to “ ochraceous buff ” after the agar is bleached.

The mycelial mat is at first soft, but later toughens somewhat in the felty areas. A

musty odour is given off by the growing colony.

Hyphal characters.

Advancing zone. — (a) Hyphae narrow, unbranched, with simple clamps over the

septa, 1-2 p. wide, (b) Thin- walled hyphae 6-10 ju. wide, with numerous, very con-

spicous clamps arranged in whorls at the septa, branched, sometimes with branches

from the clamps, often with narrow or wide side branches similarly nodose-septate

or with simple septa or simple clamps at the septa.

Aerial mycelium. — {a) Hyphae as in the advancing zone but slightly wider, up to

12 p. (b) Fibre hyphae with thick, refractive walls, sometimes faintly yellow, 1-5-3 p.

(c) Narrow, helicoid hyphae, thin-walled, 2-3 p wide.

Submerged mycelium. — Hyphae as in the advancing zone but more narrow hyphae,

1-2 p wide and much branched, are present.

Stereum hirsutum has been described in pure culture by Cartwright and Findlay

(3), but these authors do not mention the coiled hyphae which occur quite frequently

in cultures, whose features otherwise agree closely with their descriptions and which

were made from sporophores of undoubted 5. hirsutum. This common fungus is

probably one of the easiest to recognise in culture as the scattered compact balls of

mycelium and the coarse hyphae with whorled clamp connections are characteristic.

12. Stereum purpureum (Pers. ex Fr.) Fries (PI. 2, Fig. 6; PI. 5, Fig. 4.).

Key pattern: 1 1 1 1 1 2 2 2 1 1 (2, 3).

Growth characters. — Growth is rapid, the mat attaining 70 mm in seven days and

the plate being covered in about ten days. The advancing zone is even or slightly

bayed and appressed in a zone about 1 mm wide, but the hyphae are then raised. The

95

texture is coarse, silky at first, becoming plumose to farinaceous in patches. The

mycelium finally turns felty with nodules of compact mycelium against the glass sides

of the dish. The colour remains white but a “ pale buff ”, thin rind of fructification

may appear on the side of the dish. The reverse remains unchanged or may be slightly

bleached before six weeks. The mat is at first soft and adherent but later becomes

tougher. There is a slight, musty odour.

On gallic and tannic acid agar, there is good growth, the colony attaining 63 mm

diameter on gallic acid and 42 mm on tannic acid. The diffusion zones are fairly

strong but much smaller than the colony, being about 45 and 35 mm for the two media,

respectively.

Hyphal characters.

Advancing zone. — Hyphae are smooth, thin-walled, simple or branching, nodose-

septate with simple clamps, 4-6 p wide.

Aerial mycelium. — (o) Hyphae as in the advancing zone but some are thinner,

2-6 p wide, (b) Fibre hyphae narrow, thick-walled, sometimes branching, (c) Gloeocy-

stidia vesicular, globose to almost pyriform, thin-walled, 8-12 /n diameter.

Fructification. — (a) Gloeocystidia as above, (b) Basidia clavate with four sterig-

mata, 3-5 x 18-24 /x. (c) Spores elliptical to almost cylindrical, hyaline, smooth,

obliquely apiculate, 3 x 5 ju..

Submerged mycelium. — Hyphae as in the advancing zone, crystals octahedral or

amorphous and numerous.

S. purpureum has been described in pure culture by Cartwright and Findlay (3)

who noticed the swollen hyphae or vesicles which are characteristic of this fungus.

Talbot (20) considers these structures to be gloeocystidia so that they are listed under

the numeral for gloeocystidia in the key. S. purpureum has a key pattern unlike that

of any other species so that it should be readily identifiable in this way. The rather

conspicuous gloeocystidia are a valuable guide in this matter.

This fungus is best known in South Africa as a parasite of fruit trees where it

causes a disease with readily recognizable symptoms. It has, however, been reported

from Populus sp. and Quercus sp. (10), so that its inclusion here may aid its recognition

on hosts other than fruit trees.

13. Stereum sanguinolentnm (Alb. and Schw. ex Fr.) Fr. (PI. 3, Fig. 1 ; PI. 5, Fig. 5.).

Key pattern; 21 14922222 2.

Growth characters. — Growth moderately rapid, the colony reaching a radius of

50 mm after fourteen days. The advancing zone is even, appressed, with sparse, thin,

cottony mycelium extending to the limit of growth. The mat is at first white and downy

and may remain so for some weeks, then turning “ chamois ” while it thickens to become

cottony-floccose. At six weeks, the oldest pares are felty and coloured “ light buff ”

to “ capucine buff” or “ yellow ochre ”. In some parts, especially near the edge, the

mat is thin, skin-like and farinaceous or minutely granular. The mat remains soft

and adherent to the agar and gives off a pleasant, sweet, mushroomy odour. The

reverse remains unchanged or may turn “ honey yellow ”.

On gallic and tannic acid media the diffusion zones are strong and about 30 mm

in diameter, while the colonies grow to diameters of about 15 and 30 mm on gallic and

tannic acid media, respectively.

96

Hyphal characters.

Advancing zone. — Hyphae hyaline, thin-walled, branched with simple septa and

occasional large, simple clamps or paired clamps on the wider hyphae, 2-6 /x.

Aerial mycelium. — {a) Narrow hyphae with deeply staining contents and incon-

spicuous, simple septa or simple clamps, 1 • 5-5 /x wide, {b) Wide hyphae with granular

contents, often with simple septa and multiple clamps present on the same hyphae,

3-6 /X wide, (c) Helicoid hyphae are fairly numerous, 2-3 y..

Submerged mycelium. — (a) Hyphae as in the aerial mycelium, (b) Crystals large,

octahedral.

This fungus causes collar and root rot of Pinus taeda in South Africa (14) but is

also well known in North America as a cause of heart rot in coniferous trees. The

South African strain fits the description by Nobles (16) very well, but the “ conducting

hyphae ” with the swollen tips were not seen in culture, although they are present in

the sporophores.

The most striking characteristics in culture are the presence of simple septa and

multiple clamps in the same hyphae, the slow rate of growth and the thin mycelial mat.

Its exclusive occurence on coniferous trees, should aid in its identification from culture.

14. Trametes cingulata Berk. (PI. 3, Fig. 2; PI. 5, Fig. 6.).

Key pattern: 1 1 1 1 9 1 2 2 1 (1, 2) 3.

Growth characters. — Growth is moderately fast, the colony reaching a radius of

38 mm in seven days and covering the plate after two weeks. The advancing zone is

even and appressed for 1 mm behind the extreme limit of growth, the hyphae then

becoming raised. The mat is white and remains so. Texture is at first thin, downy

or downy-cottony, becoming sub-felty or chamois near the inoculum. Later it becomes

appressed, downy, cottony, with farinaceous areas near or round the inoculum.

Fruiting areas first appear as white or “light buff” farinaceous or granular specks

which soon turn dull pasty and grow together to form irregular masses which develop

irregular pores near the inoculum or over the youngest growth. The mat becomes

tough, remains free of the agar and does not emit any odour. The agar is quickly

bleached to clear, transparent, milky white.

On gallic and tannic acid media, there is no growth but a strong diffusion zone

about 50 mm diameter is found on gallic acid and a small, weak one on tannic acid.

Hypha! characters.

Advancing zone. — Hyphae narrow, branched, nodose-septate with simple clamps

of “ eyelet ” type, often branching from the clamps, 2-4 /x.

Aerial mycelium. — (a) Hyphae as in the advancing zone, {b) Fibre hyphae with

thick, refractive walls, unbranched and without lumina, numerous, 1 • 5-5 /x. (c) Chlamy-

dospores ovate elliptical to sub-globose, thick-walled, terminal, 4-8 x 6-12 /x.

Fruit body. — Basidia pyrifoiTn with four sterigmata. Basidiospores hyaline, globose,

3-7-4 /X in diameter.

Submerged mycelium. — Hyphae as in the advancing zone.

Trametes cingulata could possibly be confused in culture with Poly poms zonatus

which has the same key pattern in some forms. It appears, however, from Nobles’

description that P. zonatus has a thicker, denser mat than T. cingulata in cultures of

six weeks old or less. T. cingulata is distinguished from other forms in nature by

possessing a matt, black, upper surface to the sporophores. Cultures of this fungus

were seen to develop this matt, black colour after 10-12 weeks in minute spots on the

mycelium but visible only under 25 X magnification.

97

15. Tramentes meyenii (Klolzsch) Lloyd. (PI. 3, Fig. 3; PI. 5, Fig. 7.).

Key pattern: 1 1 1 1 9 2 2 1 1 2 3.

Growth characters. — Growth is rapid, a radius of 65 mm being reached in seven

days, while the plate is covered in ten days. The advancing zone is even, appressed

for about 1 mm, then raised. The mat is at first cottony to woolly but soon becomes

•more dense and compact to felty, until at six weeks it is very tough, felty, slightly lacunose

over parts of the surface and faintly striate with fine grains on some parts of the surface.

By this time, it is leathery in appearance and consistency. At first, the mycelium is

translucent white but turns pure white as the mat becomes denser. After 3-4 weeks,

a “ pale cream color ” band, about 1 5 mm wide, appears over the surface about half-way

across the Petri dish. The agar is rapidly bleached as growth proceeds, becoming

colourless. The cream coloured reverse of the mat is visible through the agar after

4-5 weeks.

Good growth takes place on gallic and tannic "acid, a diameter of 40 mm being

reached on the former and-65 mm on the latter medium in seven days. Strong diffusion

zones are present in both media.- - - • ■ '

Myphal c/iaracters.

' Advancing' zone.— Hyphae thin-walled, simple or with short side branches or

unbranched, often branching from.clarnps, nodose-septate, clamps simple, “eyelet”

type, 2-5 /x wide. "

Aerial mycelium. — (a) Hyphae as in advancing zone, (b) Fibre hyphae with thick,

refractive walls, no lumina, branched, without clamps or septa, 2-5 /x wide, very nume-

rous. (c) Oidia cylindrical with rounded ends, hyaline, 2-4 x 3-6 ju.

Submerged mycelium. — Hyphae branching profusely, thin-walled, nodose-septate

with simple clamps as in advancing zone, 2-5 /x.

The key pattern of this fungus is unique so that there should be no difficulty in

the identification of an otherwise featureless culture. The rapidly growing aerial

mycelium, which becomes compacted into the white, tough mat, should hint at its

identity. Although characteristic features are lacking, there should be no difficulty

in recognizing the fast-growing, tough mat of this fungus if it has been seen previously.

In old cultures in tubes, fructifications were seen when the mat was placed in a

vertical position for 2-4 months but no fructifications formed in less than six weeks.

T. meyenii has been reported on broad leaved trees and is a common saprophyte

on hardwoods (10).

16. Trametes proteus (Berk.) Fr. (PI. 3, Fig. 4; PI. 5, Fig. 8.).

Key pattern: 1 2 1 1 0 2 2 2 1 (1, 2) 3.

Growth characters. — Growth is rapid, the colony covering 55 mm in seven days

and the entire plate in less than two weeks. The advancing zone is even, raised, white

and the mycelium cottony. The mat is at first cottony but soon becomes appressed

and sub-felty with irregular translucent places on the surface. The sub-felty areas

later turn farinaceous. About half-way across the mat, a raised band of cottony

mycelium may form which turns brownish after 3-4 weeks. On the sides of the dish

and other places, fructifications form which start as areas of compacted, white mycelium,

which later turn brown, and develop large, incomplete pores. Some areas may remain

white or the entire culture may become overgrown with the cottony, brown hyphae.

The agar is quickly bleached and the mat gives off a pleasant, fragrant, mushroomy

odour.

98

On gallic and tannic acid media, there are strong diffusion zones after one week.

No growth takes place on gallic acid in seven days. On tannic acid medium, the

colony may reach a diameter of 35 mm in seven days, or no growth may occur.

Hyphal characters.

Advancing zone. — Hyphae simple or branched, hyaline, septate with simple clamp

connections of the “ eyelet ” type occasionally branching from the clamps, 2-5 p. wide.

Aerial mycelium. — {a) Hyphae as in the advancing zone, {b) Fibre hyphae brown

with thick walls and narrow lumina, aseptate, often branching but mostly simple,

numerous, long, 2-5 y. wide, (c) Hyphae encrusted with crystals but not very numerous;

4-6 y.

Fruit body. — (a) Basidia clavate to obpyriform with four sterigmata, 18-24 X 6-10

y. (b) Basidiospores hyaline, ovate to almost elliptical, obliquely apiculate,

3-^-5 X 7-5-10^.

Submerged mycelium. — Hyphae as in the advancing zone but more richly clamped

and branched, often with numerous, short, lateral branches.

The key pattern for T. proteus is unique so that cultures may be identified by

direct consultation of the key. Indeed, the rapid growth of the felty mat, with its

brown, tangled overgrowth of fibre hyphae that soon tend to form fructifications,

are so characteristic, that cultures may be identified without recourse to microscopic

examination, if this species has been seen in culture previously.

Discussion.

It is evident that in identifying these fungi from culture, a very important consi-

deration must always be their micromorphology. In the introduction to this paper,

it was stated that identification from cultures must go directly to the species, because

species in the same genus do not show common generic characters in culture. In the

Polyporaceae, this is very noticeable if the fungi are named according to the traditional

classification of Fries and others, based on external morphology. However, Cunning-

ham (5), basing his work on the original observations of Corner (4), has recently proposed

a classification of the polypores which groups the fungi according to the colour and

morphology of the hyphae, and the type of basidium present. When Cunningham’s

system is applied, it is found that cultures of polypores do show certain microscopic

features which characterise the genus. It was observed during the course of this work

that the fungi described here as Polyporus sanguineus Polystictus hirsutus, Trametes

cingulata and Trametes meyenii all have hyaline, generative hyphae with clamps at the

septa, hyaline, long, narrow, thick-walled fibre hyphae (or “ skeletal ” hyphae) and

clavate basidia. These species are all grouped by Cunningham (6) under the genus

Coriolus. On the other hand, Trametes proteus differs from these in possessing brown

fibre hyphae. These brown hyphae distinguish the genus Trametes from Coriolus in

Cunningham’s classification.

Binding hyphae, which characterise those polypores with trimitic hyphal systems,

were never seen in culture so that it is impossible to distinguish between dimitic and

trimitic genera with hyaline hyphae. Nevertheless, Cunningham’s system of classifi-

cation does help to group cultures of polypores with common generic features and

thus aids in the identification of an unknown polyporoid culture.

In the lower Hymenomycetes, a similar system of classification based largely on

microscopic features, is being worked out, (Cunningham 7). As in the Polyporaceae,

fungi grouped according to this system should show common generic features in culture.

It appears, however, that many of the structures present in the fructifications of wood-

rotting fungi, are not formed in the cultures. For this reason, the work of Cunningham

is unfortunately of limited use in the identification of unknown cultures.

99

More recently, Pinto-Lopes (17) also proposed a classification of the Polyporaceae,

based on their hyphal characteristics. He stated (p. 116), that the same types of hyphae

are produced in culture as are present in the fructifications. His descriptions are,

unfortunately, not sufficiently detailed for use in the identification of unknown poly-

poroid cultures.

In this paper, the traditional nomenclature for the Polyporaceae was adhered to

for the sake of uniformity with earlier work and because the South African species

have not yet been revised in the light of the new approach.

Literature Cited.

1 . Bavendamm, W tiber das Vorkommen iind den Nachweis von Oxydasen bei

holzzerstorenden Pilzen. Z. Pflanzenkrank. Pflanzen-

schutz, 38: 257-276, 1928.

2. Bose, S. R Identity of Polystictus (= Polypoius = Trametes) cinna-

barinus (Jacq.) Fr. with Polystictus sanguineus (L.) Fr.

Nature 170: 1020, 1952.

3. Cartwright, K. St G. and Decay of Timber and its Prevention. His Majesty’s Stationery

Findlay, W. P. K. Office, London, 1946.

4. Corner, E. J. H The fruit body of Polystictus xanthopus Fr. Ann. Bot. 46:

71-111, 1932.

5. CirNNiNGHAM, G. H Notes on classification of the Polyporaceae. New Zeal.

Jnl. Sci. and Technol. 28: (Sec. A.) 238-251, 1946.

6. Cunningham, G. H Australian Polyporaceae in herbaria of Royal Botanic Gar-

dens, Kew, and British Museum of Natural History. Proc.

Linn. Soc. N.S.W. 75: 214-249, 1950.

7. Cunningham, G. H Thelephoraceae of New Zealand Part I : Sub-family Cyphelloi-

deae. Trans. Roy. Soc. New Zealand 81, Part 2: 165-188,

1953.

8. Da Costa, E. W. B., Matters, C. N. The Identification of Basidiomycete fungi in culture. Progress

and Tamblyn, N. Report No. 2 C.S.I.R.O. (Aust.) Division of Forest Pro-

ducts. Sub-Project p. 11-12, 1952.

9. Davidson, R. W., Campbell, W. A. Differentiation of wood-decaying fungi by their reactions on

AND Blaisdell, D. J. gallic or tannic acid medium. J. Agr. Res. 57: 683-695,

1938.

10. Doidge. E. M The South African Fungi and Lichens. Bothalia 5: 1094

p.p. 1950.

11. Fritz, C. W Cultural criteria for the distinction of wood-destroying fungi.

Trans. Roy. Soc. Can. 17: 191-288, 1923.

12. Long, W. M. and Harsch, R. M. Pure cultures of wood-rotting fungi on artificial media. J.

Agr. Research, 12: 33-82, 1918.

13. Lloyd, C. G Mycological notes 6: 1270, 1924.

14. Luckhoff, H. a Two hitherto unrecorded fungal diseases attacking pines and

eucalypts in South Africa. Jnl. S.A. For. Ass. No. 26:

47-61, 1955.

15. Matters, C. N., E. W. B. Da The identification of Basidiomycete fungi in Culture. Mor-

CosTA AND N. Tamblyn. phological characters of 14 cultures of Basidiomycetes

C.S.I.R.O. (Aust.) Division of Forest Products Sub-

project p. 11-12 Progress Report No. 2, 1952.

16. Nobles, M. K Studies in Forest Pathology. VI. Identification of cultures of

wood-rotting fungi. Can. Jnl. Res., C. 26: 284-431, 1948.

17. Pinto-Lopes, J “Polyporaceae” Contribuicao para a sua bio-taxonomia.

Memorias da Sociedade Broteriana 8: 5-191, 1952.

18. PuTTERiLL, V. A The biology of Schizophyllym commune Fries with special

reference to its parasitism. Union of S. Africa Dept.

Agr. Sci. Bull. 25, 1922.

19. Ridgway, R Color Standards and Nomenclature, Washington D.C. 1912.

20. Talbot, P. H. B Micromorphology of Lower Hymenomycetes. Bothalia 6:

279, 1954.

3871936-4

fr,

1-.

101

Plate I. — Fig. 1, Amauroderma rude, 2 weeks; Fig. 2, Coniophora arida, 3 weeks;

Fig. 3, Ganoderma colossum, 3 weeks; Fig 4, Lentinus sajor-caju, 2 weeks;

Fig. 5, LenzUes palisoti, 2 weeks; Fig. 6, Polyporus arcularius, 3 weeks.

3871936

102

Plate 2. — Fig. 1, Polyporus sanguineus, 2 weeks; Fig. 2, Polystictus lu'rsutus, 2 weeks;

Fig. 3, Poria vaillaniii, 3 weeks; Fig. 4, Schizophyl/uni commune, 3 weeks;

Fig. 5, Stereum hirsutum, 2 weeks; Fig. 6, Stereum purpureum, 2 weeks.

103

Plate 3. — Fig. 1, Stereum sanguinoleutuni, 3 weeks; Fig. 2, Trametes cingulata,

3 weeks; Fig. 3, Trametes meyenii, 2 weeks, Trametes proteus, 2 weeks.

Plate 4.

Fig. \ .—Amauroderma rude: (a) Advancing hypha. (b) Aerial hypha. (c) Fibre

hyphae with swellings, (d) Wide hypha with thick walled fibre hyphae branching

off. (e) Terminally inflated fibre hyphae from crustose areas.

Fig. 2. — Coniophora arida: (a) Advancing hyphae with multiple clamps and whorl

of branches from clamps, (h) Aerial hypha with multiple clamps, (c) Aerial

hypha with simple and multiple clamps, (d) Submerged hypha.

Fig. 3. — Ganodenna colossum: (a) Advancing and aerial hyphae. (b) Widened

hyphae with simple septa and deeply staining contents, (c) Thin-walled narrow

hyphae without clamps and few simple septa.

Fig. 4. — Lentinus sajor-caju: (a) Advancing and aerial hyphae with clamps, {b) Fibre

hyphae. (c) Chlamydospores. (b) Thick-walled, swollen hyphae from crustose

layer, (e) Submerged hypha with chlamydospore.

Fig. 5. — Lenzites palisoti: (a) Advancing and aerial hyphae. (b) Fibre hyphae.

(f) Fibre hyphae with wide lumen at tips, (d) Oidia. (c) Chlamydospore.

(/) Submerged hypha.

Fig. 6. — Polyporus arcularius: {a) Advancing and aerial hyphae. (b) Fibre hyphae.

(c) Thick-walled brown hyphae from skinlike areas, (d) Basidia and basidio-

spores.

Fig. 7. — Polyporus sanguineus: (a) Advancing and aerial hyphae. {b) Fibre hyphae.

(c) Chlamydospores. (d) Submerged mycelium with chlamydospores.

Fig. 8. — Polystictus hirsutus: (a) Advancing and aerial hyphae. (b) Fibre hyphae.

(c) Submerged hyphae. (d) Oidia. (e) Basidium and basidiospores.

105

PLATE 4

Fig. 7

Fig. 8

Plate 5.

Fig. 1. — Poria vaillantii: (a) Advancing hyphae. (b) Aerial hyphae. (c) Wide

irregularly distended hyphae. (d) Fibre hypha. (e) Basidia and basidiospores.

Fig. 2. — Schizophyllwn commune: {a) Advancing hypha. (h) Chlamydospores.

(c) Hyphae with short, narrow side branches, {d) Submerged mycelium.

(e) Basidia and basidiospores.

Fig. 3. — Stereum hirsutum: {a) Advancing and aerial hyphae with multiple clamps.

(b) Fibre hyphae. (c) Helicoid hyphae.

Fig. 4. — Stereum purpureum: (a) Advancing hyphae. (b) Wide hypha from aerial

mycelium, (c) Vesicular gloeocystidia. (d) Basidia and basidiospores.

Fig. 5. — Stereum sanguinolentum: (a) Advancing hyphae. {b) Wide hyphae with

simple septa and paired clamps, (c) Helicoid hyphae.

Fig. 6. — Trametes cingulata: (a) Advancing and aerial hyphae. (b) Fibre hyphae.

(c) Chlamydospores. (d) Submerged hypha. (e) Basidia and basidiospores.

Fig. 7. — Trametes meyenii: (a) Advancing and aerial hyphae. (b) Fibre hyphae,

(c) Submerged hypha. (d) Oidia.

Fig. 8. — Trametes proteus: (of Advancing hyphae. {b) Fibre hyphae. (c) Encrusted

hypha. (d) Submerged mycelium, (c) Basidia and basidiospores.

107

PLATE 5

Fig. 8

109

New and Interesting Records of South African Fungi.

Part hi.*

By

P. H. B. Talbot.

1. Schizostoma laceratutn Ehrenb. ex Lev.

This species has previously been recorded only from equatorial Africa, Australia,

India (Cunningham, 1944) and California (Lloyd, 1923). Mr. J. P. H. Acocks, who

has a remarkable flair for finding rare Gasteromycetes, recently made a fine collection of

twenty-two fructifications of this species on a sandy slope below Jan Swartsberg, 25

miles west of Williston, 30/vii/1956 (Acocks, A. 18917; PRE. 41592). The fungus

was stated to be frequent in that locality.

The species is readily identifiable from Cunningham’s description. Published

photographs of the species show only one or two specimens, which are sometimes

incomplete. Thus it seems desirable to publish the accompanying photograph (Pig. 1)

which, in the original at least, gives a very good idea of the structure of this attractive

and unmistakeable fungus.

The exoperidium is a sand-hyphal layer, fugitive except at the base. The endoperi-

dium is umber brown, hard, papyraceous, polished, pitted, and dehisces by means of

irregular splits from the apex downwards. The stipe, somewhat paler than the endo-

peridium, is socketed to the base of the peridium, and has a small semi-volvate mycelial

pad at the base and also longitudinal striae and ill-defined appressed scales. The gleba

is coloured deep umber with a purplish tinge. The capillitium threads are short,

yellowish, non-septate, thin-walled, with characteristic short lateral branches with

rounded ends. The spores are brown, smooth, subglobose, 4-5-6 /x diameter.

2. Aseroe rubra La Bill, ex Pries.

Although this species was described by Miss Bottomley (1948, p. 526), its occurrence

in South Africa has hitherto been doubtful. The only known record was a specimen

in Herb. Kew. collected by W. T. Saxton in Cape Town, which Miss Bottomley thought

might have been confused with Anthurus archeri. It was therefore most interesting

to receive indubitable specimens of Aseroe rubra collected by Mr. M. J. Howell in a

plantation of Acacia mollissima at Otto’s Bluff, Natal, on 2/ii/1957 (PRE. 41735),

and again at Richmond, Natal, on lO/x/1957 (PRE. 41746). The first of these collec-

tions is illustrated in Fig. 3.

Miss E. L. Stephens informs me that 15 to 20 specimens of A. rubra were found

growing close together on the farm Die Hoek, Swellendam C.P., by Miss D. Hermans

and Mr. C. Steytler on 16/iii/1958 (Herb. E. L. Stephens No. 1845).

* Parts I and II of this paper appeared in Bothalia 6 (1951) 183-204 and Ibid. 6 (1956) 489-500

respectively.

110

3. Chlamydopus meyenianus (Klotzsch) Lloyd.

There is only one previous record of this species for South Africa (Bottomley,

1948, p. 625), for which no details of locality and collector were available. The species

has been found again by Mr. J. P. H. Acocks on the ground at Pampoenpoort, C.P.,

29/x/1954 (Acocks, 17842; PRE. 41007). It is illustrated in Fig. 2.

4. Battarea stevenii (Liboschitz) Fries.

This species is well illustrated and described by Miss Bottomley (1948, p. 620)

who examined several South African collections. Four unusually fine specimens

emerged in the grounds of the Division of Botany, Pretoria, in October 1950, (leg.

P. H. B. Talbot; PRE. 39098) and provided an opportunity to take the accompanying

photograph (Fig. 4). This supplements Miss Bottomley’s illustrations by showing

clearly the nature of the gleba before dehiscence of the peridium. The specimens

appeared on a roadside which had been filled in a few months previously with very

hard soil containing large pieces of shale, which were lifted and pushed aside by the

fungus as it broke through the surface. About half of the stipe was underground.

The fungus has not been seen there since.

5. Helicosporium aureum (Corda) Linder.

This attractive species was collected by Dr. A. L. James on a beam of imported

southern pine which had been stacked for sixteen months in the open at Brakpan Mines

Ltd., 8/v/1952 (T.R.L. 2900; PRE. 41738). It is superbly illustrated by Linder (1929,

p. 279).

The fungus forms an olive-yellowish, loose, cottony layer which is separable from

1he substratum. The conidiophores are erect, brown, septate, 300-400-(700) /a in

length and 6-7 /x wide at the base, tapering and becoming paler towards the apex.

They are simple at first but later branch at right angles in the upper half to two-thirds

of the conidiophore. Conidia are borne pleurogenously on hyaline bladder-like

swellings, 6-9 /x long, which arise laterally on the lower part of the conidiophores.

The conidia are hyaline to very pale yellowish, indistinctly multiseptate, coiled 3} times

in one plane. The coiled conidium is up to 22 fj. in diameter and its filament is 1-2

yx wide.

Among the helicosporous fungi imperfect i this species is easily recognised by its

erect brown conidiophores which branch above at right angles, and by its coiled conidia

borne on terminal spicules of bladders which occupy the lower part of the conidiophore.

6. Tilletia ehrhartae sp. nov.

Sori nigri in ovariis Ehrhartae calycinae. Sporae verrucosae fuscae globosae

vel subglobosae, 19-24 yu. diam. ; verrucae rolundae ad 1-5 yix altae. Typus PRE.

41604, leg. R. Story (6006).

Sori black, occupying the ovaries of Ehrharta calycina; spores dark-coloured,

globose to subglobose, 19-24 fx diam., including the verrucose epispore composed of

rounded warts up to 1 • 5 yx high. Type R. Story (6006), on Ehrharta calycina. Nature

Reserve, Lambert’s Bay, Clanwilliam distr., Oct., 1956.

Dr. M. B. Ellis of the Commonwealth Mycological Institute has favoured me by

comparing this collection with the type of Tilletia verrucosa Cooke & Massee, the only

other Tilletia yet reported on Ehrharta (Zundel, 1953, p. 301). In T. verrucosa the

spores are 20-24 yx diam. (including the warts); they are very much paler than those of

T. ehrhartae and the warts are 3-4 yx long and acute at their tips. The contents of the

sori are also much paler in T. verrucosa.

Ill

7. Thecaphora bulbinellae sp. nov.

Sori in floribus celati ovaria perdentes; Cumulus sporarum granulosus niger;

Columella nulla; Glomeruli sporarum stabili fusci subglobosi vel ellipsoidei, 24-54

/X diam., sporis 20 vel multis; Sporae globosae vel subglobosae vel ellipsoideae vel

nonnihil angulalae, (4-6)-7-7-12-3 /x maximo diametro, leves fuscae crasse tunicatae;

Epispora crassitudine 0-5-1 -5 /x.

Sori enclosed by the flowers and destroying the ovaries. Spore mass granular,

black. Columella absent. Spore-balls permanent, brown, subglobose to ellipsoid,

24-54 /X diam., containing from 20 to numerous spores. Spores firmly united, globose,

subglobose, ellipsoid or somewhat angular, (4-6)-7-7-12-3 jj. in greatest diameter,

yellow-brown to reddish-brown, thick-walled, the epispore 0-5-1 -5 /x thick, smooth

on both the united and free faces of the spores. On Liliaceae: Type PRE. 41745,

leg. J. Toxopeus on Bidhinella setosa (Willd.) Dur. & Schinz, Struben’s Valley, Pretoria,

24/vi/57.

Literature Cited.

Bottomley, a. M (1948). The Gasteromycetes of South Africa. Bothalia 4, p. 473.

Cunningham, G. H (1944). The Gasteromycetes of Australia and New Zealand, p. 190.

Linder, D. H (1929). A Monograph of the Helicosporous Fungi Imperfecti.

Ann. Mo. Bot. Gard. 16, p. 279.

Lloyd, C. G (1923). Mycological Notes 7, p. 1173.

ZuNDEL, G. L (1953). Ustilaginales of the World, p. 301.

113

Fig. 1. — Schizostoma laceratum X 5 6.

[Photo : H. King.

114

Above; Fig. 2. — Chlamyclopus meyenianus X

Below: Fig. 3. — Aseroe rubra X 1.

[Photo: J. Reyhurn.

[Photo: J. Reyburn.

Fig. 4. — Battarea steveini X 1/2.

[Photo: H. King.

117

The Genera

and

Craterellus^ Cymatoderma (Cladoderris)

Thelephora in South Africa.

By

P. H. B. Talbot.

CRATERELLUS Persoon.

1. Craterellus cornucopioides (Linn, ex Fr.) Pers., Myc. Eur. 2 (1825) 5; Doidge in

Bothalia 5 (1950) 483.

Peziza cornucopioides Linn., Sp. Plant. (1753) 1181.

Cantharellus cornucopioides Linn, ex Fr., Syst. Myc. 1 (1821) 321.

Fig. 1.

Fructifications single or caespitose, about 4-5 cm high and 2-5 cm diam. at the

apex, thin, membranous, drying brittle, trumpet-shaped, tapered to the base, hollow

to the base of the stipe. Hymenium on the outer side, cinereous when moist, drying

brownish to yellow-brown and black at the base, smooth, becoming longitudinally

wrinkled, subgelatinous, drying corneous. Abhymenial surface deep smoky brown,

becoming almost black, squamulose. Margin often darker than the rest, usually

curved downwards, sometimes erect, lobed. Stipe hollow, black, smooth, glabrous,

up to 5 mm diam.

Basidia: long clavate or subcylindrical, 6-9 X 50-70 p., with 2-4 sterigmata.

Spores: hyaline, smooth, elliptic-oblong, with a small lateral apiculus, 6-5-8-(9-6) x

12-14-(16-5) p.

Specimens examined: 10456 (J. M. Wood, 4108), J. Blake, Harrison, Natal.

This is the only species of the genus to have been recorded in South Africa, and

the above specimen is the only one yet found. The basidia and spores were clearly

seen and were quite typical of this species. Other microscopic characters were indis-

tinct.

The systematic position of Craterellus is likely to remain speculative until the

Aphyllophorales have been more critically examined by hyphal analysis and by compari-

son of other microscopic features. A recent view of Craterellus is to place it in the

Cantharellaceae (Singer in Lilloa 22, 1949, 730), a family supposedly related rather to

the Clavariaceae than to the Agaricaceae or Thelephoraceae where it has formerly

been classed.

I have here considered Craterellus with the Thelephoraceae because it has been

so recorded in previous South African literature and because my present object is to

examine the validity of these specific and generic records.

118

CYMATODERMA Jungh.

The name Cladodenis Pers. ex Berk., a later synonym of Cymatoderma Jungh.,

has almost exclusively been used for this genus in the past, and was proposed for conser-

vation against Cymatoderma (Donk in Bull. Bot. Card. Buitenzorg ser. iii, 18, 1941,

pp. 156 and 163). This proposal was defeated at Congress (Taxon 2, 1953, 31). The

type species of the genus is Cymatoderma elegans Jungh.

The little that is known of the genus is mainly summarised in Fries’ “ Fungi

Natalenses ” (1848) and in Lloyd’s “Synopsis of the Genus Cladoderris ” (Lloyd

Myc. Writ. 4, 1913).

Cymatoderma is evidently closely related to Stereum, especially through such

species as Stereum involutum (Klotzsch) Fr., but differs from Stereum in the possession

of a radially ribbed and usually papillate hymenium. With the exception of Cladoderris

funalis, the species of Cladoderris or Cymatoderma are rather uniform in microscopic

characters, having a dimitic hyphal system, hyaline spores, and usually cystidia and/or

gloeocystidia showing respectively little variation from species to species. But macro-

scopically the species are bewilderingly variable. The specific distinctions have always

been based almost entirely on the external appearance without adequate investigation

of the microscopic features and without sufficient regard for the natural variations and

intergradations that are possible in such characters as hairyness, colour and habit.

Owing to variation these characters are undoubtedly of minor importance and the

emphasis laid on them originally resulted in an unnecessary number of species being

proposed. Recognising this fact to some extent, Lloyd was able to accept only five

‘ good ’ species out of some twenty-five that had already been described. Lloyd noted

that the mode of insertion of the stipe was of no specific value, and stated that the

characters of value were: (1) the nature of the hymenial folds, whether broad and

obtuse or narrow and sharp, (2) the presence or absence of papillae (but then admitted

that these were not a constant feature), and (3) the tomentose upper surface, whether

densely or scantily clothed with hairs.

It is instructive to quote fairly extensively from Lloyd’s comments on some of

the species which he differentiated, which show clearly that the above characters are

actually of little taxonomic value owing to variation and intergradation, and secondly

that any attempt to key out species entirely on such characters is likely to fail: —

* Cladoderris dendritica: “Papillae usually none, but many specimens occur with

a few.” ... “I believe that . . . C. elegans is in reality only an excessively

warty form (of C. dendritica) for while the type forms are so different (apparently)

there are many connecting specimens in the museums.”

^Cladoderris elegans: “ As to colour, variation as to fonn and stipe characters,

and very often as to the even, thick tomentum pad on the pileus, elegans is similar

to dendritica, but elegans has the narrow folds of the hymenium densely covered

with papillae ”... “ And the hymenium folds, while narrow, are more the

nature of those of Cladoderris spongiosa, and specimens occur connecting it with

spongiosa, rather than with dendritica.'''’

^Cladoderris spongiosa: “ Nor is the distinction between it and Cladoderris elegans

strongly marked, for the nature of the (hymenial) folds is not an absolute character,

and many specimens occur that appear to be intermediate.”

^Cladoderris infundibuliformis: “All the preceding species might be broadly con-

sidered as forms of the same species, but this is widely different, in having a darker

colour, thin, ridged pileus, slight tomentum ”... “ Usually the tomentum is

but slightly developed, or almost none, but it is a varying character as shown in

specimens at Kew of the same collection.”

119

Table 1.

Microscopic organs of various species of Cymatoderma {Cladoderris): Measurements in /l

The taxonomy of this genus must undoubtedly be based primarily on microscopy

(see Table 1 ; cfr. Figs. 2-7). If this is done, C. dendritica may be separated from the

other species mentioned above by its distinctively smaller and rounder spores and by

its lack of cystidia, features mentioned also by Burt (in Ann. Mo. Bot. Card. 11, 1924, 3).

All the other species quoted above have spores, cystidia and gloeocystidia of essentially

the same size and shape respectively, and cannot be separated microscopically. The

precise configuration of the hymenium, as Lloyd showed, is variable and intergrades,

and in any case this character has frequently been demonstrated as unreliable in taxo-

nomy. The development of the surface hairs varies considerably in specimens of a

single collection, and hairs may easily rub off with age or be destroyed by insects. The

thinner pileus and darker colour of C. infiindibuliformis is mainly due to the fact that

the hairs are scanty. If the hairs be removed from a specimen of C. spongiosa, the

surface is seen to be dark in colour and composed of radiating, acute ridges not essen-

tially different from C. infundibidiformis. To the writer only one macroscopic character

so far appears to be of definite specific value, namely whether as in C. funalis P. Henn.

the pileus is deeply dissected into narrow radiating segments. But as will be seen later,

this species is probably better referred to the Clavariaceae under Clavulina or Aphelaria.

The species that have been studied may be grouped as follows: —

(A) Pileus deeply dissected into narrow radiating segments. Hyphae monomitic.

Cystidia and gloeocystidia absent. Basidia bisporous to quadrisporous.

Spores 4 -8-6 -4 x 6-4-8 /x. C. funalis.

Pileus entire to shortly incised or laciniate at the margin. Hyphae dimitic.

Cystidia and/or gloeocystidia present (B).

(B) Gloeocystidia present, cystidia absent. Spores ± subglobose, 3 X 4 /x or 3-4 p

diam. C. dendritica.

Both gloeocystidia and cystidia present. Spores ellipsoid, 3-5 X 6-10 /x.

The C. elegans Complex (C).

(C) Pileus surface densely covered by a thick, whitish, pad-like tomentum.

(a) Hymenial ribs narrow C. elegans.

(b) Hymenial ribs broad, obtuse C. spongiosa.

Pileus surface scantily tomentose, light red-brown or yellow-brown.

C. infundibuliformis.

Pileus surface almost devoid of hairs when mature, and dark red-brown to fuscous

in colour. C. australica.

120

This last group of four names (the C. elegans Complex) has been subdivided above

in accordance with the old concept of the species concerned, where extremes in macro-

scopic form were recognised. But there is little doubt that all these species are

variations of a single species, whose earliest epithet is elegans, and that the variations

intergrade so much that it would not be practicable to recognise varieties of C. elegans

based on those macroscopic features. I propose therefore that C. spongiosa, C.

infundibuliformis, and C. australica should be regarded as synonyms of C. elegans

Jungh.

Notes on South African Records of Cymatoderma (Cladoderris).*

The species of Cymatoderma and Cladoderris that have been recorded for South

Africa will now be reviewed. Accepted species are printed in bold type, synonyms or

dubious records in italics. I am much indebted to the Director of the Royal Botanic

Gardens, Kew, and to the Director of the South African Museum, Cape Town, for the

loan of some specimens. I wish also to record my thanks to Mr. D. A. Reid of Kew

Herbarium for helpful discussion of some of the problems involved.

1. Cladoderris australis Kalchbr. f. minima Bres.

There is a specimen in Kew labelled “ Cladoderris australis Kalchbr. f. minima.

Specimen authenticum ”, in Kalchbrenner’s writing. The word “ Cape ” and Bresa-

dola’s signature also appear on the label. Although I have not been able to trace

whether Bresadola ever published his f. minima, and it is probably only a herbarium

name, yet this specimen requires comment in view of the similarity of the epithets of

C. australis and C. australica and another species Cladoderris minima C. & Br.

The specimen is not a Cymatoderma but instead a good match with Stereum

thozetii Berk. According to Lloyd (Syn. Stip. Stereum 1913, 28) a specimen which he

assumes to be the original of C. australis Kalchbr. is Stereum elegans Mey.

2. Cymatoderma elegans Jungh. in Tijdschr. Nat. Gesch. Phys. ed v. d. Hoeven & De

Vriese 7 (1840) 390; Montagne in Ann. Sci. Nat. 7 (1847) 173.

Cladoderris elegans (Jungh.) Fries, Fungi Natalenses (1848) 22; Doidge in Bothalia

5 (1950) 480.

Cladoderris spongiosa Fries (!), Fungi Natalenses (1848) 20; Saccardo, Syll. Fung.

6 (1888) 548; Lloyd, Syn. Gen. Cladoderris (1913) 5, f. 526; van der Byl in Ann.

Univ. Stellenbosch 7 (1929) 51, f. 17; Doidge in Bothalia 5 (1950) 480.

Cladoderris spongiosa Fr. var subsessilis Fr., Fungi Natalenses (1848) 21; Doidge in

Bothalia 5 (1950) 480.

Actinostroma infundibuliforme Klotzsch, Fungi in Meyen, Beitrage zur Botanik,

gesammelt auf einer Reise um die Erde (1843) 237.

Cladoderris infundibuliformis (Kl.) Fries, Fungi Natalenses (1848) 21; Doidge in

Bothalia 5 (1950) 480.

Cladoderris australiea Berk, ex Saccardo, Syll Fung. 6 (1888) 548; Cooke, Handbk.

of Australian Fungi (1892) 181, f. 76; As C. australica Berk, in Herb., Cooke in

Grevillea 8 (1879) 70, nom nud.; ibid. 9 (1880) 14, nom. nud.; ibid. 11 (1882)

28, nom. nud.

Figs. 2, 4-6.

* Deur die goedgunstigheid van die Universiteit van Stellenbosch is ek toegelaat om sekere eksem-

plare in die Herbarium P. A. van der Byl te ondersoek. Verwysing na verskillende soorte in die Van

der Byl-Herbarium word hieronder gemaak.

121

Pileus flabellate to infundibuliform, substipitate or attached by a short central,

excentric or lateral stipe. Surface radially ridged, the ridges sharp, sublamellate,

varying from nearly glabrous and somewhat zonate to thickly covered with a dense

whitish, later discoloured, pad-like tomentum and azonate; the ridges yellow-brown

to light reddish-brown to dark red-brown (near Carob brown or Chestnut of Ridgway).

Margin entire or slightly incised. Hymenium creamy to light buff, radially costate,

the ribs varying from narrow and sharp to broad and obtuse, almost smooth or strongly

papillate or tuberculate.

Basidia : Clavate, 25-45 x 6 /i..

Spores; 3-2-5 X 6 -4-9 -6 fi, hyaline, smooth, elliptical, sometimes unilaterally

depressed.

Cystidia: smooth, rarely with a minutely encrusted apical portion, hyaline, very thick-

walled, fusiform, mammiform or pyriform, often with a small apical protuberance,

(6-6)-8-17 X 20-50 ju., terminal in position.

Gloeocystidia : smooth, hyaline, with homogeneous deeply-staining contents, fusoid,

pyriform or mammiform, or sometimes elongated and irregularly subcylindrical,

(6)-8-12-(15) X 25-60 /x.

Hyphae: dimitic, hyaline. Skeletal hyphae thick-walled, smooth, 2-5-6 /x diam.

Surface hairs: scanty to abundant, thick-walled, smooth, hyaline, with clamp connec-

tions, part of the generative hyphal system.

Specimens examined: As C. elegans: Herb. Hort. Bot. Bog. Java 2361 in Herb. Kew. ;

P. IV. Richards 2258, Sarawak, in Herb. Kew. (Java is the locus typicus of this species).

As C. spongiosa: Part of Type in Herb. Kew. marked by Fries “ Cladoderris spongiosa

Fung. Nat. fragmentum Cap. Nat.”; Farquharson 47, S. Nigeria, in Herb. Kew.;

Universiteit van Stellenbosch, Herbarium P. A. van der Byl nr. 319, 2085; National

Herbarium, Pretoria Nos. 36864, 33142, 15554, 31344, 9149, 39079. As C. infun-

dibuliformis: Diimmer 2108 and 3125, Uganda, in Herb. Kew.; Ex Papua, det E. M.

Wakefield, in Herb. Kew; W. Small, 471, Uganda; Maitland 1, 30 A, Uganda; Holst

2542, Usambara; National Herb., Pretoria No. 18045 and 13032 {Diimmer 2108),

Uganda; Universiteit van Stellenbosch, Herbarium P. A. van der Byl nr. 522. As

C. australica: J. M. Wood 239, Natal (In Herb. Kew. and in Herb. S.A. Museum No.

34303) neotype. As Thelephora dendritica: ex Gripps Land, Herb. Berkeley.

Under Cladoderris elegans, Doidge lists two collections by Drege from South

Africa. It has been ascertained that neither of these is represented in Montagne’s

Herbarium in the Paris Natural History Museum.

Cladoderris spongiosa Fr. var subsessilis Fr. was described as having an excentric

stipe, but, as Lloyd has shown, the characters of stipe insertion are not of taxonomic

value in this genus.

There is a specimen in Herb. Kew. labelled “ Thelephora cali.x Kze. ” in an unknown

handwriting, and “ Thelephora dendritica Er. Africa austr. leg. Ecklon & Zeyher ” by

Berkeley. It corresponds with Cladoderris spongiosa microscopically, and macro-

scopically except that its hymenium is obviously discoloured.

Cladoderris australica Berk, ex Sacc. is a species with a curious and complicated

history. Cooke first published this name in 1879, 1880, and 1882 as a nomen nudum f

attributing the epithet to “ Berk, in Herb.” In 1888 Saccardo validated the name with

a full description, citing it as “ Cladoderris australica Berk, in Herb, et in Cooke Fungi

Austral.” The formal citation of this species should therefore be Cladoderris australica

Berk, ex Sacc. Then in 1892 Cooke published a close translation of Saccardo’s descrip-

tion and added a coloured figure. He cited the species as “ Cladoderris australica

Berk, in Herb.”, and referred to Saccardo’s description. There is no concrete evidence

that Berkeley ever used C. australica as a herbarium name, but both Saccardo and

122

Cooke referred to a Berkeley specimen from New South Wales which, if it existed,

would be the holotype. The only Berkeley specimen in the type folder of C. australica

at Kew was labelled “ Thelephora dendritica ” by Berkeley, and came from Gipps

Land. Gipps Land is now part of Victoria, but I am informed by Mr. D. A. Reid

that prior to 1851 it formed part of New South Wales. The old boundaries would be

difficult to define and no exact locality was given for this specimen, so one could perhaps

accept this specimen as being the one from N.S.W. referred to by Saccardo and by

Cooke. Berkeley, on the other hand, did not use the name C. australica for this speci-

men, and so one must assume that he used it for a specimen that is now lost; conse-

quently a lectotype or neotype should be chosen. As there is no material in existence

seen and classified by Berkeley as C. australica a lectotype is out of the question and

a neotype should be nominated. Cooke’s earlier mention of the species was in connec-

tion with the Natal collection of J. M. Wood No. 239, and there is little doubt that

both Saccardo’s and Cooke’s descriptions, and the coloured figure, fit the Natal

specimens rather than Berkeley’s specimen from Gipps Land. Lloyd (Syn. Gen.

Cladoderris 1913, 10) considered that the latter specimen matched C. spongiosa Fr.,

and I support that opinion. Consequently I consider that if another Berkeley specimen,

now lost, had existed, it must have had the appearance of J. M. Wood 239, and I nominate

the latter as the neotype of Cladoderris australica Berk, ex Sacc. I regard both C.

spongiosa and C. australica as synonyms of Cymatoderma elegans Jungh. That a

neotype for C. australica should be nominated, really only becomes a practical necessity

if this synonymy should later be disputed.

It may be mentioned that the specimen of J. M. Wood No. 239 in the National

Herbarium, labelled Cladoderris thwaitesii, is instead a species of Favolus, but that the

part of this collection in Herb. S.A. Museum 34303 matches Cladoderris australica

and the neotype in Kew.

3. Cladoderris funalis P. Henn. in Engler Jahrb. 38 (1905) 120; Saccardo Syll. Fung.

21 (1912) 385; van der Byl in Ann. Univ. Stellenbosch 7 (1929) 52, f. 18; Doidge

in Bothalia 5 (1950) 480.

Fig. 7.

Fructifications laterally stipitate, anastomosed and with somewhat fused stipes,

more or less flabellate, 3 cm diam. and 1 • 5 cm high. Stipe 1 ■ 5-2 cm X 2 mm, com-

pressed, rugulosely striate. Upper surface sterile, yellow-brown to pale reddish-brown,

covered with tangled, thick, concolorous fibrils composed of fascicles of hyphae.

Margin fimbriate to deeply dissected. Hymenium yellowish to brownish, covering

broad, obtuse, radiating, branched ribs, not papillate, decurrent on the stipes.

Basidia; cylindric-clavate, 35-50 X 6-8 /x, with two to four sterigmata up to 4 /x long.

Spores: hyaline, smooth, ovate to ellipsoid, finely apiculate, 4 -8-6 -4 X 6-4-8 p.

Hyphae: thin-walled, hyaline, without clamps, septate, readily collapsing, in parts

inflated, 3-9 /x wide. H3rphae composing the surface fibrils are similar, usually unin-

flated, 3-5 /X wide, adherent in fascicles.

Specimens examined: Universiteit van Stellenbosch, Herbarium P. A. van der Byl

nr. 2465 (Eyles 5026), Salisbury, S. Rhod.

This collection agrees very well with Hennings’ description and Lloyd’s photograph

(Syn. Gen. Clad., 1913, p. 10, Fig. 530) of the type, and appears to be correctly named

as to species.

The dissection of the pileus at the margin is partly due to the running out of the

hymenial ribs, and partly to their being continued into some of the surface fibrils.

The fibrils are thick, sterile, and resemble hydnoid spines. The hymenial ribs are

broad, obtuse and not papillate. The spores are very abundant. No cystidia or other

accessory organs are present. There is a monomitic hyphal system.

123

Writing of Cladoderris funalis, Lloyd said, “ It is so different from all other species

that it is a question if Hennings was correct in referring it to Cladoderris." Lloyd

suggested an affinity with Lachnocladium, but that can be ruled out since C. funalis

has no dichophytic hyphae.

Certainly C. funalis is far removed from other Cladoderris species by its monomitic

hyphae which become inflated, by its bisporous to quadrisporous basidia, and by the

lack of cystidia and gloeocystidia. It seems certain that the ribbed hymenium is

composed basically of flattened clavarioid branches which have fused into an unusual

dorsiventral form. Except that the basidia are not all bisporous they and the other

microscopic characters agree welt with those of Clavulina as delimited by Corner

(Monogr. of Clavaria and Allied Genera, 1950), and there are species in this genus

which show flattened branching. Flattened branching is however more characteristic

of the genus Aphelaria Corner in which the basidia are bisporous to quadrisporous;

but the hyphae in Aphelaria are uninflated and tend to have thickened walls, which is

not the case in Cladoderris funalis. Mr. D. A. Reid informs me that the type of C.

funalis has basidia with four spores. Cladoderris funalis appears to have characters

somewhat intermediate between those of Clavulina and Aphelaria but I am not in a

position to decide to which of these genera it should be transferred.

4. Cladoderris thwaitesii Berk. & Broome; recorded by Kalchbrenner in Grevillea

10 (1881) 58; Saccardo Syll. Fung. 6 (1888) 550; Doidge in Bothalia 5 (1950)

480.

The collection which Kalchbrenner assigned to this species is J. M. Wood No. 239.

This collection has been nominated above as the neotype of Cladoderris australica

Berk, ex Sacc., which has here been reduced to synonymy under Cymatoderma elegans

Jungh.

According to Fetch (Ann. Roy. Bot. Gard. Perad. 9, 1924, 134) the true C.

thwaitesii is a bleeding species of Stereum.

THELEPHORA Ehrhart ex Fries.

Many of the Thelephoraceae were originally described as species of Thelephora

and have since been combined under other genera. The object of this section is to

annotate all those species which have at some time been recorded for South Africa

under the genus Thelephora.

1. Thelephora (leijostroma) acerina (Pers.) Pers. ex Fr., Syst. Myc. 1 (1821) 453; Persoon,

Syn. Fung. (1801) 581, Myc. Fur. 1 (1822) 152; Leveille in Ann. Sci. Nat. ser

iii, 5 (1846) 150.

Stereum acerinum (Pers. ex Fr.) Fr., Epicrisis (1838) 554; Saccardo Syll. Fung. 6 (1888)

587.

As no material supporting Leveille’s and Saccardo’s records is available, it is

not known whether this species is represented in South Africa. The species is refer-

able to the genus Aleurodiscus, and a description and comment on it as a South African

record is given by the writer (Bothalia 6, 1956, 466).

2. Thelephora biennis Fries, Syst. Myc. 1 (1821) 449; recorded by Kalchbrenner in

Grevillea 10 (1881) 58.

This record is based on Kalchbrenner’s determination of MacOwan (1244), collec-

tions of which have been examined at Kew Herbarium and in Herb. S.A. Museum

(sub. Stereum fuscum. No. 34292). These specimens are Stereum bicolor (Pers. ex

Fr.) Fr. Cfr. Bothalia 6 (1954) 308, f. 21.

124

3. Thelephora {Stereum) fulva Leveille (!) in Ann. Sci. Nat. ser. iii, 5 (1846) 149.

The type of this species, Drege (9441), Cap-de-Bonne-Esperance in Herb. Mus.

Paris, was examined and is annotated under Stereum Julvum (Lev.) Sacc: See Bothalia

6 (1954) 315, f. 20.

4. Thelephora fuscoviolascens Mont, in Ann. Sci. Nat. ser. iii, 7 (1847) 174.

Collected by Drege (9429) at Port Natal, and referable to Hymenochaete fusco-

violascens (Mont.) Sacc., according to van der Byl (in Ann. Univ. Stellenbosch 7, 1929,

14). Saccardo (Syll. Fung. 6, 1888, 546) however, left the species under Thelephora,

only suggesting in a footnote that it might be a Hymenochaete. The correct citation

is thus Hymenochaete fuscoviolascens (Mont.) v. d. Byl as indicated by Doidge, loc.

cit. p. 484. Drege’s specimen has not been available for personal study.

5. Thelephora hirsuta (Willd.) Pers. ex Fr. ; Recorded for South Africa by Berkeley

in Hooker Fond. Journ. Bot. 11 (1843) 516.

= Stereum hirsutum (Willd.) Pers. ex S. F. Gray; annotated in Bothalia 6 (1954)

316, f. 11.

6. Thelephora intybacea Pers. ex Fr. ; Recorded by van der Byl in Trans. Roy. Soc.

S. Afr. 10 (1922) 285 and in Ann. Univ. Stellenbosch 7 (1929) 32; Doidge in

Bothalia 5 (1950) 495.

The writer has examined van der Byl’s specimens determined as this species in

Universiteit van Stellenbosch, Herbarium P. A. van der Byl Nos. 214, 922, 1261, 1263.

These are apparently indistinguishable from numerous South African collections named

Thelephora terrestris in the National Herbarium, and from Herb. S.A. Mus. No. 34298,

originally named Thelephora laciniata but annotated by v. d. Byl as T. intybacea.

The literature on T. intybacea does not clearly indicate how it differs from T.

terrestris and moreover there is the complication that T. intybacea Pers. ex Fr. and T.

intybacea Fr. are apparently not the same species. This confusion can obviously

not be resolved outside Europe, but the probability is strong that T. intybacea does

not occur in South Africa and that specimens recorded as such are actually the far

commoner species T. terrestris. That opinion is adhered to here, and van der Byl’s

specimens are taken to be T. terrestris.

1. Thelephora laciniata Pers. ex Fr. ; Recorded for South Africa by MacOwan in Cape

Agric. Journ. 8 (1895) 331; Davidson in Natal Agric. Journ. 12 (1909) 617;

Doidge loc. cit. p. 495.

Thelephora laciniata is generally acknowledged to be a synonym of T. terrestris

Ehrh. ex Fr. Davidson’s specimen does not exist, but I have seen MacOwan’s (No.

1445, Herb. S.A. Mus. 34298) which is in no way different from T. terrestris.

8. Thelephora palmata (Scop.) Fr., Syst. Myc. 1 (1821) 432; Wood in Rept. Natal

Bot. Gard. & Colonial Herbarium for 1898, p. 19; van der Byl in Ann. Univ.

Stellenbosch 7 (1929) 31; Doidge in Bothalia 5 (1950) 495.

This record is based on a single collection, in Herb. Kew., labelled “ J. M. Wood

(3497), Thelephora palmata, on ground in N.B. Gardens, Febr. 1885.”. The pilei are

terrestrial, erect, flabelliform, and split longitudinally for about half their length into

a number of flattened, narrow segments. The undivided base and the microscopic

structure show that these segments are merely formed by splits in the fibrous pileus and

are not clavarioid branches arising from a stem as is the case in T. palmata. Spores

and basidia were absent, the only clear microscopic feature being the hyphae, which

were pale coloured, lacking clamps, 2 -5-3 -2 p diam., with thin or slightly thickened

125

walls. Both surfaces of the pile! were heavily covered with conidiophores and conidia

of an Aspergillus. It is doubtful whether this collection is in good enough condition

ever to be determined, but certainly it is not Thelephora palmata.

9. Thelephora pedicellata Schw. ; Recorded by Wood in Rept. Natal Bot. Gard. &

Colonial Herbarium for 1898, p. 19; Bottomley in S.A. Journ. Sci. 13 (1917) 440.

J. M. Wood recorded this fungus as “ Thelephora pedicellata Schuz. (sic), on bark.

No. 532.” This collection was examined in Herb. Kew. and proved to be a species

of Septobasidium, but could not definitely be referred to S. schweinitzii Burt (in Ann.

Mo. Bot. Gard. 3, 1916, 324; Couch, The Genus Septobasidium, 1938, 112) with

which Thelephora pedicellata Schw. is synonymous.

10. Thelephora penicillata Lloyd (non Fries) in Myc. Notes 6 (1920) 989, Figs. 1759,

1760; van der Byl in Ann. Univ. Stellenbosch 7 (1929) 32, f. 12; Stevenson &

Cash in Bull. Lloyd Library 35 (1936) 60; Doidge in Bothalia 5 (1950) 495.

Fig. 8.

Fructifications beginning as resupinate, fertile, membranous, whitish growths

encrusting soil and debris, finely pubescent under the lens, smooth, following the

conformation of the substratum, later becoming greyish then fuscous-purplish and

finally drying fawn or chocolate brown. Flabellate fascicles of subulate radiating

branches are emitted usually near the margin; they are oblique or suberect and become

fertile with age at the base, white at first, then yellow-brown to fawn or chocolate brown

with the apex remaining whitish and infertile. These branches are occasionally entire

but are usually dissected into a number of penicillate or ciliate parts, and are up to

0-5-0 -7 cm long. When growing on debris the fructification may become subpileate,

but it is resupinate on soil. Microscopically the resupinate and pileate parts are the

same.

Basidia: (30)-60 x 8-5-11 /u,, proliferating by basal clamps, with four sterigmata up

to 7-2 X 1-6 /X.

Spores : forming a chocolate coloured spore print, yellow-brown under the microscope,

finely warted, irregularly angled, elliptic or subglobose, 4 -8-7 -2 X 6 -4-9 -6 /x.

Hyphae: at first hyaline, some later becoming dark brown, branched, closely inter-

twined, septate, with abundant clamps, 3 - 5-7 p. diam.

Specimens examined: 31417, J. D. Krige, Stellenbosch, June, 1919, (presumed Type

collection); 31500, A. V. Duthie, Stellenbosch, July, 1923; 40512, P. H. B. Talbot,

on soil and mosses, Fountains, Pretoria, 1951 ; 40712, P. H. B. Talbot, on soil and pine

needles. Union Buildings Grounds, Pretoria, 1954.

This rather rare species inhabits very moist situations, especially favouring debris

under pine trees but not confined to this habitat.

Lloyd (l.c.) named the earlier specimens from Stellenbosch Thelephora penicillata,

about which Stevenson & Cash commented: “ Seems to have been intended by Lloyd

as a new species though his description and comparison with T. spiculosa Fr. might

indicate that he had reference to T. penicillata Fr., Syst. Myc. 1, 434, 1821. In a note

with a specimen of another species Lloyd refers to T. penicillata ‘ which I had named

from South Africa.’ ”

Van der Byl and Doidge both attributed the specific epithet to Lloyd. If this is

correct then it is a later homonym of T. penicillata Fr., and Lloyd’s name has no nomen-

clatural standing. Lloyd regarded the South African specimens as differing from

T. spiculosa Fr. in having the hymenium only on the resupinate portion and not on the

erect subulate parts. However, further specimens show quite clearly that at least the

basal parts of the penicillia become fertile with age.

126

From the literature alone there is very great difficulty in differentiating between

Thelephora mollissima Pers. ex. Fr., T. spicutosa and T. peniciUata, for these names

have been used in different senses by different authors. However, it seems fairly

certain that our material is referable to the species described by Burt (Ann. Mo. Bot.

Card. 1, 1914, 225, PI. 4, f. 2), after reference to an authentic specimen, as T. spiculosa

Fr., and by Bourdot & Maire (in Bull. Soc. Myc. de Fr. 36, 1920, 29) as T. spiculosa

Fr. forme B. mollissima. Used in this sense, T. spiculosa Fr. is regarded by several

authors (Bourdot & Galzin, Hym. de Fr. 1928, 467; Bourdot & Maire, loc. cit.; Rea,

Brit. Basid., 1922, 654; Lundell & Nannfeldt, Fung. Exsicc. Suecici No. 77, 1934)

as synonymous with Thelephora mollissima Pers. ex Fr., and this is the name which I

think should be adopted for T. peniciUata Lloyd. Further work is necessary before

this can be confirmed.

Dr. John Eriksson of Varnamo, Sweden, has kindly examined these specimens

and has concluded that they are Thelephora mollissima Pers. ex Fr.

11. Thelephora (Stereum) pulverulenta Lev. (!) in Ann. Sci. Nat. ser. iii, 5 (1846) 149;

Doidge in Bothalia 5 (1950) 491; Talbot in Bothalia 6 (1954) 323.

Corticium {Coniophora) pulverulentum (Lev.) Cooke in Grevillea 8 (1880) 89.

Coniophora pulverulenta (Lev.) Massee in Journ. Linn. Soc. Bot. 25 (1889) 129; Saccardo,

Syll. Fung. 6 (1888) 649; Doidge in Bothalia 5 (1950) 480.

The type of this species, Drege 9442, as noted previously (Talbot, loc. cit.) is a

Hymenochaete, probably H. luteobadia (Fr.) Hohnel & Litsch.

12. Thelephora punicea Alb. & Schw. ; Recorded for South Africa by Wood in Rept.

Natal Bot. Gard. & Colonial Herbarium for 1898, p. 19; Kalchbrenner in Grevillea

10 (1881) 58.

J. M. Wood recorded this fungus as “ Thelephora punicea Alb. & Schw., on bark.

No. 190 ”. Van der Byl (in Ann. Univ. Stellenbosch 7, 1929, 19) quotes Wood’s

record and gives a description, taken from Rea’s “ British Basidiomycetes ”, under

Hypochnus puniceus (A. & S.) Sacc. Bourdot & Galzin (Hym. de Fr., 1928, 769)

describe this species from Europe under the genus Tomentella as T. punicea (A. & S.)

Schroet. Wood’s collection is no longer in existence and so the record cannot be

confirmed.

13. Thelephora sinuans Pers. ; Recorded by Leveille in Ann. Sci. Nat. ser. iii, 5 (1846) 146.

The record refers to a collection by Drege (944), Caput Bonae Spei, but no material

is available from Herb. Mus. Paris and so the record has not been confirmed. Fide

Lentz (U.S.D.A. Agric. Monogr. No. 24, 1955, 35) T. sinuans is a synonym of Stereum

frustulatum (Pers. ex Fr.) Fuckel.

14. Thelephora terrestris Ehrh. ex Fr., Syst. Myc. 1 (1821) 431 ; Ehrhart, Crypt. Exsicc.

No. 178 (1785); Doidge in Bothalia 5 (1950) 495.

Thelephora laciniata Pers. ex Fr., Syst. Myc. 1 (1821) 431; Persoon, Syn. Fung. (1801)

567.

Fig 9.

Fructifications effuso-reflexed and encrusting, more usually dimidiate and sessile,

imbricate or confluent, occasionally shortly stipitate. Surface fawn to chocolate or

127

dark fuscous colour, usually strigose. Hymenium smooth to finely radiately rugose

and papillate, concolorous. Texture coriaceous, soft, thin. Margin whitish then

concolorous, entire to shortly laciniate and strigose-fibrous.

Basidia: not seen (fide Bourdot and Galzin: 40-90 X 9-12 fx).

Spores : coloured, fuscous, finely and sparsely verrucose, subglobose to broad ellipsoid

and irregularly angled, (7)-9-ll x 7-8 ju,.

Hyphae: coloured brownish, thin-walled, some inflating and then collapsing, with

occasional clamps, not encrusted, 3-5-5-8-(ll) /x diam.

Specimens examined: As T. terrestris: Nos. 27560, 30616, 40206, 40670, 40657, 13020,

19852, 27329, 27330, 1067, 10049.

As T. intvbacea: No. 25863; Universiteit van Stellenbosch, Herbarium P. A. van der

Byl Nos.'l263, 214, 1261, 922.

As T. laciniota: Nos. 11107, 34406, 30712; Herb. S.A. Mus. No. 34298 {MacOwau

1445), Table Mountain.

EXPLANATION OF THE ILLUSTRATIONS.

The following lettering has been used throughout the illustrations: —

128

Fig. 1. — Craterellus cornucopioides (J. M. Wood No. 4108). Fig. 2. — Cladoderris

spongiosa. Fig. 3. — Cladoderris dendritica. Fig. 4. — Cladoderris australica

(J. M. Wood No. 239), neotype. Fig. 5. — Cladoderris elegans. Fig. 6. — Clado-

derris infundibuliformis. Fig. 7. — Cladoderris funalis (Eyles No. 5026).

129

Fig. 8. — Thelephora penicillata Lloyd. Fig. 9. — Thelephora terrestris.

j

i

131

Studies of some South African Resupinate

Hymenomycetes.

Part II.*

By

P. H. B. Talbot.

The following descriptions and notes are based on material seen since the first

paper in this series was published. I am indebted to the Transvaal and Orange Free

State Chamber of Mines Timber Research Laboratory for submitting several interesting

collections and particularly for allowing Miss V. C. Green to collaborate in the prepara-

tion of descriptions and illustrations of three of the species discussed here. My sincere

thanks are also due to the Directors of the following institutions for the loan of

specimens: Herbarium, Royal Botanic Gardens, Kew; Herbarium, British Museum

(Natural History); Museum National D’Histoire Naturelle, Paris; U.S. National

Fungus Collections, Beltsville; Herbarium, S.A. Museum, Cape Town; Herbarium,

Dept, of Agriculture, Southern Rhodesia; Herbarium Len Verwoerd, Stellenbosch-

Elsenburg Agricultural College.!

PLATYGLOEA Schroet.

1. Platygloea opalina sp. nov.

Fig. 1.

Resupinate, effused, gelatinous, forming a thin opalescent pellicle, drying to an

extremely thin, shiny, light grey, subpruinose film, occasionally cracked across. Thick-

ness in section 130-146 p.

Basidia: probasidia arising as lateral branches of hyphae, clavate-cylindrical almost

from the beginning, elongating and changing into a cylindrical metabasidium divided

by transverse septa into 3-4 metabasidial cells each with a sterigma. Metabasidia

6-4-9 X 60-80 p.

Spores: hyaline, smooth, elliptical with one side flattened, or cylindric-depressed, with

a lateral apiculus and often with one or more conspicuous guttules, (7-2)-8-9-6 x

(14-4)-16-18'4 p, germinating to form secondary spores on a lateral germ tube.

Hyphae: 2 -4-5 -6 p wide, branched, with numerous septa, lacking clamp connections,

thin-walled, erect, possessing large guttules in the wider hyphae, loosely intertexed.

Specimens examined: Type, 36951, Talbot, on Acacia moUissima, Atholl Expt. Stn.,

1949.

This species does not correspond with any previously described species that I

have been able to trace, including those in Bandoni’s survey of the genus (in Mycologia

48, 1956, 821-840).

* Part I of these studies appeared in Bothalia 6 (1951) 1-116.

t Vir die geleentheid om eksemplare in die Herbarium P. A. van der By! te besliideer en vergelyk

is ek dank verskiddig aan die Universiteit van Stellenbosch.

132

Platygloea opalina sp. nov.

Fungus resupinatus, effusus, gelatinosus, opalinus, tenuissimus, nitidus, pallido-

griseus, 130-146 /x crassus, ut siccus subpruinosus vel membranaceus. Probasidia

ab initio cylindraceo-clavata; metabasidia cylindracea, transverso-septata, cellulis

3-4, 6-4-9 X 60-80 p. Sporae hyalinae, leves, ellipticae, uno latere compressae,

laterale apiculatae, (7-2)-8-9-6 X (14-4)-16-18-4 p. Hyphae 2 -4-5 -6 /x diam.,

ramosae, septatae, non nodoso-septatae, tenue tunicatae, erectae, laxe intertextae.

Typus No. 36951, leg. P. H. B. Talbot, in ramis Acaciae moUissimae.

EICHLERIELLA Bresadola.

1. Eichleriella macrospora (Ell. & Everh.) Martin in Univ. Iowa Stud. Nat. Hist. 18

(1944) 48, Ibid. 29 (1952) 65, Figs. 14, 36.

Corticiwn macrosporum Ell. & Everh. in Bull. Torrey Bot. Club. 27 (1900) 49.

Sebacina macrospora (Ell. & Everh.) Burt in Ann. Mo. Bot. Card. 2 (1915) 759.

Fig. 2.

Resupinate, composed at first of small, very thin, suborbicular, pruinose to byssoid

growths of a light brown colour with white, cottony margin, becoming confluent to

form a coriaceous-membranous patch, 6 X 1 cm, with a narrow, determinate, white

margin, adherent in the present specimen. Hymenium light brown, to greyish-white

or pale dirty grey-brown, pruinose under the lens, beset with small, blunt tubercles

which are fertile. Thickness in section 130-230 p. Adnate, becoming cracked with

deep fissures down to the substratum, revealing a white, cottony trama.

Basidia ; tremelloid, the metabasidia clavate to elongated obovate, cruciately longitudi-

nally divided, 9-11-5 x 21-23 p, sometimes showing a basal clamp connection.

Sterigmata four, 3 -5-4 -5 x 11-5-21 p.

Spores: hyaline, smooth, oblong or broad-cylindric, sometimes very slightly depressed

on one side, with a lateral apiculus, 0-1-2 guttulate, 5-7-(10) x 11-4-14-8 p.

Paraphyses: (1) surrounding the basidia and projecting beyond them, 2-3-4- 5 x 22-45

p, flexuous, tortuous, hyaline; (2) more clavate and not tortuous, arising from a clamp

at the base of a basidium branch.

Hyphae and Tissue Differentiation: Some hyphae appearing to be next to the substratum

are colourless, much branched, 2 -8-3 -4 p wide, (thin) to thick-walled; also present

are plentiful brownish concretions which take phloxine stain readily and are apparently

mixed with gelatinised hyphae in an intermediate layer.

Specimens examined: 39090, Talbot, On Acacia sp., Buffelspoort.

Except for the margin, which in this specimen is not free or reflexed, there is good

agreement with Martin’s description of E. macrospora. From McGuire’s monograph

of Sebacina (in Lloydia 4, 1941, 23) it appears that Sebacina calcea (Pers.) Bres. is close

in characters to the present specimen, but may be distinguished by its longer spores,

absence of tubercles, and paraphyses loaded with granules.

HETEROCHAETE Patouillard.

1. Heterochaete byliana sp. nov.

Fig. 5.

Resupinate, obicular then widely effused, closely adnate, crustose-farinaceous.

Hymenium whitish to pale straw-coloured or slightly greyed, somewhat pruinose,

bearing scattered, sterile white (occasionally pale tinted) hyphal pegs which are not

perfectly cylindrical and are sometimes forked, arising near the surface of the hymenium

133

and projecting 90-160 ii, measuring 32-48 ix. wide, composed of practically hyaline

hyphae 1 • 8-3 • 2 /x in diameter, somewhat dendroid, without appreciably thickened

walls. Margin indeterminate, pruinose, concolorous, without a conspicuous bordering

zone free from pegs. Thickness in section about 65 /x, excluding the pegs.

Basidia: probasidia pyriform to ovate, 12-8-19 x 7-7-11 /x; metabasidia cruciately

longitudinally divided; sterigmata two, of variable length (up to 18 /x seen) and 3 -2-3 -7

/X wide.

Basidiospores : broad elliptic-depressed to suballantoid, hyaline, smooth, occasionally

apiculate, 11-4-17-4 x 5-8-3 /x.

Hyphae: hyaline, subgelatinous, difficult to distinguish except at the base where they

are light brown with thin, rigid walls, 1 -8-2-3 /x diam. ; all hyphae lacking clamps.

Tissue differentiation: white, subgelatinous throughout except for the darker basal

seam of more or less horizontal hyphae sometimes distinguishable.

Conidial stage: seen in part of a specimen, consisting of white, pruinose tufts with

globose, hyaline conidia 2 - 3-3 - 6 jjl diam. ; conidial hyphae hyaline, thin-walled, 1 - 4-3 - 2

/X diam.

Chlamydospores terminal or intercalary, sometimes in short chains.

Specimens examined: Type, 41049, Talbot, Umtentweni, Natal, 1955; As Heterochaete

andina Pat. & Lagerh. : Universiteit van Stellenbosch, Herbarium P. A. van der Byl

No. 629, No. 698, on Euphorbia pulcherrima. No. 785, on Plectronia, all from Durban.

Van der Byl (in Ann. Univ. Stellenbosch, 1, 1923, 5), classed his specimens as

H. andina, but thanks to Bodman’s account of this species (in Mycologia 41, 1949,

529, f. 2) it is possible to differentiate it. H. andina has a pinkish tint to the fructifi-

cation and possesses gloeocystidia and cystidia, but otherwise appears to be close to

the new species H. byliana. The latter differs from Heterochaete grandispora, which

was collected in the same locality, in its appreciably smaller spores and lack of gloeocys-

tidia, and in being totally resupinate.

Heterochaete byliana sp. nov.

Fungus resupinatus, orbiculatus deinde late effusus, adnatus, crustaceo-farinaceus.

Hymenium albidum vel pallido-stramineum vel griseum, pruinosum, fasciculos

hypharum ferens albidos, pallidos, 32-48 p diam., 90-160 p emergentes, ex hyphis

hyalinis, dendroideis, 1-8-3 -2 p diam. formatos. Margo indeterminata, concolor.

Probasidia pyriformia vel ovata, 12-8-19 X 7-7-11 p. Metabasidia sterigmatibus

2 usque ad 18 /x longa et 3 -2-3 -7 p diam. Basidiosporae ellipticae vel suballantoideae,

hyalinae, leves, 11-4-17-4 X 5-8-3 p. Hyphae hyalinae, subgelatinosae ; hyphae

basales pallido-brunneae, 1-8-2 -3 p diam., tenue tunicatae, non nodoso-septatae.

Conidia hyalina, globosa, 2- 3-3 -6 p diam. Chlamydosporae adsunt. Typus No.

41049, leg. P. H. B. Talbot.

2. Heterochaete grandispora sp. nov.

Fig. 4.

Resupinate or with reflexed margin, at first orbicular then confluent up to 1-5 cm

diam., coriaceous. Hymenium white to pale straw colour, beset with abundant light

brown, sterile, hyphal pegs originating in the trama and projecting 187-295 p, 40-65 p

wide, composed of yellow-brown, parallel, interwoven hyphae 3 - 2 ^x wide, with thickened

walls. Margin concolorous, not adnate, determinate, often very slightly reflexed, free

of hyphal pegs up to 1 mm from the edge. Context pale brown below a colourless

hymenial layer. Thickness in section 280-380 p.

134

Basidia: probasidia ovate-pyriform, becoming longitudinally cruciately divided,

18-20 X 10-11 -8 !i, embedded below the surface; metabasidia of the same dimensions;

sterigmata of variable length, 2-7 /x wide.

Spores: cylindric-curved to allantoid, smooth, hyaline, (15)-23-30 X (5-9)-7-3-9 /x,

with a prominent truncate apiculus, germinating by repetition.

Gloeocystidia : in the hymenium, hyaline with a homogeneous content, not readily

seen, 34-63 X 4 • 5-8 • 2 ju.

Paraphyses ; the hymenial hyphae rarely run out into rather simple hyaline dendrophyses,

forked shortly once or twice, 1 -4-2 -7 /x diam.

Hyphae: lightly coloured, densely intertexed, rarely branching or septate, thick-walled,

1-5-3 -2 /X diam., with undulating walls, lacking clamps. Hymenial hyphae similar

but colourless.

Tissue differentiation: There is a colourless, subgelatinous hymenial layer above a

light brown context.

Specimens examined: Type, 41050, Talbot, Umtentweni, Natal, 1955.

Martin [in Univ. Iowa Stud. Nat. Hist. 19 (1952) 62] estimates that about thirty

species of Heterochaete have been described. Descriptions of most of these have been

traced without finding any species approaching H. grandispora in spore size. Its

margin, which is free of hyphal pegs and tends to be reflexed, and the widely spaced

hyphal pegs, are characteristic.

Heterochaete grandispora sp. nov.

Fungus resupinatus vel margine reflexa, orbiculatus deinde confluens, coriaceus.

Hymenium albidum vel pallido-stramineum, fasciculos hypharum ferens pallido-

brunneos, 40-65 /x diam., 187-295 p emergentes, ex hyphis parallelis, luteo-brunneis,

3-2 p diam., formatos. Margo concolor, determinata, sine fasciculis hypharum.

Probasidia ovato-pyriformia ; metabasidia 18-20 X 10-11-8 p, sterigmatibus 1-1 p

diam. Sporae cylindraceo-curvatae, leves, hyalinae, apiculo truncato, (15)-23-30 X

(5 - 9)-7 • 3-9 p. Gloeocystidia hyalina, 34-63 x 4 - 5-8 - 2 p. Paraphyses 1 - 4-2 - 7 p

diam., hyalini, simplices, dendrophytici. Hyphae pallidae, dense intertextae, crasse

tunicatae, 1-5-3 -2 p diam., undulatae, non nodoso-septatae. Typus No. 41050, leg.

P. H. B. Talbot.

PELLICULARIA Cooke sensu Rogers.

M. A. Donk (in Reinwardtia 2, 1954, pp. 425-434; ibid. 3, 1956, p. 369) argues

with great justification that Pellicularia koleroga Cooke, the type species of the genus

Pellicularia, is a nomen confusum, and that accordingly the generic name to be used for

this group of species is Botryobasidium Donk. In the latter paper Donk also proposed

two generic segregates, Uthatobasidium and Thanatephorus. Several “ Pellicularia ”

species have already been combined under Botryobasidium, but there are still some for

which a new combination would be necessary if the genus Pellicularia is to be dropped.

It is to be regretted that Donk did not formally propose such combinations in his

recent papers cited above. Without studying a much greater range of material than

was available to me, I do not wish to propose the new combinations that might be

necessary, and for the sake of uniformity have treated all these species under Pellicularia.

One new combination that Donk did make (in Reinwardtia 3, 1956, 376) may

be noted: This is Thanatephorus cucumeris (Frank) Donk [— Hypochnus cucumeris

Frank; Hypochnus solani Prill. & Delacr. ; Pellicularia filamentosa (Pat.) Rogers].

135

1. Pellicularia vaga (B. & C.) Rogers ex Linder in Lloydia 5 (1942) 170; Rogers in

Farlowia 1 (1943) 110, f. 9.

Corticium vagwn Berk. & Curt, in Grevillea 1 (1873) 179. Further references and

synonymy are given by Rogers (1943, l.c.).

Fig. 6.

Resupinate, thin, discontinuous, arachnoid later becoming hypochnoid, dirty

whitish to pale tawny-yellowish.

Basidia: collapsing rather readily, 6-4 x 17-6 fx, cylindrical or somewhat inflated at

the base or apex [fide Rogers “ 13-22-(27) x 6-5-10-(15) //,, bearing rarely 4 or 5,

mostly 6-8 stout, divergent recurved sterigmata (3)-4-5-6 X 1-5-2 p”].

Spores: hyaline, smooth, navicular or asymmetrically fusiform, obliquely tapered.

4-5-6 X 9-6-12-8 p [fide Rogers “7-5-12-(17) x (2-5)-3-5-5-(5-5) /x ”]

Hyphae: loosely intertexed, smooth, lacking clamps, branching at right angles, with

thin, rigid walls and a wide lumen, the basal hyphae slightly yellowish and up to 13 /x

wide with long internodes, the superior hyphae narrower, colourless, short-celled.

Specimens examined: 41048, Talbot, on dead wood in contact with the soil, Umtentweni,

Natal, 1955.

Corticium vagwn B. & C. has been recorded previously for South Africa by Doidge

(in Bothalia 5, 1950, 483) who cites Corticium solani (Prill. & Delacr.) Bourd & Galz.

as a synonym. These records require confirmation by reference to the cited specimens

since the name C. vagum has been used in conflicting senses. Rogers (1943, loc. cit.)

identifies the type material of C. vagum with Pellicularia vaga (B. & C.) Rogers ex

Linder, and points out that C. vagum sensu Burt (in Ann. Mo. Bot. Gard. 13, 1926, 295)

is synonymous with Pellicularia filamentosa (Pat.) Rogers, and that Corticium solani

is another synonym of this parasitic species.

The collection described above was saprophytic and corresponds well with Pellicu-

laria vaga, whose distinguishing features, according to Rogers are its saprophytic habit,

its smooth navicular spores, its lack of clamp connections and cystidia, and its wide

hyphae. Pellicularia filamentosa, on the other hand, is parasitic, and its spores are

ellipsoid or oblong-ellipsoid, flattened on the inside, with an abruptly truncate apiculus.

2. Pellicularia fodinarum Talbot & Green, sp. nov.

Fig. 7.

Resupinate, thin, easily separable, pruinose-pellicular, with a similar margin.

Hymenium yellowish, discontinuous, beset with long, emergent, hairlike septocystidia.

Basidia: in botryose clusters, very easily collapsing, cylindric-clavate, 4-6 X 15-22 p;

adventitious groups of basidia arise as lateral offshoots of some of the septocystidia.

Spores: hyaline, smooth, subglobose to broad ovate-elliptical, 5-7-5 X 4-5-6 ju.

Cystidia: very long, cylindrical, hyaline, septate, lacking clamps at the septa, with

rounded apex, encrusted their full length with small wartlike granules (probably derived

from a mucilaginous investment), up to 360 p. long and 6-9 fx wide.

Hyphae: loosely intertexed, collapsed in the hymenium otherwise distinct, much

branched, anastomosed, septate, lacking clamps, thin-walled, encrusted with small

wartlike granules, 4-6 fx wide.

Minerals: abundant throughout the tissues.

Specimens examined: Type, 40679 (T.R.L. 2635), A. L. James, on underground timber.

Brakpan No. 1 Shaft, Dec., 1950.

136

In its yellow colour, possession of septocystidia and spore characters this species

is closely allied to Pellicularia zealandica G. H. Cunn. (in Trans. Roy. Soc. N.Z. 81,

1953, 322, Figs. 1, 2). P. zealandica does not, however, produce adventitious basidia

from the septocystidia, possesses clamps in the hyphae and septocystidia, lacks encrusted

hyphae and has the encrustation of the septocystidia confined to near the apex. Some

rather similar fungi are placed by Bourdot & Galzin (Hym. de Fr., 1928) in the section

Hyphales of the genus Peniophora but none of these corresponds exactly with this new

species.

Pellicularia fodinarum Talbot & Green, sp. nov.

Fungus resupinatus, tenuis, non adnatus, pruinoso-pelliculatus. Hymenium

luteolum, septocystidiis longis, emergentibus, ad pilos accedentibus. Basidia cylin-

draceo-clavata, 4-6 X 15-22 /x; basidia adventitia lateralia ex quibusdam septocystidiis

oriuntur. Sporae hyalinae, leves, subglobosae vel ovato-ellipticae, 5-7-5 x 4-5-6 p.

Cystidia cylindracea, hyalina, septata, non nodoso-septata, apicibus rotundis, undique

granis verrucosis incrustata, usque ad 360 p longa, 6-9 p diam. Hyphae laxe intertextae,

ramosae, anastomosae, septatae, non nodoso-septatae, 6 p diam., granis verrucosis

incrustatae. Typus No. 40679 (TRL 2635), leg. A. L. James, in ligno in aurifodinae.

3. Pellicularia filamentosa (Pat.) Rogers in Farlowia 1 (1943) 113, Fig. 11.

Hypochnus filamentosus Pat. in Bull. Soc. Myc. de Fr. 7 (1891) 163, PI. 11, Fig. 2.

Corticium solani (Prill. & Delacr.) Bourd. & Galz. in Bull. Soc. Myc. de Fr. 27 (1911)

248.

Corticium vagum sensu Burt in Ann. Mo. Bot. Gard. 5 (1918) 128, Fig. 3, a. Ibid. 13

(1926) 295, Fig. 3, pro parte (nec. C. vagum Berk. & Curt.).

Corticium vagum var solani Burt ex Rolfs in Science n.s. 18 (1903) 729. For further

synonymy see Rogers, loc. cit.

Fig. 8.

Fructifications resupinate, hypochnoid to pellicular, delicate, thin, drying white

or buff-coloured.

Basidia: in botryose clusters, often discontinuous, subcylindrical to clavate or widest

in the middle, 13-22 X 9-11 p, with four long sterigmata.

Spores: hyaline, smooth, ellipsoid, flattened on one side, with a truncate apiculus,

8-12 X 5-7 p.

Hyphae: hyaline, thin-walled, except at the base where they have slightly thickened

walls and are pale coloured, lacking clamps, not encrusted, much branched at right

angles, septate, 5-12 p wide (up to 17 p wide fide Rogers).

Specimens examined: Myc. Herb. Dept. Agric. S. Rhodesia, No. 5151, /. C. Hopkins,

on Solanum tuberosum, Inyanga (as Corticium solani)', 41434, Martin, on Amarantus

paniculatus, Kempton Park.

Doidge (in Bothalia 5, 1950, 483) cites this specimen of Hopkins under Corticium

vagum Berk. & Curt., but it is in excellent condition and a very good match with P.

filamentosa. The former confusion existing between Pellicularia vaga (B. & C.) Rogers

ex Linder and P. filamentosa (Pat.) Rogers is noted under the first of these species.

Two other specimens cited by Doidge viz. Myc. Herb. Dept. Agric. S. Rhodesia

Nos. 4863 and 1783 as Corticium vagum were examined, but these have apparently

deteriorated and no Pellicularia was found upon them. Another specimen. Pons,

29964, is sterile and indeterminable, though its wide hyphae and their arrangement

suggest a species of Pellicularia.

137

4. Pellicularia asperula Rogers in Farlowia 1 (1943) 100, f. 2.

Fig. 11.

Fructification resupinate, delicate, pruinose to subpellicular, discontinuous, with

conspicuous basal hyphae showing under the lens, drying near Pale Olive Buff or Smoke

Gray (Ridgway).

Basidia: 9 -5-17 -2 X 7-8 /A, subcylindric, with a more or less truncate, slightly expanded

apex, with 6-(8) curved sterigmata up to 4-5 \i long.

Spores: 4 • 5-5 • 2-5 • 6-(6) x 3 • 0-4 • 0-(4 • 5) jn, ellipsoid, a little flattened on one side,

apiculate, minutely asperulate, abundant, often coherent, hyaline to faintly coloured.

Hyphae: 4-15-5 wide, without clamps, branched at right angles, the basal hyphae

widest, long celled and with lightly coloured thickened walls up to 1-7 \i thick; the

superior hyphae hyaline, narrower, shorter celled, thin-walled and with cruciform

cells.

Specimens examined: 41545, P. H. B. Talbot, on fallen twigs and litter of Acacia

moUissima, Byrne, Natal, Apr., 1956. (Part also in New York Bot. Garden Herbarium).

This species was abundant when collected, forming a thin greyish covering to

fallen twigs and humus, but only seen when the top layer of humus was disturbed.

Microscopically the spores are very distinctive and differ from those of all other known

species of Pellicularia. I am greatly indebted to Dr. D. P. Rogers for confirming my

identification of this material and comparing it with the type collection. The species

was previously known only from the type collection from Cuba.

CONIODICTYUM Har. & Pat. emend. G. Malengon.

1. Coniodictyum chevalier! Har. & Pat. in Bull. Soc. Myc. de Fr. 25 (1909) 13-14,

Figs, a-c; Malen9on in Bull. Soc. Myc. de Fr. 69 (1953) 77-100, Figs. 1-8 (as

Coniodyctium).

Hyalodema evansii P. Magnus in Ber. Deutsch Bot. Gesellsch. 28 (1910) 377-380,

PI. 11.

Coniodictyum evansii (P. Magn.) P. Magn. in Ber. Deutsch Bot. Gesellsch. 29 (1911)

1-2; Doidge in Bothalia 5 (1950) 685.

Fig. 3.

Mycelium parasitic in tissues of Zizyphus, filamentous, intercellular, hyphae 2-4 /x

diam., branched, septate, hyaline, with short coralloid, poorly differentiated haustoria,

forming galls on the fruits, leaves, petioles and twigs. Hymenium subcortical, covered

then erumpent, pulverulent, whitish-yellow, composed of homobasidia bearing basidio-

spores.

Basidia: clavate, 7-8 x 23-36 /x, often deformed or atrophied, sometimes thick-walled

especially near the apex, asterigmate.

Basidiospores : borne in a corona of (2)-4-8 sessile, ovoid, smooth spores at the apex

of the basidium, hyaline, often deformed, coherent, becoming falsely septate through

the development of a number of hyaline, spherical to compressed polygonal internal

chlamydospores, the basidiospores finally coalescing into an irregular, hyaline, multi-

cellular, dictyosporous ball, 15-20 X 18-28 /x, which is detached as a whole from the

basidium.

Specimens examined: All on Zyzyphus mucronata: 92, J. Burtt Davy, Zoutpansberg

(presumed part of type of H. evansii); 1006, Pole Evans, Fountains; 30667, L. Krause,

Johannesburg; 30223, W. G. Rump, Pietermaritzburg; 20611, E. M. Doidge, Harte-

beestpoort; 15019, J. M. Sim, Pietermaritzburg; 11812, T. Pallister, Marikana;

138

11240, J. M. Sim, Buccleuch, Natal; 10095, A. Jansen, Dundee; 41019, S. Truter,

Pietermaritzburg; 1214, P. J. Pienaar, Garstfontein; 2537, H. L. Hall, Nelspruit;

5648, Pole Evans, Cramond; 8789, A. Pegler, Kentani.

The genus Coniodictywn was at first classed as a member of the Hyphomycetae

(Mucedineae-Hyalodictyae), and only recently has Malengon (loc. cit.) shown in an

excellent paper on the nature and affinities of C. chevalieri that this and three other

puzzling genera are actually homobasidiomycetes and may best be accommodated in

a new family, the Cryptobasidieae G. Malengon, coming near to the Exobasidieae.

So far as is known C. chevalieri is purely African in distribution, and in South Africa

it is recorded from parts of the Transvaal, Natal and Cape Provinces.

CONIOPHORA DC. ex Pers.

1. Coniophora arida (Fr.) Karsten, Finska Vet.-Soc. Bidrag Natur och Folk 37 (1882)

161; Saccardo Syll. Fung. 6 (1888) 648; Burt in Ann. Mo. Bot. Gard. 4 (1917)

244, f. 3; Bourdot & Galzin, Hym. de Fr. (1928) 359.

Thelephora {Coniophora) arida Fries, Flenchus Fung. 1 (1828) 197.

Corticium (Coniophora) aridwn Fries, Hym. Fur. (1874) 659.

Fig. 9.

Resupinate, effused, adherent, floccose then pellicular or arid, or thinly submem-

branous, not readily separable. Hymenium smooth, not tubercular, pulverulent,

bright sulphur yellow dulling to ochraceous or chamois then fawn olive to brown.

Margin indeterminate, widely fibrillose to byssoid, eventually much reduced, whitish,

composed of cordons of hyphae.

Basidia: subcylindrical to clavate, 25-50 x 6-8 ju., with four sterigmata.

Spores: smooth, pale yellow to brown, ovate to ellipsoid, 10-13 x 6-7 p..

Hyphae: thin- walled, non-encrusted, septate, hyaline or pale coloured, 3-6 p wide,

the basal and marginal hyphae often up to 12 p wide; ordinary clamp connections

rare, whorled clamps present on some of the wide marginal hyphae, with branch hyphae

often arising in whorls from these clamps.

Specimens examined: 41051, Forest Res. Officer, on Pinus taeda. Border Plantation,

Natal; 39075 and 39076, Talbot, on Acacia mollissima, Atholl. Fxpt. Stn., F. Tvl.

Coniophora arida differs from C. puteana in the brighter colour of its fructification,

which is also thinner, drier, less fleshy-membranous, and whose hymenium is not

tuberculate and not easily separable from the substratum. The hyphae of C. arida

tend to become pale coloured and are less compactly arranged than those of C. puteana.

The characteristic whorled clamps which often give rise to whorls of branches on the

wide marginal hyphae are not mentioned in descriptions available to me, but I have

found them also on reliably determined exsiccati from England and Europe.

2. Coniophora atrocinerea (Karst.) Karst.; Recorded by van der Byl in Ann. Univ.

Stellenbosch 7 (1929) 17; Doidge in Bothalia 5 (1950) 480.

The record of this species is based on two specimens on Pinus in Universiteit van

Stellenbosch, Herbarium P. A. van der Byl Nos. 2451 and 2433. Both specimens have

been examined and are referable to Coniophora olivacea [(Fries) ex Pers.] Karst., a

species dealt with previously [in Bothalia 6 (1951) 35]. Rogers & Jackson (in Farlowia

1, 1943, 273) state that C. atrocinerea is synonymous with C. olivacea, and they give

a very full synonymy and useful taxonomic notes on the species.

3. Coniophora betulae (Schum.) Karst.; Recorded by Doidge in Bothalia 5 (1950) 480.

The specimen of this record is No. 30633, K. Morgan, which is here made the type

of Coniophora incrustata Talbot, sp. nov.

139

4. Coniophora cerebella Pers.; South African records cited by Doidge in Bothalia

5 (1950) 481.

Doidge (loc. cit.) indicates this species as a synonym of Coniophora puteana (Schum.

ex Fr.) Karst., which it is generally accepted to be. However, none of the South

African specimens cited are the latter species, (see under C. puteana).

5. Coniophora incrustata sp. nov.

Fig. 10.

Resupinate, membranous, not adnate, readily detached from the substratum in

large sheets. Margin not notably differentiated. Hymenium light brown to “ Warm

sepia ” or “ Bister ” (Ridgway) from the accumulated spores. Thickness in section

about 650 p.

Basidia; not seen.

Spores: deep red-brown in a mass, yellow-brown in the microscope, smooth, ovate-

elliptical, (5 • 6)-6-6 ■ 5-(8) X (8-8)-9-6-ll p, very copiously produced, some embedded

in the tissues.

Hyphae : hyaline to pale straw-colour, the upper ones more or less indistinct, the basal

ones distinct, very thin-walled and soon collapsing, 3-6-(8) p wide when not collapsed,

coated very heavily with mineral granules.

Specimens examined: Type, 30633, K. Morgan, on walls in dairy, Hopevale, Donny-

brook, 1939.

This species has heavily encrusted, readily collapsing hyphae like those of Conio-

phora betulae (Schum.) Karst, or C. suffocata (Peck) Massee, but differs from these in

several respects. It is appreciably thicker, more membranous, separable in large

pieces, lacks a notably whitish margin, and above all has much darker spores which

colour the hymenium a warm sepia or bister colour in contrast to the yellowish-olive-

brown (tawny olive or umber) of the other two species. It is felt that these differences

are sufficient to warrant the proposal of a new species. It is not entirely certain whether

or not C. betulae and C. suffocata are distinct from one another, and in Coniophora

as a whole there are few differential characters and numerous intermediates occur.

C. fumosa Karst, has somewhat the same colour as C. incrustata, but rather smaller

spores and non-encrusted hyphae, which are darker and distincter.

Coniophora incrustata sp. nov.

Fungus resupinatus, membranaceus, facile separabilis. Hymenium pallido-

brunneum vel sepiaceum. Sporae in cumulo atro-badiae, in microscopo luteo-brunneae,

leves, ovato-ellipticae, (5 • 6)-6-6 • 5-(8) x (8 • 8)-9 • 6-1 1 p. Hyphae hyalinae vel pallido-

stramineae; hyphae basales distinctae, tenue tunicatae, mox collabentes, 3-6-(8) p

crassae, mineralibus forte incrustatae. Typus No. 30633, leg. K. Morgan, in muro.

6. Coniophora pulverulenta (Lev.) Massee in Journ. Linn. Soc. Bot. 25 (1889) 129;

Saccardo Syll. Fung. 6 (1888) 649; Doidge in Bothalia 5 (1950) 480.

Thelephora {Stereum) pulverulenta Lev. ( !) in Ann. Sci. Nat. ser. iii, 5 (1846) 149; Doidge

in Bothalia 5 (1950) 491; Talbot in Bothalia 6 (1954) 323.

Corticium {Coniophora) pulverulentwn (Lev.) Cooke in Grevillea 8 (1880) 89.

The type of this species, Drege 9442, has been examined and, as noted in Bothalia

6 (1954) 323, proved to be most probably Hymenochaete luteobadia (Fr.) Hohnel &

Litsch.

140

7. Coniophora puteana (Schum. ex Fr.) Karst.; Doidge in Bothalia 5 (1950) 481.

The records of this species in South Africa are summarised by Doidge (loc. cit.)

All the specimens cited, except that collected by Laughton at Tokai, have been seen,

and none of them is C. puteana. The three collected by Mrs. R. Brown in gold mines

are all Coniophora fodinarum Talbot; No. 29723 on imported timber, is a species of

Coniophora but not C. puteana-. No. 30799 is a specimen of wood showing a cubical

rot but the fungus is so scanty that it is indeterminable, and it bears basidiospores which

are hyaline, 3 -5-8 -8 n, and are certainly not those of a Coniophora.

PUNCTULARIA Patouillard.

1. Punctularia tuberculosa (Pat.) Pat. (!) apud Patouillard & Lagerheim in Bull. Herb.

Boiss. 3 (1895) 57, PI. 2, Figs. 1, a-g; Patouillard, Essai Taxon sur les Hym.

(1900) 57, f. 60; Saccardo Syll. Fung. 14 (1899) 223.

Corticium ? tuberculosum Pat. apud Patouillard & Lagerheim in Bull. Soc. Myc. de

Fr. 8 (1892) 118; Saccardo Syll. Fung. 11 (1895) 126.

Punctularia atropurpurascens (B. & Br.) Petch in Ann. Roy. Bot. Gard. Perad. 6 (1916)

160.

Thelephora atropurpurascens B. & Br. in Journ. Linn. Soc. Bot. 14 (1875) 64; Saccardo

Syll. Fung. 6 (1888) 546.

Punctularia affinis (B. & C.) Talbot in Bothalia 6 (1951) 25, PI. 17.

Reticularia affinis B. & C. (!) in Journ. Linn. Soc. Bot. 10 (1869) 347; Saccardo Syll.

Fung. 7 (1888) 418.

Reticularia venulosa B. & C. (!) in Journ. Linn. Soc. Bot. 10 (1869) 347; Saccardo Syll.

Fung. 7 (1888) 419 (as R. venosa).

Reticularia atro-rufa B. & C. (!) in Journ. Linn. Soc. Bot. 10 (1869) 347; Saccardo

Syll. Fung. 7 (1888) 419.

Ceriomyces venulosus (B. & C.) Torrend in Bull. Soc. Portug. Sci. Nat. 4 (1910) 9.

Trichosporium curtisii Massee in Journ. Mycol. 5 (1889) 185, t. 14, f. 3; Saccardo

Syll. Fung. 10 (1892) 583.

Corticium conigenum Shear & Davidson (!) in Mycologia 36 (1944) 296, f. 1-2.

Figs. 14-25.

Conidial Stage: loose, floccose, pulvinate or irregular tufts of hyphae, forming

a growth several mm thick, powdery with innumerable conidia, later collapsing into

a tangled mass of hyphae and conidia, coloured various shades of reddy-brown to

purple-brown or vinaceous-brown (carob brown, burnt umber, cameo brown, vinaceous

brown, seal brown, warm blackish brown — Ridgway) or sometimes partly violaceous

to lavender blue or greyish blue (greyish violet-blue, dark plum-purple, bluish violet-

black — Ridgway), the margin whitish when young.

Hyphae ; often adhering in strands, 2-3 fi wide, lightly coloured, thin-walled, branched,

with clamp connections and septa, often minutely sculptured or encrusted with small

mineral granules.

Conidia; purplish-brown in a mass, globose, oval or ellipsoid with much variation in

shape, (3-6)-4~4’5-(6) /i. diam., or 3 -6-4 -5 x (5)-5 • 5-6 • 8-(8) /x. In culture the

conidiophores are much branched and modified above into an acropetal chain of

141

alternately swollen and clamped portions. The specialised terminal and intercalary

swellings round off into amerosporous conidia which are set free by dissolution of

the adjoining parts of the conidiophore, fragments of which may remain attached to

one or both ends of the conidia. The liberated conidia are smooth or very minutely

punctate at first, later becoming minutely verrucose, and finally strongly verrucose

and shrunken in old cultures.

Basidial Stage: effused, resupinate, loosely adnate, rarely narrowly but distinctly

reflexed, subgelatinous, drying waxy to horny and crustose. Margin inconspicuous

or shortly byssoid, indefinite or definite. Hymenium developed on small, discoid to

elongated pulvinae which are seated on a common foundation tissue and separated

by sterile fissures containing amorphous mineral matter which occasionally forms a

conspicuous pale buff-coloured fringe to the pulvinae. When moist the pulvinae are

irregularly hemispherical, wider at the apex than the base, and coloured a medium to

rich reddish-brown; when dry they are flattened, irregularly areolate, coloured deep

purple-brown or blue-black to fuscous (deep slaty brown, aniline black, fuscous —

Ridgway), with a minutely dark-punctate surface which is not conspicuously convoluted.

Dark brown abhymenial hairs are present in both resupinate and reflexed parts of the

fructification. Thickness in section averages 400-500 /x, excluding the abhymenial

hairs.

Basidia: long, flexuous, cylindric-clavate, 27-51 X 4-7 /x, with 4 sterigmata, often

with basal clamps. The basidia are seldom mature when collected, but fresh specimens

may be induced to spore after a few hours in a damp chamber.

Basidiospores : hyaline (? finally very dilutely coloured), smooth, ovoid to ellipsoid,

often with one flattened side, (2-7)-3- 5-4-(5) x 5-5-7-3-(8) /x.

Dendrophyses : in hymenium, composed of a hyaline, thick-walled stem bearing a

dark brown, irregularly swollen, shortly branched apex. Apical part up to 5-4 /x

wide and 29 g. long; stem 2- 3-4-1 /x diam.

Context hyphae: subgelatinous, hyaline, smooth, mostly rather indistinct, about 2 /u,

wide, with clamp connections.

Abhymenial hyphae: dark brown, fibrillose, densely interwoven, branched, smooth,

thick-walled, with clamps, (2)-2-5-4-5 /x wide, some penetrating the substratum.

Tissue differentiation: the abhymenial hairs arise from a dark basal seam. The middle

layer is hyaline, subgelatinous, containing scattered mineral concretions and somewhat

indistinct hyaline hyphae. The hymenium is subgelatinous, containing abundant

lightly coloured mineral granules and dark dendrophyses giving it a general light brown

colour.

Specimens examined: As P. tuberculosa ex Herb. Mus. Paris — Type, leg. de LagerheUn,

Puente de Chimbo, Ecuador, Aug., 1871; C. Torrend, Mycotheca Lusitanica, on

Olea europea, Lumiar, 1-08; Rick, Fungi Austro-Americani 54, Sao Leopoldo, Brazil;

As Corticium conigenum — Type, Shear (1405), on Quercus sp., Weikiwa Spa, Florida,

ex U.S. National Fungus Coll.; Also specimens listed in Bothalia 6 (1951) 26, and

No. 40150, Talbot, Pretoria.

In a previous publication (Talbot, loc. cit.) I regrettably overlooked the Rule

(Art. 69, Stockholm Code) that the first valid name or epithet applied to the perfect

state must take precedence as the name for the fungus, and incorrectly combined the

earliest epithet for the conidial state under Punctularia.

Corticium conigenum Shear & Davidson, was suspected to be a Punctularia from

the authors’ description and illustrations. Comparison of its type with that of P.

tuberculosa has not revealed any specific differences. In interesting cultural studies

142

Shear & Davidson found that single basidiospore cultures produced no clamps and

consisted of two sex strains, which when grown together would produce clamps along

the line where the two opposite strains met in culture. Clamps were not produced by

combinations of cultures within the same sex strains. Basidiospore cultures gave

rise to the conidial stage.

The method by which the conidia are formed in P. tuberculosa is particularly

interesting, and approximates to that shown by the mould fungi grouped in Section 1 B

of Hughes’ classification of the types of conidiophore and conidium development (Hughes

in Canad. Journ. Bot. 31, 1953, 577-659), except that here it is modified by the clamp

connections.

In his description and illustrations of P. tuberculosa, Patouillard exaggerated the

discreteness of the hymenial pulvinae. In rather small areas of the type they are discrete

and surrounded by the pale coloured amorphous mineral matter, but mostly they are

so close together, and separated by such narrow fissures, that the mineral fringe is not

really conspicuous, and the whole appears more or less areolate and flattened, especially

when dry. The dendrophyses occur in the hymenium, and not as a fringe to the pulvinae

as stated by Patouillard. The pulvinae are seldom perfectly discoid, but are usually

elongated and irregular in shape. My impression is that they dry more or less flat,

and not into the depressed, saucer-like bodies illustrated by Patouillard.

There are most impressive and detailed similarities between Punctularia tuber-

culosa and Phlebia strigoso-zomta (Schw.) Lloyd, which has recently been excellently

redescribed and made the type species of a monotypic genus, Phaeophlebia W. B. Cooke

(Cooke in Mycologia 48, 1956, 401).

Punctularia tuberculosa and Phaeophlebia strigoso-zonata agree in the purple-brown

to blue-black colour of the hymenium when dry, and its red-brown colour when moist.

They have essentially the same kind of hymenium. It tends to be pulvinate in Punctu-

laria and smooth to veined in Phaeophlebia, but the pulvinae of the former are often

elongated and parts of the latter are often almost pulvinate. Both have a hymenium

which, when dry, appears minutely punctate under a lens and is full of a cementing

substance composed of lightly coloured granules forming a darkish zone around and

above the hymenial elements. In each, the hymenial convolutions are separated by

narrow deep fissures which are sterile and contain amorphous mineral granules, some-

times forming a fringe to the papillae. Each has a subgelatinous texture, drying horny

to waxy. The basidia in each are rarely found in a mature condition and are long,

narrow and flexuous, with basal clamps. Sections of these two species are extraor-

dinarily alike and no significant difference in the arrangement of the tissues can be

detected, though in Phaeophlebia the context hyphae are perhaps a little more distinct.

In both, the hymenium contains dark dendrophyses. Phaeophlebia is usually resupinate-

reflexed or pileate; Punctularia is usually resupinate, but (as shown in local specimens

matched with the type and bearing also the conidial stage) can become reflexed as

much as 1 cm from the substratum. In each species there are hyaline, clamped hyphae

forming the middle layer and there is a dark basal seam and a trichoderm of dark,

thick-walled, clamped hyphae. I regard this trichoderm as an indicator of potential

reflexion even in species or specimens which are normally resupinate. The margin

of Phaeophlebia is definite; that of Punctularia is usually indefinite, but not always so.

Punctularia tuberculosa has a known conidial stage and lacks gloeocystidia ;

Phaeophlebia has no known conidial stage and possesses gloeocystidia. These features

are not, in my opinion, sufficient to hold the two genera apart, while in all other respects

I regard them as congeneric. Accordingly I propose to sink the genus Phaeophlebia

W. B. Cooke under Punctularia Pat. and to make the combination:

143

PuDCtuIaria strigoso-zonata (Schw.) Talbot comb. nov.

= Merulius strigoso-zonata Schw. in Trans. Axner. Phil. Soc. n.s. 4 (1834) 160.

Phlebia strigoso-zonata (Schw.) Lloyd in Lloyd Myc. Writ. 4 (1914) L. 53, p. 15;

Talbot in Bothalia 6 (1951) 28.

Phaeophlebia strigoso-zonata (Schw.) W. B. Cooke in Mycologia 48 (1956) 401.

For a description of this species see W. B. Cooke (loc. cit.).

The systematic position of the genus Punctularia is still doubtful. P. tuberculosa

has been placed by various authors in Corticium and Thelephora, and has been likened

to Grandinia, Aleurodiscus or Porotheleum. P. strigoso-zonata has previously been

classified as a Phlebia, Stereum, Auricularia or Phaeophlebia. Most of these can be

eliminated as not being near relatives. The nature of the hymenium, basidia, sterile

structures and perhaps the spores, set Punctularia apart from other genera. Its holo-

basidia remove it from the Heterobasidiomycetes. Its structure is certainly not

hydnaceous, nor cyphelloid. It has some brown hyphae, but does not accord with

the hymenochaetoid or xanthochroic fungi. The distribution of the tissues in Punctu-

laria is decidedly Stereum-Wke., though the texture and hymenium are different. Martin

(in Proc. Iowa Acad. Sci. 50, 1943, 167) remarked that Phlebia strigoso-zonata had

little in common with the Hydnaceae but a closer resemblance to the more fleshy species

of Merulius. There might be a good case for the proposal of a new family to accommo-

date Punctularia, but this seems premature until the majority of the species and genera

of the Thelephoraceae are known in greater detail. I would therefore place Punctularia

provisionally in the subfamily Meruloideae of the Thelephoraceae, realising however

that both the family and subfamily as now constituted are heterogeneous and artificial.

I wish to acknowledge the courteous and stimulating assistance given to me by

Professor G. W. Martin and Dr. W. B. Cooke in correspondence about the genus

Punctularia. This helped greatly in the formation of my ideas on this subject, for which,

however, I accept full responsibility.

CORTICIUM Pers. ex Fries.

1. Corticium abeuns Burt in Ann. Mo. Bot. Gard. 13 (1926) 250; van der Byl in Ann.

Univ. Stellenbosch 7 (1929) 29, f. 10.

This record for South Africa rests on the specimen in Universiteit van Stellenbosch,

Herbarium P. A. van der Byl, No. 1495. It was cited by Burt as a paratype of C. abeuns.

Rogers & Jackson (in Farlowia 1, 1943, 280) have examined all Burt’s paratypes of

this species and concluded that the description was misleading and the paratypes

confused. They identified some of the paratypes but, being mainly concerned with

North American species, did not mention van der Byl’s collection. This collection

has been examined and is here referred to Corticium porosum Berk. & Curt.

2. Corticium bombycinum (Sommerf.) Bres.; Recorded by Doidge in Bothalia 5 (1950)

481.

One specimen is cited in support of this record, namely No. 33185, Doidge &

Bottomley, Wolhuterskop, Boschfontein. This specimen is undoubtedly Peniophora

arenata Talbot, and C. bombycinum should be omitted from South African records.

3. Corticium lacteum Fr.; Recorded by Saccardo, Syll. Fung. 6 (1888) 610; van der

Byl in Ann. Univ. Stellenbosch 7 (1929) 30; Doidge in Bothalia 5 (1950) 482.

No South African specimens under this name have been located and the record

is based on Saccardo’s mention of “ Capite Bonae Spei ” as one of the localities for

this fungus, without giving any other details.

3871936-5

144

Rogers & Jackson (Farlowia 1, 1943,. 294) give a long discussion on the status,

of the name C. lacteum Fr. and reject it as a nomen dubium.

4. Coriicium laetum (Karst.) Pers.; Recorded by Lloyd, Myc. Writ. 6 (1920) 952;

Doidge in Bothalia 5 (1950) 482.

Presumably the specimen sent by van der Byl to Lloyd, which is the basis of this

record, is one of those in the National Herbarium, since there are none in the van der

Byl Herbarium. All the specimens assigned to this species in Doidge (loc. cit.) have

been examined and are referable to Corticium salmonicolor B. & Br., except No. 9154

which is Helicobasidium compactum (Boedijn) Boedijn. The material of No. 27267

is poor but may be the Necator decretus state of C. salmonicolor. This conidial stage

has recently been found on Pyrns malus in Natal (No. 41053).

5. Corticium moniliforme sp. nov.

Fig. 11.

Resupinate, effused over soft, dead wood, adnate, deeply and abundantly cracked

down to the substratum when dry, the cracks not usually connected by subicular strands.

Hymenium smooth, white, chamois or biscuit colour when dry, with a paler white,,

indefinite, pruinose or pellicular margin. Texture membranous-waxy. Context

somewhat stratified, concolorous, 400-450 fj. thick in section.

Basidia; clavate, often with a basal clamp connection, (27)-32-40 X 6 -5-8 -5 /x, with

four sterigmata.

Spores: cylindrical to elliptical, hyaline, smooth, non-amyloid, with rounded ends.

7-4-10 X 3-4-8 /X.

Gloeocystidia : arising in the middle layers or below and extending into the hymenium

but not emerging, originating from a system of tortuous hyphae, thin-walled, with

homogeneous contents, with a cylindric-clavate basal part or cell and the superior

part divided by as many as 12 constrictions into a moniliform structure with constric-

tions often so deep that a chain of superimposed cells is formed, with a scarcely per-

ceptible isthmus between each. The globose cells get progressively smaller towards

the apex of the gloeocystidium. Gloeocystidia 85-100 fj. long and 5 -7-9 -6 /x wide,

sometimes with a basal clamp connection.

Hyphae: (1) the system from which the gloeocystidia arise has hyaline, thin-walled

hyphae, very much branched and anastomosing in a wide network, tortuous, with

nodular swellings and irregular outline, septate, easily stained with phloxine, 2 -8-4 -3

jLx wide, found throughout the context; (2) a ground tissue of hyaline, septate, thin-

walled, branched hyphae, with occasional clamp connections, which do not stain and

are always indistinct as they collapse easily.

Minerals: present throughout the tissues.

Specimens examined: Type, 36899, Talbot, Hennops River; 36936, Talbot, Hennops

River.

The moniliform gloeocystidia suggest an Aleurodiscus but the relatively small

basidia and small non-amyloid spores and lack of other paraphysoid structures, dis-

tinguish the present species from Aleurodiscus. In the genus Corticium I have found

only two species described with moniliform gloeocystidia: Corticium radiosum Fr.,

according to Cunningham (Trans. Roy. Soc. N.Z. 82, 1954, 290) has moniliform

gloeocystidia, but its spores are oval or subglobose, 8-11 X 7-8 jix. Corticium sep-

tentriotwle Burt, which is a synonym of Corticium litschaueri Burt (fide Rogers & Jackson

in Farlowia 1, 1943, 300), also has gloeocystidia of the same type, but according to

Cunningham (loc. cit. p. 293) these may project sometimes up to 14 /lx and its spores

are smaller, 6-8 x 2-5-3 /ix. It appears to differ from the new species in hyphal

characters, the hyphae being much more regular, not tortuous nor collapsing.

14^:

Corticiwn moniliforme iLdilboi Xicxv,

Fungus resupinatus, effusus, adnatus, membranaceo-ceraceus, ut siccus profunde,

fissus. Hymenium leve, albidum, bubalinum vel isabellinum. Margo indefinitus,

pruinosus vel pelliculosus, albidus. Basidia clavata, (27)-32-40 x 6 -5-8 -5 /x, saepe

ad basem nodoso-septata, sterigmatibus 4. Sporae cylindraceae vel ellipticae, hyalinae,

leves, non amyloideae, 7-4-10 x 3-4-8 /j.. Gloeocystidia cellulis basalibus cylin-

draceo-clavatis, partibus superioribus moniliformibus, interdum ad basem nodoso-

septata, non emergentia. Hyphae (1) hyalinae, tenue tunicatae, tortuosae, anastomosae,

septatae, tumoribus nodulosis inaequalibus (2) hyalinae, tenue tunicatae, ramosae,

nodoso-septatae, obscurae. Typus No. 36899, leg. P. H. B. Talbot.

6. Corticium pelliculare Karst.; Recorded by van der Byl in Ann. Univ. Stellenbosch

7 (1929) 29; Doidge in Bothalia 5 (1950) 482.

The only specimen on which this record is based is Universiteit van Stellenbosch,

Herbarium P. A. van der Byl No. 2225, Knysna. This specimen is certainly not C.

pelliculare, but instead is readily identifiable with Peniophora arenata Talbot (in Bothalia

4, 1948, 945, f. 4 and Ibid. 6, 1951, 22). Corticium pelliculare should be eliminated

from South African records.

7. Corticium porosum Berk. & Curt, apud Berk. & Br. in Ann. Mag. Nat. Hist. ser. v,

3 (1879) 211 ; Rogers & Jackson in Farlowia 1 (1943) 300; Cunningham in Trans.

Roy. Soc. N.Z. 82 (1954) 295, f. 13. Further synonyms and literature references

given by Rogers & Jackson, loc. cit.

Fig. 13.

Resupinate, effused, thin, membranous, separable in small pieces when moist,

white to ivory yellow, drying chamois. Hymenium cracked occasionally and smooth,

or faintly tuberculate. Margin whitish, thinning out. About 220-250 /x thick in section.

Basidia: clavate, 15 x 4 jix.

Spores: ellipsoid to subglobose, 3 x 4 /x (4-6-7 X 3-4 /x, fide Bourdot & Galzin),

hyaline, amyloid, appearing smooth under low magnification but actually minutely

roughened under oil immersion.

Gloeocystidia: mostly originating in the basal layer but some small ones are hymenial,

thin-walled, sometimes the wall slightly thickened at the base, subulate to ventricose or

fusiform, usually becoming narrow towards the apex, flexuous, possessing granular

yellowish contents, not becoming brown in Iodine solutions, abundant, 8-13 /x wide at

the base and tapering to 3-6 /x wide higher up, 40-186 p. in length.

Hyphae: 2-3 /x diam, branched, interwoven in a more or less vertical direction, not

encrusted.

Specimens examined: As Corticium abeuns Burt: Universiteit van Stellenbosch,

Herbarium P. A. van der Byl Nr. 1495, Houtbos, Tvl.

Although determined as Corticium abeuns by Burt, this specimen is not that species.

Rogers 8c Jackson (loc. cit.) found that three other paratypes of C. abeuns were actually

Corticium porosum. The present collection has the general structure of the C. porosum

group and corresponds with this also in having amyloid, minutely roughened spores,

and gloeocystidia which do not turn brown with Iodine. Its spores are small, being

at the minimum range of size for this species. The gloeocystidia of typical examples

of C. porosum are frequently, but not invariably, bifurcate at the base, while their apex

is possibly a little less flexuous than those in the present specimen. Nevertheless the

species is known to very variable in the nature of its gloeocystidia. Cunningham

(loc. cit.) notes that “ Practically every section examined shows a somewhat different

microstructure, consequently it is advisable to regard these as forms of this variable

species, as Rogers & Jackson have done. In general, thick specimens have longer

gloeocystidia and more of them in the lower portions of the context

146

It seems likely that this species will find general acceptance as Gloeocystidiellum

porosum (B. & C.) Donk (see Donk in Fungus 26, 1956, 9).

8. Corticium punctulatum Cooke in Grevillea 6 (1878) 132; Further references and

synonymy, see Rogers & Jackson in Farlowia 1 (1943) 320.

Hypochnus eylesii van der Byl (!) in S.A. Journ. Sci. 22 (1925) 168, in Ann. Univ.

Stellenbosch 7 (1929) 18.

Fig. 12.

Resupinate, widely effused, hypochnoid to submembranous, not cracked, adherent,

light yellowish to creamy. Hymenium pruinose, pubescent from the projecting gloeocy-

stidia. Margin entire or pruinose. Thickness in section about 230 p.

Basidia: clavate to cylindrical, 5 -6-6 -5 p wide.

Spores: Hyaline, finely asperulate, subglobose to broad elliptic, 4-8 x 5-6 p or

4 -8-6 -4-8 p diam., often uniguttulate, thin to thick-walled, very abundant, free or

embedded in the context.

Gloeocystidia : embedded in all parts of the trama, some also projecting 22-64 p above

the hymenium where they are cystidioid and usually lightly encrusted with large mineral

particles; the embedded gloeocystidia are smooth, thin- walled, with hyaline, homo-

geneous contents, ventricose to subulate or cylindrical, with a blunt, rounded apex,

13-(19) p wide at the base, 70-115 p long.

Hyphae: basal hyphae clearly seen, hyaline, with frequent clamps, septate, smooth,

branched, with thickened walls, 4-6-4 p wide. Hyphae of middle layer compacted

and indistinct, but apparently like the basal ones only with thinner walls.

Specimens examined: 40684 (T.R.L. 2937), M. Mrost et ah, on underground timbers,

Venterspost Mines; 40678 (T.R.L. 2617), A. L. James et al., on underground timbers,

Venterspost Mines; 40682, (T.R.L. 2758) A. L. James, on pitch pine sills, Blyvooruitzicht

Mine; As Hypochnus eylesii (Isotype) in Herbarium Len Ven\'oerd (30), Stellenbosch-

Elsenburg Agric. Coll., leg F. Eyles, Stellenbosch, May, 1923.

These collections come clearly in the group of species which includes Corticium

albostramineum, Peniophora sphaerospora, Gloeocystidium cremicolor and G. eichleri.

These four are regarded by Bourdot & Galzin (Hym. de Fr., 1928, 262) as forms of

Gloeocystidium albostramineum, and by Rogers & Jackson (Farlowia 1, 1943, 320) as

synonyms of Corticium punctulatum Cooke, which we accept here as the name for this

complex. Bourdot and Galzin distinguish these forms on differences in spore size,

compactness of the tissues, and so on, but evidently Rogers and Jackson regard these

as minor variations about a central species.

The isotype of Hypochnus eylesii van der Byl, apparently the only material of this

species in existence, is an excellent match with C. punctulatum. Its tissues are more

laxly intertexed than those of the specimens from Transvaal mines, while more of its

spores have the walls thickened. The Transvaal material corresponds most closely

with the description of C. cremicolor in spore size and in the compact trama, but

undoubtedly the watersoaked habitat has had an effect on the appearance of this material.

9. Corticium solani (Prill. & Delacr.) Bourd. & Galz.; Records cited by Doidge in

Bothalia 5 (1950) 483.

— Pellicularia filamentosa (Pat.) Rogers, q.v.

10. Corticium vagum Berk. & Curt.; Records cited by Doidge in Bothalia 5 (1950) 483.

= Pellicularia vaga (B. & C.) Rogers ex Linder, q.v.

11. Corticium vagum var solani Burt ex Rolfs; Records cited by Doidge in Bothalia

5 (1950) 483.

= Pellicularia filamentosa (Pat.) Rogers, q.v.

147

PENIOPHORA Cooke.

1. Peniophora carnosa Burt in Ann. Mo. Bot. Gard. 12 (1925) 325; Doidge in Bothalia

5 (1950) 486.

The specimen on which this record is based is No. 27761 {Rump, 107). This is

definitely not P. carnosa, which has hairlike, non-encrusted cystidia, a bright yellow

hymenium and context, and occurs on coniferous wood. Instead this specimen has

large, conical, encrusted cystidia and occurs on frondose wood. Though poor in

condition and lacking spores it may be identified by its general structure and cystidia

as Peniophora roumeguerii Bres., which was noted in Bothalia 6 (1951) 22, pi. 14.

2. Peniophora cinerea (Pers. ex Fr.) Cooke in Grevillea 8 (1880) 20, pi. 123, f. 8.; Sac-

cardo Syll. Fung. 6 (1888) 643; Burt in Ann. Mo. Bot. Gard. 12 (1925) 353; van

der Byl in Ann. Univ. Stellenbosch 7 (1929) 16; Doidge in Bothalia 5 (1950) 486.

Corticium cinereum Pers. ex Fries, Epicr. Syst. Myc. (1838) 563.

Doidge cites three collections in support of this record. I have examined Univer-

siteit van Stellenbosch, Herbarium P. A. van der Byl Nr. 326, on Pyrus malus, determined

by Burt, loc. cit. This is a resupinate, grey member of the Coloratae group of Penio-

phora, and is probably referable to P. cinerea. But as it is very young, consisting only

of primordia just emerging from the lenticels, T would wish to see further specimens

before being certain that this species is represented here. In this specimen the margin

is free but not reflexed, and the spores are 11-11-5 X 4-5 p.

No. 22044 (Fungi MacOwaniani No. 1054, Boschberg, Somerset East, as Corticium

cinereum Fries) is a typical specimen of Stereum umbrinum B. & C. The copy of

MacOwan 1054 in Herb. S.A. Museum No. 34248 resembles the Pretoria copy super-

ficially but is a true species of Hymenochaete. It is however fragmentary and indeter-

minable.

3. Peniophora cremea Bresadola, Fungi Trid. 2 (1898) 63; Recorded for South Africa

by Burt in Ann. Mo. Bot. Gard. 12 (1925) 263; van der Byl in Ann. Univ. Stellen-

bosch 7 (1929) 16, f. 3; Doidge in Bothalia 5 (1950) 486.

All the specimens noted by Doidge in support of this record have been examined,

and none is considered to represent P. cremea. The record rested originally on Burt’s

determination of van der Byl No. 612, but he was evidently in some doubt about it

since this collection is annotated by van der Byl (translated), “ Species unknown by

Burt. In all probability P. cremea." The three specimens in Universiteit van Stellen-

bosch, Herbarium P. A. van der Byl, Nos. 612, 1468, 1488 are Peniophora pelliculosa

Talbot, No. 1468 being the most typical. Nos. 30231 and 34359 in the National

Herbarium are also P. pelliculosa, while No. 28288 is the type of Corticium gloeosporum

Talbot. On present knowledge jp. cremea is not acceptable as a South African species,

4. Peniophora gracillima Ell. & Everh. ex Rogers & Jackson in Farlowia 1 (1943) 317.

Peniophora glebulosa sensu Bresadola, Fung. Trid. 2 (1898) 61, PI. 170, f. 2; Saccardo

Syll. Fung. 16 (1902) 195; Rea, Brit. Basid. (1922) 688; Bourdot & Galzin, Hym.

de Fr. (1928) 288; Burt in Ann. Mo. Bot. Gard. 12 (1925) 282; van der Byl in

Univ. Stellenbosch 7 (1929) 15; van der Byl in S.A. Journ. Sci. 23 (1926) 288;

Doidge in Bothalia 5 (1950) 486.

Fig. 28.

Resupinate, effused, membranous, rather adherent, whitish to cream or yellowish,

pubescent on account of the projecting cystidia, becoming cracked into small clods

when dry. Margin thinner, similar or pruinose. Up to 200 p thick in section.

148

Basidia: small, clavate, about 10 x 3-5/u,.- -

Spores; hyaline, allantoid or cylindric-curved, smooth, 1-5 x 6-A-l-ljx. [(5)-9-10 x

1*5-3 /X fide Bourd. & Galz.; 6-9 x 1*5-2 ju. fide Burt].

Cystidia: abundant, arising deep in the trama and transversing it, embedded or pro-

jecting 40-64 fju beyond the hymenium (up to 110 in European material); tubuliferous,

cylindrical, hyaline, 60-120 /x long or longer, 7-12 /x wide, very thick-walled with a

canalicular lumen which is dilated brusquely at the apex so that the apex is thin-walled.

Apex rounded, smooth or often covered with granular mineral matter. Walls of the

cystidia soluble in KOH, but not in lactic acid, leaving only the lumen visible. Cystidia

borne laterally on basal hyphae, sometimes bifurcate at the base.

Hyphae: indistinct. The basal cystidiophores in European material are 3-5 fjb wide,

thick-walled, and the tramal hyphae are indistinct, 2-3 /x wide, thin-walled.

Specimens seen; Universiteit van Stellenbosch, Herbarium P. A. van der Byl, Nr.

1477,Woodbush, Transvaal.

This most interesting species has cystidia characteristic of the section Tubuliferae

of Peniophora, whose walls are dissolved by KOH. The allantoid spores are also

distinctive.

Though the species is better known as P. glebulosa Bres., Rogers & Jackson, loc.

cit., have pointed out that this name is invalid, and they published the new species

P. gracillima Ell. & Everh. to take its place. Malengon (in Bull. Soc. Myc. de Fr.

70, 1954, 139) gives useful notes on the taxonomy and nomenclature of this species.

According to Malengon the cystidia are amyloid. Donk (in Fungus 26, 1956, pp.

13-16) retains the epithet glebulosa for this species and places it in a new genus Tubuli-

crinis Donk, which appears to be a distinctive and well-founded genus. He charac-

terises the cystidia as “ lyocystidia

5. Peniophora incarnata (Pers. ex Fr.) Karsten; Recorded for South Africa by Doidge

in Bothalia 5 (1950) 486.

The specimen for this record, Scott No. 33768, is a member of the Coloratae

group of Peniophora, but differs from P. incarnata in colour and in its lack of gloeocys-

tidia. It is close to Peniophora versicolor (Bres.) Sacc. & Syd., as described and figured

by John Eriksson (in Symb. Bot. Uppsal. X; 5, 1950, 18) but differs in colour and in

having larger cystidia. It may represent a new species, but requires further study.

6. Peniophora longispora (Pat.) Hdhnel var brachyspora Talbot & Green var. nov.

Fig. 26.

Resupinate, very thin, easily separable, pruinose-pellicular, with similar margin.

Hymenium white, discontinuous, beset with emergent, hairlike, encrusted cystidia.

Basidia; clavate, usually with a basal clamp, 13-17 x 3-4 /x with four sterigmata up

to 4 /X long.

Spores; elliptic-fusoid, 2 *2-3 *2 x 6*4-8 p, hyaline, smooth, sometimes with a faint

band about the middle.

Cystidia ; hyaline, cylindrical to acicular, tapering to a sharp subulate apex, thin-walled,

with a wartlike, denticulate, incrustation or relatively large mineral granules, arising

laterally from the subhymenial hyphae and emergent for most of their length, 3 • 2-3 • 5 X

67-88 /X.

Hyphae; hyaline, loosely intertexed, thin-walled, smooth or a little encrusted, branched,

septate, with abundant clamp connections, 2 *5-3 *2 /x wide.

Specimens examined; Type, No. 40683 (T.R.L. 2901), M. Mrost, on underground

timbers, Venterspost Mine, 27/vi/52.

149

The general aspect of this variety is very like that of P. but The latter

is distinguished by having spores twice as long and somewhat narrower 0-3 x .12^18 /j.

fide Bourdot & Galzin), and also in having a bulbous base to ruiny of the cystidia. The

pruinose-pellicular fructifications, loose hyphae with clamps, characteristic cystidia

and banding of the spores are distinctive of both. The faint banding of the spores is

probably due to the formation of two polar guttules.

Another very similar fungus, possibly to be included under var brachyspora is

No. 40979, Talbot, on Populus deltoides, Piet Retief. This has bulbous-based cystidia

like P. longispora, but its spores measure 1-6-2 X 6-4-8 /x and are cylindric-curved.

That is, the spores correspond in length but not in shape with those of var brachyspora.

Bourdot & Galzin describe several varieties of P. longispora with varying shapes and

sizes of spores. None had spores as short as the present collections, and their length

is rather constant.

The spores of P. longispora var brachyspora correspond in length with those of

Peniophora soraria G. H. Cunn. (Cunningham in Trans. Roy. Soc. N.Z. 83, 1955,

280, f. 18) but the latter differs in having wider, less fusoid spores, 4-5 /x wide, and wider

cystidia, 5-7 /x wide.

Peniophora longispora (Pat.) Hohnel var brachyspora Talbot & Green var. nov.

Fungus resupinatus, albidus, pruinoso-pelliculatus, raargine simili. Basidia

clavata, 13-17 x 3^ /x, ad basem nodoso-septata, sterigmatibus 4, usque ad 4 /x longis.

Sporae elliptico-fusoideae, hyalinae, leves, interdum virga pallida circum mediam

sporam, 2 -2-3 -2 X 6-4-8 /x. Cystidia hyalina, cylindracea vel acicularia, apicibus

subulatis, tenue tunicata, emergentia, granis verrucosis complanatis incrustata,

3 *2-3 -5 X 67-88 /x. Hyphae hyalinae, laxe intertextae, tenue tunicatae, leves vel

parum incrustatae, nodoso-septatae, 2 -5-3 -2 /x diam. Typus No. 40683 (T.R.L.

2901), M. Mrost.

1. Peniophora nuda (Fries) Bresadola in I. R. Accad. Agiati Atti iii, 3 (1897) 114; van

der Byl in Ann. Univ. Stellenbosch 7 (1929) 17; Doidge in Bothalia 5 0950) 486.

•Corticium nudum Fries, Epicrisis Syst. Myc. (1838) 564.

Thelephora nuda Fries, Syst. Myc. 1 (1821) 447.

The only collection of this species recorded for South Africa is Medley Wood No.

161. I have not been able to trace this collection in any herbarium.

8. Peniophora papyrina (Mont.) Cooke in Grevillea 8 (1880) 20; South African Records

cited by Doidge in Bothalia 5 (1950) 486.

Stereum papyrinum Montagne.

Welwitsch, Iter Angolense No. 433, cited by Doidge is a specimen of Stereum

fulvum (Lev) Sacc., with pseudosetae derived from the ends of skeletal hyphae. Penio-

phora papyrina has a rather similar external appearance, but its skeletal hyphae are

modified apically into wider, more peniophoroid cystidia resembling those of Stereum

umbrinum.

The other specimen of P. papyrina mentioned by Doidge has not been traced.

It seems dubious whether the species is represented in South Africa.

150

9. Peniophora pruinata (B. & C.) Burt in Ann. Mo. Bot. Card. 12 (1925) 340; Doidge

in Bothalia 5 (1950) 486.

Stereum pruinatum Berk. & Curt. (!) in Joum. Linn. Soc. Bot. 10 (1868) 332.

This species is discussed under Stereum pruinatum B. & C. in Bothalia 6 (1954),

323. The on’y specimen recorded for South Africa under this name, MacOwan (1227)

in Herb. Kew., is not determinable, but lacks cystidia and does not match the type of

S. pruinatum in several other respects. P. pruinata must be omitted from South African

records.

10. Peniophora pruinosa (Pat.) Jackson in Canad. Journ. Res. 28 (1950) 530, f. 2.

Corticium pruinosum Pat., Cat. rais. pi. cell. Tunisie (1897) 60; Saccardo, Syll. Fung.

14 (1899) 222.

Peniophora chordalis Hohn. & Litsch. in K. Akad. Wiss. Wien Sitzungsb. 115 (1906)

1598; Bourdot & Galzin, Hym. de Fr. (1928) 280; Wakefield in Trans. Brit.

Myc. Soc. 35 (1952) 58, f. 34.

Peniophora subgelatinosa Litsch., Osterr. Bot. Zeitschr. 77 (1928) 128. Above synonymy

after Jackson loc. cit.

Fig. 30.

Resupinate, effuse, inconspicuous, gelatinous, to mucoid when fresh, hyaline or

blue-grey, drying as a greyish vernicose film which is white-pruinose under the lens

showing protruding cystidia, adnate, not more than about 50 p thick.

Basidia: not seen (fide Jackson “ cylindric or cylindric-clavate, 15-25 X 7-5-8 p

with (4)-5-6-(7) short sterigmata ”).

Spores: hyaline, appearing smooth in KOH but actually minutely punctate all over

as seen under oil immersion in lactic acid mounts, not amyloid, ovate-elliptical often

with one flat side, laterally apiculate, 5-6-(7) x 3-4 p.

Cystidia: projecting for most of their length, hyaline, smooth, not encrusted, tapering

from base to rounded apex, the base thick- walled and 8-11 /u wide often showing one

or two short rootlike branches, the apex thin-walled and 4-5-6 p wide. The cystidia

are 60-120 p long and in old collections may break off leaving the thick-walled basal

part behind as a short tubule.

Cystidioles: rare and not well shown in our material, but 45-50 x 4 p with a simple

or shortly lobed capitate apex. (Fide Jackson, “ Cylindric or tapering gradually from

base to apex, 25-45-(60) X 2 -5-3- 5 /x, wall slightly thickened below, apex occasionally

simple or capitate when young, more commonly with 3-5 short, blunt lobes ”).

Hyphae and Tissues: gelatinised and obscure.

Specimens examined: 39051, W. G. Rump, Town Bush Valley, Pietermaritzburg.

There is very little material of this collection but fortunately just enough to see

the characteristic texture of the fructification and the spores and cystidia. No basidia

and only two cystidioles were found.

This species differs from Peniophora rimicola (see next entry) in the smaller spores,

in the apically lobed cystidioles, and in having basidia with a greater number of sterig-

mata.

151

1 1 . Peniophora rimicola (Karst.) Hohnel & Litsch. in K. Akad. Wiss. Wien Sitzungsb-

115 (1906) 1556; Bourdot & Galzin, Hym. de Fr. (1928) 281; Rogers in Univ.

Iowa Stud. Nat, Hist. 17 (1935) 31; Jackson in Canad. Journ. Res. 28 (1950)

527, f. 1.

Corticium rimicolum Karsten in Hedwigia 35 (1896) 45; Saccardo, Syll. Fung. 14

(1899) 221.

Fig. 29.

Resupinate, effused, inconspicuous, gelatinous to mucoid when fresh, hyaline or

grey-blue, drying to a grey-blue or whitish vernicose film which is white-pruinose under

the lens showing numerous protruding cystidia, adnate but sometimes lifting as a

vernicose flake, not more than about 50 /x in thickness.

Basidia: not seen (fide Jackson, “ cylindric or somewhat clavate, 15-25 x 7-5-8 ju,

base usually obscure, sometimes bifurcate, bearing 2, 3 or 4 short arcuate sterigmata

3-4 /X long ”)•

Spores : hyaline, appearing smooth in KOH but actually minutely punctate all over as

seen under oil immersion when mounted in lactic acid, not amyloid, broad ovate-

elliptical, often with one side flattened, laterally apiculate, (5 ■ 6)-6 • 4-7-(8 • 8) x 8- 1 0-( 1 2)

Cystidia: projecting for most of their length, hyaline, smooth, not encrusted, tapering

from base to rounded apex, the base thick-walled and 8-11 /x wide often showing one

or two short rootlike branches, the apex thin- walled and 4-5-6 /x wide. The cystidia

are 60-120 n long and in old collections may break off leaving the thick-walled part as

a short tubule.

Cystidioles: present but not well shown in our material. (Fide Jackson, “ cylindrical

or slightly clavate, 25-35 X 2-3 yx, simple, or more commonly capitate at the apex

to 3-5 /X ”).

Hyphae and Tissues: gelatinised and obscure.

Specimens examined: 30952, P. H. B. Talbot, on Eucalyptus, Buffelspoort Expt. Stn.;

41024, P. H. B. Talbot, on Populus deltoides, Piet Relief; also collected on dead w'ood

at Fountains, Pretoria, 24/X/1948.

These dried collections were not in prime condition for examining and describing.

The basidia and hyphae were not clearly seen and the cystidioles were scanty and

inconclusive, but in other essentials they agree very well with Jackson’s excellent account

of this species.

Jackson characterises the P. rimicola group by their mucous-gelatinous fructifi-

cations which are often bluish and pruinose, and dry to a vernicose film. They tend to

produce marginal or superficial cordlike mycelial strands. Their tissues are gelatinised

and rather featureless, but the cystidia and cystidioles are distinctive, as also are the

non-amyloid minutely roughened spores. The basidia in P. rimicola are categorised

as “ pleurobasidia ” by Donk (in Reinwardtia 3, 1956, 370-371).

12. Peniophora tenuis (Pat.) Massee in Journ. Linn. Soc. Bot. 25 (1889) 149; Burt in

Ann. Mo. Bot. Gard. 12 (1926) 317; Rogers and Jackson in Farlowia 1 (1943) 322,

Corticium praetermissum (Karst.) Bresadola in Ann. Mycol. 1 (1903) 100.

Sebacina africana Burt (!) in Ann. Mo. Bot. Gard. 13 (1926) 338; van der Byl in Ann,

Univ. Stellenbosch 7 (1929) 20, f. 4; Doidge in Bothalia 5 (1950) 475; McGuire

in Lloydia 4 (1941) 43.

For an extensive synonymy and discussion see Rogers & Jackson loc. cit.

Fig. 27.

Resupinate, effused, adnate, whitish, becoming light creamy or isabelline,

Hymenium smooth, becoming vaguely granular to almost poroid in parts, not cracking.

152

Margin thin, indeterminate; Thickness in section about 100-130

Basidia ; subcylindric, closely compacted, 5-6-5 ^ wide.

Spores : cylindrical, oblong or ellipsoid with one side flattened, not amyloid, hyaline,

sometimes guttulate, 3-5 x 7-10 /^.

Gloeocystidia or Cystidia: embedded in all parts of the trama but especially arising

from the basal tissues, ventricose, fusiform or subcylindrical, with pale yellowish or

hyaline, homogeneous contents, yellow in Iodine, very variable in size, 8 • 5-12 X 39-135

p.;. a few of these bodies may be emergent, subcylindrical, with colourless contents,

smooth or slightly encrusted, cystidioid.

Hyphae: colourless, not encrusted, very compact and difficult to distinguish, forming

a dense tissue, mostly 3-5 /a wide.

Specimens examined: 39088 (T.R.L. 2436), James 8l Breyer, on underground timbers,

C.M.R., Johannesburg, 1949.

The identification of the above specimen was confirmed by Dr. D. P. Rogers. We

have followed Rogers & Jackson, loc. cit., in their interpretation of P. tenuis as a variable

complex of species. In some of its forms it might easily be taken for a gloeocystidiate

Corticium ; in others the emergent cystidia are more conspicuous and it is readily placed

in Peniophora. The cystidioid organs are especially variable in size and degree of

exsertion, but the spores are rather characteristic of the group. The Transvaal specimen

had a denser trama than is often seen in this group. The same tendency was noted in

specimens of Corticium punctulatum recovered from underground timbers.

The type specimen of Sebacina africana Burt was examined in Universiteit van

Stellenbosch, Herbarium P. A. van der Byl No. 1342, Knysna. The spores measured

9-13 X 3-5 /X, and the other characters agree with the P. tenuis group, to which S.

eifricana is here assigned.

13. Peniophora velutina (DC. ex Fr.) Cooke; Recorded by Doidge in Bothalia 5 (1950)

487.

The specimen on which this record is based. No. 27764 {Rump 117), is Peniophora

roumeguerii Bres.

MERULIUS Haller ex Fr.

1. Merulius confluens Schweinitz, Naturforsch. Ges. Leipzig 1 (1822) 92; Nel in Ann.

Univ. Stellenbosch 20 (1942) 77.

Nel’s record of this species for South Africa is based on the specimen in Univer-

siteit van Stellenbosch, Herbarium P. A. van der Byl Nr. 753. I am unable to differen-

tiate this specimen from Merulius corium (Pers. ex Fr.) Fr. (Compare Bothalia 6,

1951, 30).

Cunningham (in D.S.I.R. New Zealand Bull. No. 83, 1950, 6) states that M.

confluens differs from M. corium in having a darker hymenium with deeper pores,

smaller spores, and context hyphae which are thinner-walled, often collapsing, and

usually (but not always) coated with crystals.

2. Merulius lacrymans Wulf. ex Fries, Syst. Myc. 1 (1821) 328; Doidge in Bothalia

5 (1950) 502.

Boletus lacrymans Wulfen in Jacquin, Misc. Austr. 2 (1781) 111.

Mycelium loosely intertexed in a thick felted mat. fibrous, often with conspicuous

strands, whitish to pale fawn coloured. Fructification resupinate, widely effused and

sometimes reflexed, thick, spongy to fibrous-fleshy, readily separable from the sub-

stratum. Surface light brown to greyed-brown. Hymenium gyrose, somewhat poroid

or becoming flattened-dentate or raduloid. Pores 1-2 mm diarr. and about half as

153

deep, rust coloured, ferruginous yellow to Brussels brown or Sepia. Margin widely

sterile, pale to fawn colour, more or less tomentose. Thickness in section 2 mm or

more.

Basidia: clavate, 6-4-8 x 50 /i, with 4 sterigmata; accompanied by narrow paraphy-

soid hyphae about 2 jj. diam.

Spores: ellipsoid with one side often flattened, bright ochraceous or honey yellow,

4-8-6-4 X 9-6-12-8 n, frequently guttulate.

Hyphae: (1) coloured, thick-walled, with a narrow or invisible lumen, 5-6-8-(ll-2)

H wide; (2) hyaline, thin-walled, with clamps, 3-6-4 /x wide, sometimes with the wall

thickening slightly.

I have examined all the specimens listed under this name by Doidge (loc. cit.),

and am not quite convinced that any of them is M. lacrymans, since they are either

sterile mycelium or, if fertile, differ in being considerably darker in colour than typical

M. lacrymans. This may be due to the way these specimens have been preserved, and

I have thought it advisable to give the above description drawn up from freshly collected

English material (Nos. 35772, 35927, 36685, P. H. B. Talbot, Richmond, Surrey, on

timber in a house).

M. lacrymans is usually confused with M. himantioides in South Africa. The

hymenium of the latter is not rusty-coloured but has a darker, sordid colour. Its

hyphae, both hyaline and coloured, are thin-walled and they often become inflated and

then collapse readily.

3. Merulius molluscus Fries, Syst. Myc. 1 (1821) 329; Saccardo, Syll. Fung. 6 (1888)

416; Bourdot & Galzin, Hym. de Fr. (1928) 351.

Merulius fugax Fries sensu Burt in Ann. Mo. Bot. Gard. 4 (1917) 352, f. 33.

Merulius laeticolor Berk. & Br. in Ann. Nat. Hist. n. 1681; Saccardo Syll. Fung. 6

(1888) 417.

Fig. 31.

Resupinate, effused, membranous, very soft, readily separable. Hymenium tan,

creamy, pinkish buff to orange tinted, gyrose-plicate. Margin byssoid, whitish to pale

buff coloured. Context soft, byssoid to fibrillose, less than 300 fj. thick in section.

Basidia: clavate, sometimes with a basal clamp, 5-9 X 23-35 /x, with four short sterig-

mata.

Spores: smooth, hyaline, becoming very pale yellow, oblong-ellipsoid, 5-6 X 3-4-5

/X, commonly with 1-2 small guttules.

Hyphae: 3-5-7 /x wide, thin-walled, with abundant clamps which are often ansiform,

much branched, hyaline and smooth in the upper parts but granule-encrusted and rarely

pale-tinted in the basal tissues, all loosely intertexed.

Specimens examined: No. 41054, E. L. Stephens (679), on fir log, Cecilia Forest Plan-

tation, Cape; Rabenhorst Fungi Europaei (No. 3034), E. Marchal, Belgium.

The notes given by Burt and by Bourdot & Galzin show that there is some doubt

as to the typification of M. fugax and M. molluscus. I have followed Bourdot &

Galzin in adopting the name M. molluscus for this species since it conforms closely with

the good description given by Fries, whereas Fries’ description of M. fugax is poor and

could apply to several different species.

M. molluscus is distinctive but could possibly be confused with M. corium on

account of its bright colour. The latter is distinguished by its hyaline spores, thick-

walled hyphae lacking clamps, and effused, narrowly reflexed habit. M. molluscus is

resupinate and at least some of its spores, sometimes the majority, are coloured a pale

yellow.

154

4. Merulius pinastri (Fr.) Burt in Ann. Mo. Bot. Card. 4 (1917) 356, f. 36; Doidge in

Bothalia 5 (1950) 502.

Hydnum pinastri Fries, Obs. Myc. 1 (1815) 149, Syst. Myc. 1 (1821) 417.

Doidge records No. 27559, R. Lurie, New Modder Mine, Johannesburg, as Merulius

pinastri. The fructification appears somewhat deformed, probably owing to the habitat,

and is not determinable to species. It is not M. pinastri however, from which it differs-

in its colour and in the characters of its mycelium and spores.

5. Merulius rufus Pers. ex. Fr.; Recorded by Doidge in Bothalia 5 (1950) 502.

One specimen only is cited in support of this record, viz. No. 27790 (W. G. Rump

86). This is quite clearly eflfuso-reflexed and possesses the colour and spores of Merulius

corium. According to Burt (in Ann. Mo. Bot. Gard. 4, 1917, 338), M. rufus is fawn

to carob-brown or Natal Brown in colour, and its spores are 4-4-5 x 1-5-2 -5 /x, and

it is truly resupinate. M. rufus should be omitted from South African records.

6. Merulius serpens Tode ex Fries, Syst. Myc. 1 (1821) 327; Kalchbrenner in Grevillea

10 (1881) 57; Doidge in Bothalia 5 (1950) 502.

Kalchbrenner published a collection of MacOwan, Somerset East, as this species.

This collection has not been traced in any herbarium and the record cannot be confirmed.

7. Merulius tremellosus Schrad. ex Fr., Syst. Myc. 1 (1821) 327; Doidge in Bothalia

5 (1950) 503.

This record is based on No. 1350, E. M. Doidge, Garstfontein, Pretoria. The

specimen is clearly M. corium and not M. tremellosus, which differs in habit, the posses-

sion of very small allantoid spores, and in having a hymenium composed of radially

elongated shallow pores which become transversely ridged.

TOMENTELLA Patouillard.

]. Tomentella punicea (A. & S. ex Fr.) Schroet; Bourdot & Galzin, Hym. de Fr. (1928)

491.

Hypochnus puniceus (A. & S. ex Fr.) Saccardo, Syll. Fung. 6 (1888) 661; van der Byl

in Ann. Univ. Stellenbosch 7 (1929) 19; Doidge in Bothalia 5 (1950) 479.

Thelephora punicea A. & S. ex Fries, Elenchus Fung. (1828) 199; Alb. & Schw. Consp.

Fung. Lusatiae (1805) 278; Kalchbrenner in Grevillea 10 (1881) 58; Wood in

Kept. Natal Bot. Gard. (1898) 19.

This record for South Africa is based on J. M. Wood No. 190 which has not been

traced in any herbarium. The record is therefore doubtful. The following is a trans-

lation of the description of this species given by Bourdot & Galzin, loc. cit.

Formed of powdery granules, then confluent into a soft, floccose membrane,

loosely felted, slightly adherent, red-vermilion colour rarely persistent but passing into

a red-brown or brick red. Margin clear, arachnoid, or absent. Hyphae hyaline, thin-

walled, clamped, 3-6-(9) p, the basal ones tinted bistre-yellow. Basidia 30-42-50 x

6-7-10 /X, with 2-4 more or less arcuate sterigmata 6-8 /x long. Spores globose or

ovoid, rather regularly sinuolate and asperulate, 7-8-12 x 5-7 -5-9 p-, 1-guttulate,

hyaline, containing, as do the basidia and subhymenial hyphae, a red material which

is soluble in KOH and becomes brown.

155

2. In Bothalia 6 (1951) 63 the writer listed Hypochnus eylesii van der Byl (in S.A. Journ.

Sci. 22, 1925, 168 and in Ann. Univ. Stellenbosch 7, 1929, 18) under Tomentella, without

having seen material of this species.

Part of the type collection of H. eylesii was eventually located in Herbarium Len

Verwoerd (30) at the Stellenbosch-Elsenburg College of Agriculture. On study, this

proved to be Corticium punctulatum Cooke sensu Rogers & Jackson (in Farlowia

1, 1943, 320).

HYMENOCHAETE Leveille.

My studies of Hymemchaete in South Africa are incomplete and must remain so

pending a wider study of the genus. The collections can quite readily be arranged in

taxonomic species and it is often possible to say that a particular collection does not

represent the species under which it has been recorded. Its true identity is far more

difficult to establish, and in many instances this has not yet been done.

1. Hymenochaete dregeana (Berk.) Massee in Journ. Linn. Soc. Bot. 27 (1890) 114;

Doidge in Bothalia 5 (1950) 484.

Corticium dregeanum Berk. (!) in Hooker’s Lond. Journ. Bot. 5 (1846) 3.

In Bothalia 6 (1954) 344, the writer showed that the type of this species, Dvege

9451 c, should be referred to Irpex dregeanus (Berk.) Talbot.

2. Hymenochaete fulva Burt in Ann. Mo. Bot. Card. 5 (1918) 354; Doidge in Bothalia

5 (1950) 484.

The record rests on Doidge’s citation of Nos. 34356, 34386, and 35325 as this

species. The first two of these are co-specific but neither they nor the last-mentioned

specimen correspond with H. fulva. No. 35325 is close to H. fulva in structure, but

lacks cystidia and has much smaller setae.

3. Hymenochaete fusco-violascens (Mont.) van der Byl in Ann. Univ. Stellenbosch

7 (1929) 14; Doidge in Bothalia 5 (1950) 484.

Thelephora fusco-violascens Mont, in Ann. Sci. Nat. ser. iii, 7 (1847) 174; Saccardo

Syll. Fung. 6 (1888) 546.

This record is based on the collection of Drege (9429) from Durban. There does

not appear to be a specimen in existence. The short description, apart from the dark

violaceous colour of the hymenium, could fit many different species.

4. Hymenochaete luteobadia (Fr.) Hdhnel & Litsch. in K. Akad. Wiss. Wien Sitzungsb.

116 (1907) 754; Burt in Ann. Mo. Bot. Gard. 5 (1918) 323, f. 8; van der Byl in

Ann. Univ. Stellenbosch 7 (1929) 13, f. 2; Doidge in Bothalia 5 (1950) 485.

Stereum luteobadium Fr., Epicr. Syst. Myc. (1838) 547; Saccardo, Syll. Fung. 6 (1888)

571; Kalchbrenner in Grevillea 10 (1881) 58; Bottomley in S.A. Journ. Sci.

13 (1916) 440.

Fig. 34.

Pileate, dimidiate; surface of pileus coloured Argus Brown to Sayal Brown

(Ridgway), velutinate to tomentose. Margin entire, undulating, thin, concolorous.

Hymenium coloured Buckthorn Brown to Tawny Olive (Ridgway), smooth, concen-

Irically furrowed like the abhymenial surface. Thickness in section 300-4()0 /j,.

Basidia & Spores : not seen, ' ~

Setae: 35-50 x 4-6 /x, scanty, in a single layer projecting above the hymenfum.

Hyphae : brown, smooth, 2 /x wide.

Abhymenial hairs : similar to hyphae.

Tissue differentiation: there is a compact darker zone between the hyphae of the context

and the abhymenial hairs.

Specimens examined: No. 15558, fV. Haygarth, Ngoye (as Stereum villosum); 11974,

R. Duinmer, Kyague, Uganda; 14884, T. D. Maitland, Gongoni, Kenya; Universiteit

van Stellenbosch, Herbarium P. A. van der Byl Nrs. 146, 692, 523.

5. Hymenochaete nigricans (Lev.) Bresadola; Doidge in Bothalia 5 (1950) 485.

Stereum nigricans Leveille in Ann. Sci. Nat. ser iii, 2 (1844) 212; Saccardo, Syll. Fung.

6 (1888) 561.

Fig. 35.

Pileate, dimidiate; surface of pileus coloured Mars Brown to Chestnut Brown

(Ridgway), thickly matted with a soft, spongy, tomentum, concentrically furrowed.

Margin concolorous, undulating to slightly lobate, thin. Hymenium coloured near

Mummy Brown (Ridgway), smooth, somewhat furrowed. Thickness in section

(excluding surface hairs) about 150 /x.

Basidia & Spores: not seen.

Setae: in a single layer, projecting from the hymenium, (20)-35-40 X (6)-8-10 /x,

spaced well apart.

Hyphae: brown, smooth, 2-4-5 /x wide, arranged mostly parallel to the hymenium,

running out into abhymenial hairs with no dark or compact zone in between.

Abhymenial hairs : brown, smooth, 3-5 /x wide.

Specimens examined: MacOwan, Cap. b. sp. (sub Stereum percome in Herb. Kew.).

The specimen cited by Doidge as H. nigricans. No. 15558, Haygarth, is instead

Hymenochaete luteobadia. The latter differs from H. nigricans in having longer, thinner

and more delicate setae, which are also sparser, in the presence of a basal seam, in

colour, and in having a thinner tomentum on the surface.

The MacOwan specimen cited above corresponds in detail with Hymenochaete

spadicea Berk, which, as indicated by Bresadola (in Ann. Mycol. 14, 1916, 233) is

a synonym of H. nigricans.

6. Hymenochaete rubiginosa (Dicks, ex Fr.) Leveille in Ann. Sci. Nat. ser. ii>, 5 (1846)

151; Burt in Ann. Mo. Bot. Card. 5 (1918) 332, f. 11; Massee in Journ. Linn.

Soc. Bot. 26 (1890) 97; van der Byl in Ann. Univ. Stellenbosch 7 (1929) 14; Doidge

in Bothalia 5 (1950) 485.

Thelephora rubiginosa Dicks, ex Fr., Syst. Myc. 1 (1821) 436.

Stereum rubiginosum (Dicks, ex Fr.) Fr., Epicr. (1838) 550; Montague in Ann. Sci.

Nat. ser. iii, 7 (1847) 174.

The original South African record of this species was based on Drege No. 9450,

which I have not been able to locate in any herbarium. No. 30260 cited by Doidge

as this species is referred to Hymenochaete ochromarginata Talbot.

7. Hymenochaete labacina (Sow. ex Fr.) Lev. in Ann. Sci. Nat. ser. iii, 5 (1846) 152;

Doidge in Bothalia 5 (1950) 485.

Thelephora tahacina Sow. ex Fr., Syst. Myc. I (1821) 437.

Auricularia labacina Sowerby, Brit. Fung. (1797) pi. 25.

Of the specimens cited by Doidge for South Africa, I have been unable to locate

Cheesman, Groot Schuur. No. 27627, Rump, is a specimen of Stereum umbrinum.

No. 28287, Rump, is a species of Hymenochaete but not H. tabacina.

The MacOwan specimen cited by Doidge under H. tabacina var australis has not

been located. This variety is held to be a synonym of H. tabacina.

8. Hymenochaete tenuissima Berk (!) in Journ. Linn. Soc. Bot. 10 (1868) 333; Doidge

in Bothalia 5 (1950) 485.

Doidge cites two specimens, of which that collected by Cheesman has not been

located. No. 15596, Haygarth, is a polypore, Polystietus tabacinus Mont.

9. Hymenochaete tristicula (B. & Br.) Massee; Doidge in Bothalia 5 (1950) 485.

= Duportella tristicula (B. & Br.) Reinking. See Bothalia 6 (1951) 46.

ODONTIA Pers. ex S. F. Gray.

1 . Odontia bicolor (Alb. & Schw. ex Fr.) Bresadola in Ann. Mycol. 1 (1903) 87 ; Bourdot

& Galzin, Hym. de Fr. (1928) 429, f. 115; Miller in Mycologia 26 (1934) 27, PI. 3,

f. 5; Brown in Bot. Gaz. 96 (1935) 658, f. 14; Nobles in Canad. Journ. Bot. 31

(1953) 745-749, Figs. 1-7.

Hyclnum bicolor Alb. & Schw. ex Fr., Syst. Myc. 1 (1821) 417.

Fig. 32.

Resupinate, widely effused, pruinose to ceraceous in parts, white, becoming creamy.

Margin pruinose, white to creamy, indeterminate. Teeth variable, conical, subulate

or cylindrical, up to 0-4 mm long, often divided above into several points, the apices

usually fimbriate, composed of sterile hyphal ends with rounded or acute tips, projecting

about 50-60 /x and each 2-5-3 y. diam. Forming a profuse mycelium and white rot

within the substratum.

Basidia: clavate, 15-20 x 4-5 p, with 2-4 sterigmata up to 5 ;u long.

Spores: (not seen in South African material) elliptic-oblong with the base obliquely

attenuated, hyaline, (5-5)-7-(8) x (2- 5)-3 • 3-(4) p.

Cystidia: capitate (1) composed of an axis (2)-4-6 p wide and 15-20 p long whose

globose apex is surrounded by a smooth, thin-walled, globose envelope, sometimes

with yellowish contents, easily collapsing, 10-20 p diam. In some of these cystidia

the end of the axis is blown out into the envelope and no columella-like part is seen;

(2) other cystidia composed of a thinner axis, usually about 3 p diam. whose enlarged

end is surmounted by a crest of radiating acicular crystals about 8-12 p diam.

Hyphae: subicular hyphae are hyaline, branched, septate, with clamps, thin-walled,

1-5-2 p wide; the hyphae in the teeth are 2-3 p wide, easily collapsing, hyaline, thin-

walled, sometimes with clamps.

Crystals; radiating acicular crystals are abundant in all parts of the fructification.

Specimens examined: 36792, N. Eaton, on worked Pinus timber, Potchefstroom;

39046, P. H. B. Talbot, on Eucalyptus, Lions River, Natal.

The peculiar capitate and mineral-encrusted cystidia, the radiating groups of

crystals, and the dense white rot of the substratum, are characters by which this species

is easily recognised. Nobler ' describes \this fungus in culture and the important

rot of conifers and frondos: nees that it produces.

158

GRANDINIA Fries.

1. Grandinia rosea P. Hennings in Engler Jahrb. 38 (1905) 108; Saccardo, Syll. Fung,

21 (1912) 379; van der Byl in Ann. Univ. Stellenbosch 12 (1934) 8.

Fig. 33.

Resupinate, widely effused, with rather soft to membranous or spongy texture,

sometimes becoming crustose, coloured rosy to terra cotta or brick red. Margin

fibrillose to byssoid. Hymenium covered with small (180-200 /x diam., fide Hennings)

half-round tubercles or with short raduloid or more or less cylindrical teeth. The

tubercles and teeth are fertile and without setoid hairs or cystidia.

Basidia: cylindric-clavate, about 19-29 x 3-4 /x, with usually 2 very short sterigmata

(14-18 X 3-5-5 /X fide Hennings).

Spores: subglobose to ovate, or elliptical with one side flattened, hyaline, smooth,

3-2 X 4-8 /X (4-5 X 4 /X fide Hennings).

Hyphae: (1) layer next to substratum composed of compact horizontal adherent

hyphae with firm walls, septate, colourless, tending to collapse, 3-7 /x wide; (2) middle

and upper tissues of colourless, branched, septate, collapsing hyphae, 4-6-5 /x wide,

arranged more or less vertically and more loosely than the basal layer. No hyphae

are very clear. Airspaces common in the middle layers.

Specimens examined: Universiteit van Stellenbosch, Herbarium P. A. van der Byl

Nr. 706.

This is a distinctive species on account of its colour, soft and membranous to spongy

texture, and hyphae which are mostly widely spaced. Grandinia may not be the most

suitable genus for this species, but a better one cannot yet be suggested.

SISTOTREMA Fries emend. Donk.

1. Sistotrema muscicola (Pers.) Lundell apud Lundell & Nannfeldt in Fung. Exsicc.

Suecici Fasc. xxix-xxx (1947) 11, No. 1415 a.

Hydnwn muscicola Pers., Myc. Eur. 2 (1825) 181.

Grandinia muscicola (Pers.) Bres. apud Bourd. & Galz, in Bull. Soc. Myc. Fr. 30 (1914)

252, Hym. de Fr. (1928) 411, f. 112.

Trechispora muscicola (Pers.) Rogers in Mycologia 36 (1944) 83, f. 3.

Fig. 36.

Fructifications resupinate, consisting of delicate spines up to 1 mm long borne on

a separable pellicular to membranous subiculum, fragile, cream to ochraceous in colour,

with a pellicular, paler, sterile margin. The fructification is occasionally poroid.

Basidia: Obovate when immature, later becoming urnigerous, with a short, subcylin-

drical apical protrusion, not expanded at the apex, 13-20 X 6 -4-6 -8 /x, with (4)-6

short sterigmata up to 3-5 p long.

Spores: subglobose to broad ovate or ellipsoid, shortly attenuated at the base,

2 - 7-3 - 6 X 2 - 3-2 - 7 /x, smooth, hyaline (4-5 - 5 X 3-3 - 5 /x, fide Rogers ; 3-4 - 5 X 2 - 5-4

P, fide Bourdot & Galzin; 3-4*5 x 3-5 /x fide Lundell).

Hyphae: thin-walled, smooth, hyaline, readily collapsing in parts, with large, inflated

clamp connections up to 7-10 p wide, the rest of the hyphae being only 2- 5-5*5 p wide.

Specimens examined: 40671, Distr. Forest Officer, on dead Pinus pinaster near soil

level, Knysna, Woodville Plantation.

159

In the specimen examined, the spores are somewhat smaller than as usually described

for this species, but there appears no doubt as to its identity. Rogers (Loc. cit.) separated

Trechispora for resupinate species with urnigerous basidia, from Sistotrema which he

limited to pileate species. This treatment is not supported by Lundell, nor by Eriksson

(Svensk. Bot. Tidskr. 43, 1949, 312). I have here followed the Swedish view, since it

seems in this instance that the distinction between resupinate and pileate species is

too slight to be of generic significance. (See also Donk in Fungus 26, 1956, 4.).

2. Sistotrema brinkmanni (Bres.) J. Eriksson in K. fysiogr. Sallsk. Lund Forh. 18

(1948) 17.

Odontia brinkmanni Bres. in Ann. Mycol. 1 (1903) 88.

Trechispora brinkmanni (Bres.) Rogers & Jackson in Farlowia 1 (1943) 288; Rogers in

Mycologia 36 (1944) 88.

Corticium coronilla Hohn. apud Hohn. & Litsch. in Ann. Mycol. 4 (1906) 291; Biggs

in Mycologia 29 (1937) 686.

Fig. 38.

Fructification resupinate, very thin and delicate, pruinose to waxy or later thin

membranous, drying inconspicuous, white to yellowish, the hymenium even or minutely

papillate, the margin pruinose or finely fibrillose, sometimes accompanied in culture

by dark sclerotia or bulbils.

Basidia; 13 •5-16- 5 p. long, formed in clusters, urniform, composed of a swollen base

4-5 p wide, often with a subtending clamp connection, and a long narrower apical

part 2-1-3 -S p wide expanded somewhat below the corona of (4)-6-8 sterigmata.

Sterigmata curved, about 3-5 p long.

Basidiospores: ellipsoid to subcylindric, usually unilaterally depressed or flattened,

apiculate, smooth, 4 -4-5 -4 x 1-9-2 -6 p.

Hyphae : thin-walled, smooth, with abundant large clamps, (1 • 9)-4 • 5-7 p wide becoming

inflated into ampoullar hyphae at the clamps.

Bulbils: deep brown or light yellowish-brown, subspherical, 67-122 p diam., composed

of aggregated yellow-brown subspherical cells.

Other organs: Spherical cells 8-10 p diam., hyaline, eventually detached. These are

in the nature of chlamydospores and are produced in some cultures. Cylindrical,

hyaline oidia are found in other cultures.

Specimens examined; 41739, V. C. Green, on imported Southern Pine beams in storage

at West Rand Consolidated Mines, 22/3131 .

An extensive synonymy and description of this species is given by Rogers (1944)

and Biggs (1937).

This collection was noted in the form of bulbils on timber intended for use in the

mines, and the above description is drawn up from cultures made from the bulbils

by Miss V. C. Green.

Miscellaneous Addenda.

1. Irpex dregeanus (Berk.) Talbot and Irpex vellereus B. & Br.

D. A. Reid (in Kew Bull., 1955, pp. 631-648) holds Irpex vellereus apart from

/. dregeanus, while acknowledging a very close relationship between the two species.

He differentiates them as follows: —

Irpex vellereus. — Spores 4-5-6 x 2-5-3 p', Cystidia 3-5-5 p wide.

Irpex dregeanus. — Spores 6-5-7 x 4-5-5 p; Cystidia (5)-6-8 p wide.

3871936-6

160

2. Synonymy of Various Stereum Specier.

P. L. Lentz (in U.S. Dept. Agric. Agriculture Monograph No. 24, 1955, pp. 1-74)

gives the following synonyms : —

Stereum diaphanum (Schw.) Cooke = Cotilydia diaphana (Schw.) Lentz.

Stereum bicolor (Pers. ex Fr.) Fr. = Laxitextum bicolor (Pers. ex Fr.) Lentz.

Stereum umbrinum B. & C. = Laxitextum crassum (Lev.) Lentz.

Stereum fasciatum (Schw.) Fr. ^

Stereum lobatum (Kunze ex Fr.) Fr. I - Stereum ostrea (Blume & Nees ex Fr.)

Stereum concolor (Jungh.) Mont. | Fr.

Stereum perlatum Berk. J

3. Lopharia.

G. H. Cunningham (in Trans. Roy. Soc. N.Z. 83, 1956, 621) has emended the genus

Lopharia Kalchbr. & MacOwan to contain pileate species of the Thelephoraceae bearing

pedicellate cystidia. He typifies the emended genus by Thelephora cinerascens Schw.,

disregarding the fact that the type species of Lopharia was, and must remain, L. lirellos

Kalchbr. & MacOwan as this genus was monotypic when it was proposed. In my

opinion the structure of Lopharia is stereoid rather than peniophoroid, despite the

presence of cystidia, and I would therefore retain Stereum cinerascens as a Stereum.

However, if it should be necessary to differentiate it from Stereum because of the presence

of cystidia, 5. cinerascens is the type species of Lloydella Bres. which was differentiated

from Stereum precisely because it possessed “ metuloid ” cystidia. On the other hand,

Lopharia and Thwaitesiella Massee were originally erected on the nature of the hymenial

configuration, which is variable and unreliable, and they would be better sunk under

Stereum or Lloydella depending upon the emphasis given to the presence of the cystidia.

It appears that Cunningham was not justified in emending Lopharia when Lloydella

stands for exactly the same generic concept. Cunningham treats two species under

his emended Lopharia, viz., L. cinerascens (Schw.) G. H. Cunn. and L. vinosa (Berk.)

G. H. Cunn. (Syn. Stereum umbrinum B. & C.; Thelephora crassa Lev.). He thus

associates a monomitic, typically resupinate species lacking a basal seam, with a dimitic,

typically effuso-reflexed species possessing a basal seam. It is true that Stereum

umbrinum is ill-placed in Stereum, and for that reason I would rather follow Lentz

(U.S.D.A. Agric. Monogr. No. 24, 1955) in placing it in Laxitextum Lentz, if it is not

to be retained in Stereum.

4. Laeticorticium Donk.

In proposing the new genus Laeticorticium, Donk (in Fungus 26, 1956, 16-17)

makes the combinations L. roseum (Pers. ex Fr.) Donk and L. polygonioides (Karst.)

Donk. These two species were treated previously under Aleurodiscus (in Bothalia 6,

1956, 470), but Laeticorticium appears to be a well-founded genus. Donk refers to

the nodular dendroid hyphae as “ dendrohyphidia ”.

5. Scytinostroma Donk.

In two previous papers (Bothalia 4, 1948, 939 and Ibid. 6, 1951, 54, PI. 38) the author

erred in using the generic name Asterostromella Hbhn. & Litsch. (1907), since this is

based on the same type as the earlier genus Vararia Karst. (1903). Donk has now

proposed (in Fungus 26, 1956, pp. 19-22) the genus Scytinostroma Donk, to which he

refers Asterostromella duriuscula (B. & Br.) Talbot, Asterostromella rumpiana Talbot

and Corticium portentosum B. & C., at the same time differentiating this genus from

Vararia.

161

Summary of the Recorded South African Thelephoraceae.

This paper concludes a series of studies designed to revise the chaotic records of

the South African Thelephoraceae principally at the specific and generic levels. The

revision, carried out over a period of ten years, was based on a study of virtually every

species, and the vast majority of preserved specimens, of the Thelephoraceae recorded

for South Africa. Of necessity the results have been published in a number of apparently

disconnected papers, and so it seems appropriate at this stage to summarise them by

means of a key to the accepted species and a species-index of all the recorded South

African Thelephoraceae. This is, however, only a beginning; there must still be a

great number of other Thelephoraceae in this country awaiting collection and identifi-

cation.

A Key to the Accepted Genera and Species of South African Thelephoraceae.

1.

2.

Homobasidiomycetae, Aphyllophorales, Thelephoraceae:

Basidiocarp pezizaeform, patelliform, cupulate or tubular, sessile or stipitate, the

interior of the basidiocarp lined by the hymenium which may be smooth or wrinkled.

Basidiocarps solitary to gregarious or confluent, at times seated on or embedded

in a common floccose to membranous subiculum or stroma. . . . Subfamily Cyphelloideae

Basidiocarp resupinate, effuso-reflexed or pileate. Hymenium spread over shallow

veins, anastomosing ridges, or sometimes imperfectly poroid from anastomosis

of the veins. Texture usually soft, fleshy or subgelatinous Subfamily Meruloideae

Basidiocarp resupinate, effuso-reflexed or pileate, sessile or stipitate. Hymenium

smooth or finely rugose or somewhat tuberculate, or bearing sterile hyphal pegs

Subfamily Thelephoroideae

2.

3.

4.

(Hymenium corticioid, hydnoid or polyporoid. Basidia urniform, i.e. with a bulbous

base and a narrower tubular prolongation with a crown of (4)-5-8 sterigmata

1. Sistotrema.

Subfamily Cyphelloideae:

)

Subiculum present as a membranous stroma in which are seated contiguous cupular

basidiocarps 2. Porolheleiim.

Subiculum absent. Basidiocarps discoid, patelliform or effused, coriaceous to

fleshy-gelatinous. Hyphae usually gelatinised. Basidia and spores usually

large. Paraphysoid structures common 3. Cytidia.

Subiculum sometimes absent, if present then composed of floccose interwoven

hyphae. Basidiocarps usually cupulate to tubular, sometimes patelliform. . . 4. Solenia Sind

Cyphella.

(Compare also 16. Aleiirodiscus and 18. Dendwthele.)

3. Subfamily Meruloideae:

Hymenium covering plane to convex, discoid to elongated pulvinae separated by

narrow sterile clefts containing mineral matter, and seated on a fleshy or sub-

gelatinous stroma. Resupinate or reflexed to dimidiate. Hymenium containing

brown or yellow dendrophyses and cemented by brownish granules. Tissue

distribution StereumAik.^. Gloeocystidia present or absent. One species produces

abundant conidia 5. Punctularia.

Hymenium covering shallow anastomosing veins or pleats, alveolar or irregularly

poroid; resupinate or pileate, often rather gelatinous 6. Menilius.

4. Subfamily Thelephoroideae :

Hyphae brown, spores brown, setae or pseudosetae (setoid hyphae) absent 5.

Hyphae brown, spores hyaline, setae or pseudosetae present (guard against taking

intrusive skeletal hyphae or coloured cystidia as setae or pseudosetae) 6.

Hyphae hyaline or only faintly coloured, if the latter then the spores are hyaline;

setae or pseudosetae absent; other ancillary organs may be present or absent 7.

5. Spores smooth; hymenophore resupinate, more or less membranous 7. Coniophora.

Spores verrucose or echinulate; hymenophore resupinate, floccose (Hypochnus) =

8. Tomentella.

Spores verrucose to echinulate; hymenophore compact, subresupinate to usually

pileate 9. Thelephora.

162

6. Setae present 10. Hymenochaete.

Setae absent; pseudosetae present; gloeocystidia and cystidia also found at suitable

stages of development, but often not seen 11. Duportella.

Asterosetae present, see 17. Asterostroma.

7. Hymenophore effuso-reflexed to pileate, occasionally resupinate but then showing

tissues differentiated into a hymenium separated from the basal mycelium by a more

or less horizontal middle layer. Hyphae hyaline to faintly coloured, rarely dark.

Spores hyaline, smooth, rarely stippled-punctate. Texture coriaceous to sub-

ligneous 8.

Hymenophore resupinate with margin rarely free or only narrowly reflexed. Hyphae

usually hyaline, loosely interwoven or if compact then more or less vertically arranged

and lacking a differentiated middle layer. Spores hyaline, smooth. Texture arach-

noid, pellicular, byssoid, membranous, or fleshy, not coriaceous to subligneous 9.

8. Hymenium smooth or irregularly rugose; Hyphae normal, capillary 12. Sterewn.

Hymenium smooth or irregularly rugose; Hyphae dendrophytic to dichophytic.

Spores mostly subglobose and smooth or minutely stippled 14. Asterostromella.

Hymenium covering radiating costate ribs and generally also coarsely papillate. Hyphae

normal, capillary {Cladoderris) =13. Cymatodenna.

9. Basidia not in a palisade, formed in loose botryose clusters from lateral branches of

loosely interwoven repent hyphae. Texture arachnoid to pellicular, with generally

wide, distinct hyphae, usually branching at right angles. Septate cystidia occa-

sionally present 15. Pelliciilaria.

Basidia forming a palisade hymenium above a more compact trama 10.

10. Basidia usually very large; spores usually large and often amyloid. Paraphysoid

structures of various sorts (Acanthophyses, moniliform gloeocystidia, dendrophyses,

pseudophyses) present 16. Aleiirodisctis.

Basidia and spores not usually large and seldom amyloid. Ancillary organs present or

absent 11.

11. Asterosetae present \1. Asterostroma.

Asterosetae absent 12.

12. Hyphal pegs present, composed of dendrophytic hyphae 18. Dendrothele.

Hyphal pegs absent 13.

13. Hyphae dendrophytic to dichophytic \A. Asterostromella _

Hyphae normal, capillary, not prolifically branched 14.

14. Possessing cystidia and sometimes cystidioles and/or gloeocystidia as well. ... 19. Peniophora.

Lacking cystidia; sometimes possessing gloeocystidia and/or cystidioles 20. Corticium.

THFXEPHORACEAE (Pers.) Saccardo.

Saccardo, Syll. Fung. 11 (1895) 115.

Thelephoreae Persoon, Myc. Eur. 1 (1822) 109.

Hymenomycetes with inferior hymenium; hymenophore resupinate effuso-reflexed

or pileate, sessile or stipitate. Hymenial layer smooth or rugulose or papillate, with

continuous or discontinuous basidia, homobasidiate, frequently possessing various

ancillary organs. Spores unicellular, smooth or sculptured, hyaline or coloured.

Hyphae hyaline or coloured, with or without clamp connections, forming mono- or

dimitic hyphal systems.

Genus 1. Sistotrema Pers. ex Fr., Syst. Myc. 1 (1821) 426, emend. Donk apud Rogers

in Univ. Iowa Stud. Nat. Hist. 17 (1935) 19.

Fructification resupinate or reflexed to pileate. Hymenium smooth, granulose,

toothed or poroid. Texture arachnoid, pellicular or membranous. Hyphae hyaline,

with clamps. Basidia urniform, with a bulbous base and a narrower tubular prolonga-

tion bearing apically a crown of (4)-5-8 sterigmata. Spores smooth. Gloeocystidia

present in some species.

163

The species of this genus show hymenial characters which are associated with

several different Friesian families. The genus is included here because some of its

species may be sought in the Thelephoraceae.

1 . Resupinate, hydnoid with delicate spines up to 1 mm long. Basidia not closely clustered.

Spores subglobose or ellipsoid Sistotrema muscicola.

Resupinate, even or at most granular-papillate, lacking spines. Basidia closely clustered.

Spores ellipsoid to subcylindric. Bulbils, swollen cells (chlamydospores) or oidia

often present in culture Sistotrema hrinkmanni.

Genus 2. Porotheleum (Fr. ex Fr.) Fries, Syst. Orb. Veg. (1825) 80.

Fructifications composed of sessile, closely aggregated and confluent cupules,

partly immersed in a membranous stromatic subiculum. Hymenium smooth, lining

the interior of the cupules. Spores hyaline. Hyphae monomitic, hyaline, with clamps.

A single South African species, Porotheleum incanum.

Genus 3. Cytidia Quelet, FI. Myc. de Fr. (1888) 25.

Hymenophores cupulate, sessile, centrally attached, coriaceous to fleshy-gelatinous,

discrete, gregarious or confluent, sometimes effused. Hymenium lining the interior

of the cupules, smooth, sometimes becoming wrinkled or veined. Spores hyaline or

lightly coloured, amyloid. Hyphae usually gelatinised, with clamps, monomitic.

Paraphysoids or cystidia present or absent.

1. Spores large, ovate or broad elliptical, at first hyaline, later pale coloured with the wall

thickening to 2-5 ft and penetrated by peg-like projections of the lumen. Fusoid,

encrusted cystidia present Cytidia hahgal/ae.

Spores otherwise; Cystidia absent 2.

2. Spores of one kind only, cylindrical to allantoid, 1-5-3 X 7-10 //. Cupular hairs

long and floccose Cytidia fiocciilenta.

Spores of two kinds {a) cylindrical or allantoid, 2-2-5-(3-2) x 6-4-9 -6-( 1 1 -2) //,

ib) ovate to elliptic-depressed, 4-8-6 4 X 9-6-12-8 //. Cupular hairs short and

scanty Cytidia simtilans.

Genus 4. Solenia Pers. ex Fr., Syst. Myc. 2 (1823) 200; Persoon in Roemer Neues

Mag. Bot. 1 (1794) 116, Syn. Fung. (1801) 675, Myc. Fur. 1 (1822) 334.

Cyphella Fries, Syst. Myc. 2 (1823) 201.

Pilei cupulate, tubular or discoid, discrete or gregarious, sometimes confluent,

sessile or stipitate, naked or covered with hyaline or coloured hairs, seated sometimes

upon a hyphal subiculum. Hymenium smooth, lining the concave interior of the

cupules. Spores hyaline or coloured. Hyphae monomitic, hyaline or coloured, with

clamp connections.

Cyphella is an untenable name. The old basis of separating these genera on the

presence or absence of the subiculum and the discreteness of the basidiocarps cannot

stand; these characters vary in a single collection. For species see the following

combined key for both genera.

1. Fructifications cupular to tubular, sessile or shortly stalked 2.

Fructifications discoid with reflexed margin (i.e. patelliform). Encrusted villose

hairs on reflexed part. Spores cylindrical, straight or curved, 3-2-4 8 X (6-4)-8-l 1

Cyphella applanata.

2. Fructifications villose, with long hairs on the exterior 3.

Fructifications pruinose with short, sparse hairs or mineral encrustations, or nude, or

with hairs mostly appressed 4.

164

3. Spores smooth, broad ovate to pip-shaped, 11-2-14-4 X 9-9-6-(ll-6) /r. Cupules

1-3 mm diam Cyphella alboviolascens.

Spores finely asperulate, ovoid, 12-15 x 7-8 //. Cupules 1 mm diam., later becoming

flattened Cyphella cheesmani.

4. Hairs appressed, spirally twisted in the outer half. Cupules tubular, adnate in sheaves.

Spores 3-4 X 2-5-3-5/t Solenia mtalensis.

Hairs, if present, not spirally twisted 5.

5. Spores hyaline 6.

Spores lightly coloured 7.

6. Spores ovate to subglobose, 4 -8-6 -4 // diam., smooth or minutely punctate. Cupules

0-3-1 -5 mm diam., not hairy, pruinose with mineral deposits, greyish-white.. . Cyphella farinacea.

Spores ovate, smooth, somewhat flattened on one side, 3-4 X 2 -5-3 -5 p. Cupules

up to 0-5 mm diam., creamy yellow, with solid, parallel, appressed surface hairs. . Solenia rhoina.

Spores subglobose to broad ovate, smooth, 3-4 x 2-5-3 /<. Cupules 130 p diam.,

250 p long, yellowish when dry, lacking surface hairs Solenia minima.

7. Cupules up to 1 mm high, lacking a subiculum. Hyphae thick-walled, 3-4-8-(6-4) p

wide Cyphella pelargonii.

Cupules up to 600 p high, commonly seated on a reddish-brown hyphal subiculum

which may be scanty or absent. Hyphae thin-walled, 1 -6-2-4 // wide. . . . Cyphella variolosa.

Genus 5. Punctularia Patouillard apud Pat. & Lagerh. in Bull. Herb. Boiss. 3 (1895)

57, pi. 2, Figs. 1, a-g.

Hymenophore resupinate, rarely narrowly reflexed, or dimidiate, subgelatinous,

drying ceraceous to corneous. Hymenium covering discoid to elongated, hemispherical

pulvinae seated upon a common foundation tissue and separated by sterile fissures

containing mineral matter. Basidiospores hyaline to pale coloured, smooth. Dark

dendrophyses present in hymenium. Hyphae subgelatinous, hyaline, with clamps.

Abhymenial hairs dark brown, with clamps. Gloeocystidia present or absent. Conidial

stage often present; conidia coloured, sculptured, formed in an acropetal chain sepa-

rated by narrower, clamped, portions. Hymenial tissues cemented together by brownish

granular material.

1. Mostly resupinate or only narrowly reflexed, usually associated with a prolific conidial

state. Hymenium containing dark dendrophyses composed of a hyaline stem and

an irregularly swollen, shortly branched, coloured apical part up to 5-4 p wide.

Gloeocystidia absent Punctularia tuberculosa.

Resupinate-reflexed to dimidiate, lacking any known conidial state. Hymenium

containing yellow, irregularly branched dendrophyses 1-2 p wide, whose ultimate

branches are thread-like. Yellow gloeocystidia present, 3-5-5 p wide... Punctularia strigoso-

zonata.

Genus 6. Merulius Haller ex Fries, Syst. Myc. 1 (1821) 326.

Fructifications resupinate, eflfuso-reflexed or pileate, fleshy-gelatinous, coriaceous

or membranous. Hymenium at first plane then gyrose-plicate, porose-reticulate, or

sometimes dentate or raduloid, fertile on the edges of the folds or teeth. Hyphae

monomitic or dimitic, coloured or hyaline, with or without clamp connections. Spores

hyaline or coloured, smooth.

1. Hymenium pale to pinkish or orange 2.

Hymenium dark, some shade of yellow or brown 3.

2. Spores oblong-ellipsoid, hyaline to pale yellow, 5-6 X 3-4-5 //. Hyphae abundantly

clamped Merulius moUuscus.

Spores subcylindrical, 2-5-3 5 X 6-7-5 //, hyaline. Hyphae lacking clamps.. Merulius corium.

3. All hyphae thin-walled, hyaline or coloured 4.

Some hyphae thick-walled and coloured, lacking clamps; others hyaline with thin

or slightly thickened walls, with occasional clamps. Hymenium ferruginous-yellow

then darker Merulius lacrymans.

165

4. Hymenium more or less smooth and Dresden Brown colour when dry, changing rapidly

to warm russet and becoming convoluted when moistened. Basal hyphae often

granule-encrusted. Young parts lacking a lilaceous tint Merulius gelatinosiis.

Hymenium convoluted when dry, umber to warm Brussels Brown or darker. No

marked colour change when wetted. Young parts often with a pale lilaceous tint.

Basal hyphae not granule-encrusted MeruHus himantioides.

Genus 7. Coniophora DC. ex Pers., Myc. Eur. 1 (1822) 153; De Candolle FI. Fr.

6 (1815) 34.

Coniophorella Karsten, Finl. Basidsv. (1889) 438.

Fructifications resupinate, pellicular, fleshy or membranous. Hymenium smooth

or somewhat tubercular, usually some shade of yellow or brown. Spores coloured

yellowish to brown, smooth. Hyphae hyaline or coloured, with or without clamp

connections, forming mono- or dimitic systems. Septate cystidia sometimes present.

1. Coloured, septate, encrusted cystidia present. Spores (5)-6-7-(8) x (10)-12-(14) /i

Coniophora olivacea..

Cystidia absent 2.

2. Basal hyphae not encrusted, narrow (±2 p), dark-coloured, the remaining hyphae lightly

coloured or hyaline. Hymenium papillate, dark drab colour. Spores 4-6 x 7 -5-10

/( Coniophora papiUosa.

Basal hyphae not dark, or if slightly so then 3-(6) /t wide, forming a soft spongy abhy-

menial surface 3.

3. Basal hyphae thickly encrusted with minerals, hyaline to pale coloured, inflated then

collapsing. Hymenium light brown to warm sepia or bister. Spores (5 ■ 6)-6 -6-(8) X

(8 ■ 8)-9 • 6-1 1 /< Coniophora incriisiala.

Basal hyphae not encrusted, not inflating then collapsing 4.

4. Fructification fleshy-membranous, easily separable from the substratum. Hymenium

tubercular 5.

Fructification floccose then pellicular-membranous, arid, not readily separable.

Hymenium smooth. Spores 10-13 X 6-7 //. Hyphae pale coloured or hyaline, the

marginal ones with whorled clamps Coniophora arida.

5. [Spores (5-3)-6-5-8 X 9-12-14-7 p. Basal hyphae not specially dark... Coniophora puteana].

Spores S-6-6-6 X 6-6-7-5-8-3-10 /(. Basal hyphae dark, forming a spongy layer,

the rest of the hyphae hyaline Coniophora fodinanan.

Genus 8. Tomentella Patouillard, Hym. d’Eur. (1887) 154.

Hypochnus Fries, Syst. Myc. 3 (1829) 289 emend Karsten, Rev. Myc. 3 (1881) 23.

Fructifications resupinate, soft, floccose to membranous, or adnate and crustose.

Hyphae laxly intertexed, hyaline to coloured, monomitic, with clamps. Hymenium

smooth or papillate. Spores coloured, verrucose, asperulate, angular.

A single species, Tomentella pimicea, recorded from South Africa.

Genus 9. Thelephora Erhart ex Fries, Syst. Myc. 1 (1821) 428, pr. parte; Erhart,

Crypt. Exsicc. No. 178 (1785).

Fructifications effuso-reflexed and encrusting to dimidiate, imbricate, confluent,

sessile or shortly stipitate, sometimes the pileus divided into somewhat clavarioid

branches, membranous to soft coriaceous. Hymenium inferior, or amphigenous in

parts, smooth or somewhat papillate. Spores coloured, echinulate or verrucose,

angular. Hyphae monomitic, with clamp connections, mostly dark coloured.

1 . Fructifications effused, encrusting, resupinate then emitting oblique or suberect flabellate

fascicles of penicillate branches, especially at the margin. Spores4-8-7-2 x 6-4-9-6

p “ Thelephora penicillata ” Lloyd.

Fructifications effuso-reflexed, or dimidiate, imbricate, confluent, sessile or shortly

stipitate. Surface usually strigose. Spores 7-8 X (7)-9-ll p Thelephora terrestrls.

166

Genus 10. Hymenochaete Leveille, Ann. Sci. Nat. ser. iii, 5 (1846) 150.

Fructifications resupinate, reflexed or pileate, usually sessile, rarely stipitate, annual

or perennial, arid-crustose to coriaceous. Hymenium smooth, penetrated by setae.

Hyphae coloured, darkening in colour on application of alkali, dimitic or monomitic,

lacking clamp connections. Spores hyaline, smooth.

1 . Setae found throughout the tissues and commonly arranged in fascicles. Fructification

resupinate Hymenochaete fasciculata.

Setigerous layer(s) differentiated from hyphal layer(s). Setae not fasciculate. Fructifi-

cations resupinate or dimidiate 2.

2. Setae rather scanty, confined to the hymenium 3.

Setae abundant, forming a thick setigerous layer above the rest of the hyphal layers 4.

3. Setae 35-50 X 4-6 fi. Ffymenium Buckthorn Brown to tawny olive (Ridgway).

Tomentum thin. Compact zone differentiated at the base of the trama. Dimidiate

Hymenochaete liiteobadia.

Setae (20)-3 5-40 X (6)-8-10//. Hymenium near Mummy Brown (Ridgway). Tomen-

tum thick. No compact basal zone differentiated. Dimidiate. . . . Hymenochaete nigricans.

4. Resupinate. Setigerous layer(s) ± 260 /< thick, sometimes arranged in strata. Setae

45-70 X 5'6-8 -4/<. Hyphae moniliform in parts Hymenochaete semistupposa.

Usually effuso-reflexed to dimidiate and imbricate, sometimes resupinate. Setigerous

layer ± 160 /« thick. Setae 30-45 x 6-8 /<. Hyphae not at all moniliform. Margin

conspicuous ochre colour. Abhymenial surface soft, rubbing ochraceous Hymenochaete

ochromarginata.

Genus 11. Duportella Patouillard in Philippine Journ. Sci. 10 (1915) 87.

Fructifications resupinate, brown. Hymenium velutinate at first, later becoming

tubercular and cracked. Pseudosetae (modified ends of skeletal hyphae) present.

Gloeocystidia and cystidia present at some stage of development, sometimes not seen.

Hyphae dimitic, almost hyaline to brown, with clamps, not forming a notably differen-

tiated basal layer. Spores hyaline, ellipsoid to subcylindric-curved, smooth.

A single South African species, Duportella tristicula. But Stereum fulvum, which

differs in having no gloeocystidia or cystidia, is possibly congeneric and should be

compared.

Genus 12. Stereum Persoon ex S. F. Gray, A Natural Arrangement of British Plants

1 (1821) 652; Persoon in Roemer Neues Mag. Bot. 1 (1794) 110, Obs. Myc. 1

(1796) 35.

Fructifications coriaceous, membranous or subligneous, sessile, stipitate, effuso-

reflexed or resupinate, or dimidiate, or infundibuliform, simple or branched. Stipe

lateral, central or absent. Hymenium inferior, smooth, sometimes rugose or excep-

tionally tubercular. Context with an intermediate layer of ± horizontally arranged

hyphae. Spores hyaline, smooth. Cystidia, gloeocystidia or vesicles present or absent.

Annual or perennial, lignicolous or terrestrial. Hyphae monomitic or dimitic, usually

hyaline, with or without clamps.

1. Pilei infundibuliform and centrally stipitate 2.

Pilei not infundibuliform nor centrally stipitate 7.

2. With gloeocystidia (some may be interpreted as smooth cystidia) 3.

Without gloeocystidia. Here may be located some rare forms of Stereum lobatum

which are infundibuliform by fusion and thus often partially split down one side.

3. Without surface hairs. Sections do not show any well-marked, denser coloured zone

beneath the abhymenial surface 4.

With (scanty) surface hairs. Sections show a well-marked denser coloured zone beneath

the abhymenial surface. Here are located some unusual forms of Stereum affine

which are only infundibuliform by fusion and are often split down one side.

167

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

Spores small, ovate, broad-elliptic or subglobose, not larger than 3-4 X 4-5 // 5.

Spores larger, in the range of 3-6 X 6-5-9 (jl 6.

Gloeocystidia up to 12-8 X 144 in size Stereum ravenelii.

Gloeocystidia smaller, up to 10-5 X 80 /i Stereum nitidulum.

Gloeocystidia usually rather narrow (3 -2-9 -6 wide) and often hyphoid. Hyphae

of two types, some with clamp connections Stereum thozetii.

Gloeocystidia usually wider (8-9-6-16 wide) and clavate. Hyphae of only one

type and without clamps Stereum diaphanum.

Pilei laterally stipitate, or flabellate or spathulate, or subsessile and cuneate attached

by a markedly reduced base 8.

Pilei dimidiate or effuso-reflexed or resupinate 13.

Fresh pilei bleeding red when bruised. Conducting vessels present in fresh and dried

plants, seen microscopically Stereum australe.

Fresh pilei not bleeding. Conducting vessels absent 9.

Without gloeocystidia or cystidia 10.

With gloeocystidia or cystidia, or both together 11.

Pilei small (up to 1 cm in any direction) soft, whitish, without colour zones on the surface.

Spores pip-shaped or later distorted and angled. Hyphae monomitic. . Stereum cyphelloides.

Pilei large, coriaceous or tough, surface coloured with zones of grey, brown, chestnut.

Spores cylindric-depressed. Hyphae dimitic Stereum lobatum.

With gloeocystidia but no cystidia; Pilei with a definite stipe, flabellate, spathulate

or infundibuliform Stereum affine.

With both gloeocystidia and cystidia 12.

Pilei usually merismatoid, i.e. a compound fructification consisting of a number of

smaller pilei growing together in a bush. Gloeocystidia 7-10 X 40-66 // . . Stereum involutum.

Pilei not merismatoid, but single or sometimes dimidiate or fused laterally. Gloeocys-

tidia smaller, 7-12 x (15)-25-(40) p- Here is located the species represented by

Stereum bellum and Stereum friesii in the sense used by van der Byl.

Without cystidia, cystidioles, gloeocystidia, vesicles or conducting organs (distinguish

carefully between skeletal hyphae which intrude into the hymenium and conductors

or cystidia) 14.

With any of the following organs: cystidia, cystidioles, gloeocystidia, vesicles or con-

ducting vessels (avoid locating here species which have intrusive skeletal hyphae unless

these are much swollen like cystidia at the apex) 18.

Mature pilei small (1 cm or less) soft, whitish, azonate. Spores pip-shaped becoming

angularly distorted. Hyphae monomitic Stereum cyphelloides.

Mature pile! larger, or if immature then either not whitish or possessing more than

one type of hypha 15.

Skeletal hyphae in context brown. The skeletal hyphae which curve up into the

hymenium are brown, rugose or encrusted. Hymenium usually dark-coloured, only

rarely yellowish or light coloured. (Compare also Stereum umbrinum where the

skeletal hyphae are much expanded and resemble cystidia in the hymenium). . Stereum fulvum.

Skeletal hyphae in context not brown, but hyaline or at most only pale straw-coloured.

Hymenium not dark, usually creamy, yellow, orange, fawn or sometimes changing

to cinereous 16.

Pileus with multicoloured zones on the surface, usually flabellate, or if a uniform brown

colour then the pilei are relatively large and flabellate Stereum lobatum.

Pileus without multicoloured zones on the surface, or zoned in shades of brown; usually

smaller than 5. lobatum and effuso-reflexed or dimidiate, not flabellate 17.

Pileus thin, coriaceous, effuso-reflexed or dimidiate with a shortly villose or matted

hairy surface Stereum hirsutum.

Pileus thicker, usually more than 1 mm thick, corky or subligneous, effuso-reflexed,

with a thick padlike tomentum of ochraceous to golden hairs Stereum durbanense.

168

18. Fresh pile! bleeding red when bruised. Fresh or dried specimens possessing conductors

in the hymenium 19.

Fresh pilei not bleeding; lacking conductors 21.

19. Pilei generally dimidiate or cuneate with a reduced base, rarely widely effuso-reflexed.

Hyphae dimitic. Hymenium cinereous smooth Stereum australe.

Pilei mostly resupinate-reflexed, rarely dimidiate 20.

20. Hymenium rimose, i.e. blistered and cracking into small rough areas, yellow, tan or

cinereous. Not on conifers. Hyphae dimitic. Usually more than 700 /r thick

Stereum rimosum var africanum.

Hymenium smooth, not rimose, cinereous to light brown. Occurring on conifers.

Hyphae monomitic. Usually less than 600// thick Stereum sanguinolentum.

21. Species possessing pyriform or subglobose vesicles embedded deep in the trama (some

of the vesicles are sometimes elongated and must be differentiated from gloeocys-

tidia) 22.

Species without vesicles 23.

22. Fructifications more or less resupinate, sometimes narrowly reflexed, stratose with

a veined or marbled subligneous context and a glabrous black abhymenial surface

showing as a black line in wholly resupinate specimens. Hymenium yellowish

Stereum murraii.

Fructifications effuso-reflexed or dimidiate, not stratose or veined, with a hairy brownish

surface. Hymenium purple to purple-brown Stereum purpureum

23 Species with cystidia but lacking gloeocystidia 24.

Species with gloeocystidia and sometimes cystidia as well 25.

24. Cystidia large (12-24 // wide) conical or fusoid, encrusted, hyaline or only dilutely

coloured. Spores averaging 6x11 //. Hymenium light coloured Stereum cinerascens.

Cystidia dark yellow-brown, subhyaline where emergent, actually only apically swollen

and encrusted or rugose skeletal hyphae (rarely smooth at apex). Spores 3-4 x 6-8

//. Hymenium usually umber-brown or purplish, rarely a light sandy brown. (Com-

pare Stereum fulvum, whose skeletal hyphae in the hymenium are less like cystidia,

being not much expanded and roughly cylindrical) Stereum umbrinum.

25. Species with gloeocystidia but no cystidia 26.

Species with both gloeocystidia and cystidia 27.

26. Context pale creamy to pale yellow-brown, usually stratose. Spores subglobose, 6-7 //

diam. (Gloeocystidia sometimes seen with difficulty). Hyphae dichophytic (See

under Asterostromella duriuscula) Stereum duriusculum.

Context brown contrasting with a hyaline hymenial layer. Spores 3-4 -5 x 2-3 //.

Gloeocystidia abundant, sometimes fragmented and refractile like cystidia. Hyphae

capillary, normal Stereum bicolor.

11. Cystidia and gloeocystidia clearly differentiated. Context pale coloured throughout.

Hyphae hyaline. Here may be located the species represented by Stereum bellum and

Stereum friesii in the sense of van der Byl.

Only gloeocystidia present, the older ones fragmented and highly refractile thus resem-

bling cystidia or mineral aggregations. Context brown, contrasting with a hyaline

hymenial layer. Basal hyphae mostly brown Stereum bicolor.

Genus 13. Cymatoderma Junghuhn in Tijdschr. Nat. Gesch. Phys. ed. v. d. Hoeven &

De Vriese 7 (1840) 390.

Cladodeiris Pers. ex Berk, in Hooker Lond. Journ. Bot. 1 (1842) 152.

Fructifications coriaceous, flabellate to infundibuliform, sessile, substipitate or

stipitate. Surface radially sublamellate with scanty hairs or covered with a thick

tomentum. Hymenium radially costate, frequently coarsely papillate. Context with

a basal layer of horizontally arranged hyphae and a middle layer of more or less vertical

ones. Hyphae dimitic, hyaline, with clamp connections. Spores hyaline, smooth.

Gloeocystidia and/or cystidia present.

169

1. Pileus deeply dissected into narrow radiating segments. Hyphae monomitic. Cystidia

and gloeocystidia absent. Basidia with 2-4 sterigmata. Spores 4 -8-6 -4-8 //.

(This is a species of Clavulina or Aphelaria) Cladodenis fiinalis.

Pileus entire to shortly laciniate at the margin. Hyphae dimitic. Cystidia and/or

gloeocystidia present 2.

2. (Gloeocystidia present, cystidia absent. Spores ± subglobose, 2 x 4 or 3-4 /t

diam Cladodenis dendritica).

Both gloeocystidia and cystidia present. Spores ellipsoid, 3-5 x6-10/r.. Cymatoderma elegans.

Genus 14, Asterostromella Hohnel & Litsch., K. Akad. Wiss. Wien. Sitzungsb. 116

(1907) 773.

Fructifications resupinate, annual or perennial (becoming stratose) coriaceous to

subligneous. Hymenium smooth. Hyphae monomitic, hyaline to faintly coloured,

with clamp connections, dichophytic to dendrophytic. Context with a basal layer of

horizontal hyphae and a middle layer of it vertical hyphae. Gloeocystidia present

in most species. Spores smooth or minutely stippled, hyaline or faintly coloured, often

subglobose.

1. Context white or light yellowish; hyphae dendrophytic, hyaline. Sections not darkening

in KOH. Spores 6-6-8 /« diam., subglobose Asterostromella duriuscida (= Scytino-

stroma duriuscidum fide Donk; see p. 160).

Context not light coloured; hyphae dendrophytic, hyaline or very dilutely coloured.

Sections darkening slightly in KOH. Spores elliptic-oval, 5-5 X 8-3 // Astero-

stromella rumpiana. (Referred to Scytinostroma by Donk: p. 160).

Genus 15, Pellicularia Cooke in Grevillea 4 (1876) 116 sensu Rogers in Farlowia

1 (1943) 96.

Botryobasidium Donk in Nederl. Myc. Ver. Med. (1931) 116.

Fructifications resupinate, arachnoid, pellicular or loosely membranous. Hyme-

nium discontinuous. Hyphae monomitic, with or without clamp connections, hyaline

or dilutely coloured, wide, short-celled, branching at right angles. Basidia in terminal

botryose clusters, bearing 4-6-8 sterigmata. Spores smooth or sculptured, hyaline

or pale yellowish. Cystidia or septocystidia present or absent. Saprophytic or

parasitic.

1. Spores rough-walled, ellipsoid, minutely asperulate. Clamps lacking.... Pellicularia asperula.

Spores smooth 2.

2. Encrusted septocystidia present. Hyphae encrusted with wartlike granules, about 6 /«

wide. Spores subglobose to broad ovate-elliptical, 5-7 • 5 X 4-5-6/t... Pellicularia fodinarum.

No septocystidia present. Hyphae not encrusted. Spores otherwise 3.

3. Saprophytic. Spores navicular or asymmetrically fusiform, 9-6-12-8 X 4-5-6 //.

Hyphae up to 13 /< wide Pellicularia vaga.

Parasitic. Spores ellipsoid, flattened on one side, with a truncate apiculus, 8-12 X 5-7

p. Hyphae 5-12-(17) /< wide Pellicularia filamentosa.

Genus 16, Aleurodiscus Rabenhorst ex Schroeter, Krypt.-FI. Schlesien 3 (1888) 429.

Fructifications resupinate, or narrowly reflexed, or pileate, discoid, pezizaeform

or flabellate, coriaceous. Hymenium smooth, pulverulent. Hyphae hyaline, with or

without clamp connections. Basidia usually rather large to very large. Spores hyaline,

smooth or echinate. Simple paraphysoids, dendrophyses, acanthophyses and gloeocys-

tidia frequently present singly or in combination.

1. Both yellowish gloeocystidia and encrusted dendrophyses present together. Amyloid

granules encrusting the dendrophyses. Spores broad elliptic to ovate, 13-5-15 X

7-5-8 p Aleurodiscus botryosus.

Either gloeocystidia or dendrophyses present, or both absent. Spores otherwise 2.

170

2. Spores semilunate, biapiculate, smooth or minutely echinate, about 14 x 23 //. Acan-

thophyses and simple unbranched paraphysoids present Aleurodiscus mirabilis.

Spores otherwise 3.

3. Fructifications whitish or light coloured but not rosy, purplish, violet, lilac or brownish

to clay-coloured 4.

Fructifications rosy, purplish, violet, lilac or brownish to clay-coloured 6.

4. Spores ovate-subglobose, 18-22 x 12-5-16 /<. Simple cylindric paraphysoids and

gloeocystidioid submoniliform paraphysoids present Aleurodiscus disciformis.

Spores not more than 8 // broad. Encrusted dendrophyses and bulbous pseudophyses

with a short to long apical outgrowth present 5.

5. (Spores ovoid, 5-6-6-4 X 8-12 /(, ratio length/width less than 2-0 Aleurodiscus acerinus.)

Spores broad elliptic to ovate, 4-8 X 9-17-6 //, ratio length/width more than 2-0

Aleurodiscus acerinus var alliaceus.

6. Fructifications rose-coloured, paling to a fleshy tint on storage Aleurodiscus roseus.

Fructifications purplish, violet, lilac or partly changing to brown or clay-coloured

Aleurodiscus polygonioides.

Genus 17. Asterostroma Massee in Journ. Linn. Soc. Bot. 25 (1889) 154.

Fructifications resupinate. Hymenium somewhat pulverulent, smooth. Hyphae

hyaline to pale coloured, loosely intertexed, some being modified into asterosetae

which are deep brown in colour. Gloeocystidia commonly present. Spores hyaline,

smooth, or sculptured. Hyphal system monomitic, lacking clamp connections.

A single South African species, Asterostroma cervicolor.

Genus 18. Dendrothele Hohnel & Litsch., in K. Akad. Wiss. Wien Sitzungsb. 116

(1907) 819.

Fructifications resupinate or narrowly reflexed, effused or discoid, membranous.

Hymenium pulverulent, smooth except for emergent hyphal pegs composed of delicate

dendrophytic hyphae which are also present in the hymenium. Hyphae hyaline, with

clamp connections. Spores hyaline, smooth.

A single South African species, Dendrothele duthieae.

Genus 19. Peniophora Cooke in Grevillea 8 (1879) 20.

Fructifications resupinate, usually pellicular to membranous or crustose, occa-

sionally gelatinous-mucoid. Hyphae hyaline, A vertically arranged, with or without

clamp connections. Spores smooth or rarely minutely punctate under oil immersion,

hyaline. Cystidia present, simple or septate, encrusted or smooth. Gloeocystidia

and/or cystidioles present or absent.

] . Context yellow-buff to isabelline colour, turning strong violet colour immediately on

application of KOH solution Peniophora fUamentosa.

No violet colour reaction with KOH solution 2.

2. Texture gelatinous-mucoid, drying to a very thin greyish or bluish vernicose film.

Spores smooth in KOH but minutely punctate in lactic acid under oil immersion lens 3.

Texture otherwise; spores otherwise 4.

3, Cystidioles simple or with a capitate apex composed of a few short lobes. Spores

5_6_(7) X 3-4 n Peniophora pruinosa.

Cystidioles simple or with a unbranched inflated capitate apex. Spores (5 - 6)-6 -4-7-(8 - 8)

X 8-10-(12) // Peniophora rimicola.

4. Cystidia septate, often with clamp connections at the septa Peniophora aspera.

Cystidia not septate 5.

171

5. Cystidia long cylindrical, acicular or subulate, sometimes with a bulbous base, coated

with scale-like, flattened granules. Spores 2 -2-3 -2 X 6-4-8 //, elliptic-fusoid or

somewhat allantoid Peniophora longispora var brachyspora.

Cystidia and spores otherwise 6.

6. Cystidia occupying a relatively narrow zone confined to the hymenium and subhymenium 7.

Cystidia basal or staged throughout the trama, or at least originating in mid-trama, but

not confined to the uppermost layers in a distinct zone 8.

7. Cystidia in a zone 70-100 /< thick above a byssoid trama of thick-walled hyphae.

Cystidia encrusted, fairly thin-walled, 20-36-(40) X 10-13 [i Peniophora pelliculosa.

Cystidia thick-walled, heavily encrusted, 36-80 x 10-11 -5 /t; Tramal hyphae indistinct,

not loose and byssoid Peniophora gigantea.

8. Cystidia smooth, thin-walled or thick-walled 9.

Cystidia encrusted, thick-walled ]0.

9. Cystidia thick-walled, the lumen canalicular below but brusquely dilated at the apex.

Walls of the cystidia soluble in KOH but not in lactic acid. Spores allantoid

Peniophora gracillima.

Cystidia thin-walled, gloeocystidioid. Spores oblong or cylindrical or ellipsoid-

flattened, 3-5 X 7-10 fi Peniophora tenuis.

10. Hyphae distinct, 3-6 /t wide. Spores 2-5-3 x 6-8 //. Cystidia 6-10 x 50-120 p

Peniophora arenata.

Hyphae indistinct, agglutinated, 3 /( wide. Spores 3 X 5 -5/<. Cystidia 10-16 X 50-70

p Peniophora roumeguerii.

Genus 20. Corticium Persoon ex Fries, Epicr. Syst. Myc. (1838) 556; Persoon, Myc.

Eur. 1 (1822) \2%, pr. parte.

Eructifications resupinate, effused, pellicular, byssoid, membranous. Hymenium

smooth. Hyphae monomitic, hyaline, with or without clamp connections, arranged

more or less vertically. Spores smooth (or sculptured), hyaline (or pale coloured).

Cystidia absent. Cystidioles, gloeocystidia, simple paraphysoids present or absent.

Embedded vesicles or ampoule hyphae present or absent.

1 . Species with gloeocystidia 2.

Species with cystidioles 6.

Species without gloeocystidia or cystidioles 7.

2. Gloeocystidia moniliform, originating from tortuous hyphae. Spores cylindrical to

ellipsoid, 7-4-10 X 3-4 -8 p Corticium moniliforme.

Gloeocystidia not moniliform 3.

3. Spores amyloid, minutely roughened under oil immersion, 3-4 x 4-6-7 p. Gloeocy-

stidia frequently bifurcate at the base (but not always), not turning brown in Iodine

solution Corticium porosum.

Spores not amyloid; smooth or finely asperulate 4.

4. Spores finely asperulate, subglobose to broad elliptical, 4-8 x 5-6 p or 4 8-6-4-8 p

diam., often embedded in the context. Texture hypochnoid Corticium punctulatum.

Spores smooth 5.

5. Gloeocystidia numerous, immersed, with bright yellow contents. Spores 4-5 X 9-11-5

p Corticium luteocystidiatum.

Gloeocystidia immersed, hyaline, often capped by amber coloured resinous globules;

spores 3-4 X 7-10 /< Corticium pallidum.

6. Cystidioles rare, immersed or slightly emergent, hyaline, sometimes with a subapical

septum; spores pip-shaped, usually agglutinated, 3-4 X 5 -5-6 5 /<.... Corticium gloeosporum.

1. Fungus bright blue colour throughout Corticium caeruleum.

Fungus not at all blue 8.

172

8. Basal hyphae thick-walled, very distinct, 6-11-5 jx wide. Spores 6 -5-8 -3 x 10-13

Hymenium rosy, paling to flesh colour, smooth or cracked into areoles connected by

whitish subiculum Corticium salmonicolor.

Basal hyphae less than 5 /< wide, may be indistinct 9.

9. Spores large, in the range of 10-16 x 6-8 n 10.

Spores smaller, in the range of 4-5-9 X 3-5 11.

10. Hymenium waxy, smooth, apricot to cinnamon-buff colour. Spores 6-6-7 x 10-16

/< Corticium armeniacum.

Hymenium waxy, whitish, like candle grease when fresh, smooth, seldom cracked,

drying buff to light pinkish-buff. Spores oblong-subspherical, 7-8 X 10-11-5 /<,

with prominent apiculi Corticium confluens.

1 1 . Hymenium cracking into flaky areoles with whitish subiculum, dirty white to light tan

colour. Spores 5-7-(9) X 3 - 5-(5) /t Corticium scutellare.

Hymenium tuberculate, reddish-ochre with liver-brown tubercles and pale margin.

Context dingy yellow. Spores 3-4 - 2 X 4 - 5-5 - 5 Corticium tumulosum.

Species Index of Recorded Sooth African Thelepfioraceae.

In the Index presented below, all the species of the Thelephoraceae that have been

recorded for South Africa are listed alphabetically under their specific epithets, followed

by the genus and author citations. Names printed in capitals are of species accepted

as occuring in South Africa; those printed in small letters are rejected as not occuring

here, or as synonyms, or as dubious records which can neither be confirmed nor denied.

Following each name there is a reference to the paper or papers in Bothalia in

which the writer has described, discussed or annotated the species, each paper being

cited by the volume and page number of this journal. In the case of rejected or dubious

species a brief reason is given for their disposal as such.

abeuns, Corticium Burt: 6, p. 19; 7, p. 143.

Specimen not C. abeuns, but instead is Corticium porosum.

acerina, Thelephora (Pers.) Pers. ex Fr. : 6, p. 62; 7, p. 123.

= Aleurodiscus acerinus or its variety aUiaceus.

acerinum, Stereum (Pers. ex Fr.) Fr. : 7, p. 123.

= Aleurodiscus acerinus or its variety aUiaceus.

acerinus, Aleurodiscus (Pers. ex Fr.) Hohnel & Litsch. : 6, pp. 26, 28, 466. No material traced.

ACERINUS var ALLIACEUS, ALEURODISCUS (Quel.) Bourd. & Galz.: 6, pp. 26, 466.

acerinus var longisporus, Aleurodiscus Hbhn. & Litsch. : 6, pp. 26, 467.

= Aleurodiscus acerinus var aUiaceus.

adnatum, Stereum Lloyd: 6, pp. 39, 45, 304.

= Stereum rimosum var africanum.

AFFINE, STEREUM Lev.: 6, p. 304.

affinis, Punctularia (B. & C.) Talbot: 6, p. 25; 7, p. 140.

= Punctularia tuberculosa.

africana, Sebacina Burt: 7, p. 152.

Type is referable to Peniophora tenuis group,

albobadium, Stereum (Schw. ex Fr.) Fr. : 6, pp. 45, 305.

Material not this species.

ALBOVIOLASCENS, CYPHELLA (Alb. & Schw. ex Fr.) Karst.: 6, p. 471.

amoenum, Stereum Kalchbr. & MacOwan: 6, p. 305.

= Stereum kalchbrenneri ; = Stereum hirsutum.

APPLANATA, CYPHELLA Talbot : 6, p. 472.

ARENATA, PENIOPHORA Talbot: 4, p. 944 ; 6, p. 22.

argillaceum, Corticium Hohnel & Litsch.: 6, p. 13.

nomen nudum; = Corticium pallidum.

ARIDA, CONIOPHORA (Fr.) Karst.: 7, p. 138.

173

ARMENIACUM, CORTICIUM Sacc. : 6, p. 16.

ASPERA, PENIOPHORA (Pers.) Sacc. : 6, p. 21.

ASPERULA, PELLICULARIA Rogers: 7, pp. 137.

atrocinerea, Coniophora (Karst.) Karst.: 7, p. 138.

= Coniophora olivacea.

atrocinerea, Coniophorella Karst.: 6, p. 35.

= Coniophora olivacea.

atrocinerea, Peniophora Massee: 6, pp. 43, 44.

= Stereum fulvum.

atrocinereum, Corticium Kalchbr. : 6, pp. 19, 44.

nomen nudum', = Stereum fulvum.

atrocinereum, Stereum (Massee) van der Byl: 6, pp. 43-45, 306.

= Stereum fulvum.

AUSTRALE, STEREUM Lloyd : 6, p. 306.

australica, Cladoderris Berk, ex Sacc.: 7, p. 121.

= Cymatoderma elegans.

australis Kalchbr. f. minima Bres., Cladoderris: 7, p. 120.

The material is Stereum thozetii.

bellum, Stereum (Kunze) Sacc. : 6, p. 307.

Doubtful as to species. The species is the same as van der Byl’s material of “ Stereum

friesii ”.

betulae, Coniophora (Schum.) Karst.: 6, p. 36; 7, p. 138.

The specimen is the type of Coniophora incrustata.

BICOLOR, STEREUM (Pers. ex Fr.) Fr.: 6, pp. 39, 308.

[Syn. Laxitextum bicolor (Pers. ex Fr.) Lentz: 7, p. 160].

biennis, Thelephora Fr. : 6, pp. 62, 309, 316; 7, p. 123.

MacOwan’s specimens are Stereum bicolor.

bombycinum, Corticium (Sommerf.) Bres.: 6, p. 19; 7, p. 143.

The specimen is Peniophora arenata.

BOTRYOSUS, ALEURODISCUS Burt: 6, p. 467.

bresadoleanum, Stereum Lloyd: 6, p. 309.

= Stereum involutum.

BRINKMANNI, SISTOTREMA (Bres.) J. Eriksson: 7, p. 159.

CAERULEUM, CORTICIUM (Schrad. ex Pers.) Fr.: 6, p. 15.

calceum, Corticium Fr. sensu Romell & Burt: 6, pp. 19, 27.

nomen confusum.

calceum Fr. var lacteum Fr., Corticium: 6, pp. 19, 27.

nomen confusum', MacOwan’s material is the type of Aleurodiscus acerinus var longisponis.

calix, Thelephora Kze. : 7, p. 121.

Specimen corresponds with Cymatoderma elegans.

Candida, Solenia Pers.: 6, p. 480.

Material is the type of Solenia natalensis.

capensis, Aleurocystus Lloyd ex Stevenson & Cash: 6, p. 465.

nomen provisiorum ; = Cytidia habgallae.

capensis, Aleurodiscus Lloyd: 6, p. 468.

= Cytidia habgallae.

capensis, Gloeosoma Lloyd (as ‘ McGinty ’): 6, p. 479.

nomen provisiorum', = Cytidia habgallae.

caperatum, Stereum Lloyd: 6, pp. 309, 340.

= Stereum turgidum; = Stereum cinerascens.

carnosa, Peniophora Burt: 7, p. 147.

Material is Peniophora roumeguerii.

ceraceum, Corticium Berk. & Ravenel ex Massee: 6, pp. 16, 19.

nomen nudum', = Corticium armeniacum.

cerebella, Coniophora Pers. : 7, p. 1 39.

= Coniophora puteana, but the South African specimens are not that.

cerussatus, Aleurodiscus (Bres.) Hbhnel & Litsch.: 6, pp. 28, 468.

Material is Aleurodiscus botryosus.

174

CERVICOLOR, ASTEROSTROMA (B. & C.) Massee: 6, p. 54.

CHEESMANNI, CYPHELLA Massee: 6, p. 472.

CINERASCENS, STEREUM (Schw.) Massee: 6, pp. 40, 309, 339-346.

(= Lopharia cinerascens fide Cunningham: 7, p. 160.)

cinerea, Peniophora (Fr.) Cooke: 6, p. 24; 7, p. 147.

Doubtful. Some specimens are Stereum umbrinum, some Hymenochaete spp., while others

may be Peniophora cinerea in an immature condition,

cinereum, Corticium Pers. ex Fr. : 6, p. 19.

= Peniophora cinerea.

cinereum, Stereum Lev. : 6, p. 310.

Doubtful. Material not traced,

concolor, Stereum Jungh. : 6, p. 310.

= Stereum lobatum (= Stereum ostrea fide Lentz: 7, p. 160.).

CONFLUENS, CORTICIUM (Fr.) Fr.: 6, p. 18.

confluens, Merulius Schw.: 6, pp. 30, 33; 7, p. 152.

Specimens not distinguished from Merulius corium.

CORIUM, MERULIUS (Pers. ex Fr.) Fr.: 6, p. 30.

cornea, Cytidia Lloyd: 6, p. 476.

= Cytidia habgaltae.

corneus, Aleurodiscus Lloyd: 6, p. 468.

= Cytidia habgallae.

CORNUCOPIOIDES, CRATERELLUS (Linn, ex Fr.) Pers.: 7, p. 117.

cremea, Peniophora Bres.: 6, p. 25; 7, p. 147.

Specimens are Peniophora pelliculosa and Corticium gloeosporum.

curreyi, Cyphella B. & Br. : 6, p. 472.

= Cyphella alboviolascens.

CYPHELLOIDES, STEREUM B. & C. : 6, p. 3 1 1 .

decretus, Necator Massee: 6, p. 18; 7, p. 144.

Conidial stage of Corticium salmonicolor.

dendritica, Thelephora Fr. : 7, p. 121.

South African specimen corresponds with Cymatoderma elegans.

DIAPHANUM, STEREUM (Schw.) Cooke ex Sacc.: 6, p. 31 1.

[Syn. Cotylidia diaphana (Schw.) Lentz: 7, p. 160.].

DISCIFORMIS, ALEURODISCUS (DC. ex Fr.) Pat.: 6, pp. 28, 468.

dregeana, Hymenochaete (Berk.) Massee: 6, pp. 57, 344; 7, p. 155, 159.

= Irpex dregeanus.

dregeana, Lopharia (Berk.) Talbot: 6, pp. 57, 344.

= Irpex dregeanus.

dregeanum, Corticium Berk.: 6, pp. 19, 57, 344.

= Irpex dregeanus.

DREGEANUS, IRPEX (Berk.) Talbot: 6, p. 344; 7, p. 159.

DURBANENSE, STEREUM van der Byl: 6, p. 312.

DURIUSCULA, ASTEROSTROMELLA (B. & Br.) Talbot: 6, p. 51.

[Syn. Scytinostroma duriusculum (B. & Br.) Donk: 7, p. 160.].

duriusculum, Stereum Berk. & Br. : 6, pp. 45, 51, 313.

= Asterostromella duriuscula (B. & Br.) Talbot.

DUTHIEAE, DENDROTHELE Talbot : 6, p. 478.

ELEGANS, CYMATODERMA Jungh.: 7, p. 120.

elegans, Stereum Mey.: 6, pp. 45, 313.

Some material is Stereum thozetii, other is Stereum nitidulum.

eylesii, Hypochnus van der Byl: 6, p. 63; 7, pp. 146 155.

Isotype is Corticium punctulatum.

eylesii, Tomentella (v. d. Byl): 6, p. 63; 7, pp. 155.

Isotype is Corticium punctulatum.

FARINACEA, CYPHELLA Kalchbr. & Cooke: 6, p. 472.

fasciatum, Stereum (Schw.) Fr.: 6, p. 314.

Doubtful. Most specimens appear to be Stereum lobatum

(= Stereum ostrea fide Lentz: 7, p. 160.).

175

FASCICULATA, HYMENOCHAETE Talbot: 4, p. 943; 6, p. 48.

FILAMENTOSA, PELLICULARIA (Pat.) Rogers: 7, p. 136.

[Syn. Thanatephorus cucumeris (Frank) Donk.: 7, p. 134.].

FILAMENTOSA, PENIOPHORA (B. & C.) Burt.: 6, p. 23.

FLOCCULENTA, CYTIDIA (Fr.) Hohnel & Litsch. : 6, p. 476.

FODINARUM, CONIOPHORA Talbot: 6, pp. 34, 36, 64.

FODINARUM, PELLICULARIA Talbot & Green: 7, p. 135.

friesii, Cyphella Quel.: 6, pp. 473, 476.

Material too poor for determination,

friesii, Stereum Lev.: 6, p. 314.

Doubtful. Van der Byl’s material under Slereum helium is the same species,

fulva, Hymenochaete Burt: 7, p. 155.

No specimens correspond with this,

fulva, Thelephora Lev.: 6, p. 315; 7, p. 124.

= Stereum fulvum.

fulvodisca, Cyphella Cooke & Massee: 6, p. 473.

= Cyphella variolosa.

FULVUM, STEREUM (Lev.) Saccardo: 6, p. 315.

FUNALIS, CLADODERRIS P. Henn.: 7, p. 122.

Accepted as to species but closer to Clavulina or Aplielaria than to Cladoderris.

fuscoviolascens, Hymenochaete (Mont.) v. d. Byl: 6, p. 50; 7, p. 155,

No specimens in existence.

fuscoviolascens, Thelephora Mont.: 6, p. 62; 7, p. 155.

= Hymenochaete fuscoviolascens.

fuscum, Stereum (Schrad.) Quel.: 6, pp. 39, 40, 45, 316.

= Stereum bicolor.

GELATINOSUS, MERULIUS Lloyd: 6, p. 32.

GIGANTEA, PENIOPHORA (Fr.) Massee: 6, p. 24.

glabrescens, Stereum Berk. & Curt.: 6, p. 316.

Material is Stereum affine.

glebulosa, Peniophora (Bres.) Sacc. & Syd.: 6, p. 25; 7, p. 148.

= Peniophora gracillima.

GLOEOSPORUM, CORTICIUM Talbot: 4, p. 940; 6, p. 14.

GRACILLIMA, PENIOPHORA Ell. & Everh.: 7, p. 147.

HABGALLAE, CYTIDIA (B. & Br.) Martin: 6, p. 477.

HIMANTIOIDES, MERULIUS Fr.: 6, p. 31.

hirsuta, Thelephora (Willd.) Pers. ex Fr. : 7, p. 124.

= Stereum hirsutum.

HIRSUTUM, STEREUM (Willd.) Pers. ex S. F. Gray: 6, p. 316.

hirsutum forma kalchbrenneri, Stereum Simpson & Talbot: 6, p. 317.

nomen nudum\ = Stereum hirsutum.

INCANUM, POROTHELEUM (Kalchbr.) Sacc.: 6, p. 479.

incanum, Stigmatolemma Kalchbr.: 6, p. 481.

= Porotheleum incanum.

incarnata, Peniophora (Pers. ex Fr.) Karst.: 7, p. 148.

Specimen is close to P. incarnata but lacks gloeocystidia and may be undescribed.

INCRUST ATA, CONIOPHORA Talbot : 7, p. 139.

infundibuliformis, Cladoderris (Klotzsch) Fr.: 7, p. 123.

= Cymatoderma elegans.

intybacea, Thelephora Pers. ex Fr.: 7, p. 124.

Specimens accepted as Thelephora terrestris,

INVOLUTUM, STEREUM (Klotzsch) Fr.: 6, p. 317; 7, p. 118.

kalchbrenneri, Stereum Sacc.: 6, p. 319.

= Stereum hirsutum.

laciniata, Thelephora Pers. ex Fr.: 7, p. 124.

= Thelephora terrestris.

176

lacrymans, Merulius (Wulf. ex Fr.) Fr.: 6, pp. 32, 33; 7, p. 152.

Doubtful. Usually confused with Merulius himantioides.

lacteum, Corticium Fr. : 6, p. 19; 7, p. 143.

nomen dubium'. No specimens seen,

laetum, Corticium (Karst.) Pers. : 7, p. 144.

Specimens are Corticium salmonicolor.

laxum, Stereum Lloyd: 6, pp. 45, 319.

= Stereum bicolor.

lirellosa, Lopharia Kalchbr. & MacOwan: 6, pp. 309, 340.

= Stereum cinerascens.

LOBATUM, STEREUM (Kze. ex Fr.) Fr.: 6, p. 319.

(Syn. Stereum ostrea fide Lentz: 7, p. 160.).

lobatum var cinereum, Stereum Lloyd ex Doidge : 6, p. 320.

nomen nudum', = Stereum australe.

LONGISPORA (Pat.) Hbhn. var BRACHYSPORA Talbot & Green, PENIOPHORA: 7, p. 148.

LUTEOBADIA, HYMENOCHAETE (Fr.) Hohnel & Litsch.: 6, p. 50; 7, p. 155.

luteobadium, Stereum Fr. : 6, pp. 45, 320, 323; 7, p. 155.

= Hymenochaete luteobadia. But some specimens seen are Stereum lobatum.

LUTEOCYSTIDIATUM, CORTICIUM Talbot: 4, p. 941; 6, p. 13.

membranaceum, Stereum Fr. : 6, pp. 45, 320.

Material is Stereum fulvum.

MINIMA, SOLENIA Cooke & Phil. : 6, p. 480.

MIRABILIS, ALEURODISCUS (B. & C.) Hohnel : 6, p. 469.

mirabilis, Lopharia (B. & Br.) Pat. : 6, pp. 56, 340.

= Stereum cinerascens.

MOLLUSCUS, MERULIUS Fr.: 7, p. 153.

MONILIFORME, CORTICIUM Talbot: 7, p. 144.

MURRAII, STEREUM (Berk. & Curt.) Burt: 6, p. 320.

murrayi, Stereum (B. & C.) Burt: 6, p. 320.

Orthographic variant of Stereum murraii.

MUSCICOLA, SISTOTREMA(Pers.) Lundell: 7, p. 158.

natalensis, Femsjonia Cooke: 6, p. 476.

A species of Cyphella, but the material is in poor condition.

NATALENSIS, SOLENIA W. B. Cooke: 6, p. 480.

NIGRICANS, HYMENOCHAETE (Lev.) Bres.: 7, p. 156.

nigricans, Stereum Lev. : 7, p. 1 56.

= Hymenochaete nigricans.

NITIDULUM, STEREUM Berk. : 6, p. 321.

notatum, Stereum B. & Br. : 6, p. 322.

Dubious. South African material is probably a young Stereum hirsutum.

nuda, Peniophora (Fr.) Bres.: 6, p. 25; 7, p. 149.

Doubtful. No specimens traced.

nudum, Corticium Fr. : 6, p. 19.

= Peniophora nuda.

ochraceoflavum, Stereum Schw. ex Peck : 6, p. 322.

Specimen is not this species but its true identity is doubtful.

OCHROMARGINATA, HYMENOCHAETE Talbot: 4, p. 944; 6, p. 49.

OLIVACEA, CONIOPHORA (Fr.) Karst.: 6, p. 35.

olivacea, Coniophorella (Fr.) Karst.: 6, p. 35.

= Coniophora olivacea.

OSTREA, STEREUM (Blume & Nees) Fr.: 6, p. 322; 7, p. 160.

No specimens available under this name.

PALLIDUM, CORTICIUM Bres. : 6, p. 13.

palmata, Thelephora (Scop.) Fr. : 7, p. 124.

Material indeterminable but certainly not T. palmata.

PAPILLOSA, CONIOPHORA Talbot: 4, p. 939; 6, p. 33.

177

papyrina, Peniophora (Mont.) Cooke: 7, p. 149.

Doubtful. One specimen is Stereum fulvunv, one other specimen not seen.

pedicellata, Thelephora Schw. : 6, p. 63; 7, p. 125.

The material is a species of Septobasidium but not S. schweinitzii with which T. pedicellata

is synonymous.

PELARGONII, CYPHELLA Kalchbr. : 6, p. 473.

pellicula, Hymenochaete B. & Br. : 6, p. 329.

MacOwan (1055) in Herb. S.A. Mus. No. 34315 is Stereum umbrinum.

pelliculare, Corticium Karst.: 6, p. 19; 7, p. 145.

The material is Peniophora arenata.

PELLICULOSA, PENIOPHORA Talbot: 6, pp. 20, 63.

penicillata, Thelephora Lloyd: 7, p. 125.

Invalid as a later homonym of T. penicillata Fr. The specimens are possibly Thelephora

mollissima.

percome, Stereum B. & Br. 6, pp. 45, 322.

The material is Hymenochaete nigricans.

perlatum, Stereum Berk. : 6, p. 322.

The material is Stereum lobatum (= S. ostrea fide Lentz: 7, p. 160).

pinastri, Merulius (Fr.) Burt: 7, p. 154.

Specimen deformed and indeterminable but does not correspond microscopically with

M. pinastri.

POLYGONIOIDES, ALEURODISCUS (Karst.) Pilat : 6, p. 470.

[Syn. Laeticorticium polygonioides (Karst.) Donk. : 7, p. 160.].

POROSUM, CORTICIUM Berk. &Curt.: 7, p. 145.

portentosum, Corticium Berk. & Curt.: 6, pp. 19, 52.

South African material is Asterostromella duriuscula.

praetermissum, Corticium Karst.

= Peniophora tenuis.

proximum, Stereum Lloyd: 6, p. 322.

= Stereum involutum.

pruinata, Peniophora (B. & C.) Burt: 7, p. 150.

= Stereum pruinatum; but the South African material is different and indeterminable,

pruinatum, Stereum B. & C. : 6, pp. 45, 322; 7, p. 150.

Indeterminable material but not this species.

PRUINOSA, PENIOPHORA (Pat.) Jackson: 7, p. 150.

pulverulenta, Coniophora (Lev.) Massee: 7, pp. 126, 139.

= Thelephora pulverulenta Lev., whose type is a species of Hymenochaete, probably H.

luteobadia.

pulverulentum, Stereum Lev.: 6, p. 323; 7, p. 139.

The type specimen is a species of Hymenochaete, probably H. luteobadia.

punctiformis, Cyphella (Fr.) Karst. : 6, p. 474.

One specimen is a minute agaric. There is no material of the other which was recorded as

C. punctiformis var strigosa.

punctiformis (Fr.) Karst, var strigosa Kalchbr. & Cooke, Cyphella: 6, p. 474. Not accepted as

a good variety; no specimen available.

PUNCTULATUM, CORTICIUM Cooke: 7, pp. 146.

punicea, Thelephora Alb. & Schw. ex Fr. : 6, p. 63; 7, pp. 126, 154.

= Tomentella punicea, but no South African collections in existence,

puniceus, Hypochnus (Alb. & Schw. ex Fr.) Sacc. : 6, p. 63; 7, p. 154.

= Tomentella punicea.

punicea, Tomentella (Alb. & Schw. ex Fr.) Schroet.: 6, p. 63; 7, p. 154.

No South African collections in existence.

PURPUREUM, STEREUM (Pers. ex Fr.) Fr. : 6, p. 323.

pusillum, Stereum Berk. : 6, p. 324.

Some of the material is Stereum thozetii, other is Stereum diaphanum.

puteana, Coniophora (Schum. ex Fr.) Karst.: 6, pp. 34, 36; 7, p. 140.

Most records are Coniophora fodinarum, others are indeterminable. None of C. puteana

yet seen in South Africa.

178

radicans, Stereum (Berk.) Burt: 6, p. 324.

Poor specimen but very doubtful if it matches Stereum radicans.

RAVENELII, STEREUM Berk. &Curt.: 6, p. 325.

retiruga, Lloydella (Cooke) Bres. : 6, p. 43.

= Stereum fulvum.

retirugum, Stereum Cooke: 6, p. 325.

= Stereum fulvum.

RHOINA, SOLENIA W. B. Cooke: 6, p. 481.

RIMICOLA, PENIOPHORA (Karst.) Hdhnel & Litsch.: 7, p. 151.

rimosum, Stereum Berk. : 6, pp. 45, 325.

Material is referred to Stereum rimosum var africanum.

RIMOSUM Berk, var AFRICANUM Talbot, STEREUM: 4, p. 945; 6, pp. 38, 325.

ROSEUS, ALEURODISCUS (Pers. ex Fr.) Hohnel & Litsch.: 6, p. 470.

[Syn. Laeticorticium roseum (Pers. ex Fr.) Donk: 7, p. 160.].

ROUMEGUERII, PENIOPHORA Bres. : 6, p. 22.

rubiginosa, Hymenochaete (Dicks, ex Fr.) Lev.: 6, p. 50; 7, p. 156.

Original specimen not seen; rest of South African specimens referred to Hymenochaete

ochromarginata.

rubiginosum, Stereum Dicks, ex Fr.: 6, pp. 45, 326; 7, p. 156.

= Hymenochaete rubiginosa.

rufus, Merulius Pers. ex Fr. : 7, p. 1 54.

Material is Merulius corium.

rugosLim, Stereum (Pers.) Fr. : 6, p. 326.

Material is mostly Stereum rimosum var africanum. Some specimens are Stereum durbanense.

RUMPIANA, ASTEROSTROMELLA Talbot: 4, p. 939; 6, pp. 53, 54.

(Associated in genus Scytinostroma by Donk: 7, p. 160.).

SALMONICOLOR, CORTICIUM B. & Br.: 6, p. 17.

SANGUINOLENTUM, STEREUM (Alb. & Schw. ex Fr.) Fr. : 6, pp. 37, 326.

schomburgkii, Stereum Berk. : 6, pp. 43, 327.

= Stereum fulvum.

SCUTELLARE, CORTICIUM B. & C.: 6, p. 16.

SEMISTUPPOSA, HYMENOCHAETE Petch: 6, p. 49.

serpens, Merulius Tode ex Fr. : 6, p. 33; 7, p. 154.

No material seen; Doubtful.

setigera, Peniophora (Fr.) Bres. ex Bourd. & Galz.: 6, p. 21.

= Peniophora aspera.

SIMULANS, CYTIDIA Lloyd: 6, p. 478.

sinuans, Thelephora Pers. : 7, p. 126.

No specimens available. Said by Lentz to be a synonym of Stereum frustuiatum.

solani, Corticium (Prill. & Delacr.) Bourd. & Galz.: 7, pp. 136, 146.

= Pellicularia fiiamentosa.

spadiceum, Stereum Fr. : 6, p. 327.

Some material is Stereum australe and some Stereum hirsutum.

spongiosa, Cladoderris Fr. : 7, p. 120.

= Cymatoderma elegans.

spongiosa Fr. var subsessilis Fr., Cladoderris: 7. p. 121.

= Cymatoderma elegans.

squalidus, Merulius Fr. : 6, pp. 31-33.

= Merulius himantioides.

strigoso-zonata, Phlebia (Schw.) Lloyd: 6, p. 28.

= Punctularia strigoso-zonata.

STRIGOSO-ZONATA, PUNCTULARIA (Schw.) Talbot : 7, p. 143.

subpileatum, Stereum Berk. : 6, p. 327.

Material is Stereum durbanense.

tabacina, Cyphella Cooke & Phill.: 6, p. 474.

= Cyphella variolosa.

179

tabacina, Hymenochaete (Sow. ex Fr.) Lev.: 6, p. 327; 7, p. 157.

One specimen not located. The others are Stereum umbrinum and a species of Hymenochaete

which is not H. tabacina.

tabacina (Sow. ex Fr.) Lev. var australis Mont., Hymenochaete: 6, p. 327; 7, p. 157.

= Hymenochaete tabacina.

tabacinum Sow. ex Fr. var australis Mont., Stereum: 6, p. 327.

= Hymenochaete tabacina.

tenebrosum, Stereum Lloyd: 6, p. 327.

= Stereum australe.

TENUIS, PENIOPHORA (Pat.) Massee: 7, p. 151.

tenuissima, Hymenochaete Berk.: 6, p. 50; 7, p. 151.

One specimen is Polyst ictus tabacinus; the other has not been located.

TERRESTRIS, THELEPHORA Erhart ex Fr., 7, p. 126.

THOZETII, STEREUM Berk. : 6, p. 327.

thwaitesii, Cladoderris B. & Br. : 7, pp. 122, 123.

South African material included a species of Favolits and Cymatoderma elegans. True

Cladoderris thwaitesii is a bleeding species of Stereum.

tomentosum, Stereum van der Byl: 6, p. 328.

= Stereum durbanense.

transvaalium, Stereum van der Byl: 6, p. 328.

To be taken either as a nomen confusum or as a synonym of Stereum australe.

tremellosus, Merulius Schrad. ex Fr. : 7, p. 154.

The specimen is Merulius corium.

TRISTICULA, DUPORTELLA (B. & Br.) Reinking: 6, pp. 45-48.

tristicula, Hymenochaete B. & Br. : 6, p. 50; 7, p. 157.

= Duportella tristicula.

TUBERCULOSA, PUNCTULARIA (Pat.) Pat. : 6, p. 25; 7, p. 140.

TUMULOSUM, CORTICIUM Talbot: 4, p. 941 ; 6, p. 17.

turgidum, Stereum Lloyd: 6, pp. 328, 340.

= Stereum cinerascens.

UMBRINUM, STEREUM Berk. & Curt.: 6, pp. 41, 329.

[Syn. Laxitextum crassum (Lev.) Lentz: 7, p. 160. Lopharia vinosa (Berk.) G. H. Cunn.:

7, p. 160.].

umbrinus, Merulius Fr. : 6, pp. 31, 33.

= Merulius himantioides.

VAGA, PELLICULARIA (B. & C.) Rogers ex Linder: 7, pp, 135, 146.

vagum, Corticium B. & C. : 6, p. 19; 7, pp. 135, 146.

= Pellicularia vaga.

vagum, Corticium sensu Burt: 7, pp. 135, 136.

= Pellicularia filamentosa.

vagum var solani, Corticium Burt ex Rolfs: 7, pp. 136, 146.

= Pellicularia filamentosa.

VARIOLOSA, CYPHELLA Kalchbr. : 6, p. 474.

vellereum, Stereum Berk. : 6, p. 329.

Probably only a form of Stereum hirsutum.

vellereus, Irpex B. & Br. : 6, p. 324.

= Irpex dregeanus (Reid differentiates I. vellereus and /. dregeanus: 7, p. 159.).

velutina, Peniophora (DC. ex Fr.) Cooke: 7, p. 152.

The specimen is Peniophora roumeguerii.

versicolor, Stereum (Swartz ex Fr.) Fr.: 6, p. 330.

Dubious. The specimens seen are not Stereum versicolor. One of them is Stereum lobattim.

villosum, Stereum Lev. : 6, p. 330.

= Hymenochaete nigricans.

vitile, Stereum Fr. : 6, p. 330.

No specimen in existence. From the description not unlike Stereum umbrinum.

vorticosum, Stereum Fr. : 6, p. 330.

= Stereum purpureum.

180

EXPLANATION OF THE ILLUSTRATIONS.

The following lettering has been used throughout the illustrations

B = Basidia.

BB = Bulbils.

C = Cystidia.

CO = Conidia.

CP = Conidiophores.

CY = Cystidioles.

D — Dendrophyses.

G — Gloeocystidia.

H = Hyphae.

HA = Habit.

S ^ Basidiospores.

SE ~ Setae.

= Surface hairs.

181

182

do (7^^ ^ ()

%oo8^

^OOO o

Pi t* 50^ —

FiG. 6. — Pellicularia vaga. Fig. 7. — PeUicularia fodinarum. Fig. 8. — Pellicularia

filamentosa.

183

184

Fig. 14. — Punctularia tuberculosa. Type, hyrnenial surface. Figs. 15, 16. — Punctu/aria

tuberculosa. Type. Fig. 17. — Corticium conigenum. Type, basidial. Fig.

18. — C. conigenum. Type, conidia from bark. Fig. 19. — C. conigenum Type,

conidia from culture. Figs. 20-25. — Punctularia tuberculosa (No. 40510). Fig.

22. — Conidia from young culture. Fig. 23. — Conidia from old dried culture.

Fig. 24. — Conidia from bark. Fig. 25. — Conidia and conidiophores in slide

culture.

185

Fig. 26. — Peniophora longispora var brachyspora. Type. Fig. 27. — Peniophora tenuis.

Fig. 28. — Peniophora gracillima. Fig. 29. — Peniophora rimicola. Fig. 30. — Penio-

phora pruinosa.

186

Fig. 31. — Merulius molluscus. Fig. 32. — Odontia hicolor. Fig. 33. — Grandinia rosea.

Fig. 34. — Hymenochacte liiteohadia. Fig. 35. — Hymenochaete nigricans. Fig.

36. — Sistotrema niuscicola.

187

L

Fig. 37. — Pellicularia asperula. Fto. 38. — Sistotrema hrinkmanni.

189

Notes on some Species of Cladophora

from South Africa.

By

R. H. Simons.

Several species of Cladophora have been collected during an investigation of Cape

seaweeds and most have been readily identified by comparison with named specimens

in the Stephenson and Tyson collections of South African seaweeds at the University

of Cape Town. Most of Stephenson’s material was identified by himself, but it includes

some specimens named by overseas authorities. The type specimens of certain species

that have particularly interested me are in the Agardh Herbarium in Lund and are

not available on loan, which has meant falling back largely on literature for checking

identifications. In general, the determination of Cladophoras by these methods has

given little dificulty. But there is confusion in the application of the names C. flagelli-

formis, C. capensis and C. virgata, and the present paper attempts to resolve the difficulty .

The Tyson specimens under these three names were determined by G. F. Papenfuss

in 1939, whereas the Stephenson material was apparently named when collected about

1937. It was obvious that most of Stephenson’s specimens of C. flagelliformis and

C. virgata did not correspond respectively in morphology with specimens so named

by Papenfuss in the Tyson collection. The following paragraph was found in a paper

by Stephenson (1944) and accounts for this discrepancy: “There are two common

species of Cladophora, quite distinct in appearance, which bear these names [C. flagelli-

formis and C. virgata}. One of them is a long, coarse, blue-green species, the other a

shorter, silkier, bright green form. The former has hitherto been known as flagelliformis,

the latter as virgata. It now proves that the two names should be interchanged, so that

the coarse plant becomes virgata, the other flagelliformis ’’. Once the suggested inter-

change of names had been made in Stephenson’s material the discrepancy between this

and Papenfuss’s (Tyson collection) was removed.

This should have resolved any further difficulty in identifying specimens of these

two species, but instead I found that according to the original literature C. flagelliformis

and C. virgata are closely similar, whereas in the material at hand there was a likeness

between plants named C. flagelliformis and some of those named C. capensis, but a

dissimilarity between those named C. flagelliformis and C. virgata.

On comparing the literature on C. flagelliformis, it was found that the plants so

named in the Stephenson and Tyson collections disagreed in several respects with the

original description but agreed with the description given by Levring (1938), thus

suggesting that Levring, Papenfuss and Stephenson (after his correction) had misapplied

the name.

Stephenson stated that C. flagelliformis and C. virgata are “ quite distinct in their

appearance ’’. This is contrary to thie views of Kiitzing (1843), who transferred these

species to Cladophora, and of de Toni (1889, p. 331). Kiitzing related the two species

closely but stated that C. flagelliformis is longer and more slender than C. virgata and de

Toni added that they are “ scarcely distinct ”. Areschoug (1851) goes further and says

they are synonymous. Moreover, Kiitzing and Stephenson disagree as to which is

the longer species, for Stephenson stated that this is C. virgata.

190

If a general comparison of Klitzing’s description of C. flagelliformis is made with

Stephenson’s material under the name C. virgata a conformity is immediately apparent.

Both have shorter and more slender branches in the lower parts and longer flagelliform

branches above. The dilation of all branches towards their apices is characteristic

of both, and cell proportions in comparable parts are essentially the same. The apical

cells in both are obtuse, de Toni (1889) gives the length of C. flagelliformis as 18-20 cm;

plants of Stephenson’s C. virgata are up to 17 cm long. Several specimens of com-

parable material collected by me are more than 20 cm. in height, whereas C. flagelli-

formis sensu Levring has a maximum height of 15 cm. The similarity of C. virgata

sensu Stephenson to the true C. flagelliformis (Suhr) Kiitz. therefore seems obvious.

If Suhr’s (1840) original description of flagelliformis is also considered, one sees its

applicability to Kiitzing’s figure (Kiitzing, 1853) of C. virgata (Ag.) Kiitz. One is

thus led to the hypothesis that Kiitzing’s C. flagelliformis is merely a longer form of

his C. virgata.

My observations of Cladophora plants in which all the branches dilate upwards,

show that there are stunted forms and elongated ones which I am unable to distinguish

on morphological grounds as distinct species. The stunted forms in my opinion are

typical of C. virgata (Ag.) Kiitz., and the elongated forms could be referred to C. flagelli-

formis (Suhr) Kiitz. Apart from Levring’s account of C. flagelliformis I can find nothing

in the earlier descriptions to show any inherent difference between C. virgata and

C. flagelliformis except one of relative size, and Kiitzing’s implication of an associated

distinction in branching. My observations are that the branching habits of smaller

and larger plants show no consistent difference such as that suggested by Kiitzing.

Stephenson’s material named C. flagelliformis on the basis of Levring’s description

diverges too far from earlier descriptions to be accepted as this species; but in my

opinion it is clearly related to Cladophora capensis (Ag.) Kiitz. and the two appear

to me to be included in the range of variation of one species.

The conclusion reached is that C. flagelliformis (Suhr) Kiitz. is a synonym of

C. virgata (Ag.) Kiitz., the latter being the prior name by which the species should be

known. C. flagelliformis sensu Levring is I suggest a synonym of C. capensis (Ag.)

Kiitz.* A third species C. isaacii, identified by an earlier investigator as C. capensis,

is distinguished.

Descriptions are appended of the three species that have been recognised.

(1) Cladophora virgata (Ag.) Kiitz., Sp. Alg. p. 388, 1843.

Conferva virgata Ag., Syst. Alg. p. 119, 1824.

Cladophora flagelliformis (Suhr) Kiitz., Sp. Alg. p. 388, 1843.

Holdfast a rather small knob-like structure which is somewhat flattened at its base to

form a disc up to 3 mm in diameter. The whole is formed from the compacting of the

main filaments at their bases and is surrounded by a hyaline sheath which is continuous

with the walls of the filaments (Fig. la). The compact holdfast is not easily dissected

owing to the presence of the sheath. Erect System, (i) Branching: Several closely

compacted filaments arise from the holdfast and these soon diverge to form elongated

filaments which produce short branches almost throughout their length. These short

branches diverge strongly from the line of the main filaments and are usually naked

(Fig. lb). Elongated branches arise occasionally in the lower parts, but are mostly

* Dr. Papenfuss has kindly confirmed in a private communication that C. virgata (Ag.) Kiitz.

and C. flagelliformis (Suhr) Kiitz. are synonymous. He states that the material seen and described by

Levring was wrongly circulated by Suhr as examples of Conferva flagelliformis Suhr, but is in fact

representative of a new species. At the moment I am not convinced that C. flagelliformis sensu Levring

is a new species.

191

produced in the upper parts; the upper branches may again branch in a di- or tricho-

tomous manner in which case they are all flagelliform (PL 1). Branching throughout

the plants is typically opposite but may be alternate or in whorls of three or four. All

branches dilate towards their apices. Many of the flagelliform branches become

truncate by loss of their terminal portions. Basal cells of branches tend to coalesce

slightly at their origins but never appear to arise from the centre of a cell, except perhaps

near the holdfast, (ii) Cells of the Filaments: Near the holdfast these are clearly defined

and are two to four times longer than broad but may occasionally be as much as six

times as long as their diameter. In the upper parts all the cells have more or less the

same length and breadth (Fig. Ic). The cell contents throughout the plant are dense

and opaque, and specimens regarded as comparable in the field are very dark green

with a bluish shade. Apical cells of branches have obtuse tips and are usually more

or less umbonate (Fig. 2).

(2) Cladophora capensis (Ag.) Kiitz. Sp. Alg. p. 421, 1849.

Conferva capensis Ag., Syst. Alg. p. 118, 1824.

Cladophora ecklonii (Rud.) Kiitz., Phyc. Gen. p. 270, 1843.

Conferva ecklonii Rud., Linnaea 6, p. 180, 1831; Suhr, Flora 17, p. 741, 1834.

Holdfast: This is a matted cushion up to 1-5 cm. in diameter and formed of branched

rhizoids which usually contain chlorophyll. The plant is easily dissected into main

filaments with rhizoids below (Fig. 3). Erect System, (i) Branching: This ranges from

near dichotomy (Fig. 4) to distinct main filaments bearing fascicles of short branches

throughout their length (PI. 2). Generally branching is alternate, but sometimes

the branches arise in whorls of two or more. Longer branches characteristically arise

in the lower parts of the plants, but they may occur in the upper parts in which case

they may become somewhat dilated and frequently dissolute towards their apices

(Fig. 5a). Such dilation frequently results in these branches becoming thicker than the

main filaments in their lower parts. The ultimate branchings are often dichotomous.

Other branches taper gradually towards their apices (Fig. 5b). Basal cells of branches

tend more or less markedly to coalesce with cells of parent filaments (Fig. 5c) especially

in the lower parts of the plants where basal cells of branches appear to originate from

the centre of a cell, (Fig. 5d). All branches tend to be appressed to their parent filaments

(PI. 2). (ii) Cells of the Filaments: Cells near the holdfast are four to six times longer

than they are broad but may be up to ten times longer. In the upper parts of the plants

the cells are two to four times longer than broad but may be almost as broad as they

are long in branches which dilate upwards. Apical cells vary somewhat in shape but

they are usually two to three times longer than broad and either umbonately acute,

especially in dilated branches (Fig. 5e & f), or more or less acuminate (Fig. 6). The

hyaline cap terminating branches is always thick and may be as much as eight times

thicker than the lateral walls. A feature not al\vays apparent is the presence of inflated

cells of a clavate or fusiform shape scattered throughout the length of the plant. Some

branches may consist of one or two such cells (Fig. 7).

Plants of this species may be up to 15 cm high, but if growing in habitats where

they are uncovered at low tide and subject to strong wave action they are generally

shorter with a strong tendency towards the dichotomous habit. Such plants tend to

be more robust and less branched than those growing in pools. Specimens with a

fasciculate habit are reminiscent of Kutzing’s (1854) figures of C. neesiorum, C. humilis

and C. ramosissima, all of which de Toni (1889) regards as synonymous. Cladophora

capensis is, however, coarser than C. neesiorum.

Levring (1938) and Papenfuss (1940) have stated that C. capensis and C. ecklonii

are synonymous. In my opinion, C.flagelliformis sensu Levring is also the same species.

192

Fig.7.

Fig.8.

Fig. 9.

Fig. 1. — C. virgaia lAg.) Kiitz. a. Hyaline disc forming holdfast; b, Short divergeni

naked branchlets; c, Cells near apex of branch.

Fig. 2. — Terminal portion of a branch of C. virgata (Ag.) Kiitz. showing the obtuse

tip of the apical cell.

Fig. 3. — Rhizoids at the base of the main filaments of C. capensis (Ag.) Kiitz.

Fig. 4. — Near dichotomy of branches of C. capensis (Ag.) Kiitz.

Fig. 5. — C. capensis (Ag.) Kiitz. a, Dilated branch with dissolute terminal portion;

b. Tapering branch with subacuminate tip; c. Coalescence of basal cell of branch

with parent filament in upper portion of plant ; d, Basal cell of branch in lower

portion of plant; e and/, Apices of dilated branches.

Fig. 6. — Forms of apical cells in C. capensis (Ag.) Kiitz.

Fig. 7. — Branchlet consisting of a single inflated cell in C. capensis (Ag.) Kiitz.

Fig. 8. — Short divergent branches of C. isaacii sp. nov.

Fig. 9. — Bluntly rounded apices of branches of C. isaacii sp. nov.

194

(3) Cladophora isaacii sp. nov.

Thallus erectus usque ad 40 cm altus, colore olivaceo-viride ; fila primaria inferne

ad 200 ju. lata; ramis elongatis, ordine tertia quoquoversum ramosis; ramulis ultimis

brevioribus, gracilioribus, subdivaricatis usque ad 80 /n latis, quoquoversum vel sub-

secundo exeuntibus; fills omnibus apices versus vix attenuatis; apicibus obtusis,

rotundatis quam muris lateralibus usque ad 3-plo crassioribus incrassatis; cellulis

inferioribus diametro 2-6-plo, superioribus 2-3-plo longioribus.

Holdfast: is formed of tightly interwoven rhizoids with dense contents of a dark

green colour when fresh. The whole is not more than about 4 mm in diameter and is

frequently higher than it is wide. There is no hyaline sheath surrounding the holdfast

but closely appressed decurrent rhizoids arise from the lowest branches of the primary

filaments resulting in some matting together of the main filaments below. Erect System,

(i) Branching: Branches arise more or less spirally or sometimes oppositely and the

ultim.ate branchlets are frequently short and divergent from the parent filaments (Fig. 8).

There is a tendency for branchlets to arise in a secund manner (PI. 3), especially in the

upper parts. Almost all branches taper slightly towards their tips. Main filaments

and primary branches are elongated and end in a comparatively long naked portion

which also tapers slightly to the apex. Some specimens have main filaments which

branch furcately in their ultimate portions. These branches may become somewhat

dilated upwards. Basal cells of elongated primary branches tend to coalesce with the

parent filament and may appear to arise from the centre of a cell, (ii) Cells of the

Filaments: In the basal parts these are four to six times longer than broad, and two to

three times longer than broad higher up. In branches that dilate somewhat, the cells

may be almost as broad as they are long. Such cells have denser contents than other

filament cells and sometimes have become hyaline due to loss of contents, an occurrence

presumably associated with reproduction. The apical cells of branches almost always

have bluntly rounded tips and are scarcely narrower than their subterminal cells

(Fig. 9).

Specimens of this species which I have found occurred typically in pools at about

mid-tide level or higher and have been up to 25 cm in height. Some plants in the Tyson

collection, which I consider to be the same species, are as much as 40 cm high. In the

past this species has been mistaken for C. capensis. Its main differences from the latter

species are given below.

Comparison of the Three Species.

C. virgata is easily identified by its bluish-green colour, its disc holdfast and its

branching habit, the branches usually being opposite or whorled with all branches

dilating upwards and terminating in distinctive umbonately thickened apices. On

the other hand, C. capensis and C. isaacii are much more similar in appearance and

more likely to be confused. The main differences are seen in their branching habits.

In C. capensis the branching is variable but the branchlets tend to be closely appressed

to the parent filaments and taper gradually to rather acuminate apices with relatively

thick end walls. The apical cell becomes more rapidly narrow towards its tip. In

C. isaacii the shorter branches diverge from the parent filaments and all branches taper

very little to bluntly rounded apices, whose end walls are thin in comparison with those

of C. capensis. Moreover the apical cells of the latter species are longer relative to

their breadth than those of C. isaacii. Inflated cells, scattered throughout the length

of the plant, are common in C. capensis but have not yet been observed in C. isaacii.

Plants of C. capensis are coarser than those of C. isaacii in the upper parts (cfr. Plates

2 & 3) and are generally much more rigid. The colour of the latter species is an olive

green whereas C. capensis is a bright green. A further distinction is apparent in that

C. capensis in drying sticks to paper only slightly but C. isaacii adheres well.

195

SUMMARY.

It is shown that C. virgata (Ag.) Kiitz. and C. flageUiformis (Suhr) Kutz. are synony-

mous, the former being the correct name. The account of C. flageUiformis by Levring

(1938) is discussed and shown to refer to material not conforming with this species.

It is suggested that C. flageUiformis in the sense used bylLevring, Stephenson andPapen-

fuss (Tyson collection) is synonymous with C. capensis (Ag.) Kiitz. A third species,

C. isaacii, is discussed and described as new.

LITERATURE CITED.

Areschoug, J. E. (1851) “ Phyceae Capensis ”, Venia Amplis Philos. Facult. Upsalien-

sis. p. 9.

DE Toni, J. B. (1889) “ Sylloge Algarum ”, Vol. 1, p. 331.

Kiitzing, F. (1849) “ Species Algarum ”, p. 421, Leipzig.

Kiitzing, F. (1853) “ Tabulae Phyceae ”, T. 77.

Kiitzing, F. (1854) “ Tabulae Phyceae ”, T. 4, 5.

Levring, T. (1938) “ Verzeichnis Einiger Chlorophyceen und Phaeophyceen von

Siid-Afrika ”, Lunds Univ. Arsskr. N. F. Avd. 2, 34, 9.

Stephenson, T. A. S. (1944) “The Constitution of the Intertidal Fauna and Flora of S.

Africa, Pt. ii ”, Ann. Natal Mus. 10, p. 341.

Suhr, J. H. VON (1840) “ Beitrage zur Algenkunde”, Flora, 23, p. 294.

197

Plate 1. — Habit of upper portion of C. virgata (Ag.) Kiitz., showing elongated flagelli-

form branches.

198

Plate 2. — Habit of some specimens of C. capensis (Ag.) Kiitz. showing several main

filaments bearing fascicles of short branches throughout their length.

199

Plate 3.— Habit of C. isoacii, showing secund arrangement of branchlets in upper

parts.

201

Notes on Aristothamnion purpuriferum

(Kiitz.) J. Ag.

By

R. H. Simons.

INTRODUCTION.

This South African seaweed is listed by Stephenson (8), and known by local phycolo-

gists, as Pleonosporium purpuriferum (Kiitz.) de Toni. In Kylin’s (6) recent publication

on the genera of the Rhodophyceae it appears as Aristothamnion purpuriferum, by

which name I shall discuss it. The purpose of the present paper is to record the

occurrence of tetrasporangia and to discuss this plant’s taxonomy.

MORPHOLOGY.

A full description of the morphology of A. purpuriferum has been given by West-

brook (9) who regarded it as a species of Callithamnion. For purposes of clarity its

salient features are given here.

It is one of the species of Ceramiaceae with a well developed cortex made up of

decurrent filaments arising at the lower ends of branches (Fig. 1). The ultimate

branches are always uncorticated and two types can be distinguished — those of limited

growth and those of unlimited growth. The apices of branches of unlimited growth

have corymbosely clustered and inflexed branchlets (Fig. 2), but branches of limited

growth have branchlets which spread and rapidly taper to acute apices (Fig. 3). The

plants are typically dioecious with a separate sporophytic phase. Female organs are

produced laterally on intercalary cells near the apices of branches of unlimited growth,

whereas male organs occur laterally on intercalary cells of branches of limited growth.

Both poly- and tetrasporangia have been recorded as lateral structures on intercalary

cells of branches of limited growth.

OBSERVATIONS.

In the many male and female plants I have examined the reproductive organs

occurred in the positions noted above. On certain plants I have found polysporangia

also occupying the position described for them. Such plants have borne no other

type of reproductive structure. The only tetrasporangia observed have occupied an

unexpected position. These arise on occasional plants bearing numerous female

structures in all stages of development up to maturity and appearing therefore to be

female plants. In these specimens the apical cell of a branch of unlimited growth is

occasionally replaced by a subglobose swelling containing four tetrahedrally arranged

spores (Fig. 4). The wall of such a sporangium is very thick and a distinct pit-connection

is visible between the subtending vegetative cell and the spore group. These sporangia

are nearly twice as wide as their subtending cells, but narrower than a mature poly-

sporangium (Fig. 5). Tetrasporangia have a diameter of 25-35 /x; an average sized

polysporangium is 50 p. in diameter.

202

Fig.4

203

Further sporangium-like bodies with undivided contents are encountered in

exactly similar positions to the tetrasporangia described above, but whether they are

mature monosporangia or immature tetrasporangia I have not been able to ascertain.

They appear to be more numerous than tetrasporangia but nevertheless occur only

occasionally. Cystocarps have been observed arising lower down on branches ter-

minating in either of these sporangial structures.

RECORDS OF “ABNORMAL” TETRASPORANGIA AND THEIR SIGNIFI-

CANCE IN THE CERAMIACEAE.

The occurrence of the tetrasporangia is “ abnormal ” in that they are produced

on apparently female plants. Records of such occurrence are not uncommon in the

Ceramiaceae, particularly in species of Callithamnion. Westbrook (10) records a single

tetrasporangium arising terminally on a branch of a female plant of C. tetricum, an

“ abnormality ” apparently similar to that now recorded for A. purpuriferum. The

normal position of tetrasporangia of C. tetricum is on separate sporophytic plants

where they are borne laterally on branches of limited growth, as are the polysporangia

of A. purpuriferum. Westbrook did not investigate the cytology of the “ abnormal ”

sporangia on C. tetricum but seemed to think that no reduction division accompanied

the formation of their spores. She referred, however, to an account of C. brachiatum

by Mathias (7) of similar “ abnormal ” sporangia in which there was meiosis. Drew

(Fritsch 5) found reduction division taking place in the formation of tetraspores produced

on plants of Spermothamnion turneri — a. related species — ^bearing functional female

organs. It is possible that conclusions similar to Drew’s may result from cytological

study of the “ abnormal ” sporangia of C. tetricum and A. purpuriferum.

The significance of the occurrence of reduction division in the production of such

“ abnormal ” spores is that the plants bearing them are diploid and the carpospores

resulting from the fertilisation of procarps arising on the same plants should be triploid.

Drew found evidence of triploid carpospores developing on plants of S. turneri which

also bore tetrasporangia. It is obvious that until the cytology of the “ abnormal ”

sporangia in A. purpuriferum is investigated no conclusions may be drawn as to whether

the plants producing them are haploid or diploid.

OCCURRENCE OF TETRASPORES IN A. PURPURIFERUM IN RELATION

TO ITS TAXONOMY.

There are apparently only two other references to the occurrence of tetrasporangia

in A. purpuriferum both of which seem attributable to de Toni (3). Baardseth (2)

questions the validity of de Toni’s claim that tetrasporangia are produced by this plant;

Fig. 1 . — Origin of cortical filaments.

Fig. 2. — Ultimate branchlets of branch of unlimited growth.

Fig. 3. — Ultimate branchlets of branch of limited growth.

Fig. 4. — Tetrasporangia terminating ultimate branchlets.

Fig. 5. — Mature polysporangium above and developing polysporangium below on a

cell of a branch of limited growth.

204

he could not find tetrasporangia in material he examined, and noted that this had also

been the experience of Papenfuss, with whom he had communicated privately on this

matter. Westbrook (9) also mentions the occurrence of tetrasporangia in this species

but she did not find any in the plants she examined. Since she did not name her authority

for this statement it seems likely that she accepted de Toni’s record. Thus the evidence

of most investigators of this plant is that normal tetrasporangia are not produced.

The absence of normal tetrasporangia in this species is taxonomically important

because this was the character used originally by Agardh to separate the genus

Aristothamnion from Callithamnion. Sporophytic plants of species of CaUithamnion

characteristically produce tetrasporangia in the asexual phase, whereas in Aristothamnion

it is only poly sporangia that are produced by plants of this phase. In all other respects

the two genera are indistinguishable. Apparently de Toni regarded the production of

polysporangia in members of the Ceramiaceae as taxonomically fundamental and

placed all species with this feature in his genus Pleonosporium. Westbrook (9), however,

did not share this view but placed more emphasis on the characters of male and female

structures. Because the position and structure of the male and female organs are

identical in A. purpuriferum and Callithamnion spp., but quite unlike those of Pleonos-

porium spp., Westbrook placed A. purpuriferum in the genus Callithamnion.

Feldman-Mazoyer (4) in her treatise on the Ceramiaceae of the Mediterranean,

concluded that the structure and development of the female organs were the only

reliable characters indicating relationships and leading to a natural classification within

the family. This is partly in agreement with Westbrook’s view and seems the most

logical. Her conclusion as to the relationship of A. purpuriferum with Callithamnion

was therefore justified. But she was under the impression that normal tetrasporangia

were produced by this plant and there seemed no reason for separating it from the

genus Callithamnion as was proposed by Agardh (1).

CONCLUSIONS.

My record of tetrasporangia in A. purpuriferum is the first that has been made

since de Toni’s (3). The cytology of these structures has not been adequately investi-

gated but they are “ abnormal ” in that they occur on plants bearing functional female

organs. It would seem from the evidence of most investigators that sporophytic plants

produce only polysporangia, and this character remains a point of difference between

the genera Aristothamnion and Callithamnion. Although polysporangia are recorded

for various species of Callithamnion they are accompanied by tetrasporangia (Fritsch 5,

p. 729). Even if the tetrasporangia that I have described above should prove to arise

on diploid plants forming a sporophytic phase of this species, their situation is not

typical of species of Callithamnion. In this connection it should be noted that the

tetrasporangia recorded by Mathias (7) as arising on sexual plants of C. brachiatum

occurred in the lateral position normal for this genus. Also, Westbrook (10) found

lateral tetrasporangia on the same female plant of C. tetricum which bore a single

terminal tetrasporangium. It therefore seems appropriate at this stage to retain the

name Aristothamnion purpuriferum (Kiitz.) J. Ag. for the present species.

SUMMARY.

A new record of tetrasporangia in Aristothamnion purpuriferum is given. These

tetrasporangia are abnormal in that they occur on individuals bearing mature female

organs. Their morphological and taxonomic significance is discussed and it is concluded

that in the present state of our knowledge this plant should be known as A. purpuri-

ferum (Kiitz.) J. Ag.

205

LITERATURE CITED.

1. Agardh, J. G Anal. Alg. ii, 44, 1892.

2. Baardseth, E The Marine Algae of Tristan da Cunha. Norske Vidensk.

Akad., Oslo 95, 1941.

3. DE Toni, J. B Syll. Alg. vi, 469, 1924.

4. Feldman-Mazoyer, G Recherches sur les Ceramiacees de la Mediterranee Occiden-

tale. Alger. 237, 1940.

5. Fritsch, F. E The Structure and Reproduction of the Algae, li, 723, 1945.

6. Kylin, H Die Gattungen der Rhodophyceen. Lund. 383, 1954.

7. Mathias, W. T The Cytology of Callithamnion brachiatum Bonnem. Publ.

of Hartley Bot. Lab. 5, 20, 1928.

8. Stephenson, T. A The Constitution of the Intertidal Fauna and Flora of South

Africa. Pt. iii. Ann. Natal Mus. xi, 307, 1948.

9. Westbrook, M. A Callithamnion pwpiiriferum J. G. Ag. Jour. Bot., 65, 161, 1927.

10. Westbrook, M. A Callithamnion tetricum (Dillw.) Ag. Jour. Bot., 68, 193,

1930.

207

Virus Diseases of Lupins.

By

Patricia J. Klesser.

Although virus infections of lupins have been kno\vn in South Africa for several

years, no previous attempt has been made to identify the causal organisms. A prelimi-

nary report was published in 1953 (3).

REVIEW OF LITERATURE.

The first record of viroses of lupins came from Germany in 1929 when Merkel (6)

reported a pock mosaic and a speckle mosaic of yellow lupins.

Mastenbroek (5) described in detail what he considered to be the same virus.

It was found to be seedborne in yellow lupin and only rarely transmissible to beans.

He named it Lupinus virus 1 .

In 1934, Neill, Brien and Chamberlain (7) described “ Sore-shin ” in New Zealand,

and Richter (9) found an unidentified disease in Germany, both of which were later

proved to be caused by the pea mosaic virus (1). Further work on this virus in lupins

was done by Norris (8).

Kohler (4) reported the Lupinenbraiine virus, later found to be the cucumber

mosaic virus. In 1936 Spierenberg (10) in Holland, described a disorder, also later

proved to be cucumber mosaic.

Weimer (11) described two viruses found in L. angustifolius, but does not confirm

their identity. Neither of them is sap transmissible to lupins.

METHODS AND MATERIALS.

Plants with suspected virus infections were collected from different parts of the

Union.

The standard test plants used were: Phaseolus vulgaris var. Canadian Wonder,

Vida faba and Pisuni sativum var. Greenfeast. In addition, about 30 other legumes,

and some plants belonging to the Solanaceae were used in susceptibility tests.

Aphis craccivora was used for the insect transmission tests, but it is not necessarily

the natural vector.

Mechanical sap transmission was aided by the use of carborundum powder, and

the methods of Johnson and Grant (2) were used for the physical property tests.

This report deals with five viruses, and at the end of each section, there is a dis-

cussion on the relationship of the viruses concerned, and names are suggested for the

new ones. These names are only tentatively proposed as no proof of virus identity

is available beyond that of symptom and physical property resemblances.

208

VIRUSES FOUND OCCURRING NATURALLY ON LUPINS.

Host Plant. Virus.

Lupinus affinis Lupin virus A.

Pea mosaic virus 4.

L. albus Lupin virus A.

Lupin virus B.

Lupin virus C.

Bean yellow mosaic virus.

Pea mosaic virus 4.

L. angustifolius Lupin virus A.

Lupin virus B.

Lupin virus C.

Bean yellow mosaic virus.

Pea mosaic virus 4.

L. luteus Lupin virus A.

Pea mosaic virus 4.

L. mutabilis Lupin virus A.

Lupins are also susceptible to most other legume viruses when inoculated artificially.

With the spotted wilt virus they develop a local necrotic reaction. (See report on pea

virus diseases.)

SYMPTOMS FOUND ON LUPINS WHEN NATURALLY INFECTED.

Lupinus affinis with:

1 . Lupin virus A.

The leaves are chlorotic and malformed, and rosetted. The plant is

stunted.

2. Pea mosaic virus 4.

The leaves are chlorotic with necrosis setting in, and they are much reduced

in size. There are necrotic stem streaks. The plant is stunted.

Lupinus albus with:

1. Lupin virus A.

A top necrosis usually causes the collapse of the growing point. Secondary

shoots have chlorotic, malformed leaves and they are rosetted and stunted.

2. Lupin virus B.

Most leaves are small, malformed and chlorotic. Many drop and there may

be a complete defoliation. There are necrotic stem streaks.

3. Lupin virus C.

The leaves have chlorotic spots and possibly necrotic specks. They remain

folded and have wavy margins. The plant is stunted.

4. Bean yellow mosaic virus.

The leaves are mottled, they remain folded, and have wavy margins and a

general puckering. There is a severe rosette and the plant is stunted. (Fig. 4a).

5. Pea mosaic virus 4.

There is a vein clearing, chlorotic ringspotting or mottle. The plant may

be rosetted.

209

Lupinus angustifolius with :

1. Lupin virus A.

The leaves are chlorotic and malformed, and the plant is stunted and

rosetted.

2. Lupin virus B.

The leaves are small and have necrotic specks. Most leaves drop. There

are necrotic stem streaks.

3. Lupin virus C.

The leaves have a chlorotic speck mosaic and they are very small and

crinkled. The plant is extremely rosetted and stunted. (Fig. 3a).

4. Bean yellow mosaic virus.

The leaves are mottled and there may be necrotic stem streaks which

cause the collapse of the growing point. The leaves of the secondary shoots

are vividly mottled, rolled and malformed.

5. Pea mosaic virus 4.

There is a mottle on the leaves with possibly a veinal necrosis which results

in the collapse of the plant.

Lupinus luteus with :

1 . Lupin virus A.

The leaves are chlorotic and malformed and the plant is rosetted and

stunted.

2. Lupin virus B.

The leaves are mottled and puckered and the plant is rosetted.

Lupinus mutabilis with :

1. Lupin virus A.

There is a general chlorosis of the leaves which are also malformed, and

rosetted. The plant is stunted.

I. Lupin Virus A.

Physical properties; Thermal inactivation point, 58-60°C. Longevity in vitro,

2-3 days. Dilution end point, 1 : 1000.

Transmission: Mechanical sap inoculation. Aphis craccivora.

Host range; Cicer arietinum L., Crotalaria juncea L., C. spectabilis Roth., Glycine

max (L.) Merr., Lathyrus odoratus L., Lupinus albus L. var. Sweet, L. albus L. var.

Bitter, L. angustifolius L., L. luteus L., L. mutabilis L., Medicago lupulina L., Melilotus

officinalis Willd., Phaseolus acutifolius Gray var. latifolius Freem., P. lunatus L.,

P. vulgaris L., Pisum sativum L., Trifolium hybridum L., T. incarnatum L., T. pratense L.,

T. repens L., Vida faba L., Vigna unguiculata (L.) Walp., Voandzeia subterranea Thouars.

Reaction of Susceptible Species.

Cicer arietinum.

Local. No reaction.

Systemic. The young leaves develop a few small chlorotic spots.

210

Crotalaria juncea.

Local. Small necrotic specks develop.

Systemic. After a vein clearing in 24 days the leaves show an irregular chlorosis

with dark green islands. The plant is stunted. Several young plants collapsed after

a severe necrosis.

C. spectabilis.

Local. In 11 days chlorotic spots with a small central necrotic speck develop.

These leaves drop.

Glycine max.

Local. No reaction.

Systemic. A diffuse chlorotic mottle develops.

Lathyrus odoratus.

Local. No reaction.

Systemic. In 21 days there is a vein clearing of the young leaves followed by a

streak mottle of leaves which are small and puckered.

Lupinus albus var. Sweet.

Local. There are chlorotic spots in 1 1 days and the veins may be necrotic in 18 days.

Systemic. There may be a vein clearing soon followed by a necrotic speckling

which spreads and results in the collapse of the growing point. Secondary shoots

have small leaves which are malformed with crinkled blistered dark green margins

and a long narrow base (Fig. lb.)

L. albus var. Bitter.

Local. No reaction.

Systemic. Necrosis sets in, in 1 1 days and the plant collapses.

L. angustifolius.

Local. No reaction.

Systemic. The young leaves may show a chlorotic spot mottle and there is a slight

malformation with irregular margins. Later necrosis sets in, in the growing point and

the plant collapses within three weeks (Fig. Ic).

L. luteus.

Local. No reaction.

Systemic. After 10 days the young leaves develop a chlorotic spotting and later

a chlorotic mottle. The leaves remain folded and are almost stringlike and malformed.

New leaves are small and rosetted (Fig. la).

Medicago lupulina.

Local. No reaction.

Systemic. After a vein clearing there is a chlorotic spot mottle and slight puckering.

Melilotus officinalis.

Local. No reaction.

Systemic. There are streaks which are almost yellow and ringspots which may

become necrotic and cause a malformation.

211

Phaseolus acutifolius.

Local. In 7 days the leaves have chlorotic veins.

Systemic. In 12 days there is a vein clearing of the young leaves. The next leaves

to develop are mottled, and still later ones have yellow specks.

P. lunatus.

A symptomless carrier.

P. vulgaris var. Canadian Wonder.

Local. No reaction.

Systemic. The first and second trifoliates have large chlorotic blotches and are

reflexed. After 14 days a chlorotic spotting develops on the young leaves which are

also rolled. Later leaves develop a mottle with short dark green veinbands. There

may be a slight malformation and puckering.

var. Haricot.

Local. In seven to eight days there is a reflexing and the leaves are generally

chlorotic.

Systemic. Similar to Canadian Wonder.

vars. Long Tom, Tendergreen and Victor show a reaction similar to Canadian

Wonder.

vars. Black Wonder, Idaho Refugee and S.A. Black and White develop

large local chlorotic blotches. There is also a severe puckering of the leaves and the

plant is stunted.

Pisum sativum.

Local. No reaction.

Systemic. A chlorotic spotting of the young leaves is followed by a mosaic

with short dark green veinbands. Later leaves show a marked white vein etching,

and they are small. The tendrils are abnormally curled.

Trifolium hybridum.

Local. No reaction.

Systemic. After three weeks a vein clearing develops into alternate chlorotic and

green streaks following the veins.

T. incarnatum.

Local. Small chlorotic spots develop.

Systemic. In 18 to 19 days the young leaves develop chlorotic spots. Later leaves

have a chlorotic mosaic and are crinkled.

T. pratense.

A symptomless carrier.

T. repens.

A symptomless carrier.

Vicia faba.

Local. No reaction.

Systemic. In 14 days there are chlorotic spots, and a clearing of the veins. Later

leaves are mottled.

212

Vigna mguiculata.

Local. No reaction.

Systemic. A diffuse chlorotic network may develop. The virus can be recovered if

symptomless.

Voandzeia subterranea.

Local. No reaction.

Systemic. There may be a diffuse chlorotic spotting, but the virus can be recovered

if symptomless.

Natural source of virus : Lupinus ajfinis (Pretoria district), L. albus, L. angustifolius,

L. luteus, and L. mutabilis (Stellenbosch district).

Identification.

Because of the characteristic reflexing of the bean leaves, this virus is compared

with others showing this symptom.

Bean yellow mosaic virus Pierce causes a local necrosis on pea, is non-infectious

to cowpea and has a shorter ageing period. Other details are similar, although the host

range differs.

A necrotic strain of bean yellow mosaic virus Klesser, has similar physical properties,

but induces a local and systemic necrosis of the pea and local necrosis on the cowpea.

Alsike clover mosaic virus 2 Zaumeyer gives a similar reaction on pea, but differs

on other hosts.

Osborn’s pea virus 2 differs in physical properties, host range and some symptoms.

As no complete correlation can be found, this is considered a new virus, related

to the bean yellow mosaic group.

It is named lupin virus A.

2. Lupin Virus B.

Physical properties: Thermal inactivation point, 62-65 °C. Longevity in vitro, 2-3

days. Dilution end point, 1 : 100.

Transmission: Mechanical sap inoculation. Aphis craccivora.

Host range: Cicer arietinum L., Crotalaria juncea L., C. spectabilis Roth., Glycine

max (L.) Merr., Lupinus albus L. var. Sweet, L. albus L. var. Bitter, L. angustifolius L.,

L. luteus L., L. mutabilis L., Medicago lupulina L., M. sativa L., Melilotus officinalis

Willd., Phaseolus acutifolius Gray var. latifolius Freem., P. lunatus L., P. vulgaris L.,

Pisum sativum L., Trifolium hybridum L., T. incarnatum L., T. pratense L., T. repens L.,

Vida faba L., Vigna unguiculata (L.) Walp.

Nicotiana glutinosa L., N. tabacum L., Solanum capsicum L.

Reaction of Susceptible Species.

Cicer arietinum.

Local. Chlorotic spots with necrotic rings develop.

Systemic. After a vein clearing of the young leaves, later ones are long, narrow and

malformed.

Crotalaria juncea.

Local. There are necrotic spots and veins.

Systemic. No reaction.

213

C. spectabilis.

Local. No reaction.

Systemic. In 11 days there is a clearing of the veins followed by small chlorotic

spots, and later, a mottle. Necrosis sets in along the veins causing a puckering of the

midrib and malformation. The plant may be severely stunted.

Glycine max.

Local. No reaction.

Systemic. There are chlorotic flecks on the younger leaves.

Lupinus albus var. Sweet.

Local. In nine days there are chlorotic spots and the leaflets collapse.

Systemic. At the same time chlorotic specks develop. On later leaves there is a

general chlorosis leaving dark green blisters. The leaflets remain folded and are mal-

formed (Fig. 2b). After a speck necrosis the plant is almost completely defoliated.

L. albus var. Bitter.

Local. No reaction.

Systemic. In 14 days there are chlorotic spots on the young leaves. Later leaves

are mottled and stringlike. There is a slight necrosis and the leaves drop.

L. angustifolius.

Local. After a general necrosis the leaves collapse.

Systemic. There is a veinal necrosis of the young leaves, soon followed by necrotic

stem streaks, and collapse of the plant (Fig. 2d).

L. luteus.

Local. No reaction.

Systemic. After 10 days there is a vein clearing of the young leaves, which remain

folded, with wavy margins. Later leaves are stringlike and chlorotic and the plant is

rosetted and stunted (Fig. 2a).

L. mutabilis.

Local. No reaction.

Systemic. Most leaves have irregular chlorotic areas.

Medicago lupulina.

Local. No reaction.

Systemic. A veinal chlorosis of the young leaves develops in 14 days. This becomes

general, leaving dark green veinbands. The leaves are reduced, crinkled and puckered

along the midrib.

M. saliva.

Local. No reaction.

Systemic. There are isolated chlorotic spots.

Melilotus officinalis.

Local. No reaction.

Systemic. Only a diffuse chlorotic mottle develops.

214

Phaseolus acutifolius.

Local. In seven days there are necrotic spots and veins on chlorotic leaves.

Systemic. After 12 days there is a vein clearing of the young leaves, and a yellow

mottle and malformation of those formed later.

P. lunatus.

Local. No reaction.

Systemic. In 10 days there is a chlorotic network and later a speck mottle. Necrosis

may set in, and the leaves become malformed.

P. vulgaris var. Canadian Wonder.

Local. Chlorotic spots with necrotic rings develop in 1 1 days (Fig. 2c.)

Systemic. There are only isolated chlorotic spots which become a diffuse mottle

on later formed leaves. The pods are mottled.

var. Haricot.

Local. There are necrotic specks on chlorotic leaves in seven days.

Systemic. The leaves show a chlorotic mottle and they are elongated. The pods

are mottled.

Pisum sativum.

Local. After a general necrosis the leaves collapse.

Systemic. There is a vein chlorosis in six days followed by a mottle and sometimes

a marked vein etching. The leaves are crinkled and rosetted. The plant is stunted and

collapses from necrosis. The tendrils are tightly curled and twisted.

Trifolium hybridum.

Local. No reaction.

Systemic. In nine days there are chlorotic vein streaks and later a mild streak

mottle on all leaves.

T. incarnatum.

Local. There may be necrotic specks.

Systemic. In two weeks the young leaves show a vein clearing. Later leaves have

an almost white mosaic and are crinkled and rolled.

T. pratense.

Local. No reaction.

Systemic. After a month most leaves show vivid yellow spots and a mosaic.

T. repens.

A symptomless carrier.

Vicia faba.

Local. No reaction.

Systemic. The young leaves develop small chlorotic spots in seven days and

those formed later are mottled. Some malformation occurs, leaving an uneven surface.

Vigna unguiculata.

Local. No reaction.

Systemic. There may be vein clearing and a chlorotic speckle. If symptomless,

the virus can still be recovered.

215

Nicotiana glutinosa.

Local. Chlorotic spots and rings appear in seven days.

Systemic. A week later the young leaves develop chlorotic spots and then a mottle.

The leaves are puckered and the plant is stunted.

N. tabacum.

Local. No reaction.

Systemic. The young leaves first develop chlorotic spots and then a mottle.

Solanum capsicum.

Local. No reaction.

Systemic. A chlorotic mottle concentrated at the base of the leaves appears after

a month.

Natural source of virus; Lupinus albus (Pretoria and Stellenbosch), L. angusti-

folius (Stellenbosch).

Identification.

Apart from the local reaction on the bean this virus shows some resemblance to

several otheis.

Lupin virus B resembles the white clover mosaic virus complex Pierce, in that

both cause a systemic chlorosis on the bean with a local and systemic necrosis on the pea.

It differs from it in the reaction on the broad bean, and the type of local lesion on the

French bean. Physical properties are also different.

Lupin virus B also resembles alsike clover mosaic virus 2 Zaumeyer as both cause

a necrotic speckling and severe stunting of the pea, but differs from it in that the latter

induces a marked reffexing of the leaves of the bean and there is no local reaction.

In his report on the sweet pea streak viruses Ainsworth mentions one which caused

an identical local reaction on the bean viz. chlorotic spots with necrotic rings. Unlike

lupin virus B, it also induced a reflexing of the bean leaves and eventual collapse of

the plant. It also caused a local and systemic necrosis of the broad bean, and only a

chlorosis of the pea.

The two types of bean yellow mosaic virus, viz. the local lesion and severe yellow

strains of Zaumeyer, are similar in some respects, too.

The local lesion strain also causes a local reaction followed by a systemic mottle,

on the bean, but only local lesions develop on the broad bean and cowpea, whereas the

lupin virus B induces a systemic mottle on these hosts. Further, the host range and

physical properties are dissimilar.

Like lupin virus B, the severe yellow strain causes a mottle on broad bean, and

a mottle and necrosis on some varieties of pea, but unlike it, a systemic necrosis usually

develops on the bean. Symptoms on other plants and the physical properties also

differ.

The systemic symptoms on Nicotiana glutinosa are not unlike those of cucumber

mosaic virus but further details are not similar.

Therefore, as there appears to be no complete correlation with any other virus,

this is considered a new one. The name suggested is lupin virus B.

216

3. Lupin Virus C.

Physical properties: Thermal inactivation point, 56-58 °C. Longevity in vitro,

1-2 days. Dilution end point, 1 : 100.

Transmission: Mechanical sap inoculation. Aphis craccivora.

Host range: Crotalaria spectabilis Roth., Glycine max (L.) Merr., Lathyrus odoratus

L., Lupinus albus L. var. Sweet, L. albus L. var. Bitter, L. angustifolius L., L. luteus L.,

Melilotus officinalis Willd., Phaseolus acutifolius Gray var. latifolius Freem., P. lunatus L.,

P. vulgaris L., Trifolium hybridum L., T. incarnatum L., T. pratense L., Vigna unguiculata

(L.) Walp.

Nicotiana glutinosa L., N. tabacum L.

Reaction of Susceptible Species.

Crotalaria spectabilis.

Local. No reaction.

Systemic. The young leaves show chlorotic spots, each with a necrotic centre.

Later this becomes a mosaic with dark green veinbands and necrosis setting in in the

chlorotic areas. The leaves are slightly reduced in size, and puckered and curled.

Glycine max.

Local. There are necrotic lesions in 12 days.

Systemic. The first trifoliates develop chlorotic flecks with necrotic rings, and

soon drop. Most other leaves have vivid chlorotic flecks and later a mosaic with

necrotic specks.

Lathyrus odoratus.

Local. No reaction.

Systemic. There is a chlorotic streak mottle.

Lupinus albus var. Sweet.

Local. In nine days chlorotic spots with necrotic centres develop.

Systemic. In twelve days the young leaves show a vein clearing and chlorotic

spotting. These leaves are rolled upwards but bent down from the pulvinus like an

umbrella (Fig. 3b). They have dark green bands or raised blisters, and are rosetted

at the top of the plant. Later necrosis sets in from the growing point, and extends

down the stem until the plant collapses.

L. albus var. Bitter.

Local. No reaction.

Systemic. Within 1 5 days necrosis has caused the collapse of the plant.

L. angustifolius.

Local. No reaction.

Systemic. The leaves show only a diffuse mottling, but they are severely curled

and folded. Later necrosis causes the leaves to drop.

L. luteus.

Local. No reaction.

Systemic. There is a vein clearing and spotting of the young leaves in 15 days.

Later leaves are small with dark green blisters, and they remain folded. The plant is

stunted.

217

Melilotus officinalis.

Local. No reaction.

Systemic. Chlorotic streaks and rings develop.

Phaseolus acutifolius.

Local. Chlorotic spots with necrotic rings develop.

Systemic. There is a chlorotic spotting.

P. lunatus.

A symptomless carrier.

P. vulgaris var. Canadian Wonder.

Local. Chlorotic spots with a necrotic ring develop in 12 to 13 days.

Systemic. There is a chlorotic mottle,

var. Haricot.

Local. There is a chlorotic spotting in seven days and later the leaves drop.

Systemic. The older trifoliates become long and narrow while the young leaves

are malformed with a mottle or broad dark green veinbands.

vars. Idaho Refugee, S.A. Black and White and Tendergreen develop symp-

toms similar to Canadian Wonder.

Trifolium hybridum.

Local. No reaction.

Systemic. After a vein clearing there are'dilfuse chlorotic blotches.

T. incarnatum.

Local. Small necrotic lesions develop in 12 days.

Systemic. After a vein clearing there are chlorotic veinbands and a mosaic on

the crinkled leaves. The plant is stunted.

T. pratense.

Local. No reaction.

Systemic. There are chlorotic streaks with necrosis setting in, which results in a

malformation of the leaves. Some leaves may collapse from the necrosis.

Vigna unguiculata.

Local. In 5 days there are small necrotic spots and the leaves drop.

Systemic. The young leaves also show necrotic spots, and the plant may be com-

pletely defoliated.

Nicotiana glutinosa.

Local. Chlorotic spots and patterns develop in 7 days, sometimes with necrotic

rings.

Systemic. In 9 days there are chlorotic spots, becoming a mottle on later leaves

which are crinkled. The plant is stunted.

N. tabacum.

Local. Necrotic lesions, which enlarge and fuse, appear in 7 days.

Systemic. No reaction.

Natural source of virus : Lupinus albus and Lupinus angustifolius from Stellenbosch.

218

Identification.

Like lupin virus B, this virus has similarities with, but cannot be related to, white

clover mosaic virus complex Pierce, alsike clover mosaic virus 2 Zaumeyer, the two

strains of bean yellow mosaic virus Zaumeyer and cucumber mosaic virus.

It has however, the same characteristic local reaction on the bean as lupin virus B,

but differs from it, in its inability to infect either broad bean or pea. The symptoms on

lupins are also dissimilar.

Despite the one link with lupin virus B, viz. the local reaction on bean, the virus

seems distinct, and is named lupin virus C.

4. Bean Yellow Mosaic Virus Pierce.

Physical properties: Thermal inactivation point, 58-60°C. Longevity in vitro,

1-2 days. Dilution end point, 1 : 1000-1 : 2000.

Transmission: Mechanical sap inoculation. Aphis craccivora.

Host range: Arachis hypogaea L., Crotalaria juncea L., C. spectabilis Roth.,

Glycine max (L.) Merr., Lathyrus odoratus L., Lupinus albus L. var. Sweet, L. albus L. var.

Bitter, L. angustifolius L., L. luteus L., Medicago lupulina L., Melilotus officinalis Willd.,

Phaseolus acutifolius Gray var. latifolius Freem., P. lunatus L., P. mungo L., P. vulgaris L.,

Pisum sativum L., Trifolium hybridum L., T. incarnatum L., T. pratense L., T. repens L.,

Vida faba L., Voandzeia subterranea Thouars.

Reaction of Susceptible Species.

Arachis hypogaea.

Local. Diffuse chlorotic blotches develop.

Systemic. Most leaves have chlorotic ringspots or patterns, and the plant is

stunted.

Crotalaria juncea.

Local. No reaction.

Systemic. The young leaves develop a chlorotic network in 9 days. On later

leaves there is a chlorotic mottle with dark green blisters. These leaves are slightly

malformed and have wavy margins.

C. spectabilis.

Local. There may be chlorotic spots surrounded by necrotic rings.

Systemic. After a vein clearing of the young leaves in 13 days, later ones develop

broad dark green veinbands. The next leaves are mottled.

Glycine max.

Local. Small necrotic lesions develop in 12 days.

Systemic. The young leaves show chlorotic spots or flecks, and on the later ones

there is a diffuse mottle.

Lathyrus odoratus.

Local. No reaction.

Systemic. In 3 weeks the young leaves show a vein clearing and chlorotic spotting.

The next leaves are rolled and have a mosaic. There is a colour break on the flowers.

219

Lupinus albus var. Sweet.

Local. There are isolated chlorotic spots in 9 days and the leaves become flaccid.

Systemic. The young leaves remain folded and develop a vein clearing and chlorotic

spotting after 16 days. Later leaves are mottled with dark green blisters. They are

elongated, malformed and the margins are rolled (Fig. 4b).

L. albus var. Bitter.

Local. No reaction.

Systemic. After a severe necrosis the plant collapses. If secondary shoots develop,

they are rosetted with small malformed leaves.

L. angustifolius.

Local. No reaction.

Systemic. The young leaves have a diffuse chlorotic spotting. Later ones are

mottled.

L. luteus.

Local. Diffuse chlorotic areas may develop.

Systemic. The young leaves remain folded and have wavy margins, and are

chlorotic. Later leaves are stringlike and malformed. The plant is severely rosetted

and stunted.

Medicago lupulina.

Local. No reaction.

Systemic. The leaves develop a vein clearing and later a mottle. They are puckered

along the midrib.

Melilotus officinalis.

Local. No reaction.

Systemic. After 3-4 weeks most leaves show chlorotic spots, streaks, rings or

concentric patterns (Fig. 4c). The old leaves have a chlorotic etching.

Phaseolus acutifolius.

Local. No reaction.

Systemic. The young leaves curl back and have a yellow network. Later leaves

are mottled.

P. lunatus.

Local. No reaction.

Systemic. Chlorotic vein slashes may develop in 2-3 weeks or the plant is a

symptomless carrier.

P. mungo.

Local. Necrotic veins develop.

Systemic. A limited amount of veinal necrosis may occur.

P. vulgaris var. Canadian Wonder.

Local. No reaction.

Systemic. The young leaves show a marked reflexing from the pulvini and they

develop a vein clearing and flecking. Later formed leaves show a mottle and they are

slightly malformed with an uneven surface. Most pods have dark green blisters.

220

var. Haricot.

Local. These leaves bend at right angles to the pulvini, and have diffuse chlorotic

areas.

Systemic. The first trifoliates show a marked reflexing and have chlorotic spots.

The young leaves are curled, small and mottled, and may have necrosis of the veins.

The pods develop dark green blisters with necrosis.

vars. Black Wonder, Idaho Refugee, Long Tom, S.A. Black and White,

Tendergreen and Victor develop the typical systemic reaction.

vars. Black Wonder, Idaho Refugee, Long Tom and Tendergreen also

show a local chlorosis.

Pisum sativum.

Local. A spreading necrosis causes the leaves to collapse in 14 days.

Systemic. After a vein clearing in 7 days there is a chlorotic spotting followed by

a mosaic. The leaves are small, the tendrils are abnormally curled and the plant is

stunted.

Trifolium hybridum.

Local. No reaction.

Systemic. After a vein clearing in 15 days the leaves have chlorotic streaks.

T. incarnatum.

Local. No reaction.

Systemic. In 12 days there is a vein clearing and chlorotic spotting. Later leaves

develop a mosaic.

T. pratense.

A symptomless carrier.

T. repens.

A symptomless carrier.

Vicia faba.

Local. The leaves may become flaccid and drop.

Systemic. Within 7 days the young leaves develop a spotting or mottle. Later

ones are slightly malformed and have chlorotic patterns.

Voandzeia subterranea.

Local. No reaction.

Systemic. A diffuse mottle develops.

Natural source of virus: Lupinus albus (Pretoria) and L. angustifolius (Stellen

bosch).

Identification.

The reflexing of the leaves of Phaseolus vulgaris is a characteristic of four viruses

or strains.

Only with bean yellow mosaic virus is this reflexing also associated with similarity

in physical properties, host range and symptoms on other hosts. This virus is therefore

considered to be bean yellow mosaic virus.

221

5. Pea Mosaic Virus 4 Zaumeyer.

Physical properties: Thermal inactivation point, 60°-65°C. Longevity in vitro,

3-4 days. Dilution end point, 1 : 5000-1: 10000.

Transmission: Mechanical sap inoculation. Aphis craccivora.

Host range: Crotalaria juncea L., C. spectabilis Roth., Glycine max (L.) Men*.,

Lathyrus odoratus L., Lupinus albus L. var. Sweet, L. albus L. var. Bitter, L. angustifolius

L., L. luteus L., Medicago lupulina L., M. saliva L., Melilotus officinalis Willd., Phaseolus

acutifolius Gray var. latifolius Freem., P. lunatus L., P. vulgaris L., Pisum sativum L.,

Trifolium fragiferum L., T. hybridum L., T. incarnatum L., T. pratense L., T. repens L.,

Vida faba L., Vigna unguiculata (L.) Walp., Voandzeia subterranea Thouars.

Reaction of Susceptible Species.

Crotalaria juncea.

Local. No reaction.

Systemic. After 2 weeks there is a mosaic with irregular chlorotic areas and dark

green blisters. Later leaves are malformed and stringlike, and the plant is stunted.

C. spectabilis.

Local. Chlorotic spots with small necrotic centres develop in 10-11 days.

Systemic. The young leaves develop chlorotic spots. The next formed are chlorotic

with almost white streaks and dark green blisters. There is a slight malformation.

Glycine max.

Local. Chlorotic spots develop in 4-5 days.

Systemic. After a month the leaves show a chlorotic spotting. The older ones

also have necrotic specks. The plant is stunted.

Lathyrus odoratus.

Local. No reaction.

Systemic. After 3 weeks there are chlorotic veinbands followed by a streak mottle

on later leaves.

Lupinus albus var. Sweet.

Local. After 11 days there is a general chlorosis leaving dark green veinbands.

The leaves become flaccid and drop.

Systemic. The young leaves develop a vein clearing and remain folded (Fig. 5b).

Later leaves show an irregular chlorosis with dark green blisters. They are malformed,

elongated or even stringlike, and rolled (Fig. 5d).

L. albus var. Bitter.

Local. No reaction.

Systemic. The plant collapses after a necrosis.

6096259—2

222

L. angustifolius.

Local. No reaction.

Systemic. A diffuse chlorotic mottle develops in 1 8 days.

L. luteus.

Local. No reaction.

Systemic. In 14-15 days the young leaves show a vein clearing and spotting.

Later leaves are small and malformed and have dark green blisters (Fig. 5c). The

plant is stunted and rosetted (Fig. 5a).

Medicago lupulina.

Local. No reaction.

Systemic. There is a vein clearing and spotting of the young leaves. Later ones

have irregular chlorotic areas, and are slightly puckered along the midrib.

M. sativa.

Local. No reaction.

Systemic. The leaves develop diffuse chlorotic spots.

Melilotus officinalis.

Local. No reaction.

Systemic. There may be chlorotic spots or streaks, or the plant may be a symptom-

less carrier.

Phaseolus acutifolius.

Local. No reaction.

Systemic. The young leaves curl down severely. They are mottled and have

dark green blisters. Later leaves have yellow specks.

P. lunatus.

A symptomless carrier.

P. vulgaris var. Canadian Wonder.

Local. No reaction.

Systemic. The centre leaflet of the first trifoliates shows a marked reflexing.

Chlorotic spots develop in 12-13 days and on later leaves there is a mottle. The leaves

are crinkled.

var. Haricot.

Local. The leaves bend at right angles to the petiole and may have necrotic specks.

Systemic. As for Canadian Wonder, but there is also a malformation of the leaves.

There may be a necrosis of the growing point, which results in the collapse of the plant.

vars. Black Wonder, Long Tom, S.A. Black and White, Tendergreen and

Victor show similar symptoms to those of Canadian Wonder.

vars. Long Tom and S.A. Black and White also develop a local chlorosis

in 3 days.

223

Pisum sativum.

Local. No reaction.

Systemic. After a preliminary vein clearing and spotting, a diffuse mottle develops

(Fig. 5e).

Trifolium fragiferum.

A symptomless carrier.

T. hybridum.

Local. No reaction.

Systemic. The vein clearing is vivid, and is followed by alternate chlorotic and

green streaks, with occasional intervenal spots.

T. incarnatum.

Local. No reaction.

Systemic. There is a vein clearing of the young leaves in 14 days. Later leaves

develop a mosaic and crinkling. The plant is stunted.

T. pratense.

Local. No reaction.

Systemic. There are chlorotic ringspots and streaks on most leaves. On the

younger leaves a veinal necrosis may develop which causes a slight malformation.

T. repens.

Local. No reaction.

Systemic. There is vein clearing after 3 weeks which is followed by chlorotic

streaks.

Vicia faba.

Local. No reaction.

Systemic. After 14 days the young leaves develop chlorotic spots, and later leaves

have a mosaic. The old leaves show a chlorotic network.

Vigna unguiculata.

A symptomless carrier.

Voandzeia subterranea.

Local. No reaction.

Systemic. There may be a diffuse chlorotic spotting.

Natural source of virus; Lupinus affinis (Pretoria and district) L. albus, L. angusti-

folius and L. luteus (Stellenbosch).

Identification.

Of the viruses causing the characteristic reflexing of the leaves of the bean, most

similarity is shown to the pea mosaic virus 4 Zaumeyer, and this virus is considered

to be the same, or very closely related to it.

224

HOST RANGES.

PHYSICAL PROPERTIES AND METHODS OF TRANSMISSION.

225

SYMPTOMS ON THREE MAIN TEST PLANTS.

Abbreviations used: —

neg. — negative,

necr. — necrosis or necrotic,

reflex. — reflexing,

ros. — rosette,

veinb.— veinbands.

O.— rings.

SUMMARY.

Lupins in South Africa are naturally infected by five viruses:

1. Bean yellovti mosaic virus was found on Lupinus albus and L. angustifolius

at Pretoria and Stellenbosch. This is the first report of the presence of this virus

in South Africa.

2. Pea mosaic virus 4 was found on L. affinis, L. albus, L. angustifolius and L. luteus

in the Pretoria district and Stellenbosch. It is also reported for the first time.

chi. — chlorosis, chlorotic,

c. mot. — chlorotic mottle,

c. sp. — chlorotic spots,

dgr. blist. — dark green blisters,

malf. — malformation,

mos. — mosaic.

226

3. Lupin virus A was found on L. affinis, L. albus, L. angiistifolius, L. luteus and

L. mutabilis. It is possibly related to the bean yellow mosaic virus group.

4. Lupin virus B was found on L. albus and L. angustifolius. This is apparently

a new virus.

5. Lupin virus C was found on L. albus and L. angustifolius, and also seems to be

a distinct entity, possibly related to lupin virus B.

LITERATURE CITED.

1. Chamberlain, E. E. (1935) Sore-shin of blue lupins ; its identity with pea mosaic. N.Z.J.

2. Johnson, J. and Grant, T. J.

(1932)

3. Klesser, P. J. (1953)

4. Kohler, E. (1935)

5. Mastenbroek, C. (1942)

6. Merkel, L. (1929)

7. Neill, J. C., Brien, R. M., and

Chamberlain, E. E. (1934).

8. Norris, D. O. (1943)

9. Richter (1934)

10. Spierenberg, Dina (1936)

11. Weimer, j. L. (1950)

Agric., 51 : 86-92.

The properties of plant viruses from different host species.

Phytophathology, 22: 741-757.

Virus diseases of lupins. Farming in S. Africa, Oct. 1953.

iibertragungs versuche mit dem Virus der Lupinenbraune.

Angew. Botanik., 17: 277-286.

Enkele veldwaarnemingen over virusziekten van lupine en

een onderzoek over haar mozaiekziekte. Tijdschr. Plantenz.,

48: 97-118.

Beitrage zur Kenntnis der Mosaikkrankheit der Familie

der Papilionaceen. Z. Pflanzenkr., 39: 289-347.

Soreshin, a virus disease of blue lupins. N.Z.J. Agric.,

49: 139-146.

Pea mosaic on Lupinus variiis L. and other species in Western

Australia. C.S.I.R. Bull. 170.

Eine noch nicht aufgeklarte Lupinenkrankheit. Nachrichten-

blatt Deuts. Pflanzen D., 14: 81.

Een virusziekte in lupinen. Tijdschr. Plantenz., 42: 71-76.

Two virus diseases of blue lupine. PI. Dis. Reptr., 34: 376-

378.

227

Fig. 1. — Lupin Virus A.

A. Lupinus hiteus. B. L. alhus. C. L. angiistifolius. Artificial infections.

MI96259

228

Fig. 2. — Lupin Virus B.

A. Lupinus luteus. B. L. aihus. C. Pliaseolus vulgaris. Artificial infections.

D. L. ougustifolius — natural infection.

229

Fig. 3. — Lupin Virus C.

A. Lupinus angustifoUus — natural infection.

B. L. albus — artificial infection.

230

Fig. 4. — Bean Yellow Mosaic Virus.

A. Liipinus albus — natural infection.

B. L. albus — artificial infection.

C. Melilotus officinalis — artificial infection.

231

Fig. 5. — Pea Mosaic Virus 4.

A. and C. Lupinus luteus. B. and D. L. albus.

E. Pisum sativum. Artificial infections.

233

Virus Diseases of Cowpeas.

By

Patricia J. Klesser.

Virus infections of cowpea, Vigna unguiculata, are common in South Africa.

There are three types of symptom found in the field viz. a vivid yellow mosaic with

dark green blisters; dark green veinbands with necrosis and severe malformation,

and a mild veinbanding or mosaic. There is no evidence of seed transmission but

Aphis species are efficient vectors.

REVIEW OF LITERATURE.

The first reports of mosaic diseases of Vigna species were by McClintock (1917),

Elliot (1921) and Elmer (1925). Smith (1924) established that a virus causing mosaic

symptoms on cowpea was transmissible by a leaf-beetle, Cemtoma trifurcata. In 1 929,

Briant & Martyn recorded the transmission of a mosaic of V. catjang in Trinidad;

and the following year Smith & Barker described a virus causing severe stunt and

chlorosis on cowpeas in Haiti. Yu reported a mosaic of cowpea in China in 1939.

Detailed descriptions of viroses of Vigna spp. are given by McLean — a seed-borne

mosaic of cowpea (19); Snyder — a seed-borne mosaic of asparagus bean (31);

Vasuveda — a mosaic of V. catjang in India (32); Yu — cowpea mosaic in China (41);

d’Oliviera — cowpea mosaic viruses 1 , 2 and 3 (1 1); Dale — a cowpea mosaic transmitted

by a leaf-beetle (8 and 9); Warid & Plakidas — cowpea mosaic viruses with extremely

high thermal inactivation points (35) and Anderson (3b). Capoor et al described

mosaic diseases of V. catjang (6) and V. cylindrica (7).

Price (21) was the first to report a strain of cucumber mosaic virus (strain Y) which

caused a systemic reaction as well as the usual local lesions on cowpea. Later Whipple &

Walker (38) described strains 14 and 17, and Fulton found strains A-F on spinach (14).

Anderson (3a) recorded yet another strain, also systemic on cowpea.

Sill & Walker (26a and b) when working on the optimum conditions for local

lesion production on cowpea with cucumber mosaic virus, found that some lines of

cowpeas became systemically infected. Sinclair & Walker (25) determined that this

factor was controlled by the mutation of a single gene pair.

Other legumes which can be naturally infected with strains of cucumber mosaic

virus are peas and beans, Whipple & Walker (38), and Hagedorn (15), sweet peas,

Ainsworth (2) and lima bean, Harter (16).

Several viruses, when inoculated artificially induce local lesions only on cowpea,

viz. tobacco ringspot, Wingard (39); tobacco necrosis. Smith & Bald (30); potato

calico. Black & Price (4); tomato bushy stunt. Smith (28); alfalfa mosaic 1, Zaumeyer

(43) and Black & Price (4); alfalfa mosaic lA, IB and 2, Zaumeyer (43) and pea wilt,

Johnson (17).

The curly top virus causes severe stunt of the plant and proliferation of the leaves

of cowpea, Severin & Henderson (24).

234

METHODS AND MATERIALS.

The following three viruses were isolated from naturally infected cowpeas found

in several districts of the Transvaal.

The standard test plants used were: Phaseolus vulgaris var. Canadian Wonder,

Vida faba and Pisum sativum var. Greenfeast; also Nicotiana tabacum var. White

Burley, N. glutinosa, Cucumis sativus, and Zinnia elegans as so many cowpea viruses

are related to the cucumber mosaic virus group. In addition, about 30 other legumes

were included in the susceptibility tests.

Carborundum powder was used for the mechanical sap inoculations, and Aphis

craccivora for the insect transmissions. (This aphis was used solely as a criterion of

whether or not the viruses were indeed aphis transmissible.)

In this report three virus diseases found naturally on cowpea are described and

compared with those already recorded in the literature.

During this survey of legume virus diseases, cowpea was found to be susceptible

to many other legume virus when inoculated artificially. It also produced local

necrotic lesions with the tomato spotted wilt virus.

1. Cowpea Mosaic Virus A.

Physical properties: Thermal inactivation point, 62-65 °C. Longevity in vitro,

2-4 days. Dilution end point, 1 : 2000.

Transmission: Mechanical sap inoculation. Aphis craccivora.

Host range: Arachis hypogaea L., Crotalaria juncea L., C. spectabilis Roth.,

Dolichos lablab L., Glycine javanica L., G. max (L.) Merr., Lathyrus odoratus L., Lupinus

albus L., vars. sweet and bitter, L. luteus L., L. mutabilis Sweet., Medicago lupulina L.,

M. sativa L., Melilotus officinalis Willd., Phaseolus acutifolius Gray var. latifolius

Freem., P. lunatus L., P. mungo L., P. vulgaris L., Pisum sativum L., Trifolium fragiferum

L., T. hybridum L., T. incarnatum L., T. pratense L., T. repens L., Vida faba L., Vigna

sesquipedalis (L.) Wight., V. unguiculata (L.) Walp., Voandzeia subterranea Thouars.

Reaction of Susceptible Species.

Arachis hypogaea.

Local. No reaction.

Systemic. The leaves are crinkled, and chlorotic with dark green veinbands.

Crotalaria juncea.

Local. There are necrotic specks with chlorotic rings. The leaves drop.

Systemic. After a vein clearing and spotting in 10 days, the leaves develop a

mottle with dark green blisters and veinbands. There are also necrotic specks. These

leaves are crinkled and malformed with constrictions of the laminae. The plant is

rosetted.

C. spectabilis.

Local. No reaction.

Systemic. After 13 days there is a vein clearing and chlorotic spotting and flecking

of the young leaves. The next leaves develop irregular chlorotic areas with dark green

veinbands and islands. These leaves are rolled, crinkled and slightly malformed.

Dolichos lablab.

Local. No reaction.

Systemic. There are diffuse chlorotic areas.

235

Glycina javanica.

Local. There are necrotic spots with chlorotic rings in 4 days.

Systemic. The leaves may have chlorotic areas or, if symptomless, the virus can

be recovered.

G. max.

Local. Chlorotic spots may develop.

Systemic. There is a chlorotic spotting and vein flecking in 9 days. The next

leaves develop a mottle with dark green blisters and irregular yellow areas. They are

elongated and malformed with a crinkled surface. The older leaves have necrotic

specks.

Lathyms odoratus.

Local. No reaction.

Systemic. There are dark green veinbands on most leaves, which are also rolled.

The plant is stunted.

Lupinus albus var. Sweet.

Local. There are dark green lesions on chlorotic leaves in 9 days and the leaves

drop.

Systemic. At the same time the young leaves remain folded and have wavy margins,

with a vein clearing and later spotting. The plant remains stunted. Necrotic specks

develop later and the plant collapses.

L. albus var. Bitter.

Local. No reaction.

Systemic. The young leaves remain small and are chlorotic with dark green blisters.

They are rosetted and have twisted tips. Necrosis sets in and many leaves drop.

L. luteus.

Local. Necrotic specks develop in 10 days.

Systemic. The young leaves remain folded and have twisted tips and chlorotic

spots. Later leaves are mottled with dark green veinbands. They are small, rolled and

malformed. Necrosis may set in and the leaves drop. The plant is stunted.

L. mutabilis.

Local. No reaction.

Systemic. The leaves are mottled with dark green blisters. The plant is rosetted

and stunted.

Medicago lupulina.

A symptomless carrier.

M. sativa.

Local. No reaction.

Systemic. There is a chlorotic spotting with necrotic specks.

Melilotus officinalip.

A symptomless carrier.

236

Phaseolus acutifolius.

Local. There are chlorotic spots in 6 days.

Systemic. After a vein clearing of the young leaves in 9 to 10 days, most later

leaves have a speck mottle.

P. luriatus.

Local. Large chlorotic blotches develop.

Systemic. The young leaves develop a vein clearing in 6 to 7 days (Fig. If). Later

leaves have chlorotic vein flecks or irregular areas which become necrotic (Fig. le).

These leaves may be crinkled.

P. mungo.

Local. There are necrotic spots, ringspots and veins in 7 days.

Systemic. In 3 weeks the young leaves may develop chlorotic specks which soon

become necrotic.

P. vulgaris var. Canadian Wonder.

Local. In 4 days there are chlorotic spots which later have necrotic rings.

Systemic. The young leaves develop a vein clearing and chlorotic spotting in

6 days. Later leaves are mottled.

var. Haricot.

Local. In 4 days there are chlorotic spots which spread and fuse. There are also

necrotic ringspots and veins.

Systemic. In 10 days the young leaves curl down and develop a vein clearing.

These veins become necrotic and there are necrotic stem streaks. The plant is stunted

and usually collapses.

Pisum sativum.

Local. No reaction.

Systemic. After a vein clearing and spotting there is a chlorotic mottle. The

leaves may be slightly malformed and rosetted and the tendrils abnormally curled.

Trifolium fragiferum.

Local. There are necrotic rings on chlorotic leaves.

Systemic. No reaction.

T. hybridum.

Local. No reaction.

Systemic. Many leaves develop chlorotic streaks and spots.

T. incarnatum.

Local. The veins become necrotic and there are necrotic spots.

Systemic. The young leaves develop a vein clearing in 10 days, and later leaves

have a chlorotic spotting, streaking, or mosaic with necrosis setting in, in the chlorosis.

These leaves are crinkled, puckered and small. The plant is stunted and may later

collapse.

Ill

T. pmtense.

Local. No reaction.

Systemic. After a vein clearing of the young leaves in 18 to 19 days, later ones

show a broad dark green midrib with chlorotic leaf margins. Some plants were sym-

tomless carriers.

T. repens.

A symptomless carrier.

Vida faba.

Local. There are necrotic ringspots on chlorotic leaves.

Systemic. In 3 weeks there is a chlorotic mottle.

Vigna sesquipedalis.

Local. There are chlorotic spots in 6 days.

Systemic. The young leaves show a vein clearing and spotting in 10 to 11 days-

The next formed leaves are mottled with dark green veinbands and they are malformed.

The old leaves become necrotic and drop.

V. unguiculata.

Local. There are necrotic spots in 5 days and there may be a slight veinal necrosis

(Fig. la). The leaves show a marked epinasty and soon drop.

Systemic. The young leaves develop a vein clearing and flecking in 12 days. The

next formed leaves have chlorotic spots and veinbands becoming a mosaic (Fig. lb).

Necrotic specks develop causing malformation and later these leaves drop (Fig. Id).

The plant is stunted and there are necrotic stem streaks (Fig. Ic).

Voandzeia subterranea.

Local. There are chlorotic spots in 6 days.

Systemic. After 13 days the young leaves develop a chlorotic network and spotting.

Later leaves are mottled with dark green blisters and are contorted. The old leaves

may develop a necrotic sheen.

Natural source of virus: Vigna unguiculata.

The plants were stunted. The younger leaves had chlorotic spots, dark green

veinbands or a mosaic; and they were small and malformed. Older leaves showed

varying amounts of necrosis in the chlorotic areas.

Identification.

According to Anderson (3b) in a classification of the known cowpea viruses, most

are seed-borne. Exeeptions are the mosaics described by d’Oliviera (11) and Dale (9)

and the cucumber mosaic virus strains.

Cowpea mosaic viruses 2 and 3 d’Oliviera cause only systemic symptoms on cow-

pea, whereas this virus induces a marked local reaction as well. The thermal inactiva-

tion point is about 70°C but this virus withstood 62-65°C only.

d’Oliviera’s cowpea mosaic virus 1 has a lower deathpoint (about 60°C) and causes

both local and systemic symptoms on cowpea, but too few details are available for a

complete comparison.

238

The cowpea mosaic of Da’e is not transmissible by aphis, but by a leaf-beetle;

and it withstands a much longer ageing period (20 days) and higher dilution (1 : 100,000)

than the virus described here.

In summing up the five groups of cowpea viruses, Anderson reports that none

are able to infect Medicago sativa. Trifolium incarnatum or Lathyrus] odoratus, all

of which are susceptible to this virus. Further, the viruses of Yu (41), McLean (19)

and Anderson (3b) are unable to infect the bean, and Snyder’s (31) virus appears to

be confined to Vigna spp.

The extremely high thermal inactivation points (98-1 04°C.) of Warid’s Vigna

viruses 1, 2, 3 and 3A and complexes 1, 2 and 3 exclude them from this comparison (35).

The virus described here is therefore considered to be a new entity belonging to

the cowpea mosaic group, and is named cowpea mosaic virus A.

2. Cowpea Mosaic Virus B.

Physical properties; Thermal inactivation point, 60-62°C. Longevity in vitro,

2-3 days. Dilution end point, 1 : 1000.

Transmission; Mechanical sap inoculation. Insect vector — probably an aphid

but results were inconclusive.

Host range; Crotalaria juncea L., C. spectabilis Roth., Dolichos lablab L., Glycine

javanica L., G. max (L.) Merr., Lathyrus odoratus L., Lupinus albus L., vars. sweet and

bitter, L. luteus L., L. mutabilis Sweet., Medicago lupulina L., Melilotus officinalis Willd.,

Phaseolus acutifolius Gray var. latifolius Freem., P. lunatus L., P. mungo L., P. vulgaris

L., Pisum sativum L., Trifolium incarnatum L., T. pratense L., Vida faba L., Vigna

sesquipedalis (L.) Wight., V. unguiculata (L.) Walp., Voandzeia subterranea Thouars.

Reaction of Susceptible Species.

Crotalaria juncea.

Local. No reaction.

Systemic. There are chlorotic spots on the young leaves which have wavy margins.

The next leaves have a mottle with irregular chlorotic areas which result in malformation.

C. spectabilis.

Local. No reaction.

Systemic. After a vein clearing and spotting in 13 days, later leaves are rolled

and have dark green veinbands.

Dolichos lablab.

Local. No reaction.

Systemic. There are scattered irregular chlorotic areas.

Glycine javanica.

Local. Necrotic lesions develop in 8 days.

Systemic. Some leaves have large chlorotic blotches on a puckered surface.

G. max.

Local. Chlorotic rings develop in 10 days and the leaves drop.

Systemic. There are chlorotic specks and irregular areas which cause a distortion

of the leaf surface. The old leaves have necrotic specks.

239

Lathyms odoratus.

Local. No reaction.

Systemic. The leaves curl down and have chlorotic spots and a dark green midrib.

Lupinus albus var. Sweet.

Local. No reaction.

Systemic. The leaves have wavy margins and a chlorotic network. Later ones

develop a dark green veinbanding.

L. albus var. Bitter.

Local. No reaction.

Systemic. The leaves are small with twisted tips. They have a chlorotic mottle

with a necrotic sheen. Many leaves drop and the plant is rosetted.

L. luteus.

Local. There are necrotic ringspots.

Systemic. Necrosis may spread rapidly into the growing point which curls to

one side, before a total collapse of the plant. If not, the leaves remain folded, with

wavy margins, and are mottled and small. The plant is stunted and rosetted.

L. mutabilis.

Local. No reaction.

Systemic. The leaves are malformed and stringlike and are chlorotic with dark

green marginal blisters. The plant is rosetted.

Medicago lupulina.

A symptomless carrier.

MeUlotus officinalis.

Local. No reaction.

Systemic. After a chlorotic spotting there is a dark green mottle.

Phaseolus acutifolius.

Local. In 10 days there are chlorotic spots with necrotic rings and veins.

Systemic. The plant collapses in 12 days after a severe necrosis.

P. lunatus.

Local. There are chlorotic spots in 10 days. These fuse to a general chlorosis.

Systemic. Diffuse vein flecks develop.

P. inungo.

Local. There are chlorotic spots in 7 days.

Systemic. The young leaves develop chlorotic then necrotic specks and veins in

1 1 days (Fig. 2c).

P. vulgaris var. Canadian Wonder.

Local. There are chlorotic spots in 4 days with a slight necrosis.

Systemic. The young leaves develop a vein clearing in 10 days and later ones

have a mottle with dark green blisters.

240

var. Haricot.

Local. There are chlorotic and necrotic spots, and the leaves drop.

Systemic. Chlorotic spots develop on the young leaves. Later ones are mottled

and reduced in size.

Pisum sativum.

Local. No reaction.

Systemic. The young leaves develop chlorotic spots and later ones a mottle.

The tendrils are abnormally curled.

Trifolium incarnatum.

Local. There are chlorotic spots with necrotic rings and veins in 10 days.

Systemic. In 14 days there is a vein clearing followed by a veinbanding or mosaic

on most leaves. They are crinkled and the plant is stunted.

T. pratense.

Local. No reaction.

Systemic. There are diffuse dark green veinbands.

Vida faba.

Local. There are necrotic lesions in 10 days which fuse and the leaves drop.

Systemic. There is a chlorotic mottle.

Vigna sesquipedalis.

Local. Chlorotic spots develop in 10 days.

Systemic. The young leaves develop a vein clearing which may become necrotic.

The next leaves are mottled and slightly malformed. (Fig. 2b.).

V. unguiculata.

Local. There are necrotic specks and a necrotic sheen in 3 days.

Systemic. The young leaves develop a vein clearing and chlorotic spotting while

later ones are mottled with veinbands (Fig. 2d). There may be necrotic stem streaks.

The plant is stunted (Fig. 2a).

Voandzeia subterranea.

Local. There is a necrotic sheen in 10 days.

Systemic. The young leaves develop a chlorotic spotting and veinbanding, and

later leaves are malformed and have a mottle with necrotic specks.

Natural source of virus: Vigna unguiculata.

There was a mosaic or dark green veinbanding on the leaves, which were normal

in size and shape. On the old leaves there was a slight necrosis in the chlorosis. The

plants were not stunted or rosetted.

Identification.

Like cowpea mosaic virus A this virus cannot be identified with any of the described

cowpea mosaics, nor with the cucumber mosaic strains. It further differs from cowpea

mosaic virus A in host range, physical properties and symptoms on many hosts. See

tables. Generally, the symptoms caused by the B virus are milder than those caused by

the A virus.

This virus is therefore named cowpea mosaic virus B.

241

3. Cucumber Mosaic Virus Strain.

Physical properties: Thermal inactivation point, 62-64 °C. Longevity in vitro,

4-5 days. Dilution end point, 1 : 5000.

Transmission: Mechanical sap inoculation. Aphis craccivora.

Host range: Arachis hypogaea L., Crotalaria juncea L., C. spectabilis Roth.,

Dolichos lablab L., Glycine max (L.) Merr., Lupinus albus L., vars. sweet and bitter,

L. luteus L., L. mutabilis Sweet., Medicago lupulina L., M. sativa L., Melilotus officinalis

Willd., Phaseolus acutifolius Gray var. latifoUus Freem., P. mungo L., P. vulgaris L.,

Trifolium hybridum L., T. incarnatum L., T. medium L., T. pratense L., Vida faba L.,

Vigna sesquipedalis (L.) Wight., V. unguiculata (L.) Walp., Voandzeia subterranea

Thouars.

Cucumis sativus L., Nicotiana glutinosa L., N. tabacum L., Solanum capsicum L.,

Zinnia elegans Jacq.

Reaction of Susceptible Species.

Arachis hypogaea.

Local. No reaction.

Systemic. There is first a chlorotic spotting and then a mottle on later formed

leaves.

Crotolaria juncea.

Local. No reaction.

Systemic. Most leaves develop a mottle with dark green islands. They are slightly

malformed.

C. spectabilis.

Local. Necrotic lesions develop in 7 days.

Systemic. The young leaves show a chlorotic spotting and flecking. Later ones

have chlorotic interveinal areas or a diffuse speck mottle.

Dolichos lablab.

Local. No reaction.

Systemic. In 3 weeks the older trifoliates show marked chlorotic line patterns

and some chlorotic spotting. The young leaves are slightly ruffled and have irregular

chlorotic blotches.

Glycine max.

Local. There may be a general chlorosis in 4 days.

Systemic. After 1 1 days there is a vein clearing of the young leaves. Later ones

are mottled with dark green blisters and are severely malformed.

Lupinus albus var. Sweet.

Local. No reaction.

Systemic. The young leaves remain folded and have wavy margins. Later ones

are mottled with dark green blisters and are malformed. The plant is rosetted.

var. Bitter.

Local. No reaction.

Systemic. The leaves become elongated and chlorotic with dark green areas.

The plant may collapse.

242

L. luteus.

Local. No reaction.

Systemic. In 3 weeks the young leaves develop small chlorotic spots. Later leaves

are very small and rosetted and the plant is very stunted.

L. mutabilis.

Local. No reaction.

Systemic. Most leaves are chlorotic with dark green marginal blisters. The plant

is stunted and rosetted.

Medicago lupulina.

Local. No reaction.

Systemic. Most leaves develop diffuse chlorotic vein slashes.

M. sativa.

Local. There are small chlorotic spots.

Systemic. The young leaves show a vein clearing and later leaves may develop

small dark green blisters. These are puckered and crinkled.

Melilotus officinalis.

Local. No reaction.

Systemic. In 3 weeks the young leaves show chlorotic streaks following the veins.

On the old leaves there are occasional chlorotic ringspots.

Phaseolus acutifolius.

Local. No reaction.

Systemic. The young leaves show a vein clearing then a mottle with yellow specks.

P. mango.

Local. No reaction.

Systemic. The first trifoliates show a spotting, and later formed leaves have a

general chlorosis with slight necrosis. The necrosis may spread until the growing

point collapses.

P. vulgaris var. Canadian Wonder.

Local. No reaction.

Systemic. In 13 days there is a mottle followed by vivid yellow spots and vein

flecks on the second and third trifoliates (Fig. 3b). The young leaves may be symptom-

less and when subinoculated there is no virus present.

var. Haricot.

Local. No reaction.

Systemic. The young leaves develop chlorotic spots and then a mottle. Later

leaves may be malformed and have yellow areas. The old leaves have large almost

white chlorotic blotches.

243

TrifoHum hybridum.

Local. No reaction.

Systemic. There are chlorotic streaks.

T. incarnatum.

Local. After 2 weeks dark green rings with chlorotic centres appear.

Systemic. The young leaves develop a chlorotic mosaic with necrotic specks

and a slight crinkle.

T. medium.

Local. No reaction.

Systemic. In 3 weeks there are chlorotic streaks along the veins.

T. pratense.

Local. No reaction.

Systemic. There is first a vein clearing and later some veins become necrotic.

This leads to a distortion of the leaves. The plant is stunted.

Vicia faba.

Local. There are necrotic rings in 6 days.

Systemic. In 1 1 days chlorotic spots develop on the young leaves and then necrotic

specks. Later leaves are mottled with raised dark green areas and necrotic ring and

line patterns. The petioles and stem may have superficial necrotic streaks.

Vigna sesquipedalis.

Local. In 7 to 8 days there are chlorotic spots with necrotic specks. These leaves

drop.

Systemic. In 1 1 days the young leaves show a vein clearing, and later ones are

vividly mottled, and rolled, curled and crinkled (Figs. 3d and f). The symptoms are

similar to, but not as severe as, those on V. unguiculata.

V. unguiculata.

Local. In 4 to 5 days there are chlorotic spots which become necrotic. The leaves

bend down and later drop.

Systemic. A week later chlorotic flecks develop on the older trifoliates which may

be slightly puckered. Young leaves become progressively more crinkled and also

malformed. They are a bright yellow and have dark green blisters and necrotic specks

(Fig. 3e). These leaves are small and the plant is stunted and may collapse completely

(Fig. 3a).

Voandzeia subtenanea.

Local. No reaction.

Systemic. After 3 weeks the young leaves develop dark green blisters on a chlorotic

background. Later leaves have vivid yellow spots and are curled and contorted.

244

Cucumis sativus.

Local. In 14 days small chlorotic spots develop (Fig. 2g).

Systemic. At the same time the young leaves show a clearing of the veins with

dark green blisters. The plant remains stunted (Fig. 2g).

Nicotiana glutinosa.

Local. In 4 days necrotic lesions develop.

Systemic. Some leaves have chlorotic ring and line patterns which cause a dis-

tortion of the surface (Figs. 2e and f.)

N. tabacum.

Local. No reaction.

Systemic. Small chlorotic spots develop in 1 1 days and on later leaves there are

chlorotic ring and line patterns, which distort the leaf surface.

Solatium capsicum.

Local. There are necrotic lesions in 4 days and the leaves drop.

Systemic. In 13 days there is a vein clearing followed by a mottle with necrotic

specks which cause a malformation of the leaf.

Zinnia elegans.

Local. No reaction.

Systemic. After a vein clearing there is a mottle (Fig. 3c).

Natural source of virus : Vigna unguiculata.

The leaves had a vivid yellow mottle, and they were puckered and blistered. The

plants were severely stunted.

Identification.

The very marked yellow mottle and severe stunt on cowpea can be compared with

only one other virus. Smith & Barker (29) in their report on the bean yellows virus,

mention the severe stunt and extreme chlorosis that virus causes on cowpea.

However, the symptoms on the bean are not similar, the host range is very limited

and that virus is transmissible by Empoasca fabalis.

Apart from this characteristic yellow mottle on cowpea, there are similarities

with several of the strains of cucumber mosaic virus, in respect of both physical pro-

perties and host range. The host range includes non-leguminous species.

On the cowpea there is some resemblance to the Y strain of Price (21), the yellow

isolate of strain 14 of Whipple & Walker (38) and strain C of Fulton (14).

However, Price does not describe the symptoms on other legume hosts, so a com-

parison can not be made.

Both the Y isolate of strain 14 and strain C cause systemic necrosis and ultimate

collapse of beans and peas, while this virus either induces a yellow fleck on beans or is

latent; and it is unable to infect peas.

Whipple & Walker report that their strain 17, the celery virus 1 strain and the type

cucumber mosaic virus all result in local lesions on cowpea, and systemic necrosis on

pea. They are unable to infect beans. Peas are susceptible to all the other strains

described by Fulton.

245

Harter’s lima bean mosaic (16), also considered a strain of cucumber mosaic

virus, causes only local lesions on cowpea, and is non-infectious to bean, lupin, soybean,

alsike etc., all of which are hosts of this virus.

The strain of cucumber mosaic virus described] by Anderson (3a), may cause

both local and systemic symptoms on cowpeas, but on most varieties they are masked —

in contrast to the severe effect of this virus. Furthermore the symptoms of Anderson’s

strain on tobacco and cucumber are transient only, whereas with this virus they are

definite and lasting. Anderson’s strain withstands only 6-24 hours ageing, and this

one lasts 4-5 days.

To sum up — although this virus shows many links with the cucumber mosaic

virus group, it can not be fully identified with any of the strains already described.

It is therefore considered a new strain of the cucumber mosaic virus.

COMPARISON OF SYMPTOMS CAUSED BY COWPEA MOSAIC VIRUSES A AND B

ON SOME HOST PLANTS.

1.

2.

3.

Host Plant.

Cowpea Mosaic Virus A.

Crotalaria juncea....

Lupin us a lb us

Melilotus officinalis.

Local — necr. specks and chi. O.

Syst. — mot., dgr. blisters and

veinb., necr. specks.

Local — dark green spots.

Syst. — chi. sp., necr. specks, col-

lapse.

Symptomless carrier.

4. Phaseolus acutifolius Local — chi. spots. !

Syst. — speck mottle. j

5. Trifoiium incarnatum... . \ Local — necr. spots and veins. 1

Syst. — chi. streaks, mosaic, necr., ]

crinkle, ros., collapse.

6. Vigna sesquipedalis Local — chi. spots. I

Syst. — mot., dgr. veinbands, necro- '

sis, malformation.

7. Vigna unguiculata Local — reflex., necr. sp. and veins.

Syst. — mosaic, necr. specks, mal-

formation, necr. stem streaks.

Cowpea Mosaic Virus B.

Local — no reaction.

Syst. — irreg. chi. areas, malforma-

tion.

Local — no reaction.

Syst. — chi. network, dgr. veinbands.

Local — no reaction.

Syst. — dgr. mottle.

Local — chi. spots, necr. O and veins.

Syst. — necrosis and collapse.

Local — chi. spots, necr. O and

veins.

Syst. — mosaic, crinkle.

Local — chi. spots.

Syst. — mottle.

Local — necr. sheen.

Syst. — mottle, dark green vein-

bands.

Abbreviations used: —

chi. — chlorosis/chlorotic,

dgr. — dark green,

irreg. — irregular,

mot. — mottle,

necr. — necrosis /necrotic.

ros. — rosette,

sp. — spots,

veinb. — veinbands.

O — rings.

COMPARISON OF PHYSICAL PROPERTIES AND METHODS OF TRANSMISSION.

j Thermal j Longevity | Dilution

inaetiva- , in end

246

Arachis hypogaea

Crotalaria juncea

C. spectabilis

Dolichos lablab

Glycine javanica

G. max

Lathyrus odoratus. . . .

Lupimis albus — sweet.

Lupinus albus — bitter.

L. luteus

L. mutabilis

Medicago lupidina

M. sativa

Melilotus officinalis. . .

Phaseolus acitlijolius. .

P. lunatiis

P. mungo

P. vulgaris

Pisutn sativum

Tri folium fragiferum. .

T. hybridum

T. incarnatiim

T. pratense

T. repens

Vida faba

Vigna sesquipedalis. . .

V. imguiculata

Voandzeia subterranea

Cucumis sativus

Nicotiana glutinosa. . .

N. tabacum

Solanum capsicum

Zinnia elegans

HOST RANGES.

SUMMARY.

Three virus diseases of naturally infected cowpeas are described in detail — their

physical properties, methods of transmission, host ranges and symptoms. None could

be fully identified with any previously recorded virus or strain.

1. Cowpea mosaic virus A. In the field the plants are stunted and the leaves are

small and malformed. They have dark green veinbands or a mosaic with necrosis.

The virus is inactivated at temperatures of 62-65 °C., at a dilution of 1 :2,000 and after

2-4 days ageing in vitro. It is sap and aphid transmissible and the host range is confined

to the Leguminosae.

2. Cowpea mosaic virus B. The only symptom in the field is a dark green vein-

banding. The thermal inactivation point of the virus is 60-62°C. The longevity in

vitro is 2-3 days and the dilution end point, 1 ; 1,000. It is sap transmissible, but results

from insect transmissions were inconclusive. The host range is confined to the legumes,

and the symptoms are generally milder than for the A virus.

3. Cucumber mosaic virus strain. The field symptoms are severe — the plants are

very stunted and the leaves show a vivid yellow mottle. Similarities in physical proper-

ties, host range (which includes the non-legumes) and symptoms, indicate a relationship

with the cucumber mosaic virus group.

247

1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

LITERATURE CITED.

Ainsworth, G. C. (1935)

Ainsworth, G. C. (1940).

Anderson, C. W. (1955)

(«)

(b)

(t)

Black, L. M. and Price, W. C.

(1940)

Briant, a. K., and Martyn, E. B.

(1929)

Capoor, S. P., Varma, P. M. and

Uppal, B. N. (1947)

Capoor, S. P. (1956)

Dale, W. T. (1943)

Dale, W. T. (1949)

Dale, W. T. (1953)

d’Oliviera, M. a. (1947).

Elliott, J. A. (1921)

Elmer, O. H. (1925)

Eulton, J. P. (1950)

Hagedorn, D. j. (1950)

Harter, L. L. (1938)

Johnson, F. (1942)

McClintock, j. a. (1917)

McLean, D. M. (1941)

Pierce, W. H. (1934)

Price, W. C. (1934)

Price, W. C. (1935)

Reddick, D. and Steward, V. B.

(1918)

Severin, H. H. P. and Henderson,

C. F. (1928)

Sinclair, J. B. and Walker, J. C.

(1955)

Mosaic diseases of the cucumber. Ann. Appl. Biol., 22:

55-67.

The identification of certain viruses found infecting legu-

minous plants in Great Britain. Ann. Appl. Biol., 27:

218-226.

Crotalaria and Vigna viruses in Florida.

I. Preliminary report on a strain of cucumber mosaic

virus obtained from cowpea plants. Plant Dis. Reptr.,

39: 346-348.

II. Notations concerning cowpea mosaic virus (Marmor

vignae). Plant Dis. Reptr., 39: 349-353.

III. Notations concerning identification difficulties, indicator

plants, possible vector relationships and virus main-

tenance. Plant Dis. Reptr., 39: 354-357.

The relationship between viruses of potato calico and alfalfa

mosaic. Phytopathology, 30: 444-447.

Diseases of cover crops. Trop. Agric., 6: 258-260.

A mosaic disease of Vigna catjang Wa\p. Curr. Sci., 16: 151-

Mosaic of Vigna cylindrica. Ind. Jour. Agric. Sci., 26: 95-

Preliminary studies of the plant viruses of Trinidad. Trop.

Agric., 20: 228-235.

Observations on a virus disease of cowpea in Trinidad.

Ann. Appl. Biol., 36: 327-333.

The transmission of plant viruses by biting insects, with

particular reference to cowpea mosaic. Ann. Appl. Biol.,

40: 384-392.

Contribuicao ao estudo dos virus causadores de mosaico

nos feijoes macassar. (Vigna spp.) Bol. Teen. Agron. Sul.

(Pelotas), 1: 1-36.

A mosaic of sweet and red clovers. Phytopathology, 11:

146-148.

Transmissibility and pathological effects of the mosaic

disease. Iowa A.E.S. Bull., 82: 39-91.

Studies on strains of cucumber virus 1 from spinach. Phyto-

pathology, 40: 729-736.

A cucumber strain with a wide leguminous host range.

Phytopathology (abstract), 40: 11.

Mosaic of lima beans. (Phaseolus hmalns macrocarpiis).

Journ. Agric. Res. (U.S.), 56: 895-906.

The complex nature of white clover mosaic. Phytopathology,

32: 103-116.

Lima bean mosaic. Phytopathology, 7: 60-61.

Studies on mosaic of cowpeas, Vigna sinensis. Phytopa-

thology, 31: 420-430.

Viroses of the bean. Phythopathology, 24: 87-115.

Isolation and study of some yellow strains of cucumber

mosaic virus. Phytopathology, 24: 743-761.

Acquired immunity from cucumber mosaic in zinnia. Phyto-

pathology, 25: 776-789.

Varieties of beans susceptible to mosaic. Phytopathology,

8: 530-531.

Some host plants of curly top. Hilgardia, 3: 339-392,

595-636.

Inheritance of resistance to cucumber mosaic virus in cowpea.

Phytopathology, 45: 563-564.

248

26. (fl) Sill, W. H. Jr. and Walker,

J. C. (1952)

26. (6) Sill, W. H. Jr. and Walker,

J. C. (1952)

27. Smith, C. E. (1924)

28. Smith, K. M. (1935)

29. Smith, R. C. and Barker H. D.

(1930)

30. Smith, K. M. and Bald, J. C.

(1935)

31. Snyder, W. C. (1942)

32. Vasuveda, R. S. (1942)

33. Walker, J. C., Whipple, O.C. AND

Virgin, W. (1937)

34. Warid, W. a. and Plakidas,

A. G. (1950)

35. Warid, W. A. and Plakidas,

A. G. (1952)

36. Weiss, F. (1945)

37. Wellman, F. L. (1934)

38. Whipple, O. C. AND Walker, J. C.

(1941)

39. WiNGARD, S. A. (1928)

40. Yu, T. F. (1939)

41. Yu, T. F. (1946)

42. Yu, T. F. (1947)

43. Zaumeyer, W. J. (1938)

44. Zaumeyer, W. J. (1939)

45. Zaumeyer, W. J. and Wade, B. L.

(1935)

Cowpea as an assay host for cucumber virus 1. Phyto-

pathology, 42: 328-330.

A strain of cowpea susceptible to systemic infection by

various strains of common cucumber mosaic virus. Phyto-

pathology, 42 : 442.

Transmission of cowpea mosaic by the bean leafbeetle.

Science, 60: 268.

A new virus disease of the tomato. Ann. Appl. Biol., 22:

731-741.

Observations on the “ yellows ” disease of beans and related

plants in Haiti. Jour. Econ. Ento., 23 : 842-847.

A necrotic virus disease affecting tobacco and other plants.

Parasitology, 27: 231-245.

A seed-borne mosaic of asparagus bean, Vigna sesquipedatis.

Phytopathology, 32: 518-523.

A mosaic disease of cowpea. Ind. Jour. Agric. Sci., 12:

281-283.

Plant diseases. Rept. Wise. Agric. Exp. Sta. 1935-36. Bull.

438.

New viruses naturally infectious to cowpea. Phytopathology,

40: 30-31.

Thermophilic viruses naturally infectious to cowpea. Plant

Dis. Reptr., 36: 380-381.

Viruses described primarily on leguminous vegetable and

forage crops. Plant Dis. Reptr., Suppl. 154: 32-80.

Identification of Celery Virus 1, the cause of southern celery

mosaic. Phytopathology, 24: 695-725.

Strains of cucumber mosaic virus pathogenic on bean and

pea. Jour. Agric. Res., 62: 27-60.

Host and symptoms of ringspot, a virus disease of plants.

Jour. Agric. Res. 37: 127-153.

A list of plant viruses observed in China. Phytopathology,

29: 459-460.

A mosaic disease of cowpea. { Vigna sinensis Endl.). Ann.

Appl. Biol., 33 : 450—454.

Spotted wilt of broad bean. Phytopathology (Notes), 37-191.

A streak disease of peas and its relation to several strains of

alfalfa mosaic virus. Jour. Agric. Res., 56: 747-772.

Two new viruses affecting pea. Phytopathology (Abstract),

29: 25.

The relationship of certain legume mosaics to bean. Jour.

Agric. Res., 51: 715-749.

249

P,Q i_ — CowPEA Mosaic Virus A.

A-D Vigna unguiculata. A. Local necrotic lesions. B. Systemic vembandin'g and

mosaic C Systemic necrotic specks. D. Malformation of young leaves.

Tf PlJeolus l.nan.s, E. Necrosis and chlorosis on older leaf. F. F,„e chlorot.c

network on young leaf.

6096259

250

Fig. 2. — A-D Cowpea Mosaic Virus B.

A and D Vigna unguiculata. B. V. sesquipedalis. C. Phaseolus nnmgo.

E-G Cucumber Mosaic Virus Strain.

Band F Nicotiana glutinosa. G. Cucumis sativus.

251

Fig. 3. — Cucumber Mosaic Virus Strain.

A and E. Vigm unguiculata. D. and F. V. sesquipedalis. B. Pliaseolus vulgaris.

C. Zinnia elegans.

253

Virus Diseases of Peas and Sweet Peas.

By

Patricia J. Klesser.

According to the Revised List of Plant Diseases occuring in South Africa (3),

the only virus disease of peas and sweet peas that has been positively identified, is the

one caused by the tomato spotted wilt virus.

Other disorders due to virus infections have long been recognised but the causal

organisms were not determined.

This report concerns eight viruses which occur naturally on Piswn sativum and

Lathyrus odoratus in South Africa.

Apart from tomato spotted wilt there are three other diseases which are commonly

found on both peas and sweet peas, viz. pea mosaic, pea stunt and a strain of lucerne

mosaic.

Further, peas are naturally infected by pea virus 2, pea mosaic virus 4, and a

strain of pea wilt virus. Sweet peas are also susceptible in the field to bean local chlorosis

virus B and a strain of bean yellow mosaic.

Although pea mosaic, pea virus 2 and pea stunt are well-known elsewhere, they

are reported for the first time in South Africa. (Pea mosaic virus 4 was discussed in

detail in the paper on lupin virus diseases, as it occurs frequently on Lupinus species.)

Details of what appear to be new strains of the pea wilt, lucerne mosaic and bean

yellow mosaic viruses are given, and a virus causing vivid local chlorotic spots on bean

is described.

Pea mosaic has long been recognised as the primary virus disease of peas, but in

this country more damage is caused by the tomato spotted wilt virus. This occurs

frequently, and results in a systemic necrosis and possible collapse of the plants.

The other viruses are found spasmodically but the effect on individual plants can

be severe. The necrotic ring strain of the lucerne mosaic virus also induces a systemic

necrosis and collapse of the plant, while the pea stunt and pea wilt viruses cause necrotic

stem streaks with a mosaic.

With the pea mosaic virus and pea virus 2 there is only a mosaic or mottle.

On the sweet pea, chlorotic spots, veinbands or a mosaic, may develop on the

leaves of plants infected with the pea mosaic virus, or strains of the bean yellow mosaic

and bean local chlorosis viruses.

Systemic necrosis occurs when the spotted wilt virus, pea stunt virus or a strain

of lucerne mosaic virus are present.

Most of these viruses can cause a colour break on the ffowers.

254

METHODS AND MATERIALS.

The standard test plants used were; Phaseolus vulgaris var. Canadian Wonder,

Vida faba var. Aquadulce and Pisum sativum var. Greenfeast. In addition about

30 other legumes and some plants belonging to the Solanaceae were used in susceptibility

tests.

Aphis craccivora or Macrosiphum pisi were used for the insect transmission tests,

although they are not necessarily the natural vectors. Carborundum powder was

used for the sap inoculations.

In this report eight virus diseases are described — their physical properties, methods

of transmission, host ranges and symptoms on susceptible hosts.

Their identification is based on information from the literature, but when no com-

plete correlation can be found, new names are provisionally proposed.

1. Pea Mosaic Virus Doolittle & Jones.

Physical properties; Thermal inactivation point, 56-58 °C. Longevity in vitro,

2-3 days. Dilution end point, 1 ; 1,000-1; 2,000.

Transmission; Mechanical sap inoculation. Aphis cracdvora.

Host range; Crotalaria juncea L., Glycine javanica L., Lathyrus odoratus L.,

Lens esculentum L., Lupinus albus L., L. luteus L., Medicago lupulina L., Melilotus

officinalis Willd., Phaseolus mungo L., P. vulgaris L., (?) Pisum sativum L., Trifolium

hybridum L., T. incarnatum L., T. pratense L., T. repens L., Vida faba L.

Reaction of Susceptible Species.

Crotalaria juncea.

Local. Necrotic rings can be seen after 6 days. These fuse and the leaf drops

Systemic. The young leaves develop chlorotic spots with necrotic centres. This

necrosis spreads until the leaf is “ scorched ” and eventually drops. Due to streak

necrosis, the stem curves at the top, and the plant soon dies.

Glycine javanica.

A symptomless carrier.

Lathyrus odoratus.

Local. No reaction.

Systemic. The leaves develop chlorotic spots and later a mosaic. They are rolled

and reduced in size. The flowers have a marked colour break (Fig. Ic).

Lens esculentum.

Local. No reaction.

Systemic. In about 11 days the young leaves show chlorotic spots or streaks.

Later necrosis may set in, in these areas and the plant collapses. If necrosis is not

present the leaves are tightly curled and the stems are rosetted, with a general stunt of

the plant.

255

Lupinus albus.

Local. In 6 days there are small chlorotic spots with necrotic centres.

Systemic. In 14 days the veins of the young leaves become necrotic and the leaflets

drop. Later formed leaves are firstly chlorotic but soon become necrotic. The growing

point is stunted and almost completely defoliated. The stem becomes necrotic and

the plant collapses. The stem necrosis occurs in streaks on one side only, resulting in

a curved stem (Fig. Id).

L. luteus.

Local. No reaction.

Systemic. The young leaves remain folded and have wavy margins. They are

mottled. The plant is stunted and no flowers are formed.

Medicago lupulina.

A symptomless carrier.

Melilotus officinalis.

Local. No reaction.

Systemic. Chlorotic vein slashes develop in one month.

Phaseolus mungo.

Local. The veins become necrotic and the leaves collapse.

Systemic. The young leaves curl down and their veins become necrotic, with the

interveinal areas quite chlorotic. A stem necrosis may develop and the plant then

collapses.

P. vulgaris var. Canadian Wonder.

Local. No reaction.

Systemic. Occasionally the first trifoliates develop large chlorotic blotches. The

young leaves may have a chlorotic network,

var. Haricot.

Local. No reaction.

Systemic. The leaves may be mottled with a slight malformation.

Pisum sativum.

Local. No reaction.

Systemic. In 7 days there is a clearing of the veins of the young leaves with small

chlorotic flecks. A week later this is followed by a mosaic mottle of the youngest

leaves, (Fig. la) but there may also be dark green veinbands or irregular chlorotic

slashes. The leaves remain slightly folded, and are puckered and reduced in size.

The growing point is rosetted and the tendrils abnormally curled.

Trifolium hybridum.

Local. No reaction.

Systemic. After a vein clearing the leaves show a chlorotic streaking.

T. incarnatum.

Local. No reaction.

Systemic. The young leaves develop a clearing of the veins in 7 days. Later

formed leaves may show a chlorotic spot mottling, veinbanding or a characteristic

chlorotic triangle at the apex. These leaves are crinkled and rolled and the plant is

stunted (Fig. lb).

256

T. pratense.

A symptomless carrier.

T. repens.

A symptomless carrier.

Vida faba.

Local. No reaction.

Systemic. After 7 days a diffuse mottle develops on the young leaves. A week

later the young leaves show a mosaic, and they are rolled and reduced in size.

Natural source of virus: Pisum sativum (Pretoria and district). The older leaves

were flecked and mottled and the younger ones had a mosaic mottle. The plants were

rosetted and the tendrils were abnormally curled.

Lathyrus odoratus (Pretoria). The leaves had a chlorotic streak mottle and there

was a marked colour break on the flowers.

Identification.

From the symptom expression, host range and physical properties, there can be

little doubt that this is the pea mosaic virus.

2. Pea Virus 2 Osborn.

Physical properties: Thermal inactivation point, 60-62°C. Longevity in vitro,

3-4 days. Dilution end point, 1 : 2,000.

Transmission: Mechanical sap inoculation. Macrosiphum pisi.

Host range: Lathyrus odoratus L., Lupinus albus L., Medicago lupulina L., Phaseolus

vulgaris L., Pisum sativum L., Trifolium hybridum L., T. pratense L., Vica faba L.

Reaction of Susceptible Species.

Lathyrus odoratus.

Local. No reaction.

Systemic. In 14 days the leaves show a mosaic mottle with chlorotic streaks.

These leaves are slightly curled downwards.

Lupinus albus.

Local. In 9 days there are large necrotic lesions which fuse, and the leaflets drop.

Systemic. 3 to 4 days later the necrosis extends up the stem and into the growing

point, and the plant collapses (Fig. 2d).

Medicago lupulina.

Local. No reaction.

Systemic. Only a fine chlorotic network of the younger leaves develops.

Phaseolus vulgaris var. Canadian Wonder.

Local. No reaction.

Systemic. In a week chlorotic spots appear on the young leaves. The second

and third trifoliates develop chlorotic flecks, and later formed leaves show only irregular

chlorotic areas. These leaves have slightly uneven surfaces (Fig. 2b). The plant is

stunted and rosetted, with pods blistered in dark green.

257

Pisum sativum.

Local. The inoculated leaves wither and drop.

Systemic. In a week the young leaves develop a vein clearing and chlorotic spotting.

Later leaves show a mosaic with dark green veinbands and they are curled upwards.

The plant is rosetted and stunted, and the tendrils are abnormally curled (Fig. 2a).

Trifolium hybridum.

Local. No reaction.

Systemic. There are alternate chlorotic and green streaks following the veins of

the leaves.

T. pratense.

Local. No reaction.

Systemic. Diffuse chlorotic vein slashes develop in one month.

Vida faba.

Local. No reaction.

Systemic. In 7 days there is a clearing of the veins of the young leaves followed

by a chlorotic spotting. Later leaves have a mosaic in which the dark green areas

are raised (Fig. 2c.) There may also be concentric chlorotic patterns on these leaves.

Natural source of virus: Pisum sativum (Pretoria and district). The plants were

rosetted and the leaves showed a chlorotic flecking and mosaic mottle.

Identification.

This virus is similar to the pea mosaic virus of Doolittle & Jones (4) in many respects,

but it regularly causes a systemic reaction on the bean.

The pea virus 2 Osborn (10), shows this ability to infect certain varieties of bean;

and therefore this virus appears to be identical with or closely related to it.

3. Pea Stunt Virus Zaumeyer.

Physical properties: Thermal inactivation point, 60-62°C. Longevity in vitro,

2-3 days. Dilution end point, 1 : 5,000-1 : 10,000.

Transmission: Mechanical sap inoculation.

Host range: Crotalaria juncea L., C. spectabilis Roth., Glydne max (L.) Merr.,

Lathyrus odoratus L., Lupinus albus L., L. luteus L., Medicago sativa L., Phaseolus

vulgaris L., Pisum sativum L., Trifolium hybridum L., T. incarnatum L., Vida faba L.,

Vigna sesquipedalis (L.) Wight., V. unguiculata (L.) Walp., Voandzeia subterranea

Thouars.

Reaction of Susceptible Species.

Crotalaria juncea.

Local. Some necrosis may occur.

Systemic. After 17 days there are dark green veinbands and a speck mottle. The

leaves are small, rolled and rosetted, and the plant is stunted.

C. spectabilis.

Local. No reaction.

Systemic. The young leaves show chlorotic veins and spots in 7 days. Later

leaves are malformed and crinkled, with irregular chlorosis. The plant is stunted.

258

Glycine max.

Local. The leaves become chlorotic and drop.

Systemic. After a chlorotic spotting the leaves develop a mottle with some yellow

specks. The old leaves have a chlorotic network.

Lathyrus odoratus.

Local. No reaction.

Systemic. After 2 weeks many leaves show dark green veinbands or a mottle.

They are rolled and slightly puckered.

Lupinus albus.

Local. There are necrotic specks with chlorotic halos.

Systemic. The young leaves have necrotic specks and veins, and soon drop. The

next leaves are chlorotic with dark green blisters, and they are rolled and crinkled.

There are necrotic stem streaks and the plant is stunted.

L. luteus.

Local. No reaction.

Systemic. The young leaves remain folded and in 9 days there is a vein clearing.

Later ones are mottled, elongated and crinkled with wavy margins. The plant is stunted

and rosetted.

Medicago sativa.

Local. No reaction.

Systemic. There is a chlorotic network followed by a chlorotic flecking on some

leaves.

Phaseolus vulgaris var. Canadian Wonder.

Local. No reaction.

Systemic. In 6 to 8 days the young leaves show a vein clearing while the first

and second trifoliates develop chlorotic blotches, spots, or vein flecks. Later leaves

are mottled and rolled with a slight puckering (Fig. 3b). The pods are blistered (Fig.

3c) and the plant is stunted.

var. Haricot.

Local. No reaction.

Systemic. The young leaves may drop after a veinal necrosis. The next formed

leaves are small and malformed with a chlorotic mottle.

Pisum sativum.

Local. No reaction.

Systemic. The young leaves develop chlorotic veins in 6 days, and then spots.

Later leaves have a mosaic, and they are malformed and reduced in size. The tendrils

are abnormally curled (Fig. 3a). Later leaves have necrosis in the chlorosis and there

are necrotic stem streaks. The pods have purple lesions.

Trifolium hybridum.

Local. No reaction.

Systemic. After 16 days the leaves show chlorotic veins, spots and streaks, and

they may be slightly crinkled.

259

T. incarnatum.

Local. No reaction.

Systemic. There is a chlorotic network on the young leaves. Later ones have

dark green veinbands or a mosaic, and they are crinkled. The old leaves may have

necrotic specks.

Vida faba.

Local. No reaction.

Systemic. After 7 days there are chlorotic spots on the young leaves. Later

leaves have dark green veinbands or a mosaic; they are long and narrow and rolled

(Fig. 3d).

Vigna sesquipedalis.

A symptomless carrier.

V. unguiculata.

A symptomless carrier.

Voandzeia subteiranea.

A symptomless carrier.

Natural source of virus: Lathyrus odoratus (Pretoria and district). Most leaves

had a chlorotic mottle, and there were necrotic stem streaks. The flowers showed a

marked colour break. Pisum sativum (Pretoria and district). The leaves had irregular

chlorotic areas with varying amounts of necrosis. There were also necrotic stem streaks

and the plants were stunted.

Identification.

This virus shows most similarity with ttie stunt virus of Zaumeyer (17), both in

respect of symptomology and physical properties. Although not identical, it is con-

sidered closely related to the pea stunt virus.

4. Pea Wilt Virus — New Strain.

Physical properties: Thermal inactivation point, 58-60 °C. Longevity in vitro,

2-3 days. Dilution end point, 1 : 10,000.

Transmission: Mechanical sap inoculation. Aphis craccivora.

Host range: Arachis hypogaea L., Crotalaria juncea L., C. spectabilis Roth.,

Dolichos lablab L., Glycine javanica L., G. max (L.) Merr., Lathyrus odoratush., Lupinus

albus L., L. luteus L., Medicago lupulina L., Melilotus officinalis Willd., Phaseolus

acutifolius Gray var. latifolius Freem., P. lunatus L., P. mungo L., P. vulgaris L., Pisum

sativum L., Trifolium hybridum L., T. incarnatum L., T. pratense L., Vida faba L.,

Vigna sesquipedalis (L.) Wight., V. unguiculata (L.) Walp., Voandzeia subterranea

Thouars.

Reaction of Susceptible Species.

Arachis hypogaea.

Local. No reaction.

Systemic. Some leaves have a chlorotic mottle with patterns and the older ones

have chlorotic blotches.

260

Crotalaria juncea.

Local. No reaction.

Systemic. In 15 days the growing point becomes stunted and may bend over.

The leaves are uniformly chlorotic with only isolated dark green areas. Later formed

leaves are chlorotic with dark green blisters and veinbands. They have crinkled surfaces

and are slightly malformed. The plant is stunted.

C. spectabilis.

Local. There are necrotic lesions in 7 days.

Systemic. After a chlorotic vein flecking of the young leaves, later ones are

chlorotic with dark green specks. These leaves are crinkled and the plant is stunted.

Dolichos lablab.

A symptomless carrier.

Glycine javanica.

A symptomless carrier.

G. max.

Local. There are chlorotic spots which later have a necrotic ring. There is also

some veinal necrosis.

Systemic. In 9 days the young leaves show a clearing of the veins. This becomes

a dark green mottle on the next formed leaves, which are also puckered. The older

leaves may have small necrotic spots.

Lathyrus odoratus.

Local. No reaction.

Systemic. The leaves remain folded and they are chlorotic with dark green islands,

and slightly malformed. Some plants collapsed from necrosis.

V fir. vs albus.

Local. Small chlorotic spots appear in 9 days and the leaflets drop later.

Systemic. The young leaves have a chlorotic network and the leaflets remain

folded and have wavy margins. There may be stem necrosis. There is little upward

growth so that the growing point becomes stunted and rosetted with many small string-

like leaves. These are chlorotic with dark green marginal blisters. No flowers are

formed.

L. luteus.

Local. No reaction.

Systemic. In 1 3 days the young leaves show chlorotic spots and they remain folded.

Later leaves are mottled or have dark green blisters on malformed, stringlike leaves.

The plant is stunted and rosetted.

Medicago lupulina.

Local. No reaction.

Systemicr A diffuse chlorotic mottle develops.

Melilotus officinalis.

Local. No reaction.

Systemic. After a vein clearing of the young leaves, there are chlorotic spots or

streaks on later formed leaves.

261

Phaseolus acutifolius.

Local. In 4 days the veins are necrotic and later chlorotic spots develop.

Systemic. After 10 to 11 days, the young leaves show a vein clearing and later a

mottle. There may be some malformation. New leaves have almost yellow specks.

P. lunatus.

Local. Chlorotic veins and irregular areas develop in 7 to 9 days.

Systemic. In 9 days there is a vein clearing of the young leaves. Later ones have

a diffuse mottle, veinbanding or flecking.

P. mungo.

Local. No reaction.

Systemic. There is a diffuse chlorotic speckling after 2 weeks. On the old leaves

there is also necrosis.

P. vulgaris var. Canadian Wonder.

Local. No reaction.

Systemic. After 9 days the first trifoliates show large chlorotic star flecks. The

young leaves have a vein clearing and later ones a spot mottle with the dark green along

the veins (Fig. 4d).

var. Haricot.

Local. The veins are slightly necrotic and later small chlorotic spots with necrotic

halos appear.

Systemic. There is a chlorosis with dark green blisters, and malformation and

twisting of the leaves. There is only a slight necrosis.

Pisum sativum.

Local. The inoculated leaves collapse.

Systemic. There is a vein clearing in 2 weeks, followed by a chlorotic mottling

of the leaves which remain slightly folded. The plant is rosetted and stunted (Fig. 4a).

The stem may have necrotic streaks and the tendrils are abnormally curled.

Trifolium hybridum.

Local. No reaction.

Systemic. In 24 days there is a clearing of the veins of the young leaves, and there-

after, alternate streaks of yellow and green develop on later formed leaves.

T. incarnatum.

Local. There are necrotic specks in 7 to 8 days.

Systemic. The young leaves develop a vein clearing. Later ones have dark green

veinbands and they are puckered and crinkled.

T. pratense.

Local. No reaction.

Systemic. There are occasional chlorotic vein streaks which become necrotic and

result in a malformation of the leaf surface (Fig. 4c).

262

Vida faba.

Local. The leaves develop dark green or necrotic rings. They become flaccid

and drop (Fig. 4b).

Systemic. No reaction, but the virus is present.

Vigna sesguipedalis.

Local. Within a week chlorotic spots appear.

Systemic. In 21 days the young leaves show chlorotic spots, and later ones a mottle

and malformation.

V. mguiculata.

Local. There are necrotic specks with chlorotic halos.

Systemic. In 2 weeks there is a vein clearing followed by a diffuse chlorotic

mottle.

Voandzeia subteiranea.

Local. No reaction.

Systemic. There are chlorotic specks and dark green veinbands.

Natural source of virus: Pisutn sativum (Pretoria). The plants were stunted and

unthrifty and the leaves had a very diffuse mottle. Phaseolus vulgaris (Pretoria and

district). On the variety Black Wonder the younger leaves had vivid chlorotic spots

and the older ones a dark green veinbanding and mottle. The plants were stunted.

Identification.

On the major host plants e.g. pea, cowpea, bean and broad bean, the symptoms

produced by this virus are very similar to those caused by the pea wilt virus Johnson (8).

But on other plants there are small differences and the physical properties differ in the

longevity and dilution end points.

Nevertheless, this virus is grouped with it as a possible strain, and is named pea

wilt virus, strain A.

5. Lucerne Mosaic Virus — Necrotic Ring Strain.

Physical properties: Thermal inactivation point, 62-65 °C. Longevity in vitro,

7-8 days. Dilution end point, 1 : 3,000-1 : 5,000.

Transmission: Mechanical sap inoculation. Aphis craccivora.

Host range: Crotalaria juncea L., C. spectabilis Roth., Glycine max (L.) Merr.,

Lathyrus odoratus L., Lupinus albus L., L. luteus L., Melilotus officinalis Willd., Phaseolus

acutifolius Gray var. latifolius Freem., P. vulgaris L., Pisum sativum L., Trifolium

hybridum L., T. incarnatum L., Vida faba L., Vigna unguiculata (L.) Walp., Voandzeia

subterranea Thouars.

Reaction of Susceptible Species.

Crotalaria juncea.

Local. In 6 days there are necrotic lesions.

Systemic. The young leaves show chlorotic spots which soon become necrotic

and cause a malformation of the leaves. The plant is stunted.

263

C. spectabilis.

Local. There are dark green rings with necrosis.

Systemic. The young leaves show chlorotic spots and later leaves have chlorotic

ring and line patterns and are crinkled. There may be a slight malformation and dark

green blisters.

Glycine max.

Local. No reaction.

Systemic. After a chlorotic spotting of the young leaves, later ones develop a

mottle. They are slightly crinkled.

Lathyrus odoratus.

Local. Small chlorotic spots develop which later become necrotic.

Systemic. In 3 weeks the stem has become necrotic and the plant collapses.

Lupinus albus.

Local. No reaction.

Systemic. The young leaves develop chlorotic spots. Later ones are crinkiea

and rolled with chlorotic veinbands.

L. luteus.

Local. No reaction.

Systemic. The young leaves remain folded and have chlorotic spots. Later ones

are crinkled or malformed, with an irregular chlorotic mottle. The plant is rosetted

and stunted.

Melilotus ojficinalis.

Local. No reaction.

Systemic. There are chlorotic areas and a slight distortion.

Phaseolus acutifolius.

Local. No reaction.

Systemic. In 15 days there is a vein clearing, followed by a mottle with slight

necrosis.

P. vulgaris var. Canadian Wonder.

Local. Within 5 days there are small chlorotic spots which later have a halo.

Severe necrosis sets in, in the chlorosis (Fig. 5c).

Systemic. No reaction,

var. Haricot.

Local. In a week there are necrotic ringspots and veins.

Systemic. On some plants the young leaves show severe rolling and curling down.

There are small chlorotic spots with necrotic veins and specks which spread until the

plant collapses (Fig. 5d).

vars. Black Wonder, Idaho Refugee and Long Tom develop only local

chlorotic rings or irregular areas in 3 to 6 days.

vars. S.A. Black and White and Tendergreen have no local reaction and

only a transient vein clearing or spotting.

6096259—3

264

Pisum sativum.

Local. There may be necrotic rings and spots which cause the leaves to collapse.

Systemic. In 11 to 12 days the young leaves show a vein clearing or chlorotic

flecks which soon become necrotic. This necrosis progresses down the stem and the

growing point collapses. Secondary shoots may develop and these have malformed

twisted and rosetted leaves with irregular chlorotic areas. The tendrils are abnormally

curled (Fig. 5a).

Trifolium hybridum.

Local. No reaction.

Systemic. After 12 days there are chlorotic spots on the young leaves and later

formed ones develop a streak mottle.

T. incarnatum.

Local. The leaves may collapse.

Systemic. In 12 days a vein clearing develops on the young leaves. Later leaves

shov\ a mosaic mottle and are crinkled and malformed. The plant is stunted.

Vida faba.

Local. In 4 to 5 days large chocolate-coloured necrotic lesions develop. The leaves

become flaccid (Fig. 5b).

Systemic. A week later small chlorotic spots or a vein clearing may occur on the

young leaves. Usually the necrosis spreads quickly up the stem and into the growing

point and the plant collapses.

Vigna unguiculata.

Local. In 5 days the leaves develop chlorotic spots with necrotic rings and veins.

These soon drop.

Systemic. Only occasionally are there similar symptoms on the trifoliates, which

collapse. If symptomless the virus can be recovered.

Voandzeia subterranea.

Local. No reaction.

Systemic. The young leaves develop a chlorotic spotting.

Natural source of virus: Pisum sativum (Northern Transvaal). The young leaves

were mottled, the stems were necrotic and the growing points flaccid. The plants were

stunted and rosetted and the tendrils abnormally curled.

Lathyrus odoratus (Pretoria and district). The leaves had a veinclearing, chlorotic

spotting or veinbanding, with necrosis setting in. Some plants collapsed after this

necrosis, while others produced secondary shoots with malformed, blistered leaves.

Flowers on these stems showed a marked colour break.

Identification.

Several viruses show similar reactions on the pea, but cannot be considered the

same as the above for the following reasons: Both pea American streak virus Zaumeyer

(16) and Wisconsin pea stunt virus Hagedorn & Walker (5) are unable to infect the

Phaseolus spp. which are hosts of this virus.

Pea Wisconsin streak virus Hagedorn «& Walker (6) does not infect Vida faba’

whereas this virus causes severe necrosis.

265

The pea stunt virus Zaumeyer (17) and the pea stunt virus strain Klesser (9) induce

only a systemic reaction on the bean, while in addition the latter causes only a

systemic chlorosis on the broad bean.

Pea New Zealand streak virus Chamberlain (2) infects severa Inon-legumes, induces

different symptoms on many hosts and has different physical properties.

On the bean the symptoms are very like those caused by the lucerne mosaic virus,

Weimer (15). Physical properties and legume host range are also similar, although

lucerne mosaic virus is able to infect the non-legumes.

On pea and broad bean the symptoms resemble those of Zaumeyer’s alfalfa mosaic

viruses lA and IB (16), but neither of these induces a lethal necrosis of the sweet pea.

Further, there is a considerable discrepancy in physical property values.

Apart from the necrotic reaction on the sweet pea this virus is almost identical

with, and can be considered the same as the lucerne mosaic virus necrotic ring strain

Klesser.

6. Bean Local Chlorosis Virus — Strain B.

Physical properties; Thermal inactivation point, 58-60°C. Longevity in vitro,

2-3 days. Dilution end point, 1 : 1,000-1: 2,000.

Transmission: Mechanical sap inoculation. Aphis craccivora.

Host range: Crotalaria spectabilis Roth., Glycine max (L.) Merr., Lathyrus odoratus

L., Lupinus albus L., L. luteus L., Phaseolus vulgaris L., Pisum sativum L., Trifolium

hybridum L., T. incarnatum L., Vida faba L., Vigna unguiculata (L.) Walp.

Reaction of Susceptible Species.

Crotalaria spectabilis.

Local. No reaction.

Systemic. After 7 days the young leaves show a vein clearing and spotting. Latei

leaves have irregular, almost white, streaks which lead to malformation of the leaf

surface (Fig. 6c).

Glycine max.

Local. No reaction.

Systemic. A diffuse chlorotic mottle develops on most leaves.

Lathyrus odoratus.

Local. No reaction.

Systemic. After 1 1 days the young leaves show chlorotic streaks. This continues

on later leaves or there is a mosaic. Necrotic streaks may occur on the stem and petioles.

(Fig. 6f). There is a colour break on the flowers (Fig. 6e).

Lupinus albus.

Local. No reaction.

Systemic. After a vein clearing and spotting of the young leaves the next formed

are crinkled with dark green blisters. There may be small necrotic speckles.

L. luteus.

Local. No reaction.

Systemic. Most leaves remain folded and have small chlorotic spots.

266

Phaseolus vulgaris var. Canadian Wonder.

Local. In 5 days vivid chlorotic spots appear (Fig. 6a).

Systemic. Shortly afterwards the young leaves also develop chlorotic spots and

vein flecks (Fig. 6d). Later leaves are malformed and mottled (Fig. 6b). The old

leaves show large chlorotic blotches.

Pisum sativum.

Local. No reaction.

Systemic. In 6 days there is a vein clearing and spotting of the young leaves. Later

leaves are small and malformed with irregular dark green areas. The plant is rosetted

and stunted and the tendrils are abnormally curled. There are necrotic stem streaks.

Trifolium hybridum.

Local. Small chlorotic spots develop.

Systemic. Most leaves have a chlorotic streak mosaic.

T. incarnatum.

Local. No reaction.

Systemic. In 16 days there is a vein clearing. Later leaves have a mosaic and are

slightly malformed.

Vida faba.

Local. No reaction.

Systemic. Chlorotic spots develop on the young leaves in a week. The next

leaves all have a mosaic.

Vigna unguiculata.

Local. Small red necrotic lesions appear in 5 days.

Systemic. No reaction.

Natural source of virus: Lathyrus odoratus (Pretoria district). The leaves had

chlorotic veinbands or streaks, and they were small and puckered.

Identification.

The vivid chlorotic spotting on the inoculated leaves of the bean is a characteristic

feature of a group of viruses recorded in England by the writer.

The only published reports of similar viruses are those by Ainsworth and Zau-

meyer.

Ainsworth (1) isolated a virus which caused a “sweet pea streak” and which

induced a local chlorotic spotting on the bean. However, necrotic rings developed

later on these leaves and the plant collapsed from a systemic necrosis, whereas this

virus causes no necrosis on the bean.

On other plants there are also differences — on broad bean and sweet pea Ains-

worth’s virus resulted in a systemic necrosis whereas this virus induces a mottle or

mosaic.

Zaumeyer’s alfalfa yellow mosaic virus (18) causes symptoms on some varieties

of bean which are very similar to those described for this virus, but on other varieties

there is both a local and systemic necrosis.

267

However, on other host plants the symptoms differ. On broad bean and pea

Zaumeyer’s virus induces a local and systemic necrosis whereas this virus causes a

mosaic on broad bean, and malformation and stunt on pea. On cowpea, there is a

local necrosis as well as a yellow mottle and malformation with the alfalfa yellow mosaic,

but only local lesions develop with this virus.

Further, Zaumeyer’s virus has a larger host range which includes solanaceous

spp., and the physical properties differ.

Although this virus was isolated from Lathyrus odoratus the disease it causes on

that host is very similar to many others, but on Phaseolus vulgaris the symptoms are

unusual.

Therefore the name chosen for this virus is bean local chlorosis virus. Of the three

strains described in England (9) there is most similarity with the B strain — the A strain

induces a local chlorosis on broad bean and a local and systemic necrosis on pea, whereas

the C strain causes a veinbanding on broad bean and a mottle only on pea. Physical

property values also differ.

This virus is therefore considered the same as bean local chlorosis virus — strain B.

7. Bean Yellow Mosaic Virus, Necrotic Strain.

Physical properties: Thermal inactivation point, 58-60°C. Longevity in vitro,

2-3 days. Dilution end point, 1: 1,000.

Transmission: Mechanical sap inoculation. Aphis craccivora.

Host range: Crotalaria juncea L., C. spectabilis Roth., Glycine max (L.) Merr.,

Lathyrus odoratus L., Lupinus albus L., L. mutabilis Sweet., L. luteus L., Melilotus

officinalis Willd., Phaseolus acutifolius Gray var. latifolius Freem., P. lunatus L.,

P. vulgaris L., Pisuni sativum L., Trifolium hybridum L., T. incarnatum L., Vida faba L.,

Voandzeia subterranea Thouars.

Reaction of Susceptible Species.

Crotalaria juncea.

Local. No reaction.

Systemic. After 2 weeks the young leaves are chlorotic with dark green specks.

Necrosis may set in, in the chlorosis and the plant may collapse. Usually the necrosis

causes an uneven surface on the leaves which are small and malformed. The plant is

stunted.

C. spectabilis.

Local. In 7 days there are chlorotic spots with a slight necrosis.

Systemic. At the same time the young leaves develop chlorotic spots or vein

flecks. Later leaves are mottled and malformed with dark green blisters and necrotic

specks. The plant is stunted.

Glycine max.

Local. No reaction.

Systemic. A diffuse chlorotic spotting develops.

Lathyrus odoratus.

Local. Some leaves become chlorotic (Fig. 7e).

Systemic. In 2 weeks the young leaves develop a vein clearing and chlorotic spotting.

Later leaves are mottled and rolled (Fig. 7d).

268

Lupinus albus.

Local. No reaction.

Systemic. In 2 weeks the young leaves show a veinal necrosis. This may extend

to the stem and the plant may collapse. Secondary shoots have small malformed

and rolled leaves which are mottled. The plant is stunted.

L. mutabilis.

Local. No reaction.

Systemic. The young leaves curl back severely; there is a stem necrosis and the

plant collapses.

L. luteus.

Local. No reaction.

Systemic. The young leaves remain folded and in 10 days there is a vein clearing

and chlorotic spotting. The next ones are small and crinkled with dark green veinbands.

Later leaves are almost stringlike. The plant is rosetted and stunted.

Melilotus officinalis.

A symptomless carrier.

Phaseolus acutifoUus.

Local. No reaction.

Systemic. A diffuse chlorotic mottle develops on some leaves.

P. lunatus.

Local. No reaction.

Systemic. Only chlorotic vein flecks develop.

P. vulgaris var. Canadian Wonder.

Local. The leaves bend at right angles to the petiole and in 5-6 days there are

chlorotic blotches and slightly necrotic veins (Fig. 7a). These leaves become leathery

and remain attached.

Systemic. The first and second trifoliates also show a marked reflexing and they

have chlorotic flecks or blotches. Later leaves develop a vein clearing, spotting or

flecking and then a mottle. The next leaves are small and malformed and the plant is

rosetted and stunted. The pods have sunken dark green areas (Fig. 7b).

var. Haricot.

Local. In 7 days there are chlorotic spots on the reflexed leaves.

Systemic. A week later, the first trifoliates bend down and show a vein clearing.

Later leaves are mottled with dark green blisters. The plant is stunted, and the stem

may have short necrotic streaks. The pods develop sunken necrotic rings.

vars. Black Wonder, Idaho Refugee, Long Tom and Tendergreen develop

symptoms similar to those on Canadian Wonder.

var. S.A. Black and White shows no local reaction but the systemic is

similar to Canadian Wonder.

269

Pisum sativum.

Local. In 6 days there may be necrotic rings or the leaves just wilt and collapse.

Systemic. Soon afterwards the young leaves develop a vein clearing and chlorotic

spotting. Necrosis sets in on the leaves, stipules and stem, and the growing point

collapses (Fig. 7f). Secondary shoots are stunted and rosetted, and the leaves are

small, malformed and mottled. The tendrils are abnormally curled.

Trifolium hybridum.

Local. No reaction.

Systemic. After 3 weeks the young leaves show a vein clearing and chlorotic

spotting. Later ones develop vivid yellow streaks and a slight puckering (Fig. 7c).

T. incarnatum.

Local. No reaction.

Systemic. After 2 weeks there is a vein clearing of the young leaves followed by

a chlorotic veinbanding. The leaves are slightly puckered and the plant is stunted.

Vida faba.

Local. No reaction.

Systemic. In 7 days the young leaves show a vein clearing, or chlorotic spots or

flecks. Later leaves develop a mosaic.

Voandzeia subterranea.

Local. No reaction.

Systemic. Diffuse chlorotic spots develop on some leaves.

Natural source of virus; Lathyrus odoratus (Pretoria district). The leaves had

chlorotic spots, veinbands or a mottle, and there was a colour break of the flowers.

Identification.

There are several viruses which cause a reflexing of the bean leaves, but only with

alsike clover mosaic virus 2 Zaumeyer (17) is this characteristic associated with a systemic

necrosis on the pea. However, as the necrosis caused by the alsike virus results in the

complete defoliation of the pea plant, and the physical properties differ, the casual

virus is not the same as the one described here.

Although the type bean yellow mosaic virus Pierce (11) does not induce a necrosis

on the pea, symptoms on many other hosts and the physical properties are very similar

to this virus.

The bean yellow mosaic virus necrotic lesion strain of Zaumeyer and Fisher (19)

is not infectious to pea, white lupin or crimson clover, all of which are hosts to this

virus. Further, it induces a local and systemic necrosis on broad bean (whereas this

virus causes a mosaic) and local lesions on tobacco, which is not susceptible to this

virus.

The severe yellow strain of Thomas and Zaumeyer (13) is excluded as there are

differences in the symptoms on several host plants and in the physical property values.

It is also able to infect tobacco.

The bean yellow mosaic virus “ Isolate I ” of Hagedorn and Walker (7) does not

cause any necrosis on the pea and other symptoms and properties also differ.

Although the virus described here does not infect the cowpea and remains latent

in sweet clover, so many details are similar to a virus found in England by the writer (9),

that it is considered to be the same, and is named bean yellow mosaic virus, necrotic

strain.

270

8. Tomato spotted Wilt Virus. Samuel et al.

Physical properties: Thermal inactivation point, 40-42°C. Longevity in vitro,

-1 day. Dilution end point, 1: 5,000-1; 10,000.

Transmission: Mechanical sap inoculation. Tlirips spp.

Host range in Leguminosae: Arachis hypogaea L., Crotalaria juncea L., C. spectabilis

Roth., Glycine max (L.) Merr., Lathyrus odoratus L., Lupinus albus L., L. luteus L.,

Medicago denticulata L., M. sativa L., Melilotus officinalis Willd., Phaseolus mango L.,

Pisum sativum L., Trifolium fragiferum L., T. incarnatum L., T. pratense L., Vida faba

L., Vigna unguiculata (L.) Walp., Voandzeia subterranea Thouars.

Reaction of Susceptible Species.

Arachis hypogaea.

Local. Usually no reaction.

Systemic. There is a veinal necrosis of the young leaves and the growing point

collapses. Secondary shoots are rosetted, with small, malformed leaves. These have

concentric chlorotic patterns or dark green veinbands. Necrotic speckling may be

severe.

Crotalaria juncea.

Local. There are necrotic spots in 5 days which enlarge by concentric bands

(Fig. 8h).

Systemic. There are concentric chlorotic patterns and possibly a lethal necrosis.

C. spectabilis.

Local. Necrotic ringspots develop in 5 days. Later there is also a veinal necrosis

(Fig. 8b).

Systemic. Some leaves are severely curled and have a chlorotic network.

Glycine max.

Local. There are necrotic specks with chlorotic haloes. The leaves later become

almost orange (Fig. 8e).

Systemic. No reaction.

Lathyrus odoratus.

Local. Large necrotic lesions develop and the leaves soon abciss.

Systemic. The necrosis spreads into the main stem causing external streaks,

and the plant eventually collapses.

Lupinus albus.

Local. Necrotic specks develop and the rest of the leaflet becomes chlorotic.

The necrosis spreads and the leaflets drop (Fig. 8 j).

Systemic. No reaction.

L. luteus.

Local. There are chlorotic rings which soon become necrotic and the leaflets

abciss.

Systemic. No reaction.

271

Medicago denticulata.

Local. A scorch necrosis^causes a defoliation.

Systemic. There is a vein clearing of the young leaves followed by a veinal necrosis.

Later leaves have necrotic ringspots (Fig. 8f).

M. sativa.

Local. There are necrotic ringspots on a vivid yellow background (Fig. 8i).

Systemic. No reaction.

Melilotus officinalis.

Local. Chlorotic ringspots develop in 10-11 days.

Systemic. No reaction.

Phaseolus mungo.

Local. Isolated necrotic lesions develop.

Systemic. No reaction.

Pisum sativum.

Local. After a general necrosis the leaves wilt.

Systemic. A veinal necrosis of the leaves follows, and young plants may collapse.

On older plants there are necrotic spots on the leaves and necrotic streaks on the stems.

(Fig. 8c). The pods may have necrotic lesions and the tendrils are abnormally curled.

Trifolium fragiferum.

Local. There are necrotic specks with chlorotic haloes in 5 days (Fig. 8g).

Systemic. No reaction.

T. incarnatum.

Local. Chlorotic spots with necrotic rings develop and there is a slight veinal

necrosis.

Systemic. No reaction.

T. pratense.

Local. There are necrotic specks which enlarge and fuse to give a “ scorch ”

effect.

Systemic. No reaction.

Vida faba.

Local. A severe veinal necrosis may develop.

Systemic. Usually this necrotic effect is first seen on the young leaves. It is soon

followed by sunken necrotic stem streaks and a collapse of the plant (Fig. 8a).

Vigna unguiculata.

Local. There are red necrotic spots with chlorotic rings.

Systemic. No reaction.

Voandzeia subterranea.

Local. Necrotic spots develop which later enlarge by concentric bands.

Systemic. A diffuse chlorotic mottle may occur.

Natural source of virus: The natural incidence of spotted wilt is high on peas,

sweet peas and broad beans. On groundnuts and the Crotalaria spp. it occurs less

frequently.

272

Identification.

Sub-inoculations to several Solanaceous hosts proved that this is the tomato

spotted wilt virus (12).

HOST RANGES.

PHYSICAL PROPERTIES AND METHODS OF TRANSMISSION.

273

SYMPTOMS ON THE MAIN TEST PLANTS.

Abbreviations used:

bio. — blotches,

chlorotic.

coll. — collapse,

fl. — flecks,

loc. — local.

1 & n — long and narrow

malf. — malformation.

mos. — mosaic.

mot. — mottle.

necr. — necrotic/necrosis,

no — no reaction.

leaves, ros. — rosette,

sp. — spots,

s.s. — stem streaks,

syst. — systemic.

O. — rings.

7 — reflexing.

SUMMARY.

The virus diseases which occur naturally on peas and sweet peas in South Africa

are described in detail.

In the field, peas develop mosaic and/or necrotic symptoms. The mosaic is caused

by the pea mosaic virus or pea virus 2; the necrosis (and ultimate collapse) by the

lucerne mosaic virus necrotic ring strain or by the tomato spotted wilt virus; and a

mosaic with necrotic stem streaks by pea stunt virus or a strain of the pea wilt virus.

Symptoms on naturally infected sweet peas are similar viz. a mosaic develops

with the pea mosaic virus, the necrotic strain of bean yellow mosaic or the bean local

chlorosis virus B. A necrosis develops with the lucerne mosaic virus necrotic ring

strain or with the tomato spotted wilt virus; and mosaic plus necrotic stem streaks

develops with the pea stunt virus.

Of these, tomato spotted wilt has already been reported as a virosis of legumes in

South Africa, but the others are new records.

Pea mosaic virus, pea virus 2 and pea stunt virus were identified by comparing with

published descriptions.

Three others were found to be very similar to, but not identical with previously

described viruses, and are considered to be strains of the type viruses, viz. a strain of

the pea wilt virus, bean yellow mosaic virus necrotic strain, and lucerne mosaic virus

necrotic ring strain.

One is apparently a new entity, and is named bean local chlorosis virus B, after

the characteristic reaction on the bean.

274

LITERATURE CITED.

1. Ainsworth, G. C. (1940)

2. Chamberlain, E. E. (1939)

3. Doidge, E. M., Bottomley, A. M.,

VAN DER Plank J. E. and Pauer,

G. E. (1953)

4. Doolittle, S. P. and Jones E. R.

(1925)

5. Hagedorn, D. J. and Walker,

J. C. (1949)

6. Hagedorn, D. J. and Walker,

J. C. (1949)

7. Hagedorn, D. J. and Walker,

J. C. (1950).

8. Johnson, F. (1942)

9. Klesser, P. j. (1951)

10. Osborn, H. T. (1937)

11. Pierce, W. H. (1934)

12. Samuel, G., Bald, J. G., and

Pittman, H. A. (1930)

13. Thomas, H. R. and Zaumeyer,

W. J. (1953)

14. Wade, B. L. and Zaumeyer, W. J.

(1938)

15. Weimer, j. L. (1934)

16. Zaumeyer, W. J. (1938)

17. Zaumeyer, W. J. (1940)

18. Zaumeyer, W. J. (1953)

19. Zaumeyer, W. J. and Fisher,

H. H. (1953)

The identification of certain viruses found infecting leguminous

plants in Great Britain. Ann. Appl. Biol., 27: 218-226.

Pea streak (Pisimi virus 3). N.Z. Jour. Sci. and Tech. 20:

365A-381A.

A revised list of plant diseases occuring in South Africa.

Dept, of Agric. Sci. Bull. 346.

The mosaic disease in the garden pea and other legumes.

Phytopathology 15: 763-772.

Wisconsin pea stunt, a newly described disease. Jour. Agric.

Res., 78: 617-626.

Wisconsin pea streak. Phytopathology, 39: 837-847.

The relation of bean virus 2 to pea mosaic in Wisconsin.

Phytopathology, 40: 684-698.

The complex nature of white clover mosaic. Phytopathology,

32: 103-116.

Ph. D. thesis. Cambridge University.

Studies on the transmission of pea virus 2 by aphids. Phyto-

pathology, 27 : 589-603.

Viroses of the bean. Phytopathology, 24: 87-115.

Investigations on spotted wilt of tomatoes. Aust. Counc.

Sci. Ind. Res. Bull., 44.

A strain of yellow bean mosaic virus producing local lesions

on tobacco. Phytopathology, 43 : 11-15.

Varietal reaction of pea to a virus from alsike clover.

Phytopathology, 28 : 505-5 1 1 .

Studies on alfalfa mosaic. Phytopathology, 24: 239-247.

A streak disease of peas and its relation to several strains of

alfalfa mosaic virus. Jour. Agric. Res., 56: 747-772.

Three previously undescribed mosaic diseases of pea. Jour.

Agric. Res., 60: 433^52.

Alfalfa yellow mosaic virus systemically infectious to beans.

Phytopathology, 43 : 38^2.

A new necrotic lesion-producing strain of yellow bean mosaic.

Phytopathology, 43 : 45^9.

275

Fig. 1. — Pea Mosaic Virus.

A. Pisum sativum. B. Trifolium incarnatum. C. Lathynis odoratus. D. Lupiuis

albus.

6096259

276

Fig. 2.— Pea Virus 2.

B. Phaseolus vulgaris. C. Vida faha. D. Lupinus alhus.

A. Pisum sativum.

Ill

Fig. 3. — Pea Stunt Virus.

A. P'lsum sativum. B. and C. Phaseolus vulgaris. D. Vida f aba.

278

Fig. 4. — Pea Wilt Virus, Strain.

A. Fisum sativum. B. Vida faha (local lesions). C. Trifolium prateuse

vulgaris.

D. Fliaseolus

279

Fig. 5 — Lucerne Mosaic Virus, Necrotic Ring Strain.

A. Pisum sativum. B. Vida faba (local lesions). C. Phaseolus vulgaris var. Canadian

Wonder (local lesions). D. Phaseolus vulgaris var. Haricot.

280

A., B. and D. Pliaseolus vulgaris A. local lesions. B. and D. systemic reaction.

C. Crotalaria spectabilis. E. and F. Latbyrus odoratus.

Fig. 6. — Bean Local Chlorosis Virus B.

281

Fig. 7.— Bean Yellow Mosaic Virus, Necrotic Strain.

A. and B. Phaseolus vulgaris. C. Trifolium hyhridum. D. and E. Lathyrus odorcitus.

F. Pisum sativum.

282

Fig. 8. — Tomato Spotted Wilt Virus.

A. Vida f aha. B. Crotalaria spectahilis. C. Pisum sativum. D. Nicotiana glutiuosa.

E. Glycine ma.\. F. Medicago denticulata. G. Trifoliunt fragifenim. FI. Crotalaria

juncea. I. Medicago sativa. J. Lapinas albas.

283

Virus Diseases of Broad Beans.

By

Patricia J. Klesser.

Broad beans are not grown extensively in South Africa. The viruses described

here were found on naturally infected Vida faba plants grown in isolated areas of the

Transvaal, and they are possibly not characteristic for the whole country.

In the field, the symptoms are: mosaic, mosaic with stunt, or severe necrosis and

collapse.

The viruses causing mosaic symptoms are found frequently. An unusual feature

of this mosaic is the fact that the chlorotic areas are depressed below the normal surface

of the leaf.

The systemic necrosis is caused by the tomato spotted wilt virus, which is described

in detail in the paper on pea virus diseases. The incidence of this disease can be fairly

high, with severe crop losses; but usually it occurs only sporadically thoughout a

planting.

REVIEW OF LITERATURE.

Although Elliot transmitted a mosaic from clover to broad bean in 1921, the

first descriptions of naturally occurring mosaic diseases of broad bean were those of

Boning 1927, van der Meulen 1928, Merkel 1929 and Fukushi 1930.

Imai 1934 successfully transmitted a virus causing mosaic on broad bean, to

broad bean, pea and sweet pea, by both aphids and sap.

Further reports of broad bean mosaics came from Gigante in Italy (9); Yu in

China (25); d’Oliviera in Portugal (5); and Quantz in Germany (17).

Murayama (14) determined the physico-chemical properties of a mosaic of broad

beans in Japan.

Pierce (16) described the broad bean local lesion virus which he isolated from red

clover, and which Weiss (22) considers to be the same as the pea wilt virus of Johnson

(12).

Weiss regards the mosaic virus of Boning as being identical with the common pea

mosaic virus. Further, he groups the viruses of Merkel, Murayama and Yu together,

and suggests a relationship with either the bean yellow mosaic or the pea mottle viruses.

Other types of viroses found naturally on broad bean are the mild mosaic and

rosette of Yu (25, 26); a vascular wilt — Stubbs (20); a mottle — ^Bawden, Chaudhuri

& Kassanis (1); a soil-borne necrotic mosaic — Fujikawa (7) and tomato spotted wilt,

Yu (27).

284

Rubio & van Slogteren studied the X-bodies associated with the mottle virus (18).

When artificially inoculated, broad beans develop systemic mosaic or necrotic

symptoms with most of the legume viruses (10, 15, 16, 28, 29, 30, 31, 32) and also with

the tobacco ringspot virus (24) and the curly top virus (19).

With celery virus 1, only local lesions develop (23).

METHODS AND MATERIALS.

Naturally infected broad bean plants were collected from several parts of the

Transvaal. The viruses were inoculated to a number of species of Leguminosae, and

also to several Solanaceous plants. (None of the latter was susceptible).

Carborundum power was used for the sap inoculations and Aphis fabae and

A. craccivora for the insect transmission tests.

1. Broad Bean Mosaic Virus A.

Physical properties: Thermal inactivation point, 55-56 °C. Longevity in vitro,

2-3 days. Dilution end point, 1: 1,000.

Transmission: Mechanical sap inoculation. Aphis fabae.

Host range: Arachis hypogaea L., Crotalaria juncea L., C. spectabilis Roth.,

Glycine max (L.) Merr., Lathyrus odoratus L., Lupinus albus L., L. luteus h.,L. mutabilis

Sweet., Medicago lupulina L., Melilotus officinalis Willd., Phaseolus acutifolius Gxdiy\a.x.

latifolius Freem., P. lunatus L., P. mungo L., P. vulgaris L., Pisum sativum L., Trifolium

hybridum L., T. incarnatum L., T. pratense L., Vida faba L., Vigna sesquipedalis (L.)

Wight., V. unguiculata (L.) Walp., Voandzeia subterranea Thouars.

Reaction of Susceptible Species.

Arachis hypogaea.

Local. No reaction.

Systemic. There is a diffuse chlorotic mottle.

Crotalaria juncea.

Local. There are necrotic spots after 8 days. These fuse and the leaves drop.

Systemic. In 12 days necrosis has spread up the stem and into the veins of the

young leaves (Fig. lb). The plant may collapse or may have malformed rolled leaves,

with dark green blisters.

C. spectabilis.

Local. No reaction.

Systemic. After 3 weeks the young leaves develop irregular chlorotic areas, which

cause a malformation of the leaves, due to uneven growth of the laminae. Soon necrosis

sets in in these areas. Later formed leaves have chlorotic line patterns which become

necrotic to produce an “ etched ” effect. Older leaves may show necrotic rings without

any chlorosis.

285

Glycine max.

Local. No reaction.

Systemic. A few leaves show a chlorotic mottle, the young leaves being symptom-

less. Some old leaves have vivid yellow spots with central necrotic specks.

Lathyrus odoratus.

Local. No reaction.

Systemic. Most leaves are rolled and have dark green streaks.

Lupinus albus.

Local. Chlorotic spots with dark green rings can be seen after 11 days. The

leaflets drop.

Systemic. The young leaves show a vein clearing, and they remain folded and

have wavy margins. Later leaves develop small chlorotic spots and then dark green

blisters. The latter leaves are crinkled and may be malformed. The growing point

is rosetted and stunted.

L. luteus.

Local. No reaction.

Systemic. In 8 to 9 days small chlorotic spots develop on the young leaves. After

3 weeks most leaves are small, malformed and mottled. The plant is stunted. There

may be enough necrosis to cause the collapse of the plant.

L. mutabilis.

Local. No reaction.

Systemic. The young leaves have chlorotic specks. Later ones are chlorotic

with dark green blisters and malformation. The plant is rosetted.

Medicago lupulina.

Local. No reaction.

Systemic. The young leaves develop chlorotic streaks and later leaves are mottled

and slightly puckered.

Melilotus officinalis.

Local. No reaction.

Systemic. The plant is stunted and the leaves have chlorotic vein streaks.

Phaseolus acutifolius.

Local. In 4 days there are chlorotic spots and the veins become necrotic.

Systemic. After 12 days the young leaves develop a vein clearing with small

chlorotic spots. There is necrosis in the chlorosis (Fig. le) and the growing point may

collapse. Later formed leaves are mottled.

P. lunatus.

Local. In 7 to 8 days necrotic lesions develop and there are large chlorotic areas.

Systemic. The necrosis spreads quickly into the stem and veins of the young leaves,

and the plant collapses.

286

?. mungo.

Local. No reaction.

Systemic. A diffuse spot mottling develops.

P. vulgaris var. Canadian Wonder.

Local. There are irregular chlorotic areas which fuse to a general chlorosis.

Systemic. In 6 days the young leaves show a vein clearing. Later leaves develop

a diffuse chlorotic spotting and the older leaves have large chlorotic blotches.

var. Haricot.

Local. Within a week there are chlorotic blotches.

Systemic. At the same time the young leaves curl down severely. They are crinkled

and chlorotic.

Pisum sativum.

Local. In 8 days the leaves show small necrotic specks.

Systemic. Soon after, the young leaves develop a vein clearing with small necrotic

specks. Later leaves and stipules are all similarly speckled and the stem develops

purple necrotic streaks which may result in the collapse of the plant. If not, the leaves

are rosetted and folded with irregular dark green areas and the tendrils are abnormally

curled.

Trifolium hybridum.

Local. No reaction.

Systemic. Most leaves show a chlorotic streak mottle and slight crinkle after

about 3 weeks.

T. incarnatum.

Local. No reaction.

Systemic. After 9 days the young leaves show a vein clearing. Later leaves have

irregular chlorotic areas and are crinkled. Some may have necrotic spots or veins

(Fig. Ic).

T. pratense.

Local. No reaction.

Systemic. A chlorotic streak mottle develops (Fig. Id).

Vida faba.

Local. No reaction.

Systemic. In 9 days the young leaves show a vein clearing. Later formed leaves

develop chlorotic spots or dark green streaks and then a mosaic. A characteristic

of this mosaic is that the chlorotic areas are sunken below the leaf surface, resulting in

the dark green streaks being apparently raised. (Fig. la). The leaves are also rolled

and the plant is slightly stunted.

Vigna sesquipedalis.

Local. Isolated chlorotic areas develop.

Systemic. Only the first and second trifoliates show a diffuse chlorotic mottle.

287

V. unguiculata.

Local. There are chlorotic spots in 4 days, and the leaves become flaccid and drop.

Systemic. After 9 to 10 days the young leaves develop a vein clearing. Later

ones are diffusely mottled.

Voandzeia subterranea.

Local. No reaction.

Systemic. After 3 weeks the leaves show a chlorotic network and irregular chlorotic

areas.

Natural source of virus: Vida faba. The plants were slightly stunted and the

leaves were rolled and elongated. They had a mosaic mottle with the chlorotic areas

depressed below the dark green or normal coloured areas.

Identification.

The characteristic symptom of this virus on broad bean, is a mosaic with the

chlorotic areas depressed. This apparently occurs with only one other described virus.

Boning (3) mentions a similar effect with his virus, which was later identified as

the common pea mosaic virus. However, the pea mosaic is not transmissible to bean

(which is a host of this virus) and causes a mosaic on pea and not a necrosis which is

usual with this virus.

Although pea virus 2 Osborn (15) is able to infect bean, the symptoms on pea

are again chlorotic and not necrotic.

This necrotic reaction on pea can be compared with several others :

(a) The alfalfa mosaic virus IB Zaumeyer (29) causes a similar local and systemic

necrosis on the pea, but only a local necrotic spotting on the bean, whereas

this virus induces a mottle only. Moreover this alfalfa mosaic results in a

necrotic collapse of the broad bean.

{b) The pea American streak virus Zaumeyer (28) is non-infectious to the bean,

causes a local reaction on broad bean and withstands a temperature of 10°

higher than this virus.

(c) The pea Wisconsin streak virus Hagedorn & Walker (10) is not able to infect

either bean or broad bean.

{d) The pea New Zealand streak virus Chamberlain (4) causes a local and systemic

necrosis of the bean (whereas this virus induces a mottle only), and broad

bean is not included in the host range. Further, there is a great difference

in the physical property values of the 2 viruses.

{e) The alsike clover mosaic virus Zaumeyer & Wade (32) results in a conspicuous

reflexing of the bean leaves, which does not occur with this virus.

(/) The white clover mosaic virus complex Zaumeyer & Wade (31) induces a

necrosis on both the bean and broad bean, both of which show a mottle with

this virus.

(g) Although the symptoms of the pea stunt virus Zaumeyer (30) on pea and bean

are similar to this virus, symptoms on other hosts, the host range and the

physical property values are different.

Pierce’s broad bean local lesion virus (16) causes only a local necrosis on broad

bean, while Stubbs’ virus (20) induces a severe systemic necrosis.

288

The mild mosaic virus of Yu (25) is unable to infect bean, cowpea and soybean

(all of which are hosts of this virus) and does not withstand more than 3 hours ageing,

whereas this virus lasts 2-3 days in vitro.

The broad bean mottle virus Bawden, Chaudhuri & Kassanis (1) induces symptoms

similar to this virus on bean, pea, sweet pea and the clovers, but differs on soybean and

broad bean. (During the initial subinoculations with the mottle virus, broad bean

developed a systemic necrosis as well, but later, only a mottle). However, Bawden’s

virus withstood heating up to 95°C., and ageing up to 20 days. Further it was not

transmissible by any of the Aphis species.

As this virus cannot be fully identified with any previously described, it may be

a new virus or strain. It is named broad bean mosaic virus A.

2. Broad Bean Mosaic Virus B.

Physical properties: Thermal inactivation point, 56-58°C. Longevity in vitro,

1-2 days. Dilution end point, 1 : 1,000.

Transmission: Mechanical sap inoculation. Aphis craccivora.

Host range: Crotalaria juncea L., C. spectabilis Roth., Lupinus albus L. vars.

sweet and bitter, Phaseolus lunatus L., P. vulgaris L., Pisum sativum L., Trifolium

hybridum L., T. incarnatum L., Vida faba L., Vigna sesquipedalis (L.) Wight., V.

iinguiculata (L.) Walp., Voandzeia subterranea Thouars.

Reaction of Susceptible Species.

Crotalaria juncea.

Local. No reaction.

Systemic. In 8 days the young leaves develop a vein clearing and spotting. The

next leaves have chlorotic streaks which lead to a malformation. Later leaves are small

and rolled and have dark green blisters. (Fig. 2e). The plant is stunted.

C. spectabilis.

Local. No reaction.

Systemic. In 13 days there are chlorotic spots and vein flecks on the young leaves.

Later leaves have broad alternate chlorotic and green veinbands (Fig. 2c). There may

be necrotic stem streaks.

Lupinus albus var. Sweet.

Local. After 13 days the leaves have chlorotic spots with necrotic veins, and they

soon drop.

Systemic. At the same time the young leaves remain folded and have small chlorotic

spots. Later formed leaves are small, malformed and mottled. The growing point is

rosetted and the plant stunted.

var. Bitter.

Local. No reaction.

Systemic. The first leaves are mottled. Later the growing point becomes rosetted

with small chlorotic leaves. There may be necrotic stem streaks.

289

Phaseolus lunatus.

Local. No reaction.

Systemic. After 3 weeks chlorotic vein flecks develop (Fig. 2b).

P. vulgaris var. Canadian Wonder.

Local. In 11 to 12 days large chlorotic blotches appear.

Systemic. The first few trifoliates develop large chlorotic areas (Fig. 2d). New

leaves are symptomless.

var. Haricot.

Local. In 10 to 12 days there are small chlorotic areas.

Systemic. Soon after, the young leaves curl down and have a chlorotic network.

Later formed leaves are small, puckered and mottled.

Pisum sativum.

Local. No reaction.

Systemic. After 11 to 12 days the young leaves show a vein clearing and spotting.

The next formed leaves develop a mosaic, and the tendrils are abnormally curled-

Necrosis of the growing point leads to the eventual collapse of the plant.

Trifolium hybridum.

A symptomless carrier.

T. incarnatum.

Local. No reaction.

Systemic. The young leaves show a chlorotic network and flecking in 11 days.

Later leaves are crinkled and have a mosaic mottle. (Fig. 2a). The plant is severely

stunted.

Vida faba.

Local. No reaction.

Systemic. In 18 days chlorotic spots develop on the young leaves. Later leaves

show a mosaic with the chlorotic areas depressed, leaving raised dark green islands.

Vigna sesquipedalis.

Local. No reaction.

Systemic. There is a chlorotic network followed by a chlorotic spotting.

V. unguiculata.

Local. No reaction.

Systemic. The young leaves show small chlorotic specks after 18 days. Later

ones are chlorotic with short dark green veinbands, and are rolled downwards.

Voandzeia subterranea.

A symptomless carrier.

Natural source of virus: Vida faba. Although the mosaic symptoms on the leaves

were similar to those on the plants from which the A virus was isolated, there was little

stunt or rolling. The general appearance of the plants was normal.

Crotalaria spectabilis. Most leaves had bands of vivid yellow and dark green.

There was also a slight veinal necrosis.

290

Identification.

This virus is very similar to the previous one, but differences in host range, physical

property values and symptoms on some plants are too marked for it to be classified

as the same.

As this virus also results in the typical sunken chlorotic reaction on the broad

bean, it is named broad bean mosaic virus B.

SYMPTOM DIFFERENCES BETWEEN BROAD BEAN MOSAIC VIRUSES A AND B.

.Abbreviations used: —

bl. — blisters.

chi. — chlorosis/chlorotic,

coll. — collapse,

d. gr. — dark green,

irreg. — irregular.

mail. — malformed.

mos. — mosaic.

mot. — mottle,

no. — no reaction,

necr. — necrosis/necrotic.

n.s.s. — necrotic stem streak,

str. — streaks,

veinb. — veinbands.

PHYSICAL PROPERTIES AND METHODS OF TRANSMISSION.

291

HOST RANGES.

LITERATURE CITED.

1. Bavvden, F. C., Chaudhuri, R. P.

AND Kassanis, B. (1951)

2. Berkeley, G. H. (1947)

3. Boning, K. (1927)

4. Chamberlain, E. E. (1939)

Some properties of the broad-bean mottle virus. Ann.

Appl. Biol., 38; 774-784.

Alfalfa mosaic on pepper in Ontario. Phytopathology

(Abstr.), 37; 3.

Die mosaikkrankheit der Ackerbohne {Vida faba L.) Ein

Beitrag zu dem Mosaik der Papilionaceae. Forsch. auf

dem Gebiet der Pflanzenkrankh. und der Immunitat im

Pflanzenreich, 4; 43-111.

Pea Streak (Pisimi virus 3). N.Z. Jour. Sci. & Tech., 20;

365A-381A.

5. d’Oliviera, Maria de L. (1946). . O “ mosaico ” de Faviera e suas relapoes com outras Legumi-

nosas. Broteria, 15; 90-94. (Abstr. R.A.M., 26; 372).

6. Elliott, J. A. (1921) A mosaic of sweet and red clovers. Phytopathology, 11:

146-148.

7. Fujikawa, T. (1955) Studies on the necrotic mosaic of broad bean. Agric. &

Hort., 30: 721-722. (Abstr. R.A.M., 35: 742.)

8. Fukushi, T. (1930) On the mosaic disease of broad bean. Jour. PI. Prot., 18:

707-712,779-784. (Abstr. R.A.M., 10: 700.)

9. Gigante, R. (1937) 11 mosaico della Fava {"Vida faba ’L.) in Italia e comporta-

mento di alcune Feguminose di fronte ad esso. Boll.

Staz. Pat. veg. Roma., 17: 497-529. (Abstr. R.A.M.

17: 645.)

10. Hagedorn, D. J. & Walker, J. C. Wisconsin pea streak. Phytopathology, 39: 837-847.

(1949).

11. Imai, Y. (1934) Studies on the transmission of broad bean mosaic. Trans.

Sapporo Nat. Hist. Soc., 13: 241-245. (Abstr. R.A.M.,

14: 4).

292

293

Fig. 1. — Broad Bean Mosaic Virus A.

A. Vida faba — showing depressed chlorotic areas. B. Crotalaria juncea.

C. Trifolium incarnatum. D. Trifolium pratense. E. Phas'eolus lunatus.

294

Fig. 2. — Broad Bean Mosaic Virus B.

A. Trifolium incarnatum. B. Phaseolus lunatus. C. Crotalaria spectabilis.

D. Phaseolus vulgaris. E. Crotalaria juncea.

295

Chromosome Morphology in Kiiiphofia.

By

J. M. J. de Wet.

ABSTRACT.

A number of species and varieties of the genus Kniphofia (Liliaceae) were studied

cytologically. The somatic chromosome number h 2n = 12 in all the species. This

is also true in Notosceptrum natalense Baker.

MATERIAL AND METHODS.

The material studied was collected in the veld and cultivated on the Prinshof

Experiment Station in Pretoria. The collectors’ numbers listed in Table 1 also refer

to the specimens filed with the National Herbarium, Pretoria. Root-tips were fixed

in Randolph’s (1935) fluid, dehydrated using normal butyl alcohol and embedded

in a paraffin-beeswax-rubber mixture. Sections were stained in Stockwell’s (1934)

fluid. Inflorescences were fixed in 3: 1 absolute ethyl alcohol, propionic acid. Anthers

were squashed in propionic-carmine after the method outlined by Swaminathan,

Magoon and Mehra (1954). Karyotypes were computed with the aid of a camera

lucida, using the average length of chromosomes from five different metaphase plates.

The magnification is X2,500.

CHROMOSOME NUMBER.

The genus was previously studied by Webber (1932), Moffet (1932) and Janaki-

Ammal (1950), demonstrating n = 6. Polyploidy appears to be absent except in a

triploid specimen of K. snowdenii from Uganda (Janaki-Ammal, 1950). The species

studied are listed in Table 1. Polyploid cells were observed in otherwise normal diploid

roots of K. sp. (Codd 6955). This is also true among pollen mother cells of this species.

In these cells quadrivalents, together with a varying number of bivalent and univalents,

were observed. Similar observations were made by Moffet (1932) in polyploid cells of

K. triangularis ( = K. nelsonii). In the diploid pollen mother cells, chromosome pairing

and movement are normal.

CHROMOSOME MORPHOLOGY.

The haploid chromosome set of « = 6 in Kniphofia may be subdivided into three

distinct groups. Two chromosomes are significantly longer than the rest. Both are

characterized by secondary constrictions. They differ from each other in the position

of the centromere.

Three chromosomes are of medium length. One of these is characterized by a

secondary constriction, whereas the other two differ from each other in the position of

the centromere. The sixth chromosome is short with a submedian centromere.

The material of Notosceptrum resembles Kniphofia in karyotype except for the sixth

chromosome which is also of medium length. The karyotype varies very little from one

species of Kniphofia to the other.

During meiotic telophase six nucleoli are always produced. This indicates the

presence of six chromosomes with nucleoli organizing regions (Pathak, 1940, Gates,

1942, de Wet, 1953). These are evidently the six chromosomes (three chromosome

pairs) with secondary constrictions.

296

Table 1. — Chromosome numbers in Kniphofia a/j^/Notosceptrum.

13I1'

14

17

297

DISCUSSION.

Polyploidy is absent in the South African species of Kniphofia. The presence of

polyploid cells in otherwise normal diploid species could give rise to viable pollen with

variable chromosome numbers. These could produce individuals such as the triploid

specimen of K. snowdenii from Uganda. Although more than 50 collections of various

species of Kniphofia from all over the Union were studied no abnormal chromosome

numbers were encountered.

Species evolution in Kniphofia must have taken place through point mutations

and small chromosomal aberrations. This perhaps is also the reason why many

collections are difficult to classify with certainty into specific units. The specific charac-

ters could become established in the newly evolved individuals through isolating mecha-

nisms such as flowering time and geographical distribution. When the species are

planted together they hybridize freely.

References.

de Wet, J. M. J. (1953). Cytologia 18: 229-234 (1953).

Gates, R. R. Jour. Linn. Soc. Lond. 51: 7-9 (1940).

Janaki-Ammal, E. K. Jour. Roy. Hort. Soc. 75: 23 (1950).

Moffet, A. A. Jour. Gen. 25: 315 (1932).

Pathak, G. N. Jour. Gen. 39: 437-467 (1940).

Randolph, L. F. Stain Techn. 10: 95-96 (1935).

Stockwell, P. Science 80: 121-122 (1934).

Swaminathan, M. S., M. L. Magoon and K. L. Mehra. Indian Jour. Genet. Plant

Breed. 14: 87-88 (1954).

Webber, J. M. Amer. Jour. Bot. 19: 411 (1932).

299

Leaf Anatomy of Six South African

Grass Genera.

By

J. M. J. de Wet.

The significance of leaf anatomy in grass taxonomy was demonstrated by Duval-

Jouve (1875), Pee-Laby (1898), Avdulov (1931) and Prat (1936). This paper deals with

the genera Entoplocamia Stapf, Fingerhuthia Nees, Lasiochloa Kunth., Plagiochloa

Adamson and Sprague, Tetrachne Nees and Urochlaem Nees. They are usually included

in the Festuceae but appear to occupy an aberrant position in this tribe. Recently,

Chippindall (1955) referred Fingerhuthia Nees to the tribe Chlorideae and the remaining

five genera to the tribe Eragrosteae. Anatomical data may provide clues as to the true

relationships of these genera.

MATERIAL AND METHODS

The specimens investigated were made available by the National Herbarium,

Pretoria. Anatomical slides were prepared as indicated by Prat (1948) and de Wet

(1956). Root-tips were collected in the veld, fixed in Randolph's (1935) fluid, dehydrated

using normal butyl alcohol as clearing agent and stained in Stockwell’s (1934) solution.

RESULTS

Anatomical, morphological and cytological observations are presented in Table 1.

The leaf anatomy of Fingerhuthia Nees was first studied by Prat (1936) who indicated

that the genus is chloridoid in this respect. Moffet and Hurcombe (1949) demonstrated

small chromosomes and 2n — 20 for Tetrachne dregei Nees and de Wet (1958) reported

small chromosomes in Fingerhuthia africana Lehm. (2«= 40) and F. sesleriaeformis

Nees (2« = 20). The chromosomes of Plagiochloa uniolae (L.f.) Adamson and Sprague

{In = 48) were found to be larger than those of the other two genera. The remaining

three genera are unknown cytologically. Morphological data, although varified, are

borrowed from Stapf (1900), Phillips (1951) and Chippindall (1955). The genus Plagio-

chloa was recently removed from Demazeria Dum. by Adamson & Sprague (1941).

This new genus is closely related to Fasiochloa Kunth. Chippindall (1955) indicates

that their distribution is similar and although the extremes in each genus are clearly

distinct, there are species in both that suggest possible hybridization between them.

Discussion

The characteristics of the main subdivisions of the Gramineae have been illustrated

and discussed by Stebbins (1956). It is evident that the genera studied do not belong

with the tribe Festuceae. This tribe is characterized by large chromosomes, mostly

in multiples of n — 1, spherical siliceous cells in an epidermis that lacks bicellular

hairs and the chlorophyll tissue of the leaf is uniformly distributed throughout in the

300

mesophyll between the vascular bundles. The genera Entoplocamia Stapf, Fingerhuthia

Nees and Tetrachne Nees are chloridoid in all respects, except that the spikelets dis-

articulate below the glumes, but this characteristic is present in other typically chlori-

doid genera such as Catalepis Stapf and Stent and Spartina Schreb. For these reasons

the classification of Chippindall (1955) who refers Fingerhuthia Nees to the Chlorideae

and the other two genera to the Eragrosteae are justifiable. The genera Fasiochloa

Kunth., Plagiochloa Adamson and Sprague and Urochlaena Nees, however, do not

belong with either of the tribes Eragrosteae or Eestuceae. Internal leaf anatomy is of the

festucoid type and epidermal traits are pancicoid. The chromosomes of Plagiochloa

imiolae Adamson and Sprague are larger than those of the panicoid group of grasses,

but smaller than those of the festucoid group. In respect of anatomical traits, chromo-

some size and basic chromosome number these genera resemble members of the tribe

Danthonieae. The rhachilla, however, disarticulates below the glumes, a character

not present in members of the Danthonieae. These genera also resemble members of

the Unioleae except for the relatively large chromosomes observed in Plagiochloa

imiolae Adamson and Sprague.

Summary

Leaf anatomy was studied in six small South African grass genera. These data

were correlated with cytology and spikelet morphology in an attempt to determine

their phylogenetic position in the family Gramineae. The genera Fntoplocamia Stapf,

Tetrachne Nees and Fingerhuthia Nees are chloridoid in respect of leaf anatomy and

spikelet morphology. Cytological data from the latter two genera point to a similar

conclusion. Fasiochloa Kunth., Urochlaena Nees and Plagiochloa Adamson and

Sprague are festucoid in respect of spikelet morphology and internal leaf anatomy,

panicoid in respect of epidermal traits and the latter genus resemble members of the

Danthonieae in respect of chromosome size and basic number.

Key based on Anatomical and morphological Characters.

1. Inner bundle sheath poorly differentiated, outer bundle sheath comprises a layer

of large parenchymatous cells which contain chloroplasts, remaining chlorophyll

tissue confined to a single layer of cells surrounding the vascular bundle; stomata

rhomboid in shape, siliceous cells saddle-shaped, bicellular hairs club-shaped;

chromosomes small, /; = 10; embryo as long as the grain; rhachilla dis-

articulates below the glumes 2

Inner bundle sheath well developed, outer bundle sheath comprises a layer of

small parenchymatous cells which do not contain chloroplasts, chlorophyll uni-

formly distributed throughout the mesophyll between the bundles; stomata

rhomboid in shape, siliceous cells dumbbell-shaped, bicellular hairs linear in

shape; chromosomes relatively large, /; = 12; grain enclosed by the lemma and

palea; rhachilla disarticulates below the glumes; flowers all bisexual or the

uppermost gradually reduced 4

2. Lower florets all bisexual, uppermost lemma subtending a male flower or

empty; grain not enclosed by the lemma and palea, lodicules 2, cuneate, fleshy,

nerved Fingerhuthia

Lower lemmas empty florets aborted, upper bisexual, uppermost subtending a male

flower or palea only; grain loosely enclosed by the lemma and palea 3

3. Lodicules absent Entoplocamia.

Lodicules 2, cuneate, fleshy Tetrachne.

4. Embryo large, i the length of the grain; lodicules 2, minute Urochlaena,

Embryo small, less than \ the length of the grain; lodicules 2, cuneate, ciliate. . 5

5. Spikelets 3-9-flowered, placed obliquely to the central axis Plagiochloa.

Spikelets 2-3-flowered, not placed obliquely to the central axis Fasiochloa,

LITERATURE CITED

Adamson, R. S. and T. A. Sprague The genus Plagiochloa. Jour. So. Afr. Bot. 7; 89-92.

(1941)

Avdulov, N. P. (1931) Karyo-systematische Untersuchung der Familie Gramineen.

Bull. App. Bot. Plant Breed. Suppl. 44; 1-428.

301

Chippindall, L. K. a. (1955) The classification of grasses. The grasses and pastures of

South Africa: 1-527.

De Wet, J. M. J. (1956) Leaf anatomy and phylogeny in the tribe Danthonieae.

Amer. Jour. Bot. 43: 175-182.

(1958) Additional chromosome numbers in Transvaal grasses.

Cytologia 23: in press.

Duval-Jouve, M. j. (1875) Histotaxie des feuilles de Graminees. Ann. Sci. Nat. Bot.

XIV. 1: 294-371.

Mofeet, a. a. and R. Hurcombe, Chromosome numbers of South African grasses. Heredity

(1949) 3: 369-374.

Pee-Laby, E. (1898) Etude anatomique de la Eeuille des Graminees de la France.

Ann. Sci. Nat. Bot. VIII. 8: 227-346.

Phillips, E. P. (1951) The genera of South African flowering plants. Bot. Survey

So. Afr. Mem. 25: 58-148.

Prat, H. (1936) La systematique des Graminees. Ann. Sci. Nat. Bot. X.

18: 165-258.

(1948) General features of the epidermis in Zea mays. Ann. Missouri

Bot. Gard. 35: 341-351.

Randolph, L. F. (1935) A new fixing fluid and a revised schedule for the paraffin

method in plant cytology. Stain Techn. 10: 95-96.

Stapf, O. (1900) Gramineae. Flora (Tapensis 7 : 310-750.

Stebbins, G. L. (1956) Cytogenetics and evolution of the grass family. Amer. Jour.

Bot. 43: 890-905.

Stockwell, P. (1934) A stain for difficult plant material. Science 80: 121-122.

303

Leaf Anatomy and Morphology in South African

Species of Danthonia.

By

J. M. J. de Wet. *

The genus Danthonia is variable and this is particularly true of the South African

species (de Wet, 1954a and 1956). On the basis of anatomical and morphological

evidence Nevski (1934)t removed D. glauca Nees and D. forskalii (Vahl.) R. Br. from

Danthonia to form the basis of a new genus, Asthenatherum. It was suggested by

Hubbard (1937) that some other South African species may be equally worthy of generic

rank. A new genus Alloeochaete was described by Hubbard (1940) to include D. andon-

gensis Rendle. The South African species were studied in detail in order to determine

whether any other species should be removed from Danthonia.

MATERIAL AND METHODS

Anatomical slides were prepared from Herbarium material as indicated by Prat

(1948) and de Wet (1956). Chromosome numbers were counted in root-tip sections.

Root-tips were collected in the veld, fixed in Randolph’s (1935) fluid, dehydrated and

sectioned as indicated by de Wet (1953) and stained in Stockwell’s (1934) solution.

Drawings of chromosomes and vascular bundles were made with the aid of a camera

lucida.

RESULTS

The anatomical study is confined to mature leaves. These data are presented in

the form of a key in Table 1. Only the lower epidermis surface has been studied. In

D. forskalii (Vahl.) R. Br. the siliceous cells are in the shape of short dumbbells which

often appear spherical.

Table 1 .—ANATOMICAL TRAITS

1. Chlorophyll localized around the vascular bundles; inner bundle sheath scleren-

chymatous, outer bundle sheath parenchymatous; bicellular hairs present;

siliceous cells dumbbell-shaped, localized above nerves; unieellular hairs form

an arch over stomata. (Figs. 2 and 5) D. forskalii, D. glauca, D. mossamedensis,

D. suffrutescens.

Chlorophyll not localized around the vascular bundles 2

2. Chlorophyll uniformly distributed throughout the mesophyll of the leaf (Fig. 3). . 3

Chlorophyll not uniformly distributed throughtout the mesophyll of the leaf... . 4

3. Inner bundle sheath sclerenehymatous 5

Inner bundle sheath colenchymatous 6

5. Outer bundle sheath composed of thin parenchyma; bicellular hairs present ... . 7

Outer bundle sheath composed of thick parenchyma; bicellular hairs absent. ... 8

* For reading the manuscript and for his valuable suggestions, the author wishes to thank

Dr. G. L. Stebbins.

t Ex, Hubbard (1937), page 135.

304

7. Unicellular hairs do not form an arch over the stomata 9

Unicellular hairs form an arch over the stomata 10

9. Epidermis cells thin parenchyma; dumbbell-shaped siliceous cells localized above

the nerves (Fig. 6). D. brachyphylla, D. cwva, D. macrantha, D. purpurea, D,

tenella.

Epidermis cells thick parenchyma; spherical siliceous cells localized above the

nerves (Fig. 7) D. dura, D. stricta.

10. Epidermis cells thick parenchyma; dumbbell-shaped siliceous cells localized

above nerves D. pumila.

8. Epidermis cells thick parenchyma; halfmoon-shaped siliceous cells distributed

all over the epidermis D. cincta, D. papposa.

6. Outer bundle sheath composed of thin parenchyma 11

Outer bundle sheath composed of thick parenchyma 12

11. Epidermis cells thin parenchyma; bicellular hairs absent; siliceous cells, spherical,

localized above nerves D. lanata, D. lupulina.

Epidermis cells thick parenchyma; bicellular hairs absent; siliceous cells spherical,

distributed all over the epidermis D. arundinacea.

12. Phloem divided into two groups by sclerenchymatous cells 13

Phloem in one group 14

13. Epidermis cells thin parenchyma; bicellular hairs absent, siliceous cells spherical,

localized above nerves D. zeyheriana, D. zeyheriana var. trichostachya.

Epidermis cells thick parenchyma ; bicellular hairs absent ; siliceous cells spherical,

distributed all over the epidermis D. drakensbergemis, D. macowanii.

14. Epidermis cells thick parenchyma; bicellular hairs absent, siliceous cells spherical,

distributed all over the epidermis (Fig. 8) D. disticha.

4. Leaves more or less spherical: chlorophyll localized in a continuous band 3-5

layers of cells thick directly below the epidermis; mesophyll in the middle lacks

chloroplasts; inner bundle sheath sclerenchymatous, outer bundle sheath com-

posed of thin parenchyma (Fig. 1); bicellular hairs absent; epidermis cells thick

parenchyma; siliceous cells spherical, localized above nerves D. rangei.

The morphology of the South African species of Danthonia has been discussed

by Stapf (1900), Hubbard (1937) and Chippindall (1955). It will suffice to discuss the

main characteristics of each group into which Danthonia was subdivided on the basis

of anatomical traits. The species D. mossamedensis Rendle and D. sujfrutescens Stapf

share with Aslhenatherum (D. glauca Nees and D. forskalii (Vahl.) R. Br.) the following

prominent characteristics of external morphology. The lower culms are shortly pube-

scent to tomentose; innovation shoots are extravaginal; glumes strongly chartaceous,

the lower distinctly 7-1 1-nerves and the upper 5-9-nerved; lemmas 7-10-nerved with

the hairs arranged in rows between the nerves, the outer rows ending in tufts of longer

hairs just below the lobes and the inner rows ending just below the insertion of the awn;

lodicules 2, cuneate, nerved and glabrous; callus acute. These characteristics are also

descriptive of D. pumila Nees, except that in the latter species the callus is obtuse.

Danthonia glauca Nees and D. suffrutescens Stapf resemble each other so closely that

the latter species could perhaps be regarded as a variety of D. glauca Nees. As indicated

by Stapf (1900) they differ from each other only in two characters. The spikelets of

D. suffrutescens Stapf are slightly larger and more acuminate than those of D. glauca

Nees and the plant is suffrutescent with the lower sheath coriaceous. In contrast D.

glauca Nees has smaller spikelets, is herbaceous and has a thin lower sheath. The genus

Asthenatherum is further characterized by the panicoid type of internal leaf anatomy

and epidermis.

The typical Danthonia species with dumbbell-shaped siliceous cells may be sub-

divided into three groups on the basis of external morphological traits. Danthonia

macrantha Schrad. and D. brachyphylla Nees resemble Fentameris in superficial ap-

pearance but are more typical of Danthonia in spikelet morphology (Chippindall, 1955).

They differ from other species of Danthonia conspicuously only in having two florets per

spikelet and in that the hairs of the stigma bend down to join over the top of the ovary.

305

Fig. 1-4. — Internal leaf anatomy x 750. 1. D. rangei; 2. D. forskalii ; 3. D. arundi-

nacea; 4. D. cincta.

Fig. 5-6. — Epidermis X 1000. 5. D. forskalii; 6. D. ciirva; 7. D. stricta; 8. D. disticha.

6096259—4

306

Glumes that are 5-nerved; 9-nerved lemmas with a transverse fringe of tufts of

long hairs below the insertion of the awn and several submarginal tufts below this

fringe; distinctly 3-nerved lemma-lobes and 2 lodicules which are small, cuneate-

obovate and ciliate are characteristic of D. purpurea Beauv. and the closely related

annual D. tenella Nees.

Danlhonia curva Nees with dumbbell shaped siliceous cells resemble D. lanata

Schrad., D. tupulina (Thunb.) Roem. and Schult. and D. zeyheriana Steud. in spikelet

morphology. These species have in common: glumes 3-5-nerved; lemmas 9-nerved

with the hairs arranged in rows between the nerves ending just below the middle in a

transverse fringe of long hairs; lobes of the lemma partly adnate to the awn; lodicules 2,

obovate, ciliate, nerved.

Danthonia cincta Nees and D. papposa Nees differ from all other species of Danthonia

in having 1-3-nerved glumes. Anatomically they are also quite distinct from other

species of Danthonia (Figure 4). The cells between the bundles are sclerenchymatous

except for two or more layers of cells along grooves of the upper epidermis. The

chloroplasts are localized in these layers of parenchymatous cells.

Morphologically D. dura Stapf D. stricta Schrad., D. drakensbergensis Schweick.

and D. macowani Stapf form a natural group with the following salient features:

glumes 3-5-nerved; lemmas 7-9-nerved, with the hairs variously tufted; lemma-

lobes partially adnate to the awn; callus obtuse; lodicules 2, obovate, ciliated, nerved.

Fig. 9. — Diagram showing the relationships between species of Danthonia. The numbers

represent groups of species as follows: Hypothetical ancestor; i. D. curva, D.

purpurea, D. tenella; ii. D. pumila; iii. Asthenatherum (D. forskalii, D. glauca,

D. mossamedensis, D. suffrutescens); iv. D. rangei; v. D. arundinacea; vi. D. cincta,

D. papposa; vii. D. lanata, D. lupulina; viii. D. dura, D. stricta; ix. D. drakens-

bergensis, D. macowanii, D. zeyheriana; x. D. disticha, xi. D. brachyphylla, D.

macrantha.

Fig. 10-11. — Chromosome numbers X 3000. 10. D. forskalii, 11. D. tenella; 12.

D. arundinacea; 13. D. macowanii.

307

Danthonia arundinacea (Berg.) Schweick. is a robust, tufted perennial with stout

culms from 80 cm to more than 100 cm high; glumes 1-nerved; lemmas 7-9-nerved,

villous all over and the lobes taper into a fine, short awn. Danthonia disticha Nees

differs from all other species of Danthonia in having the inflorescense shaped into a

distichous spike. An analysis of the spikelet reveals typical Danthonia characteristics.

Glumes 3-nerved, lemmas 9-1 1-nerved, bilobed, with the hairs arranged in a tuft

at the base and with a line of similar hairs on each side extending to the base of the

lobes; lodicules 2, obovate, ciliated, nerved.

Danthonia rangei Pilger with its peculiar internal leaf anatomy resembles

Asthenatherum in habit and superficial appearance. Spikelet morphology, however, is

more typical of Danthonia, except that the florets are reduced to 2 per spikelet. Glumes

are 3-nerved; lemmas 5-7-nerved, hairy all over, with three tufts of hairs, one each at

the base of the lobes and the third at the insertion of the awn; lodicules 2, cuneate,

minute; callus obtuse.

Three species, D. macowanii Stapf (2n = 12), D. arundinacea (Berg.) Schweick.

(2/7 = 12) and D. tenella Nees {In = 24), were studied for the first time cytologically.

A number of other species were previously studied by Calder (1937), Stebbins and

Love (1941) and de Wet (1954 b). Basic chromosome numbers appear to be « = 6

and n = 1. It was also indicated by de Wet (1954b) that the chromosomes of species

belonging to Asthenatherum [D. forskalii (Vahl.) R. Br.] are smaller than those of

more typical members of Danthonia (compare Figures 10-11).

DISCUSSION.

The trends of evolutionary progress in the family Gramineae were pointed out by

Stebbins (1956). From this discussion it becomes evident that the complex panicoid-

eragrostoid type of epidermal traits (siliceous cells variously shaped, bicellular hairs

present) and the festucoid type of chlorophyll distribution (uniformly distributed

throughout the mesophyll of the leaf), are the more primitive leaf anatomical characters.

Members of the genus Danthonia exhibit a combination of primitive and advanced

anatomical traits.

The known diploid species are confined to southern Africa. On the basis of

stomatal size it would appear as if these, D. curva Nees, D. disticha Nees, D. arundinacea

(Berg.) Schweick. and D. macowanii Stapf, are the only diploid representatives of the

genus. The stomata of these species vary in size from 23-30 // to 25-00 //. Other

species with relatively small stomata are present in South America: D. lechleri Steud.

(26-40 //) and D. secimdiflora Presl. (27-60 //), but they resemble in stomatal size the

known tetraploid species D. tenella Nees (26-30 /<) and D. purpurea Beauv. (26-40 ,w)

more closely (de Wet, 1954b).

The known diploids exhibit the primitive type of chlorophyll distribution. The

epidermis is of the primitive type in D. curva Nees, but more advanced (bicellular hairs

are absent and the siliceous cells are spherical) in the remaining three diploids. The

lemma-lobes are well developed in these diploids, with the lemmas hairy all over in

D. arundinacea (Berg.) Schweick. and variously tufted and fringed in the others.

In respect of spikelet morphology the North American species, together with some

from South America and the European species, D. calycina Roem. and Schult., appear

to be the most primitive. These are characterized by relatively poorly developed

lemma-lobes and lemmas are hairy along the margins only, or also sparsely so on the

back. The North American species with these characteristics are hexaploids (2n = 36)

according to Stebbins & Love (1941). South American species which exhibit these

primitive floral characteristics: D. chilensis Desv. has 2n = 36, D. oresigena Phil.

308

has 2n ^ 48, D. cirrata Hack, and Arech. has 2n — 72 and the remaining species

D. malacantha (Steud) Pilger, D. montana Doell., D. montevidensis Hack, and Arech.,

D. secundiflora Presl. and D. tandilensis Kotze are unknown cytologically. All these

presumed primitive species are characterized by an epidermis with dumbbell-shaped

siliceous cells and linear bicellular hairs. The chlorophyll is uniformly distributed

throughout the mesophyll of the leaf and the outer bundle sheath is parenchymatous.

From this discussion it would appear as if the most primitive characters exhibited

by present day species of Danthonia are as follows: Lemma-lobes poorly developed;

hairs on the lemma not tufted or fringed and confined to the margins and the back;

inner bundle sheath of the vascular bundles of the leaf sclerenchymatous, outer bundle

sheath composed of parenchymatous cells; chlorophyll uniformly distributed throughout

the mesophyll of the leaf; epidermis with dumbbell-shaped siliceous cells and linear

bicellular hairs. No South African species exhibit all of these assumed primitive

characters.

Two dominant lines of phylogenetic specialization are evident in the South African

species of Danthonia. The first line gave rise to the genus Asthenatherum. The latter

genus is regarded in this discussion to include the following species: D.forskalii (Vahl.)

R. Br., D. glauca Nees, D. mossamedensis Rendle and D. siiffrutescens Stapf. Morpho-

logically they resemble the more typicai species of Danthonia in many respects (Hubbard,

1937). Anatomically they retained the primitive epidermal characters but became

specialized in respect of chlorophyll distribution. Only one of these species, D. forskalii

(Vahl.) R. Br. is known cytologically, but it has been pointed out by de Wet (1954b)

that although the basic chromosome number is, as in Danthonia, n = 6 or 12 the

chromosomes are significantly smaller than those of typical representatives of Danthonia.

It is of interest to note that all the species here included in Asthenatherum are adapted

to grow exclusively in dry sandy areas. The second line of phylogenetic development

retain the primitive type of chlorophyll distribution, but the outer bundle sheath became

sclerenchymatous, bicellular hairs are absent from the epidermis and the siliceous cells

are spherical. This group includes the species D. disticha Nees, D. drakensbergensis

Schweick., D.,fnacowanii Stapf and D. zeyheriana Steud. Three more species, D. lanata

Schrad., D. hipidina (Thunb.) Roem. and Schult., and D. arundinacea (Berg.)

Schweick., resemble the above species in leaf anatomical characters except that the

outer bundle sheath is parenchymatous.

Danthonia stricta Schrad. and D. dura Stapf combine in their epidermis a peculiar

combination of primitive and advanced characters. The siliceous cells are spherical,

but linear bicellular hairs are always present.

Three species with the festucoid type of epidermal traits appear to represent specia-

lized lines of development in respect of internal leaf anatomy. The chlorophyll bearing

tissue is localized but not as in the highly developed panicoid type. These are D. rangei

Pilger on the one side and D. cincta Nees and D. papposa Nees on the other side.

Among the species which retained the primitive type of epidermal traits, D. brachy-

phylla Nees and D. macrantha Schrad. exhibit a reduction in the number of fertile

florets to two per spikelet. In these species the glumes are much enlarged and could

serve as a means of seed dispersal.

The relationships of the species groups into which Danthonia may be subdivided

on the basis of leaf anatomical and morphological characters are indicated in Figure 9.

The actual connections between the species will only be known after a complete cyto-

genetical study of the genus has been undertaken. For this reason the type of diagram-

matic presentation discussed by Stebbins (1956) has been adopted.

309

The various theories about centre of origin and plant migration have been discussed

by Cain (1944). Although the greatest concentration of species is found in the south

western tip of Africa, it would appear as if the genus migrated from the North. The

species D. calydna Roem. and Schult. from Eastern Europe was shown to have re-

tained many primitive characters and could perhaps be regarded as a relic. The tropical

African species are characterized by the primitive type of anatomical characters and

could perhaps be regarded as relics which have adapted themselves to their mountain

existence in a tropical climate.

Small colonies of one or more species are also localized along the mountain ranges

of the summer rainfall areas of southern Africa with a greater concentration of species

in the winter rainfall area. The genus Asthenatherum extends from south east Asia

into North Africa and along the arid areas of the west coast to South Africa. The

presence of a species like D. pumila Nees along the arid west coast of South Africa is

difficult to explain. As was pointed out earlier, this species resembles Asthenatherum

very closely except for chlorophyll distribution of the leaf. In this respect it appears

to form a direct link between Danthonia and Asthenatherum. If this species were present

in south eastern Asia or North Africa it could have been regarded as a left-over of the

original Asthenatherum type. Its presence in South Africa could be explained in any

one of many ways. The most obvious although unlikely explanation is that Asthena-

therum originated in South Africa, with D. pumila Nees forming a direct link between

Danthonia and Asthenatherum. It is also possible that D. pumila Nees represents such

a link and migrated south with Asthenatherum. It could also be regarded as a specialized

line of development from Asthenatherum. The latter statement is hardly likely because

it is difficult to imagine that the advanced type of chlorophyll distribution should

change back to the more primitive one, especially as there is no obvious advantage in

such a process. Another possibility is that D. pumila Nees represents a hybrid between

a species of Asthenatherum and Danthonia or some other closely related genus. To the

author the most likely explanation appears to be that D. pumila Nees represents a

specialized line of phylogenetic development which came about after Danthonia had

reached South Africa. It would appear as if the genera Danthonia and Asthenatherum

had their origin in a common ancestor. The latter genus, being better adapted to dry

conditions migrated south along the arid regions of the west coast of Africa. Danthonia

migrated along the mountain ranges of the tropics and subtropics to reach the winter

rainfall area of southern Africa.

SUMMARY

Leaf anatomy and morphology of the South African species of Danthonia were

studied in detail. These data were correlated to trace the relationships of species.

The problem of what is primitive and what is advanced has been related to leaf anatomy.

Species with a combination of the panicoid type of epidermis and the festucoid type of

internal leaf anatomy appear to represent the more primitive types. The possible

direction of migration of the South African species of Danthonia was discussed. The

chromosome numbers of D. macowanii Stapf (2n = 12), D. arundinacea (Berg.)

Schweick. (2n = 12) and D. tenella Nees (2n = 24) are reported for the first time.

Three species, D. mossamedensis Rendle, D. pumila Nees and D. suffrutescens Stapf

should be transferred to Asthenatherum. The latter species could be regarded as a

variety of A. glauca (Nees) Nevski.

LITERATURE CITED

Cain, S. A. (1944) Foundations of plant geography. Harper and Brothers,

New York. 556 pp.

Calder, W. J. (1937) A cytological study of some New Zealand species and varieties

of Danthonia. Jour. Linn. Soc. London 51: 1-9.

310

Chippindall, L. K. a. (1955) The classification of grasses. Grasses and pastures of South

Africa: 1-527.

DE Wet, J. M. J. (1953) Nucleoli numbers in Danthonia polyploids. Cytologia

18: 229-234.

DE Wet, j. M. J. (1954a) The genus Danthonia in grass phylogeny. Amer. Jour. Bot.

41: 204-211.

DE Wet, j. M. J. (1945b) Stomatal size as a cytological criterion in Danthonia. Cvto-

logia 19: 176-181.

DE Wet, j. M. J. (1956) Leaf anatomy and phylogeny in the tribe Danthonieae.

Amer. Jour. Bot. 43 : 176-182.

Hubbard, C. E. (1937) Graminea. Flora tropical Africa 10: 123-147.

Hubbard, C. E. (1940) AUoeochaete andongensis (Rendle) C. E. Hubbard. Hook.

Icon. Plant. 5: tab. 3418.

Prat, H. 1948 General features of the epidermis in Zea mays. Ann.

Missouri Bot. Gard. 35: 341-351.

Randolph, L. F. (1935) A new fixing fluid and a revised schedule for the paraffin

method in plant cytology. Stain Tech. 10: 95-96.

Staff, O. (1900) Gramineae. Flora Cap. 7: 310-750.

Stebbins, G. L. (1956) Cytogenetics and evolution of the grass family. Amer. Jour.

Bot. 43: 890-905.

AND R. M. Love, (1941) A cytological study of California forage grasses. Amer.

Jour. Bot. 28: 371-382.

Stockwell, P. (1934) A stain for difficult plant material. Science 80: 121-122.

311

Culm Anatomy in Relation to Taxonomy.

By

J. M. J. de Wet.

The problem of phylogenetic relations is not always to be solved by a study o

external characters only. This is particularly true in the family Gramineae. A survey of

characters that are useful in the classification of the grasses is presented by Stebbins

(1956). During an investigation of anatomical characters in Danthonia it was noted

that those species which could be referred to Asthenatherum Nevski differ from typical

representatives of Danthonia in culm anatomy. These two species-groups also differ

from each other in leaf anatomy (de Wet, 1954). To test the taxonomic significance of

this character, a representative group of grasses from different tribes were studied.

MATERIAL AND METHODS

Specimens studied were obtained from the National Herbarium, Pretoria. Pieces

of the culm taken just below the inflorescence were placed in a softening medium con-

sisting of equal parts 50 per cent ethyl alcohol and glycerol for 24 hours. Sections were

cut by hand and stained in a 1 per cent aquous safranine solution. Drawings were made

with the aid of a camera lucida.

RESULTS

The genera studied are listed in Table 1 . Two or more species of each genus were

investigated. The concepts of tribes are similar to those of Avdulov (1931), Rrat (1936)

and Chippindall (1955). Leaf anatomical data are mostly from Avdulov (1931) and

Prat (1936). The letters F (festucoid), E (eragrostoid-chloridoid), P (panicoid) and

B (bambusoid) refer to the types of characters discussed and illustrated by Stebbins

(1956).

Sections were cut transversely through the stem at about 1 cm below the inflo-

rescence. Below the epidermis a more or less continuous circular layer of sclerenchyma

and collenchyma is present, carrying bundles which are solitary or in pairs. Other

bundles lay irregularly scattered in the pith. The bundles included by this circular

sclerenchymatous strand are usually connected with the epidermis by a strand of collen-

chyma cells. The chlorenchyma is entirely situated between these strands and the

epidermis. In other words, these small vascular bundles are always girdered by chloren-

chyma cells. The layer of chlorenchyma cells directly bordering onto the vascular

bundle usually distinguish themselves clearly by their size from the rest of the chloren-

chyma. It is this layer of cells (bundle sheath) that appears to be of use in grass taxonomy.

In the festucoid group of grasses; tribes Festuceae, Brachypodieae, Hordeae,

Aveneae, Phalarideae and Agrosteae this parenchymatous bundle sheath is poorly

differentiated and can not be distinguished from the rest of the chlorenchymatous

tissue (Fig. 1). This is also true in genera studied belonging to the tribes Oryzeae,

Ehrharteae and Danthonieae (except for the genus Asthenatherum). This type of

anatomical character was also observed in the genera Lasiochloa, Plagiochloa and

Urochlaena which Chippindall (1955) refer to the tribe Eragrosteae.

312

in ths eragrostoid-chloridoid group of grasses (Zoiseae, Eragrosteae, Chlorideae

and Sporoboleae) the parenchymatous bundle sheath is composed of very large cells

''Figs. 3 and 4). This bundle sheath is also obvious in the Paniceae, Arundinelleae

313

and the Andropogoneae (Figs. 5 and 6) but the individual cells are often not quite as

large as is the case in the eragrostoid-chloridoid group. This is also true in represen-

tatives of the tribes Meliceae, Arundineae, Pappophoreae and the genus Stipa.

The genus Aristida is quite distinct from all the other grasses studied. A very

obvious internal bundle sheath of sclerenchyma is present, surrounded by a parenchy-

matous sheath of smaller cells, which often are also thickened and apparently lack

chlorophyll (Fig. 2).

Table 1. — Anatomical characteristics.*

* In culm anatomy: Type 1 — parenchymatous sheath not obvious; type 2 — sheath cells small;,

types — sheath cells extremely large ; type 4 — Aristida. In leaf anatomy and epidermis: F — festucoid;,

E — eragrostoid-chloridoid; P — panicoid; B-bambusoid types.

314

Genus.

Culm

anatomy.

Leaf

anatomy.

Epid.

Eragrosteae —

Diplachne

Dinebra

Eragrostis

Entoplocamia . . .

Lophacme

Leptocarydion . .

Lasiochloa

Odyssea

Plagiochloa

Pogonarthria . . .

Stiburus

Tetrachne

Trichoneura

Triraphis

Tripogon

Urochlaena

Chlorideae —

Cynodon

Chloris

Ctenium

Eustachys

Fingerhuthia . . .

Harpechloa

Microchloa

Oropetium

Rendlia

Anmdinelleae —

Arundinella. . . .

Danthoniopsis. .

Loudetia

Tristachya

Trichopteryx . . .

Paniceae —

Anthephora . . . .

Alloteropsis. . . .

Acroceras

Brachiaria

Cenchrus

Digitaria

Eriochloa

Echinochloa —

Melinis

Oplismenus

Panicum

Paspalum

Rhynchelytrum

Setaria

Tricholaena. . . .

Urochloa

A ndropogoneae —

Andropogon. . .

Bothriochloa. . .

Chrysopogon. .

Diectomis

Elyonurus

Hyparrhenia. . .

Hetevopogon. . .

Hemarthria

Ischaemum

Imperata

3

1

3

1

3

1

3

2

o

2

3

2

3

2

T

2

P

F

P

F

P

F

P

315

DISCUSSION

Comparing the anatomical data presented in Table 1, it becomes evident that there

is a close correlation between stem anatomy and leaf anatomy. The data have little

value as such, but only supply another tool in studying grass phylogeny. Other charac-

ters which were proved to be useful in the taxonomy of the Gramineae are leaf anatomy

(Duval-Jouve, 1875, Avdulov, 1931); histology of the epidermis (Prat, 1936); Cytology

(Avdulov, 1931); embryo anatomy (Reeder, 1957); root-hair development (Row and

Reeder, 1957); physiology (Harz, 1880, Cugnac, 1931, Al-Aish, 1956) and organization

of the shoot apex (Brown, Heimsch and Emery, 1957.)

The data from stem anatomy further emphasize the difference in phylogenetic

affinities between Aristida and Stipa. A further difference between Danthonia and

Asthenatherwn also becomes evident. In this respect Danthonia resembles the festucoid

group and Aslhenathenim the panicoid group of grasses. These anatomical data also

show that the genera Lasiochloa, Plagiochloa and Urochlaena occupy an aberrant

position in the tribe Eragrosteae.

SUMMARY

The culm was sectioned transversely about 1 cm. below the inflorescence. The

bundles closest to the epidermis are flanked by parenchymatous cells which contain

chloroplasts. Directly against the bundle these cells arrange themselves into a parenchy-

matous sheath which may be poorly differentiated or well developed and composed

of small or large cells. It was pointed out that these anatomical data show a close

correlation with data from leaf anatomy.

LITERATURE CITED

Al-Aish, M. (1956)

Avdulov, N. P. (1931)

Brown, W. V., C. Heimsch and W. H.

P. Emery (1957)

Chippindall, L. K. a. (1955)

Cugnac, U. de (1931)

DE Wet, J. M. J. (1954)

The effect of isopropyl-n-phenyl carbamate on the germination

of grass seeds and the bearing upon systematics of the

Gramineae. Ph.D. Thesis, University of Texas.

Karyo-systematische Untersuchungen der Familie Gramineen.

Bull. Appl. Bot. Suppl. vol. 44, 428 pp.

The organization of the grass shoot apex and systematics.

Amer. Jour. Bot. 44: 590-595.

The classification of grasses. The grasses and pastures of

South Africa: 1-527.

Recherches sur les glucides des Graminees. Ann. Sci. Nat.

Bot. 10: 1-121.

The genus Danthonia in grass phylogeny. Amer. Jour. Bot.

41: 204-211.

316

317

Notes on the Sapotaceae of

Southern Africa.

By

A. D. J. Meeuse.

INTRODUCTION.

In view of the proposed “ Flora of Southern Africa”, the Sapotaceae of the

area were studied. It soon became apparent that several South African species also

occur in tropical Africa or are closely related to tropical species, and this necessitated

the study of material from adjoining regions such as Angola, Rhodesia and Portuguese

East Africa. Several species not recorded from the Union, but found in neighbouring

areas are included in this treatment, either because they may be recorded later, or

because they were interesting in connection with related South African forms.

This revision is mainly based on material from herbaria in Southern Africa.

Several problems that could not be solved in this country were passed on to our officer

stationed at Kew, Mr. B. de Winter. My thanks are due to Mr. de Winter, not only

for his valuable assistance with taxonomical problems, such as comparing of type

specimens and other material at Kew and in the British Museum (Nat. Hist.), but

also for his kind help in obtaining abstracts of several publications which were not

available here. Without his contribution many minor difficulties and doubtful points

could never have been satisfactorily cleaied up.

The material of the National Herbarium Pretoria, was studied and, in addition,

material was kindly sent on loan by the following herbaria (abbreviations are, where

possible, those of the latest Index Herbariorum by Lanjouw and Stafleu): BOL,

COI, GRA, J, L, LM, NBG (including SAM), NH, NU, PRE (including TRV), SRGH;

and Forestry Department Herbarium, Pretoria (SAFD).

History of the Study of Sapotaceae in Southern Africa.

The first comprehensive work, De Candolle’s Prodromus Vol. 8 (1844) only men-

tions two species from Southern Africa, viz., Sideroxylon inerme L. and Mimusops

cajfra E. Mey. ex A. DC. Until the publication of Engler’s monograph of all the

African Sapotaceae in his series “ Monographien Afrikanischer Pflanzenfamilien und

Gattungen ”, Vol. 8 (1904) — cited in the following pages as “ Mon. Sapot. Afr.” — -

the study of Sapotaceae in Southern Vfrica was restricted to occasional descriptions of

new species such as by Sonder in Linnaea 23 (1850) and by N. E. Brown in Kew Bull.

1895.

The Sapotaceae of tropical Africa were treated by Baker in Flora of Tropical

Africa 3 (1877), but only very few of the species mentioned extend into South Africa.

In 1906 the family was treated in Flora Capensis, Vol. 4 (1), by C. H. Wright,

whose account was partly based on unpublished notes left by Harvey. This treatment

added very little to Engler’s monograph and is not critical, in fact, most descriptions are

practically literal translations of those given by Engler. Phillips, in Genera S. Afr.

Flow. PI. ed. 1 (1926), recognised three genera, viz. Sideroxylon (with 2 species in S.

Africa), Chrysophyllum (3 spec.) and Mimusops (11 species). In the second edition

318

of the Genera (1951), Phillips maintains the same 3 genera with 2, 3, and 12 species

respectively. As will be pointed out later, his conception of the Sapotaceous genera

cannot be maintained and more genera have to be recognised if present trends are

accepted. The late Father J. Gerstner, finally, treated the family in two papers which

are rather superficial as far as systematics and nomenclature are concerned, but most

valuable on account of important field-notes; cf. J. S. Afr. Bot. 12: 47-55 (1946),

and 14: 171-174 (1948). Gerstner’s work has clarified the status of Chrysophyllum

wilmsii Engl, and of Mahea natalemis Pierre (= Mimusops natalensis Engl, non Schinz).

He recognised 5 genera. The following Table gives a comparative analysis of the

species mentioned in the more comprehensive publications and the present author’s

interpretation (Cf. Table I).

Delimination of the genera,

The genera of Sapotaceae are not very sharply defined and the modern mono-

graphers are inclined to “ moderate splitting ”, which trend is followed here. For a

more detailed account of the history of the generic taxonomy and of the various systems

of classification, see Baehni, Candollea 7: 394-508 (1938).

The most recent systems of classification are those published by Baehni (l.c.) and

by Lam (Occ. Papers. Bern. P. Bishop Mus. Honolulu 14, no. 9: 137-141 (1938), and

Rec. Trav. Bot. Neerl. 36: 524 (1939). Although these two systems are basically entirely

different, these authors agree in several essential points and in the delimitation of

most genera. The most important feature of the most recent classifications, as far as

the S. African representatives are concerned, is the segregation of the several genera

from the large and heterogeneous genus Mimusops sensu Engler (1904), such as

Manilkara, Muriea, Lecomtedoxa, Baillonella and others.

In this revision many more genera are accepted than was done by Phillips in his

“ Genera ”, but it is considered preferable to follow modern monographers of the

family rather than one general, and consequently unspecialized, reference work.

FAMILY CHARACTERS.

Trees often attaining a large size or sometimes shrubs, occasionally climbing,

containing latex ducts in all parts (even the fruits). Leaves alternate, almost invariably

petiolate, undivided and usually with entire margin, often more or less leathery;

stipules often present but usually early deciduous, setaceous or linear-subulate. Flowers

axillary but often on older branches or stems, generally solitary or fascicled, sessile or

pedicelled, regular, almost invaiiably bisexual, usually small but often fragrant. Calyx

consisting of one or two whorls of free or nearly free sepals, usually with a rusty-brown

pubescence on the outside; sepals 2-5, occasionally more, in each whorl, usually

entire, often firm to coriaceous. Corolla gamopetalous, usually cream or white and

almost invariably glabrous, consisting of a tube and one or two whorls of 2-5 or occa-

sionally more lobes; the tube usually cylindric to campanulate, usually short; the

lobes imbricate, entire or occasionally fringed, in several genera each lobe bearing two

lateral petaloid appendages of various sizes, sometimes nearly as large as the lobe

itself, som.etimes much smaller. Alternipetalous stami nodes as many as there are petals

in one whorl, or fewer, or absent, varying from rather large and petaloid to small and

scale-like, entire, or variously fringed, lobed or dissected, glabrous or hairy. Epipe-

taloiis staminodes sometimes present, but if so they are transformed, sterile or abortive

stamens and often resemble the latter in shape. Stamens in one or two whorls

(occasionally in more whorls, not in Southern Africa), as many per whorl as the number

of corolla-lobes or twice as many or fewer, inserted in the mouth of the corolla-tube

or sometimes lower down; filaments usually present but generally short; anthers

2-thecous; thecae usually extrorse, dehiscing with longitudinal slits. Ovary usually

Table I. THE SAPOTACEAE OF SOUTHERN AFRICA (COMPARATIVE ANALYSIS)

319-320

Author,

Genus.

A.DC., Prodr. VIII (1844).

Sideroxylon. ,

Chrysophyilum.

Sideroxylon inerme L,

Poiitei ia.

Vincentclla.

Mimusops.

Austromimusops

Manilkaia.

(Under Chrysophyilum)

(Sideroxylon brevipcs Baker)

Mimusops caffra E. Mey. ex A. DC.

Mimusops kirkii Baker

Mimusops mochisia Baker

(Mimusops mochisia Baker)

Muriea.

Lecomtodoxa.

^ OUIIU.

Chrysophyilum Wilmsii Engl.

Chrysophyilum argyrophyllum Hiern

Chrysophyilum antunesii Engl.

Chrysophyilum carvalhoi Engl.

Chrysophyilum gorungosanum Engl.

I j liuiii

Chrysophyilum natalensc Sond.

Chrysopliyllum Wilmsii Engl.

(Chrysophyilum)

Pachystela brevipes (Baker) Engl.

Pachystela cinerea (Engl.) Pierre

Mimosops caffra E. Mey. ex A. DC.

Mimusops kirkii Baker

Mimusops mochisia Baker

Mimusops obovata Sond.

Minusops woodii Engl.

Mimusops oleifolia N.E.Br.

Mimusops zeyheri Sond.

Mimusops densiflora Engl.

Mimusops mcnyhartii Engl.

Mimusops fischeri Engl.

Mimusops zanzibarensis Engl.

Mimusops marginata N.E.Br.

Mimusops schinzii Engl.

Mimusops dispar N.E.Br.

Mimusops discolor (Sond.) Harlog

Mimusops natalensis (Pierre) Engl.

Mimusops henriquesii Engl, et Warb.

(Mimusops marginata)

(Mimusops dispar)

(Mimusops mochisia)

(Mimusops zanzibarensis)

(Mimusops discolor)

(Mimusops natalensis)

(Mimusops henriquesii)

(Chrysophyilum)

Mimusops caffra E. Mey ex A. DC.

Mimusops obovata Sond.

Mimusops woodii Engl.

Mimusops oleifolia nTe.Bl

Mimusops zeyheri Sond.

Mimusops marginata N.E.Br.

Mimusops schinzii Engl.

Mimusops dispar N.E.Br.

Mimusops discolor (Sond.) Hartog

Mimusops natalensis (Pierre) Engl.

Mimusops concolor Harv. ex Wright

(Mimusops marginata)

(Mimusops dispar)

(Mimusops concolor)

Chrysophyilum natalense Sond.

= Chrysophyilum magalismontanum

C hrysophyilum viridifolium VV\)od ct I ranks

(Chrysophyilum magalismontanum)

(Chrysophyilum natalense)

= Pouteria magalismontana (Sond.) A. Me. use

= Pouteria natalensis (Sond.) A. Meciisc

= Pouteria magalismontana.

Chrysophyilum gorungosiinum Engl.

Chryscphyllum viridifolium Woed et Franks

Pouteria magalismontana (Sond.) A. Meeuse

Pouteria natalensis (Sond.) A. Meeuse

Pouteria brcvijTcs (Baker) Bachni

= Pouteiia brevipes.

Vincentclla sapinii (De Wild). Brenan

Mimusops caffra E. Mey.

Mimusops kirkii Baker

Mimusops mochisia (1946) = Manilkara n'.ochisia (1948)

Mimusops obovata Sond.

= Mimusops obovata

Mimusops zeyheri Sond.

Mimusops marginata N.E.Br.

Mimusops schinzii Engl.

(Not mentioned)

Labourdonnaisia discolor

= Labourdonnaisia discolor

“ Mimusops Henriquesiana Sim ”

Mimusops concolor (1946) — Manilkara concolor (1948)

(Mimusops marginata)

Mimusops caffra E. Mey. ex A. DC.

= Mimusops zeyheri.

= Manilkara mcchisia.

Mimusops obovata Sond.

= Mimusops obovata.

Mimusops obovata.

Mimusops zeyheri Sond.

Manilkara m(K;hisia.

= Manilkara mochisia.

= Manilkara spec. (M. mochisia?).

= Manilkara zanzibarensis.

— Austromimusops marginata.

Austromimusops dispar.

Muriea discolor.

= Muriea discolor.

Lccomtcdoxa henriquesii.

— Manilkara concolor.

“ Manilkara mochisia (Baker) Gerstn.”

“ Manilkara concolor (E. Mey.) Gerstn.

(Mimusops discolor, Mimusops natalensis) Labourdonnaisia discolor Sond.

Austromimusops marginata (N.E.Br.) A. Meeuse.

Austromimusops dispar (N.E.Br.) A Meeuse.

Austromimusops sylvestris (Sp. Moore) A. Meeuse.

Manilkara mochisia (Baker) Dubard

Manilkara zanzibarensis (Engl.) Dubard

Manilkara concolor (Harv.) Gerstn.

Manilkara macaulayae (Hutch, et Cerb.) H. J. Lam

Muriea discolor (Sond.) Hartog

6096259

(Mimusops henriquesiana Sim)

Lecomtedoxa henriquesii (Engl, et Warb.) A. Meeuse

of

re

ca

ac

ar

Vc

ai

Wi

H

sp

in

gi

m

of

R

di

m

th

fr

Ai

fa

cc

Pt

St

a?

Pt

cc

pi

er

al

sii

lo

la

it;

in

sc

to

St

(c

ol

oi

2-

321

5- to many-locular with a single ovule in each locule; style cylindric or more or less

conical or subulate, ending in an acute or capitellate stigma; ovary superior, often

rusty-pubescent. Fruit a berry with a usually thin outer layer and a juicy or mealy,

rarely tough and leathery pulp in which the seeds are embedded. Seeds as many as

there are ovules or fewer, compressed or tumid; testa either hard (“ bony”) and in

this case usually smooth and shiny, or leathery to crustaceous and in this case often

dull, with an attachment area (cicatrix or “scar”) containing the hilum; cicatrix

basal or lateral, small and circular or large, sometimes covering about half the surface

area of the seed, of a different texture (duller and rougher than the testa and usually

thinner and softer); endosperm either copious and found on either side of the flat

foliaceous cotyledons, or scanty to absent and in this case the cotyledons thick and

fleshy.

An almost completely tropical family comprising over 600 described species in

a number of genera which are very difficult to define, because there is hardly any other

family of the Phanerogams where the characters integrate and overlap to such an extent.

For this reason the number of “ genera ” varies from author to author, some

recognising about a dozen, others up to 120, but following present trends about 40

genera can be recognised. Another difficulty is that genera which appear to be rather

sharply defined in one area may show intermediates in another region of the world

and only a world-wide study can reveal the relationships. Several genera are already

known to occur in more than one continent (Poiiteria and Manllkara are circumtropical,

Mimusops occurs in Africa and tropical Asia, Chrysophyllum is found in America

and Africa and possibly occurs in Asia as well) and more similar links may be found.

Key to the genera.

As the genera of Sapotaceae are not very sharply defined, and both characters of

the flower and the seed are used in their classification, the following key will not always

lead direct to the proper genus if only flowers or only fruits are available, but the alter-

natives are indicated. Two characters used in the key need some explanation. In

several genera the corolla-lobes occur in groups of three. The generally adopted con-

ception is that each corolla-lobe, which is the central one of each group of three, bears

two “ lateral (or dorsal) appendages ” which are often interpreted as stipules. Only

Gilly [in Trop. Woods 73 (1934), p. 1-22] developed a different theory, viz., that in

genera such as Mimusops and Manilkara the biseriate calyx of most authors represents

the calyx and the corolla, the inner whorl being the true corolla, whereas the corolla

of other authors is interpreted as an outer whorl of staminodes. It is not very likely

that Gilly’s views will be generally adopted, because there are genera having a mono-

seriate calyx and corolla-lobes with lateral appendages (Biimelia). The African genus

Lecomtedoxa often shows sub-biseriate calyces and corolla-lobes with small lateral

appendages and forms a link between the groups with biseriate calyces and (mostly)

lateral appendages to the corolla-lobes (included by Lam in his subfamily Mimuso-

poideae) and the groups with mostly monoseriate calyces and usually without lateral

appendages (Lam’s Sideroxyloidea-Pouteriinae).

The other character used is the “ scar ” or cicatrix of the seed (area derasa of

Baehni). The part of the testa by which the seed is or was attached to the inside of

original loculum of the ovary is usually very distinct from the rest of the testa by a paler

colour, a less smooth surface, a surrounding rim, location in a depression of the seed,

etc. The shape and size of this area form a very good distinguishing character.

Flowers generally 5-merous; sepals in one whorl or at least never manifestly biseriate; lateral

appendages to the corolla-lobes absent (only in Lecomtedoxa present):

Corolla-lobes without lateral appendages; flowers rather strictly 5-merous throughout:

Berry globose, 1 -seeded; seed depressed-globose, bluntly 4- or 5-angled or somewhat

ribbed, with a small circular cicatrix; alternipetalous staminodes always as

many as the corolla-lobes and about as long, petaloid, usually triangular to

lanceolate from a broad base ; corolla-lobes not strongly reflexed .... 1 . Sideroxylon.

322

Berry usually ovoid to oblong; seed ellipsoid, ovoid or oblong, not angular or some-

what ribbed, with a lateral, usually long and sometimes very large scar:

Fruits usually 3-5-seeded; seeds laterally compressed with a hard, shiny bony

testa and narrow linear scar; endosperm copious, cotyledons thin and

foliaceous; alternipetalous staminodes O (if fruit lacking and stami-

nodes O, see also 3. Pouteria) 2. Chrysophyllum.

Fruits usually 1 -seeded, sometimes 2-seeded; seeds usually not much compressed,

with a thin, more crustaceous testa; scar linear or very large and

occupying the ventral half of the seed; endosperm O, cotyledons thick

and fleshy; alternipetalous staminodes 1-5, or absent:

Corolla-tube short but distinct; petals not strongly reflexed; filaments

rather short and stoutish, stamens therefore not or but slightly

exerted; ovary in flower not conspicuous 3. Pouteria.

Corolla-tube very short to almost O; petals completely reflexed; filaments

long and slender, erect and hence stamens almost completely

exerted; ovary exposed, comparatively large and conspicuous. 4. Vincent ella.

Corolla-lobes with lateral appendages; whorls of flower 3-6-merous; alternipetalous

staminodes present, as many as the corolla-lobes; fruit large (± 4 cm. by

2-2-5 cm.), 1-seeded, with an oblong large lateral scar; seed without endo-

sperm 5. Lecomtedoxa.

Flowers generally 3^-merous, sepals always distinctly biseriate, corolla-lobes usually each

with 2 petaloid lateral appendages, very rarely appendages small or O:

Flowers generally 4-merous ; alternipetalous staminodes more or less lanceolate, entire

(except sometimes their tips), hairy outside ; lateral appendages of the corolla-

lobes always well-developed:

Seed ellipsoid, not or but slightly laterally compressed, with a pergamaceous or crusta-

ceous, not very shiny or quite dull testa and a large scar occupying most of

the ventral half of the seed; endosperm O, cotyledons thick and fleshy;

leaves distinctly crowded at the ends of the branches, usually not coriaceous

nor shiny, with a fine usually conspicuous reticulate nervation; flowers in

lower leaf-axils, often pendulous; ovules not basally attached 6. Austromimusops.

Seeds laterally flattened with a hard and shiny testa and a basal or sub-basal circular

scar; endosperm present; cotyledons flat; leaves not distinctly crowded at

the ends of the branches, usually more or less coriaceous, shiny on upper

surface, the nervation usually not finely reticulate; flowers in the leaf-axils or

sometimes also on naked branches; ovules basally attached 7. Mirnusops.

Flowers generally 3-merous; alternipetalous staminodes various, entire or more or less

divided, lobed, lacerated or fimbriate, glabrous, or absent and stamens twice as

many as there are corolla-lobes (i.e., usually 12); rarely some or all stamens

sterile and resembling staminodes; seed with endosperm and flat, foliaceous

cotyledons :

Stamens as many as there are corolla-lobes (i.e., usually 6), alternating with as many

or nearly as many alternipetalous staminodes ; in the South African species

corolla-lobes always with well-developed, lateral appendages and testa

hard 8. Manilkara.

Stamens twice as many as there are corolla-lobes (i.e. usually 12); alternipetalous

staminodes O, but occasionally some or all the stamens abortive and reduced

to sterile staminode-like organs (in the latter case the normally present,

lateral appendages to the corolla-lobes small or occasionally O) ; testa rather

thin, crustaceous and brittle when dry 9. Muriea

1. SIDEROXYLON

L., Gen. PI. ed. 5, 89 (1754); Roem. et Schult., Syst. 4: 45 (1819); Endl., Gen. 739

(1837); Harvey, Gen. S. Afr. Plants 224 (1838); A.DC., Prodr. 8: 177 (1844),

pro parte; Benth. et Hook, f., Gen. PI. 2, 2: 655 (1876), pro parte; .Baker in

Oliv., FI. Trop. Afr. 3: 503 (1877), pro parte; Engler in Engler-Prantl, Natiirl.

Pflanzenfam. 4, 1; 143 (1890), pro parte, and Mon. Sapot. Afr. 25 (1904), pro

parte; Harvey ex Wright in Dyer, FI. Cap. 4, 1 : 142 (1906); Baehni in Candollea

7: 49 (1938), pro parte; H. J. Lam in Rec. Trav. Bot. Neerl. 36: 521 (1939);

Phillips, Gen. S. Afr. FI. Plants, ed. 2, 567 (1951).

323

Calvaria Comm, sensu Dubard in Ann. Mus. Col. Marseille 20: 84 (1912); H. J. Lami

in Occ. Papers Bern. P. Bishop Mus. Honolulu 14, no. 9: 138-139 (1938);

Adamson in Adamson and Salter, FI. Cape Penins. 667 (1950).

Type Species: Sideroxylon inerme L. (see discussion below).

Trees or shrubs. Flowers normally 5-merous throughout, all whorls single.

Corolla-lobes without lateral appendages Alternipetalous staminodes petaloid, with

a broad base. Ovules basally attached. Berry 1 -seeded. Seed depressed-globose,

usually more or less 4- or 5-angled and somewhat ribbed; scar small, circular, situated

in a basal depression of the seed; testa thick and bony; endosperm copious; cotyle-

dons thin, foliaceous; embryo in the type species horizontal.

There has been a considerable amount of disagreement as regards the delimitation

of the genus. Basing the genera of Sapotaceae almost exclusively on the structure

of the flowers Bentham and Hooker and Engler included many forms with the flower

pattern: sepals 5, petals 5 without lateral appendages, alternipetalous staminodes

5, stamens 5, ovarium 5-loculated. in one lage genus Sideroxylon which is certainly

not homogeneous. Pierre, and later Dubard [see Ann. Mus. Col. Marseille 20 (1912)],

using other characters apart from the floral structure, (e.g. the seeds) came to the con-

clusion that only a few species had to be retained in Sideroxylon, excluding, for instance,

all forms which do not possess a basal scar. Unfortunately, Dubard placed the type

species S. inerme L. in the genus Calvaria Comm, emend. Dub., as distinct from

Sideroxylon L. sensu Dub. (in which he includes forms with a basal scar but not with

a horizontal embryo as in Calvaria sensu Dub.).

The generic description agrees in principle with the delimination of the genus as

understood by Baehni (op. cit.). Baehni mentions Sideroxylon inerme as the type

species, but his description of the genus contains an error; the berry is said to be 4-

seeded. This is obviously a misprint or an oversight, because he clearly agrees with

Dubard’s ideas although he points out that Dubard erred when he took up the name

Calvaria. Baehni is also inclined to include Dubard's genus “ Sideroxylon ” (Dub.,

op. cit., p. 81): “ Nous adoptons I’idee de Dubard qui consiste a resteindre le genre

aux seules especes a cicatrice basilaire, mais nous y reintegrons cependant les Calvaria

sensu Dubard ” (Baehni, op. cit., p. 492). This seems to be reasonable, because Dubard’s

genus Calvaria (=Sideroxylon sensu str.) and his genus Sideroxylon {= M astichoden-

dron Jacq.) differ only in the position of the embryo. Under this delimitation the

genus includes a few species in America, probably only two in Africa and a few in

Asia.

As regards the fixation of S. inerme as the type species, Linne mentioned two

species in Sp. PI. ed. 1, viz. S. inerme and S. spinosum. Both are properly defined

and have never been confused with any other species. S. spinosum was placed in the

monotypic genus Argania by Roemer and Schultes (Syst. Veg. 4: 46) as Argania

Sideroxylon R. et S. {^Argania spinosa (L.) Skeels], leaving S. inerme L. as the only

species in Sideroxylon and thus typifying it.

Sideroxylon inerme L., Sp. PI. ed. 1: 192 (1753); Burm. f.. Prodr. FI. Cap. 6 (1768);

Drege, Zw. Pflanzeng. Doc. 144, 153 (1843); A. DC. in DC., Prodr. 8: 182

(1844); Wood, Natal PI. 4, 1: pi. 314 (1903) Engl., Mon. Sap. Afr. 27, t.8 fig. B

(1904), inch var. schlechteri Engl.; Wright in Dyer, FI. Cap. 4, 1: 438 (1906);

Sim, For. FI. Cape Col. 252, pi. 295 (1907); Marloth, FI. S. Afr. 3: 36

pi. 10 (1932);Gerstner in J. S. Afr. Bot. 12: 47, fig. 1 (1946). S. cinereum Lamk.,

Encycl. 1:244,(1789), partim; Drege, op. cit., 222. S. diospyroides Baker in OWw.,,

FI. Trop. Afr. 3: 502 (1877); Engler, op. cit. 27, t. 27, fig. A.

Myrsine querimbensis Klotzsch in Peters, Reise Mossamb., Bot. 185 (1862).

324

Calvaria inermis (L.) Dubard in Ann. Mus. Col. Marseille 20: 86 (1912), incl. var.

zanzibarensis Pierre ex Dub.; Adamson in Adams, and Salter, FI. Cape Penins,

667 (1950). C. diospyroides (Baker) Dub., op cit. 87.

A shrub or small tree, up to 8 m. occasionally 10-20 m. high, but usually branched

from the base and not forming a clean bole. Innovations reddish-brown tomentose.

Leaves in vivo dark green and shiny above, paler below, often drying a peculiar dull

greyish-green colour above, pale greyish brown beneath, occasionally (and especially

the younger ones) drying dark olive-green above and reddish beneath, usually quite

glabrous (except when very young), but occasionally with irregular chocolate-colouied,

often powdery, patches of adpressed hairs; blade usually elliptic to obovate-oblong,

more rarely (ob-)ovate, ovate-lanceolate, obovate-spathulate or (ob)lanceolate, 4-9

(-12) cm. long and 2-4 (-5) cm. wide, with subreflexed edges, obtuse (sometimes

emarginate or retuse), its base acute or subacute, sometimes distinctly cuneate, always

more or less decurrent on the petiole; midrib prominent beneath, secundary nerves

7-10 on either side, thin, more or less immersed and not very conspicuous; tertiary

nerves not to hardly distinguishable from the fine reticulate nervations, the latter in

older (dried) leaves often inconspicuous, but quite distinct in younger ones; petiole

rather stout, often thickened towards the base, at first rusty-tomentose but soon quite

glabrous, at least near the top distinctly winged by decurrent leaf, 6-10 (-15) mm.

long. Flowers disagreeably scented, in few to many-flowered and sometimes very

dense fascicles, occasionally (a few) flowers solitary, in the axils of the lov/er leaves

on the branches, sometimes also on the naked branches just below the leaves, often

raised on short warts. Bracts minute, deciduous, rufo-tomentose. Pedicels rather

thick, slightly and gradually broadening towards the top, not very abruptly passing

into the calyx, 2-12 mm. long, sparingly whitish-pubescent, in fruit hardly lengthening

but becoming much stouter, 1-1^ mm. thick and often glabrescent. Calyx 2-2^ mm.

long, the lobes about as long as the calyx-tube, broadly ovate, subacute, erect, entire,

ciliate, with a paler edge when dry, finely and sparingly whitish-pubescent, in fruit

adpressed, scarcely accrescent. Corolla greenish-white, about twice as long as the

calyx-lobes, rotate, up to about 5 mm. across; tube short, lobes usually longer than

the tube, ovate, entire, obtuse, glabrous. Stamens inserted in the throat of the corolla

tube, longer than the corolla lobes. Alternipetalous staminodes ovate-lanceolate to

oblong or oblong-lanceolate, as long as or slightly shorter than the corolla lobes, but

always less wide than the latter, usually more or less distinctly incised, serrate, lacerate,

dentate, crenulate or with a wavy edge, acute, acuminate or obtuse, sometimes tri- or

multi-dentate at the apex. Ovary conical-semi-ovoid, covered with rather long adpressed

white hairs (except in a zone near the base), usually 5-celled but occasionally 3, 4- or

6-celled; style about the same length as the ovary, glabrous. Berry black, globose,

smooth, up to 12 mm. in diam. when fresh, when dry up to 10 mm. in diam. and wrinkled,

usually crowned with the short persistent style; pulp purple or purplish-green, with

white viscid juice; the latex long remaining sticky in dried specimens. Seed black

when fresh but drying a shiny yellowish brown, usually depressed semi-globose or

ellipsoid, rarely higher than broad, usually more or less disctinctly 4- or 5-angled and

5-ribbed, sometimes indistinctly lobed, 6-8 mm. long, 5-7 mm. wide and 5-8+ mm.

high rarely up to 9+x8x7^mm., with several (4 or sometimes more) more or less

distinct grooves which are most conspicuous near the scar, and with 2-4 small impres-

sions between the grooves close to the scar.

Cape Province. — Cape Peninsula: Marloth 584; Chapmans’ Bay, Wolley Dod

3444 (BOL); Gordon’s Bay, Gerstner 6144; Witsand, Smuts 1190. Caledon: Onrust

Riv., Schlechter 10396. Bredasdorp: Bredasdorp, Smith 2587. Riversdale: Rivers-

dale, Muir 148; Marloth 3536. Mossel Bay: leg. Town Clerk, Mossel Bay (specimen

of the “ Post Office Tree ”, declared a national monument, PRE herb. no. 28382).

Knysna: Kapp 100; Fourcade 623 (BOL). Humansdorp: Phillips 3328; Klipdrift,

325

Thode 2490 (PRE). Uitenhage: Alexander; Ecklon or Ecklon & Zeyher (L); Ecklon

& Zeyher (GRA), Zeyher 17 (BOL); Zwartkopsrivier, Drege (L). Port Elizabeth:

Redhouse, Mogg 4672 (PRE). Alexandria; Kariega, White 101 (GRA). Bathurst:

Karouga Mouth, Britten 2350; Kowie and Pt. Alfred, Tyson s.n.; Britten 1885; Barker

2106. Albany: near Riebeeck East, 3321. Alice: Acocks%9%l. Fort Beaufort:

Story 1698. Stutterheim: Fort Cunynghame, Galpin 2468. King Williams Town;

near King Williams Town, Comins 1031. East London; Smith 3816, Galpin 9843.

Queenstown: near Queenstown, Galpin 8137. Komgha: Komgha, Flanagan 777;

Kei Mouth, Flanagan 770. Kentani; Pegler 882. Port St. Johns; Galpin 11465.

Cape without precise locality: specimen in Linnaean herbarium (photo in PRE),

lectotype!

Natal. — Port Shepstone: Paddock, Oribi Gorge, McClean 267. Umzinto;

Umkomaas, Pennington s.n. (NH no. 27756). Pinetown: Isipingo, Ward 570; Aman-

zimtoti: Kotze 452 = FD Herb. 6875. Durban: near Durban, Wood 8707, 9578.

Inanda: Tongaat Beach, Hillary 375 (NU). Lower Tugela: Darnall, Schmidt 41

(NH). Estcourt: Mooi River, Wood 6306. Weenen: between Weenen and Estcourt,

Edwards 102. Msinga: near Tugeja Ferry, 935. Mtunzini; Lovwz 623 (NH).

Eshowe: Eshowe, Gerstner 1945 (NH). Hlabisa: False Bay, Gerstner 4818; Hlu-

hluwe Game Reserve, Ward 1895. Ngotshe; near Magut, Acocks 13023; Codd 1961 ;

West 2117.

Swaziland: on road to Komatipoort, Pole Evans 3463, 3467; near Stegi, Compton

26017; Rodin 4548, Usutu River, Miller S249.

Transvaal. — Nelspruit: Kruger National Park, near Pretorius Kop, Codd 6029',

near Skukuza, Codd 5735. Barberton: Komatipoort, Rogers 20800; Codd IIK).

Portuguese E. Africa. — Sul do Save, Maputo, Hornby 2668; Myre & Balsinhas

607; between Umeluzi and Porto Henrique, Myre & Carvalho 92; Louren90 Marques,

Schlechter 11710 (L, GRA, PRE, BOL; type no. of S. inerme L. var. Schlechteri Engl.);

Borle 418; Bremekamp LM60a; Rodin 4173; Porto do Oura Beach, Gomes e Sousa

3926; Inhaka Island, Mogg sm., Mrs. Moss s.n.; Goba Mts, Torre 6490; Nuanetsi-

Limpopo Valley nr. Transvaal Border, Smuts P. 322, Guifa, Pedro ScPedrogao 2135;

between Muianga and Macia: Pedrogdo 1443; near Chibuto, Pedro & Pedrogao 1534;

between Su Larrime and Ganda, Pedro & Pedrogao 1 890. Sabi River, near Meringua,

Chase 2525. Niassa: near Cabo Delgado, 2167 (LM).

Tanganyika Terr. — Morogoro Distr. ; Kwaba, Wigg 974. Kisarawe distr. :

Kisiju, Semsei 1377, Paulo 153.

Kenya Col. — Greater Kiboko River: Jarrett 518.

Mr. B. de Winter kindly examined the Linnaean Herbarium in London and informed

me that there is only one specimen which is undoubtedly Sideroxylon inerme and must

be considered to be the lectotype (Mr. de Winter matched it with a specimen Hutton

s.n. from Fish River Heights, Albany, C.P.). S. inerme was one of the first Sapotaceae

known in Europe and one of the oldest plates is found in Burman, Dec. Rar. Afr. (1738),

p. 238, t. 94, fig. 2. The seed is unmistakable and Burman also mentions: “ . . in

Cod. Wits. . . . latescens . . . vocatur ”.

This plant was described from South Africa by Linne and, in its typical form with

comparatively narrow leaves and long pedicels, is found from the Cape Peninsula

eastwards along the coast to Natal and extends into Portuguese East Africa. Baker

I

326

described a species 5'. diospyroides, from Zanguebar in tropical East Africa, which has

smaller flowers, short pedicels and obovate-cuneate leaves. Engler remarked on the

similarity between the two, but mentioned the following differences : —

(1) pedicels of S. diopyroides shorter than in S. inerme,

(2) staminodes broader and acuminate in S. diospyroides, narrower and not acu-

minate in S. inerme.

(3) seed of diospyroides smaller than that of S. inerme.

Acuminate and broad staminodes and small seeds are, however, also found in specimens

of S. inerme from the Cape and Mr. de Winter, who studied the material at Kew, in-

formed me that the only difference he could find was in the length of the pedicels.

The material from Portuguese East Africa is very often intermediate and includes

forms with narrow leaves, short pedicels and small fruits and forms with broad leaves

(as in typical S. diospyroides) with large seeds, etc. These intermediate specimens link

up the two forms so that in my opinion, S. diospyroides only represents a minor geo-

graphical variant of S. inerme and is not even worthy of varietal rank.

Doubtful localities are: “Near Pretoria’’, McLea in herb. Bolus no. 5698 (BOL,

PRE), because this species has never since been found near Pretoria, and “ Johannesburg:

Melville ’’: Moss 15906 (J). Dr. J. B. Gilliland, formerly of the Dept, of Botany, Wit-

watersrand University, has kindly informed me that there is no trace of the species

anywhere in the Witwatersrand area.

Excluded species'.

Sideroxylon argenteum Thunb., Prodr. FI. Cap. 36 {\194) = Rhus thunbergii

Hook. — Heeria argentea (Thbg.) Meissn.

Sideroxylon randii S. Moore = Pouteria magalismontana (Sond.) A Meeuse (see

p. 335).

Sideroxylon dentatum Burm. f., Prodr. FI. Cap. 6 (1768) = Curtisia dentata (Burm.

f.) C. A. Smith in J. S. Afr. Forestry Assoc. 20: 34, 50 (1951). This species was legiti-

mately published, as Burman based it on the plate and description of his father’s Sider-

oxylon foliis acuminatis dent at is, fructu monospermo flavo [Burm., Dec. Rar. Afr.

Plant. (1738), p. 235, t. 82]. The plant in question is undoubtedly the same as published

by Aiton [Hort. Kew ed. 1 (1789), p. 162] under the name Curtisia faginea. Not only

are the elder Burman’s plate and description quite adequate to recognise the species,

but he also mentioned the name “ Assagay-Boom ’’ used by the Dutch at the Cape,

under which name it is still known. Moreover, Aiton, l.c. quotes “ Burm. Afr. p.

235 t. 82 ’’ and the phrase name “ Sideroxylon foliis acuminatis etc.’’, and Harvey in

Harv. and Sond., FI. Cap. 2: 570, sub Curtisea faginea Ait., also mentioned “ Burm.

Dec. Afr. p. 235, t. 82 ’’. This identity had already been recognised by the late C. A.

Smith, who did not give any reasons, however, when he proposed the above-mentioned

change of name.

2. CHRYSOPHYLLUM

L., Gen. PI. ed. 5, 89 (1754); A. DC. in DC., Prodr. 8: 56 (1844); Benth. et Hook, f.,

Gen. PI. 2: 653 (1876); Baker in Oliv., FI. Trop. Afr. 3: 498 (1877), ex parte:

Engler in Engler & Prantl., Nat. Pflanzenfam. ed. 1, 4, 1: 147 (1890), pro majore

parte, in Nachtrage 278 (1897), and Mon. Sap. afr. 38 (1904) ex parte et exclus.

Section Zeyherella; Wright in Dyer, FI. Cap. 4: 436 (1906F pi'O parte; Pilger

in Engler & Prantl, Pflanzenf., Nachtrage 1897-1907: 288 (1908); Hutch. &

Dalz., FI. W. Trop. Afr. 2, 1: 8 (1931); Eyma in Rec. Trav. Bot. Neerl. 33: 201

(1936), (with discussion on p. 157-158); Baehni in Candollea 7: 429 (1938);

Phillips, Gen. S. Afr. Flow. PI. ed. 2, 568 (1951) pro parte.

327

Donella Pierre ex Bail!., Hist. d. PI. 11: 294 (1892).

Gambeya Pierre, Not. bot. Sapot. 61 (1891); Baillon, Hist. d. PI. 11; 296 (1892).

Type species: C. cainito L., Sp. PI. ed. 1, 192 (1753).

Trees or shrubs, rarely climbing; ultimate branches and lower surface of leaves

often densely adpressed-tomentose; lateral nerves of leaves parallel, close or distant,

usually spreading and curved near the margin; tertiary nerves usually inconspicuous.

Stipules wanting or at least very early deciduous. Flowers axillary, or on the naked

branches below the leaves, solitary or in fascicles, rarely sessile, usually isomerous.

Sepals 5 (rarely 4, 6 or 7); lobes entire, imbricate. Corolla 5 (4-, 6- or 7-) lobed;

lobes entire, imbricate; tube cylindric, urceolate or campanulate, usually short. Alter-

nipetalous staminodes O. Stamens usually short, not or but little exerted; filaments

inserted in the throat of the corolla tube at the base of the lobes and short, or inserted

lower down and longer; anthers versatile, more or less dorsifix, ovate to triangular,

often apiculate, sometimes somewhat sagittate at the base. Ovary hairy, 5- or more

rarely 4-, or 6-7-loculated; style columnar, short, thick, glabrous; ovules with lateral

or basilateral attachment. Berry 1- to few-seeded, but usually 3-5-seeded. Seeds with

long and narrow ventral scar; testa hard, smooth and shiny; endosperm copious,

cotyledons thin and foliaceous.

Between 100 and 200 species described, but undoubtedly some of them have been

or have to be referred to other genera, because the absence of presence of alternipetalous

staminodes is not a reliable character in Sapotaceae if not used in conjunction with

other characters and many species without or apparently without staminodes but with

seeds altogether different from the type species of Chrysophyllwn were at one time

or another placed in the genus Chrysophyllum. Krause, Engler and later Eyma (cf.

Eyma, op cit., p. 157-158) and Baehni (op. cit., p. 405-406) criticised the systems of

classification of Sapotaceae in which the character of the alternipetalous staminodes is

over-emphasized, so that, e.g., a separate tribe “ Chrysophyllinees ” based on the

absence of staminodes was recognised by Dubard. Eyma (l.c.) and Lam [in Rec.

Trav. Bot. Neerl. 36 (1939), p. 509-525] pointed out the relationships with the Sider-

oxyleae and indeed Chrysophyllwn is, in my opinion, to be included in this group.

The genus Chrysophyllwn - sensu lato - was split up (at least in Ms.) into a large number

of smaller genera by Pierre (Pachystela, Donella, Gambeya. Zeyherella, Malacantha,

etc.). Some of these genera were adopted by Engler (1904, l.c.) such as Pachystela

and Malacantha. The remainder, at least as far as the African representatives are

concerned, and if Engler’s section Zeyherella is excluded, is a fairly homogenous

group which is distinct from Sideroxylon s.l. and Pouteria s.l. in that the staminodes

are completely lacking and the seed characters are different.

General Distribution'. Mainly tropical America, less than 30 species in Africa,

one in Madagascar; the few species recorded from tropical Asia, Australia and the

Pacific have mostly been referred to other genera (e.g. Nesoluma).

The African representatives belong to two sharply defined subgenera:

(1) Chrysophylum L. subgenus Donella (Pierre ex Baill.) A. Meeuse, stat. nov. Donella

Pierre ex Baill., Hist. d. PI. 11; 294 (1891), pro gen. Chrysophyllum L. sect. Donella

(Pierre ex Baill.) Engl., Mon. Sapot. Afr. 41 (1904).

Leaves with numerous close, parallel lateral veins, usually dark green above and

as a rule quite glabrous; corolla with a short broad unceolate to subglobose tube.

Type species: Chrysophyllum roxhurghii Don = C. lanceolatum (Bl.) DC. (India

to New Guinea).

(2) Chrysophyllum L. subgenus Gambeya {Pierre) A. Meeuse, stat. nov. Gambeya

Pierre, Not. Bot. Sapot. 61 (1891) (pro gen.). Chrysophyllum L sect. Afrochryso-

phyllum Engl, in Engl & Prantl., Nat. Pflanzenfam., Nachtrage 272 (1897).

Chrysophyllum sect. Gambeya (Pierre) Engl., Mon. Sapot. Afr. (1904).

328

Leaves with rather distant and usually on lower surface more or less prominent

lateral veins and as a rule more or less rufo-tomentose on lower surface; corolla with

a subcylindric to campanulate tube.

Type species: Chrysophyllum subnudum Baker (West Tropical Africa).

A few species of Chrysophyllum occur in Southern Africa, of which only two are

treated here (one South African, the other occuring in Southern Rhodesia and Portuguese

East Africa fairly close to the Union border):

Leaves almost completely glabrous and green when old, with fine parallel nervation, up to

11 cm., but usually under 8 cm. long ]. C. viridifoUiim.

Leaves rusty-tomentose or sometimes greyish tomentose beneath, often 10-20 cm. long;

lateral nerves distant, very prominent beneath 2. C. gorungosanum.

1. C, viridifolium Wood et Franks in Wood, Natal PI. 6: 569 (1912); Gerstner in J.

S. Afr. Bot. 12: 48, Fig. 3 (1946) Type: Franks in herb. Wood No. 11636 from

Stella Bush, Berea, near Durban, in NH, holo! photo in PRE!, in BOL and PRE,

isos!).

A large tree, 10-30 m. high, with a girth at 2 m. from the ground of 150 cm. and

over. Trunk usually unbranched for 5 m. or more, and strongly many-ribbed to the

origin of the branches. Bark grey. Innovations, petioles, pedicels, and calyx-lobes

finely rusty velvety-tomentose, the older twigs and fruiting pedicel glabrous. Leaves

scattered on the branches, thinly coriaceous, exstipulate, 4-8 (-11) cm. long and 2-3-5

(-4-5) cm. wide (those of coppice shoots are the longest), oblong or (ob-) ovate-oblong,

more rarely ovate or elliptic, obtuse or bluntly acuminate with oblong, obtuse about

8 mm. long and about 3 mm. wide acumen, more or less rounded but always decurrent

at the base, dark glossy green above, lighter and dull beneath, glabrous when mature,

jxcept near the midrib and at the very base near the petiole beneath, with reflexed

edge ; midrib channelled above, prominent beneath and, at least when dry, discolourous,

reddish or brownish; lateral nerves numerous (about 11 per cm.), patent but not quite

horizontal, often forked, almost straight, parallel (hence the leaf appearing striate),

joining the fine intramarginal vein close to the margin. Petioles 5-10 (-12) mm. long

more or less dorso-laterally flattened, channelled above. Flowers 2-2-5 mm. long,

nearly globose, in clusters in the axils of the lower leaves and on raised warts on the

older twigs; clusters few- to many-flowered (sometimes with over 20 flowers); bracts

very minute or wanting; pedicels thin, almost capillary, 4-5 mm. long. Calyx-lobes

free nearly to the base, erect, concave, ciliate. Corolla scarcely longer than the calyx;

tube urceolate; lobes erect, ovate-oblong-rotundate, very obtuse or rounded, ciliate

(at least at the lateral margins). Stamens inserted half way down the corolla-tube o'"

even lower; filaments terete, longer than the apiculate anthers. Ovary depressed-

globose, 5-celled, densely rusty -villous, often somewhat lobed; style conical-cylindric

from a broad base, thick, obtuse or truncate, longer than the ovary. Berry depressed-

globose with a depression near the top (shaped like a small apple), often ribbed (at

least when dried), 20-30 mm. long and 20-35 mm. in diam., smooth, glabrous, yellow

when ripe with yellowish-white pulp saturated with white latex, 3-5-seeded; pedicel

under fruit much incrassate, 5-8 mm. long and 1 -5-3 mm. thick, rugose; calyx usually

not persistent in fruit. Seeds semi-circular-elliptic, compressed, 15-18 mm. long,

9-12 mm. wide and 5-6 mm. thick in centre, keeled and curved at the back, almost

straight at the ventral side with a long narrow, linear scar occupying nearly the whole

length of the seed; testa hard, shiny as if polished, bright yellowish-brown when dry.

Natal. — Durban: Berea, Franks in herb. Wood No. 11636 (NH, holo.! PRE,

BOL, isos!): Durban; Bayer 14485. Eshowe: Gerstner 2071 (NH), 2546 (NH, PRE,

BOLk Lawn 202 (NH); Kotze 34 = F.D. herb. No. 3178 (SAFD). Ingwavuma:

Bayer s.n. (NH. No. 31432). Ngoya Forest: Mehliss FD No. 2686 (SAFD).

329

Swaziland: N.N. in Forestry Herb. No. 5328 (PRE); Hlatikulu, Boocock

31 = FD herb. No. 2686 (SAFD).

Portuguese E. Africa. — Sul do Save: Gomes e Sousa 1648 (COI, PRE).

C. viridifoHum belongs to the subgenus Donella and is closely related to a

number of African species of this section. From C. prunifonne Engl. (Mon. Sapot.

Afr. p. 42, Fig. A) it differs in the shape of the fruit and the shorter seeds. C. welwit-

schii Engl. (op. cit., 41, t. 13, Fig. A) is a climbing shrub (Angola, West Tropical

Africa) and differs also in the shape of the fruit and the nervation of the leaves (in COI

represented by: Welwifsch 4830, 4831, Gossweiler s.n. from Lunda, Saurimo and

Gossweiler 1644, 4439, 4852, 5011, 6908, 8048).

C. bangweolense R. E. Fries in Schwed. Rhodesia — Kongo exp. 1: 254 (1914)

(not seen) is, judging by the description, very closely related to C. viridifoHum if not

conspecific and occurs in Rhodesia. C. pentagonocarpum Engl, et Krause in Engl.

Bot. Jahrb. 49: 387, Fig. 2 (1913) differs in the size of the fruits and seeds (E. Africa).

2. Chrysophyllum gorungosanum Engl., Mon. Sapot. Afr. 8: 44; (1904) Brenan in Mem.

New York Bot. Card. 8 (5): 498 (1954). Type: R. de Carvalho s.n. from

Gorungosa in COI, lecto!, B, destroyed. C. fulvum S. Moore in J. Linn. Soc.

Bot. 40: 13 (1911-1912); Type: Swynnerton 19 from Chirinda forest, Southern

Rhodesia, in BM, not seen, duplicate in SRGH!

A large tree, up to at least 50 m. high, with characteristic fluted bole. Buds and

young branches rufo-tomentose, young leaves silvery-strigose above and silvery-

tomentose beneath. Branchlets terete, glabrescent, densely leafy. Leaves on flowering

branches 6-15 cm. more rarely up to 20 cm. long and 2 -5-3 -5, rarely up to 6 cm. wide,

on a 7-16 mm. long petiole; those on sterile branches larger, up to 30 cm. long and 9

cm. wide on a 12-28 mm. long petiole; blade lanceolate-oblong to oblong-oblanceolate

or (ob) lanceolate, cuspidate-acuminate (the acumen itself obtuse or subacute), narrowed

and acute or sometimes cuneate at base, coriaceous, very soon glabrous and green

above, densely rusty-tomentose beneath (or more greyish-tomentose when old); midrib

impressed above, very prominent below; secondary nerves 12-17 on either side, im-

pressed above, very prominent below, 5-7 mm., more rarely up to 14 mm., apart,

parallel, patent, ascending and becoming inconspicuous well within the margin;

tertiary nerves hidden by the tomentum beneath but usually distinct above, -h perpen-

dicular to the main nerve and usually ± parallel, connecting the secondary nerves

at an angle between 60° and 90°; ultimate nervation fine, areolate. Petioles terete,

narrowly sulcate above, finely ferrugineo-pubescent or tomentose, 7-20 (-28) mm.

long. Stipules O. Flowers in the leaf axils or on the naked branches below the leaves

in, sometimes very dense, clusters, or a few of them solitary. Bracts O. Pedicels 1-3

mm. long, rather thick, shiny rusty-tomentose; some flowers almost sessile. Sepals

almost completely free, broadly ovate, obtuse or very obtuse sometimes one of the outer

ones subacute, very concave, more or less unequal,it;3 • 5 mm. long; the inner ones

pale yellowish-brown strigose outside and thinner in texture than the coriaceous rusty-

tomentose outer ones; all sepals densely pale fulvo-strigose inside, the inner ones ciliate

on one side or all round. Corolla white, ±4 mm. long, the tube cylindric-inflated,

slightly longer or about as long as the lobes and ± 3 mm. in diam., the lobes more or

less erect and concave, obovate, ovate or subrotundate, obtuse or very obtuse, some-

times with nearly straight truncate apical edge, about 1-5 x 1-5 mm., densely ciliate-

barbellate. Stamens inserted in the lower half of the corolla tube; filaments about

1-5 mm. long; anthers subsagittate, shortly apiculate, about 1 mm. long. Ovary

globose, densely hirsute, about 2 mm. in diam., slightly longer than the glabrous, cylindric,

truncate style; ovules with baso-lateral attachment. Berry globose, sometimes some-

what apiculate, finely pubescent, up to about 3-5 cm. in diam., 4-5 seeded. Seeds

330

19-21 mm. long, 11-12 mm. wide and 5-6 mm. thick in the middle, brown, ovate-

oblong, slightly attenuate and subacute at the base; the scar 12-14 mm. long and 1-2

mm. wide at the widest place.

Portuguese E. Africa. — Gorungosa: Rodrigues de Carvalho s.n. (COI, lecto)!.

Southern Rhodesia. — Chipinga: Silinda, Chirida forest, Swynnerton 19 (SRGH:

duplicate of type of C.fulvwn S. Moore); Obermeyer 2161 (PRE, BOL); Hack 151/50,

Whellan 163, McGregor 17/48, Wild 2096, 2245, Chase 620 (all in SRGH); Vdri 1826,

1865 (PRE), Jack s.n. (SRGH, herb. No. 6355); Chase 427 and s.n. (SRGH, herb..

No. 19262 and 19263); Fisher 1223 (PRE); photo of fluted bole of specimen by N. C.

Chase in SRGH. “Eastern Border": Chorley s.n. (SRGH No. 6686). Melsetter:

Eyles 5721 (SRGH), Jack s.n. (SRGH No. 5962). Umtali: Eyles 5533 (SRGH).

Nyasaland. — Kota-kota Distr. : Nchisi Mountain, Brass 17067 (SRGH).

Kenya.— S. of Mt. Kenya; Hocklijfe 1370 (PRE).

The type gathering of Chrysophyllum gorimgosamim Engl, is sterile. As the material

in Berlin was destroyed and Engler (l.c.) cited both “ Herb. Coim.bra ” and “ Herb

Berlin ”, the material in COI is taken as the lectotype. The type material consists of

coppice shoots which are a perfect match of coppice shoots of specim.ens from the type

locality of Chrysophyllum fulvum S. Moore from Rhodesia (such as Obermeyer 2161).

In addition, the only true Chrysophyllum of the section Gambeya found in the area

near the type locality is C. fulvum S. Moore (Gorungosa is the region bordering Chipinga

and Melsetter in S. Rhodesia), so that there is very little doubt that these two names

are synonyms.

Excluded species:

Chrysophyllum magalismontanum Sond. = Pouteria magalismontana (Sond.)

A. Meeuse, see p. 335.

Chrysophyllum natalense Sond. = Pouteria natalensis (Sond.) A. Meeuse, see p. 339.

Chrysophyllum wilmsii Engl. = Pouteria magalismontana.

EXPLANATION OF FIGURES.

Fig. 1. — Sideroxylon inerme, seed and fruit (the seed seen from the side, the top and the

base): (a) From a specimen Kotze 452 (from Amanzimtoti, Natal), {b) From a

specimen Galpin 2468 (from Stutterheim, E. Cape). (All figures if not otherwise

stated, natural size).

Fig. 2. — Chrysophyllum viridifolium, seed and fruit (from Gerstner 2546 Eshowe,

Zululand). In this figure, and in all subsequent figures of seeds, the seed is shown

from the lateral side and from the ventral side.

Pig. 3. — Chrysophyllum gorungosanum, seed and fruit (from Obermeyer 2161, Chipinga,

S. Rh., in PRE).

Pig. 4. — Pouteria magalismontana, seed and fruit (from Gerstner 5728, Louis Trichardt,

N. Transvaal). (In the lateral view a part of the testa has been removed).

Pig. 6. — Pouteria natalensis, seed and fruit (from Lawn 57, Eshowe, Zululand).

331

■i

332

3. POUTERIA

Aubl., Hist. PI. Guiane Fr. 1: 85 (1775); Eyma in Rec. Trav. Bot. Need. 33: 159

(1936); Baehni in Candollea 9: 149, pro parte; Herrmann-Erlee & Royen in

BlumeaS: 453 (1957).

Lucwna Molina, Saggia Chil. 186 (1782).

Sersalisia R. Br. Prodr. 529 (1810), p.p., as to type species.

Zeyherella Pierre ex Baill., Hist. d. PI. 11 : nota 3 (1892), nomen nudum.

Pachystela Pierre ex Engl., Mon. Sapot. Afr. 35 (1904).

Chrysophyllum L. sect. Zeyherella (Pierre ex) Engl. op. cit., 47.

Breviea Aubr. et Pellegr. in Bull. Soc. Bot. France 81 : 792 (1934).

Aningeria Aubr. et Pellegr., op. cit., 795.

Chrysophyllum Auct. et Sideroxylon Auct., ex parte.

Type species: P. guyanensis Aubl. (South America).

Trees or shrubs. Branches terete, the young ones often tomentose. Leaves varying

from papyraceous to coriaceous, exstipulate or occasionally stipulate, often crowded

towards the tips of the branches; often hairy on both sides or the lower side, usually

more or less glabrescent but rarely ultimately or initially quite glabrous; secondary

nerves as a rule distinctly stronger than tertiary ones; tertiary nervation mutually

parallel or reticulate, parallel with or more or less perpendicular to secondary nerves;

petioles usually distinct; stipules, if present, subulate or setaceous. Flowers fasciculate

in axils of leaves or leaf-scars, sometimes inserted on brachyblasts (raised warts, etc.),

rarely solitary, pedicelled or sometimes sessile; 2-4 bracteoles sometimes present.

Sepals generally 5, occasionally 4 or 6 in a single whorl, connate at the base only, sube-

qual or more or less unequal (innermost narrower), deciduous or persistent in fruit.

Corolla exserted, sometimes only a little so, tubular to campanulate; lobes 5 (occasio-

nally 4 or 6-8), more or less erect to somewhat spreading but never reflexed; the tube

short but distinct to occassionally rather long. Alternipetalous staminodes generally 5,

but sometimes fewer, or absent, occassionally 6-8, generally lanceolate or subulate,

but sometimes squamiform or larger and petaloid. Stamens 5 (occasionally 4 or 6-8),

inserted in or near the throat of the corolla. Ovary generally 5-loculated, sometimes

with a subcupular disc, more or less conical, gradually contracted into the, usually

short, cylindric or subulate style. Fruits 1-5-seeded but often only 1 or 2 seeds develop

in many species; pericarp various but rarely hard. Seeds with a thin or sometimes

rather thick, often brittle, crustaceous testa and a large to very large cicatrix covering

the ventral half of the seed or more, sometimes cicatrix smaller, linear or oblong;

cotyledons thick and fleshy; endosperm absent or present as a thin membranous layer.

The delimitation of this genus, as given here, agrees with the circumscription of

Herrmann-Erlee and Van Royen (l.c.), except in one character: I include some species

with more or less persistent stipules, whereas the Leiden authors state: “ Leaves . . .

exstipulate ”. As the leaves of Sapotaceae are in principle all stipulate, but apparently

often exstipulate because the stipules are so early deciduous, I cannot accept the

presence or absence of stipules in this family as a very important generic character.

This delimitation agrees also very well with Eyma’s conception of the genus based

on a study of South American species. As circumscribed here there are about 150 species,

in tropical and subtropical America, Africa, Asia, Australasia and the South-West

Pacific region.

The genus, as defined here, is considerably smaller than Baehni’s enormous genus

Pouteria of 300-500 species, which includes forms with and without endosperm and

with an enormous variation in the characters of the corolla, the stamens, etc. Mr.

.1. P. H. Brenan of Kew criticized Baehni’s very broad generic limits of Pouteria in

333

Mem. New York Bot. Gardens 8 (5): 499 (1954) as follows: “ In rejecting Baehni’s

wholesale amalgamation of African sapotaceous genera under Pouteria Aubl., I recognise

that the delimitation of genera in this family is often fiendishly difficult and very much

a matter of opinion. But at the same time I remain unconvinced that the proposed

fusion is going to clear the air and make identification easier This statement expresses

aptly the opinion of other modern authors. Van Royen has especially critisized the

“ amalgamation ” of species without endosperm (which he retains in Pouteria if not

referred to different genera for other reasons) and those with copious endosperm

(referred to Xantholis Rafin. and Planchonella Pierre), see Blumea 8: 238-239. Dr.

Van Royen kindly pointed out to me that at least some of the African species included

by Baehni in his large genus Pouteria are indeed Pouteria sensu Van Royen, but

apparently Planchonella does not occur in Africa. My studies, limited as they are,

confirm this conclusion. (^)

As far as the African sapotaceous genera are concerned, the species to be included

in Pouteria sensu mihi are sharply distinghuished from the genus Chrysopbyllum sensu

mihi in that they have usually alternipetalous staminodes and seeds with a thin testa,

a broad ventral scar and no endosperm, whereas the other genus lacks alternipetalous

staminodes and has seeds with a hard thick testa, a narrow ventral scar and copious

endosperm. This implies that several species at one time under Chrysophyllum have to

be transferred to Pouteria.

On the other hand, there are some African genera included in Pouteria by Baehni

which, although they have the same type of fruit and seed, (i.e. thin testa, large scar,

no endosperm) to my mind, do not belong here because they differ in other respects.

Vincentella Pierre with its totally refiected corolla-lobes, very short corolla-tube, long

and capillary stamens, is clearly distinct and is retained. Mr. Brenan independently

came to the same conclusion. The genus Synsepalum A. DC. of tropical West Africa

may also have to be retained on account of the strongly gamosepalous calyx.

It is difficult to say how many species of Pouteria sensu mihi there are in Africa,

because I have studied only a few representatives and there are probably more. A

count of Baehni’s species (African) under Pouteria, omitting those belonging to Vin-

centella and Synsepalum, and including some which Baehni refers to Chrysophyllum

but are better placed in Pouteria, shows a total of about 20 African species probably

to be retained in Pouteria.

Leaves with secondary veins 1-3 cm. apart and very prominent on lower surface. ... \. P. brevipes.

Leaves with secondary veins much closer (several per cm.) and not very prominent;

Leaves usually rounded or emarginate at the apex, usually rusty-tomentose rarely more

silvery, on lower surface; flowers fasciculate or solitary, often on raised warts

on the naked branches below the leaves; pedicels and calyx rusty-pubescent

2. P. magalismontana.

Leaves usually bluntly acuminate at the apex, usually silvery-white on lower surface;

flowers solitary or 2-3 together sessile in the leaf-axils, calyx with a dark tobacco-

brown pubescence 7>. P. iiataleiisis.

1. P. brevipes (Baker) Baehni in Candollea 9: 290 (1942); (for full synonymy see Baehni).

Skieroxylon brevipes Baker in Oliv., FI. Trop. Afr. 3: 502 (1877), type: Kirk s.n.

in K, from Zanguebar.

(0 Note added in proof: Aubreville and Pellegrin in Bull. Soc. Bot. France 105; 37 (1958) raised

Englet’s section Zeyherelta to generic rank (including only Chrysophyllum nmgalismontammi Sond.)

and described a genus Boivinella with 5 species including Chrysophyllum argyrophyllum Hiern, C. wilmsii

Engl, and C. natalense Sond. The first two I consider to be taxonomical synonyms of C. magalismon-

tanum (= Pouteria magalismontana, see p. 335), which these authors place in a different genus

(Zeyherella)\ This is an example of the other extreme, viz., excessive splitting of genera, resulting in

the creation of a number of (to my mind, unnecessary) synonyms and adding to the confusion instead

of clearing up the generic delimitations in the African Sapotaceae.

334

Pachystela brevipes (Baker) Baill. in Bull. Soc Linn. Paris 11: 947 (1891), noinen

nudum. P. brevipes (Baker) Engl., Mon. Sapot. Afr. 37 (1904). P. cinerea (Engl.)

Pierre ex Engl., op cit., 36, t. 12, incl. vars. ; type: Wehvitsch 4824 in Bf, isotype

BM, type number COI!

Bakeriella brevipes (Baker) Dubard and B. cinerea (Engl.) Dub. in Ann. Mus. Col.

Marseille 20; 27(1912).

A tree reaching a height of at least 10-15 m. Branches rather thick, at first thinly

pubescent or thinly brownish-tomentose, glabrescent, later longitudinally fissured and

often turning ashy-gray or almost white. Stipules coriaceous, linear-subulate, 5-15 mm.

long, brown rusty-pubescent or glabrous, very acute. Petioles stoutish, 5-10 (-15) mm.

long, 2-4 mm. thick, fiat above, when dry longitudinally sulcate. Leaves lanceolate-

oblong or oblong-oblanceolate to obovate-oblong, 5-20 cm. long and 2-8 cm. wide,

coriaceous, shiny and glabrous above, much paler and dull, shortly whitish-tomentose

or glabrous below, with obtuse or shortly and bluntly acuminate apex and narrow,

cuneate or decurrent-attenuate base, and revolute edge; midrib impressed and dis-

tinctly keeled above, prominent and when dry longitudinally fissured below; lateral

veins 8-10 on either side, distant (1-3 cm. apart), arcuate-ascending, impressed above,

very prominent below, all reaching the edge of the leaf or nearly so ; ultimate nervation

coarsely reticulate mainly more or less perpendicular to the midrib; veinlets very

delicate and inconspicuous. Pedicels short and thick, dz 3 mm. long and 1-2 mm. in

diam., covered with a pale fawn tomentum. Flowers clustered in the axils of the lower-

most leaves or on the naked wood below the leaves on raised warts, sweet-scented.

Sepals ovate-oblong, oblong or oblong-lanceolate (the inner ones narrower), 3 -5-4 -5

mm. long and 1 • 5-3 m.m.. wide, subacute or obtuse, more or less concave, pale fawny-

tomentose outside and inside. Corolla glabrous; the tube ± 2 mm. long and 1-1-5 mm.

in diam.; the lobes oblong or ovate-oblong, subacute or obtuse i 4 mm. long and

1-5-2 -25 mm. wide. Atternipetalous staminodes glabrous, lanceolate-linear, linear-

subulate, or filiform, acute, acuminate and often lacerate or incised-dentate (the filiform

ones not infrequently with a term.inal thickening or even a sm.all sterile anther), usually

shorter than the filaments, but occasionally equalling the stamens, sometimes small,

squamiform or O. Filaments linear-filiform, 3-4 mm. long; anthers pink (Mrs. Faulk-

ner), 2-2-5 mm. long. Ovary ovoid, i 2 m.m. long and 1 -5 mm. in diam., distinctly 5-

lobed-sulcate below, densely fulvo-villous, situated on a flat disc; style thick, colum-

nar, angular, widened, subcapitate-truncate and indistinctly 5-lobed at the apex, 4-5 mm.

long, covered with long hairs at the base or sometimes half way up. Fruiting pedicels

hardly changing but more or less glabrescent; the calyx persistent, ± spreading but not

reflexed. Fruit edible, yellow when ripe, ellipsoid, 15-22 mm. long and 9-12 mm. in

diam. Seed the same shape as the fruit but smaller, 12-16 mm. long and 6-9 mm. in

diam., scar occupying more than half the surface of the seed; testa smooth and shiny

light brown, the scar duller and paler, somewhat rough.

Fig. 5. — Pouteria brevipes, seed and fruit (from Zenker 4324, Cameroons, in PRE).

335

Widespread in tropical Africa, but not recorded from Northern Rhodesia. I

have seen numerous specimens from tropical East and West Africa in several herbaria

and only cite those occuring in Southern Africa:

Portuguese East Africa. — Manica e Sofala; Chipinga, Busi Drift (East of

Melsetter, S. Rhodesia), Whellan 133 (SRGH). Maribane. Gomes Pedro 4193 (LMJ,

PRE); “ Na floresta de Maronga ”, Simdo 375 (LM); Matarara do Lucite, Gomes

Pedro 4278 (LMJ. PRE). Zambezia: Quelimane Distr., Metola, Barbosa & Carvalho

4002 (LM, PRE). between Mualama and Gile. Barbosa & Carvalho 4342 (PRE);

Mocuba, Faulkner “ Kew 18 ” (PRE, SRGH, COl). Niassa: Nampula, Nova Chaves,

Barbosa & Lemos 1780 (LM); Pto. Amelia, Mueda, Barbosa 2238A (LM, PRE);

between Mueda and Chomba Barbosa 2248 (LM).

Southern Rhodesia. — Vumba: Wychwood, Ball 14 (SRGH, PRE). Melsetter:

between Hayfield and Lusitu river/Haroni, Drummond 5001 (SRGH, PRE).

Angola. — Cuanza; Golungo Alto, between Cambondo and Luinha River, Wel-

witsch 4818 (COI); Pungo Andongo: Wehvitsch 4824 (COI; type number of Pachystela

cinerea); Calemba Island in Cuanza Riv., Wehvitsch 4826 (COI); Ponta Eilomene de

Camera, nr. Cuanza Riv., Gossweiler 10649 (COI).

As regards the author of the combination “ Pachystela brevipes ”, the genus

Pachystela was only validly described in 1904, so that Baillon’s name '"Pachystela

brevipes (Baker) Baillon ”, published in 1891, is a nomen nudum and "Pachystela

brevipes (Baker) Engl.” is the correct citation under the Rules.

2. P. magalismontana (Sond.) A. Meeuse, comb. nov.

Chrysophyllum magalismontanum Sond. in Linnaea 23: 72 (1850) (sphalm. “magalis-

montana”); Engl., Mon. Sapot. Afr. 47 t. 16, f. C (1904); Wright in Dyer, El.

Cap. 4, 1: 437 (1906); Phillips in Flow. PI. S. Afr. 3, t. 98 (1923); Marloth, El.

S. Afr. 3: 36, t. 10 (1932); Gerstner in J. S. Afr. Bot. 12: 40, Fig. 4 (1946), and

14: 171, Figs. A-F (1948); Brenan in Mem. New York Bot. Card. 8 (5): 498

(1954); type: Zeyher 1849 from Magaliesberg, Transvaal in herb. Sonder nunc S,

holo, BOL and SAM, isos!). C. argyrophyllum Hiern, Catal. Afr. PI. Welw. 3: 641

(1898); Engl. op. cit., 46, t. 16, Fig. A; Brenan & Greenway, T.T. Check List 2:

(1949); type: Welwitsch 4827, 4828, 4829, syns. in BM, 4828 in COI!) C.

antunesii Engl, in Engl. Bot. Jb. 32: 137 (1903); type: Antunes 98 (B t, COI,

lecto.!). C. carvalhoi Eng., op. cit. (1904), 47; type: Rodrigues de Carvalho s.n.

in COI, lecto!, B C. wilmsii Engl., op. cit. (1904), 46, t. 16, Fig. B; Wright,

op. cit., 437; type: Wilms 1812 in B holo, K, iso. C. gossweileri De Wild.,

PI. Bequart. 4: 130 (1926); type: Gossweiler 2808 in BR, dupl. in COI!

Sideroxylon randii Sp. Moore in J. Bot. 41: 402 (1903); Wright, op. cit. 439; type:

Rand 1017 from Johannesburg, BM, holo.; photo in J!.

Pachystela magalismontana (Sond.) H. Lee. in Bull. Mus. Hist. Nat. Paris 25: 192

(1919). P. argyrophylla (Hiern) H. Lec., l.c.

A large tree when growing in forests (up to at least 15 m. high), but in its more

characteristic form, growing on stony koppies and rocky ledges in the Transvaal, a shrub,

already flowering and fruiting freely when only about 1 m. high. Leaf-bearing branches,

especially in the shrubby form, often stout, 5-10 mm. thick, with short internodes.

Innovations and twigs rufo-tomentose. Leaves often near apex of otherwise leafless

branches, stipulate and sometimes thinly papyraceous when young, coriaceous when

mature, first with a white bloom but soon glabrous and green above, rufo- or aureo-

tomentose (older ones often more greyish- or silvery-tomentose, sometimes with a pale

pink or mauve tinge) below, 4-15 cm., sometimes up to 30 cm. long, 2-5 cm., sometimes

up to 7 cm., wide, those of the characteristic shrubby form usually deflexed, rather small

(up to 12 cm. long and 5 cm. wide), oblong-obovate, obovate-elliptic or oblong, usually

336

rounded or slightly narrowed at the base, rarely obovate with cuneate base; emarginate,

retuse or at least obtuse at the apex, sometimes mucronate with short, often blackish,

mucro, sometimes more strongly tapering at the base and oblanceolate-oblong; petiole

6-12 mm. long; those of the forest form usually distinctly cuneate at the base, lanceolate,

oblong or oblanceolate-oblong to oblanceolate-cuneate, occasionally (on coppice

shoots) up to nearly 30 cm. long and 7 cm. wide on a longer (up to 24 mm. long) petiole,

but usually smaller, with obtuse, rounded, emarginate or retuse, sometimes shortly

and bluntly acuminate, occasionally mucronate apex; midrib immersed and narrow

canaliculate above, very prominent below; secondary nerves numerous (6-7 per cm.),

slender, usually inconspicuous above, partly hidden under the tomentum, but on the

older leaves often loosing their pubescence and becoming slightly prominent beneath,

patent, straight, but distinctly ascending and curved towards the margin, sometimes

joining and forming an irregularly sinuous, often very inconspicuous intramarginal

vein, more rarely bifurcate and archingly joining; tertiary nerves parallel to and thinner

than secondary ones, somewhat sinuous, mostly not reaching the margin without

branching or joining other veins, often inconspicuous; ultimate reticulate nervation

usually sparse, inconspicuous, mainly present towards the margin and i parallel to

secondary and tertiary nerves. Petioles thick, subterete but often longitudinally ribbed

or angled, at least when dry, rusty- or aureo-pubescent, later sometimes ashy-grey-

pubescent on a brown background. Stipules long-subulate, often curved, pubescent,

soon deciduous. Flowers in few- to many-flowered, sometimes very dense, fascicles,

the majority usually on the lower leaflets parts of the branches or on older wood on

sometimes rather large, raised warts, and fewer, or none, in the leaf axils. Bracts O

or very minute. Pedicels densely rufo-tomentose, varying in length from it 2 mm. to

It 5 mm., more rarely up to i 10 mm. long, rather stout and gradually or more abruptly

widening at the top into the calyx; rarely flowers sessile. Calyx 2-5-4 (-5) mm. long,

rufo-tomentose outside; sepals free nearly to the base, often unequal, ovate, obtuse or

subacute, often greyish-pubescent inside, 2-4 • 5 mm. long and 2-3 ■ 5 mm. wide. Corolla

white or whitish, turning brown (according to the labels also “ pink ” or “ red ” in

Rhodesia and East Africa, but this may refer to already wilted flowers), glabrous,

varying in length from scarcely longer than to about 2 mm. longer than the calyx;

tube cylindric-urceolate, usually 1-1 -5 mm. sometimes up to 2 mm., rarely only about

0.5 mm. long; the lobes somewhat to distinctly spreading, broadly ovate, obtuse or

subacute, 2-4-5 mm. long and about 2 mm. wide. Alternipetalous staminodes O or

sometimes 1-5, inserted just below the sinuses between the corolla-lobes, much smaller

than the latter, scale-like and minute or sometimes petaloid, ovate or suborbicular.

more or less irregularly serrate, dentate or incised in the supper half, up to 1-5 mm.

long and 0-5-1 mm. wide. Stamens subincluded, inserted at the base of the corolla-,

lobes; filaments 1-5-2 mm. long; anthers 1-2 mm. long, somewhat cordate-sagittate

at the base before dehiscence, acute, apiculate; sometimes stamens sterile, staminodial,

either resembling a stamen with a filament-like basal portion and a sagittate-cordate,

broader top which often shows two longitudinal marks (cf. Gerstner, 1948, Figs. A-F),

or more irregularly shaped, very rarely (fide Engler, op. cit., 47. t. 16, Fig C, and c)

lanceolate, petaloid. Ovary globose-ovoid, about 2 mm. in diam.. densely villous,

5-, sometimes 3- or 4-celled, gradually passing into the — 1-5 mm. long, glabrous

style. Fruit ellipsoid, i 25 mm. long ± 18 mm. wide, dark red when ripe, edible,

crowned with the persistent style, 1- or sometimes 2-seeded. Seeds of 1 -seeded fruits

compressed -ovoid, 16-20 mm. long, 14-16 mm. broad and 8-11 mm. thick; those

of 2-seeded fruits with one flattened lateral side; testa light brown, shiny, thin and

brittle when dry; scar linear-triangular, ventral, occupying about of the length of

the seed; 2 mm. wide or more in widest place.

General Distribution. — Tropical Africa, from the Congo to Tanganyika and south-

wards to Angola, Bechuanaland. the Transvaal, Swaziland and Natal. At least in the

Transvaal this species is apparently confined to quartzite and granite rocks, so that the

337

plant is as much an indication of the formations as the formations are an indication

of the occurence of P. magalismontana. In N. Zululand, for instance, where quartzite

rocks occur locally, P. magalismontana is only found in the area where the rocks come

to the surface.

Selected citations',

Bechuanaland. — Kanye; Hillary & Robertson 613; Lobatsi Govt. Farm: Miller

Transvaal. — Zoutpansberg: Louis Trichardt, Hanglip, Gerstner 5728. Sebasa:

near Sebasa, 4515. Pietersburg: Woodhush, Hoffmann 22. Warmbaths:

Warmbaths. Leendertz s.n., Burtt Davy 2616, Hutchinson 1883. Brits: Silkaatsnek,

Smuts &. Gillett 1061. Magaliesberg Range: Zey/rer 1849 (BOL, SAM, type gathering);

Burke 377 (“twin type BOL). Pretoria: Pretoria, Leendertz 322, 510; Burtt Davy

2675, Pole Evans 30, 161; Hutchinson 2314. Nelspruit: Kruger National Park,

V. d. Schijff 59, Codd 5145. Barberton: Burtt Davy, 25%. Marico: ZetXMSt, Gerstner

4413 (NH). Rustenburg: Leendertz s.n., Pegler 1033. Witwatersrand: Johannesburg,

Rand 1017 (PRE, fragment of type of Sideroxylon randii); Gerstner 6418, 6424; English

in Herb. Galpin 1486. Heidelberg: Leendertz s.n. (TRV. No. 8077); Delmas: Maude

s.n. Brits: Silkatsnek, Smuts & Gillett 1061; Hartebeespoort, Prosser 1297 (NBG).

Magaliesberg Range: Zeyher 1849 (BOL, SAM, type gathering); Burke 377 (BOL).

Pretoria: Pretoria, Leendertz 322, 510; Burtt Davy 2675; Pole Evans 30, 161; Hut-

chinson 2314.

Swaziland. — Codd 1585; Acocks 12850; Miller S/108; Bolus H. No. 12110

(BOL).

Natal and Zululand. — Ngotshe: Magut, Pongola, Gerstner 2461; Ngome

Forest Stat., Tustin = FD herb. No. 6552.

Portuguese E. Africa. — Manica e Sofala: “ Floresta do Garuso ", Simdo 552

(PRE: with lanceolate, petaloid epipetalous staminodes instead of stamens); Chimani-

mani Mts., Plowes 1250 (SRGH); Gorungosa, Rodrigues de Carvalho s.n. (COI,

sterile branch, type of Chrysophyllum carvalhoi Engl.).

There has been some misunderstanding as regards the occurrence of the reduced

sterile flowers and their proper significance. Sonder described Chrysophyllum magalis-

montanum as having normal stamens, citing Zeyher 1 849 as the type number. Engler

(1904) described C. magalismonatum as having female flowers with lanceolate epipetalous

staminodes and oblong leaves (he cites Zeyher 1849 and Burke) and distinguished

C. wilmsii, which is described as having leaves that are much narrowed at the base,

and fertile anthers. Wright (in FI. Cap. 4, 1 : 437) distinguished these two as follows:

Leaves oblong, obtuse, obtuse at the base; flowers pedicellate magalisnwnt anirn

Leaves oblong, obtuse, mucronulate, acute at the base, flowers shortly pedicellate W ihrxii

Wright made no mention of staminodes, but also cited Zeyher 1849 and Burke 377

under C. magalismonatum.

Gerstner identified the form with sterile anthers with C. wilmsii Engl., which is

not correct, but he discovered that C. magalismonatum under unfavourable conditions

produces depauperate flowers {Gerstner, op. cit. 1948, p. 171), especially after a severe

drought of several months. As soon as sufficient rain has fallen the same plants develop

complete flowers. The sterile stamens are sometimes transformed into lanceolate

epipetalous staminodes (e.g., in Simao 552 and in a specimen leg. N.N. from Bulawayo =

SRGH No. 5579), and the specimen Zeyher 1849 in the Berlin Herbarium studied by

Engler is probably a branch with such abnormal flowers, whereas other specimens

of the gathering Zeyher 1849 such as those in BOL! and SAM! (taken from other

branches or other trees ?) bear normal flowers and this explains the apparent contro-

versy. I have also found that fertile and sterile stamens can occur in one flower, and

338

these sterile anthers, sometimes appearing as subpetaloid epipetalous staminodes are

clearly abnormal, at any rate they have no taxonomic value. The difference in leaf

shape as indicated in FI. Cap. breaks down altogether as a character and C. magalis-

.montanum and C. wilmsii are clearly synonomous.

Fig. 7. — Pouteria magalismontana, opened corolla-tubes of two flowers of Gerstner

6418 (from Johannesburg, Transvaal), x 5, seen from the inside. Above: flower

with 5 developed alternipetalous staminodes, below: flower without staminodes.

The presence or absence of alternipetalous staminodes is independent of the degree

of development of the stamens and they also occur in flowers of Gerstner’s forma

depaiiperata (such as those of the specimen Gerstner 6424 cited by him). These alterni-

petalous staminodes are by no means rare: if sufficient flowers of a single specimen

are dissected, some flowers with 1-5 staminodes (at least one in 10) are found. However,

some specimens (or perhaps some individual plants) show a much higher frequency

of flowers with developed staminodes of this type, such as Gerstner 6418 in which the

majority (over 60%) of the flowers possess petaloid staminodes (see Fig. 7). Moore’s

type specimen of Sideroxylon randii {Rand 1017 in BM) is obviously such a special case.

In every other respect, the description of 5. randii and a photo of the type specimen

(in J) agree perfectly with C. magalismontanum. Mr. de Winter, who examined the

type specimen, confirmed my opinion and sent one flower of Rand 1017 which proved to

be identical with P. magalismontana, so that S. randii falls into synonymy. This is a

good example of the unreliability of the absence or presence of alternipetalous stamin-

odes as a main distinguishing character in this family. Moore, after observing distinct

staminodes, made it a Sideroxylon, whereas Bonder and Engler, who did not see distinct

staminodes, referred the same species, apparently without any doubt, to Chrysophyllum.

As regards the identity of P. magalismontana and Chrysophyllum argyrophyllum

Hiern, the differences indicated by Engler in his monograph are very slight. The dif-

ferences reported are: (1) colour of the lower leaf surface (rusty-tomentose in magalis-

montanum, grey or silvery in argyrophyllum)', (2) leaf shape (oblong to obovate-oblong

in the first, oblanceolate (ioblong) in the second); (3) petioles, pedicels and calyx rusty-

tomentose in the first, greyish-pilose in the second. These differences are not important,

because the pubescence may be rusty- or greyish-tomentose on various parts of the

same specimen, older parts tending to change from rusty-brown into grey and the leaf-

shape varies even in one of the original gatherings cited by Hiern: Welwitsch 4828

339

(in COI!). I have seen many specimens from Angola, Rhodesia and East Africa and

there is no sharp distinction between those referred to C. argyrophyllum in various

herbaria and those referred to C. magalismontanwn.

The species under discussion was redescribed several times again: ChrysophyUum

antunesii Engl, and C. gossweileri De Wild, are small-leaved forms from Angola (I

have seen types or isotypes of these in COI); C. carvalhoi Engl, is a form with narrow,

lanceolate and acuminate leaves, collected in Mozambique (type in COI!).

3. P. natalensis {Sond.) A. Meeuse, comb. nov.

ChrysophyUum natalense Sond. in Linnaea 23: 72 (1850); Engl., Mon. Sapot. Afr.

43, t. 34, Fig. C (1904); Wood, Natal PI. 4, t. 378 (1906); Wright in Dyer, FI. Cap.

4, 1 : 437 (1906); Sim, For. FI. Cape Col. 252, t. 94, (1909); Gerstner in J. S. Afr.

Bot. 12: 48, Fig. 2 (1946); type: Gueinzius 181 from Berea near Durban in herb.

Sonder nunc S.

A medium-sized tree, 8-15 m. high, but often fruiting freely without much height,

“ growing mostly in the open or gregariously ” (Sim). Stem up to 50 cm. in diam.,

with smooth bark, according to Sim and Gerstner yielding a valuable timber. Twigs

slender, terete, grey, glabrescent; internodes decreasing in length towards the apex,

so that the leaves are crowded at the tops of the twigs. Innovations and calyx densely

dark tobacco-brown-tomentose. Leaves 6-12 (-15) cm. long and 2-4 (-5) cm. wide,

obovate-lanceolate (ob) lanceolate-elliptic or (ob) lanceolate-oblong, shortly acuminate,

acute or obtuse, at the very apex canaliculate-subdeflexed. sub-emarginate and mucronu-

late (occasionally in a few leaves retuse or emarginate), gradually tapering into the

acute or somewhat acuminate base, when mature green (drying a characteristic greyish-

green), glabrous and very shiny above, very minutely greyish- or silvery-tomentose,

later sometimes glabrescent, beneath, with entire, subdeflexed and ± undulate margins;

midrib immersed and channelled above, prominent and conspicuous below; seeondary

nerves thin, raised, about 20 on either side, patent, almost straight, bifurcate usually

well within the margin and archingly joining, no distinct intramarginal vein present;

teritiary nerves inconspicuous, mostly parallel to secondary ones. Petioles 6-12 (-14)

mm. long, canaliculate above, often brown or blackish, rugose, with two short dorsal

ridges where the slightly decurrent leaf-base ends and there seemingly distinctly

narrowing into the leaf-base when seen from above (at least when dry), but in fact

continuous with the midrib. Stipules wanting or at least very early deciduous. Flowers

sessile, in clusters of 1-3 (sometimes more) together in the leaf axils; bracts small,

inconspicuous, broadly triangular, rusty-pubescent, or wanting. Calyx ± 4 mm.

long, divided more than half way down, the segments erect, ovate, subacute. Corolla

“white” or “yellowish”, glabrous; tube somewhat longer than the calyx, more

or less constricted above the middle; the lobes suberect, ovate, obtuse, about 1 mm.

long. Anthers subsessile, inserted at the very base of and about as long as the corolla-

lobes, ovate, apiculate, not or scarcely exerted. Ovary depressed-globose, more or less

5-lobed, shortly pilose, contracted into the glabrous, obtuse, cylindrical style which is

up to about twice as long as the ovary. Berry subsessile, cylindrical-ovoid or cylindric-

oblong, pointed (more or less shaped like an acorn), crowned with the persistent style,

2-2-5 cm. long and 1-1 -5 cm. in diam., deep red when ripe (? also sometimes trans-

parent white, Sim) minutely pubescent, one-seeded, edible. Seed ellipsoid-oblong,

± 20 mm. long, ± 8 mm. wide, and ± 5 mm. thick in the centre, with a long ±2-4 mm.

wide scar occupying the whole ventral side of the seed; testa thin, brittle.

Distribution. — In frostless forests in the Eastern Cape Province and Natal, from East

London and Komgha northwards, also recorded from the central Transvaal, Portuguese

East Africa and the eastern part of Rhodesia, and extending into tropical East Africa

as far as Tanganyika or perhaps Uganda.

6096259—5

340

The type (not seen) is Gueinzius 181 from Berea nr. Durban, Natal(in herb. Sonder).

This same number is cited by Harvey ex Wright in FI. Cap. 4, 1 : 437, together with

Sanderson 657, Wood 732 and a few other gatherings. Engler (1904>^ites Gueinzius 181

and Medley Wood 8950. The specimen Wood 732 and Engler’s figure leuv^ no doubt

about its identity.

Cape Province. — East London: near East London, Galpin 9284 (PRE), 9677

(PRE). Komgha: Flanagan 1138 (GRA, PRE, BOL, SAM). Pondoland: Egossa,

Sim 2374 (NU, GRA, PRE, BOL, SAM). Kentani: Pegler 859 (GRA, PRE, BOL,

SAM, NBG); Manubi Forest, Story 4475 (PRE). Willowvale: Acocks 12284, 12286

(PRE). Ngqeleni: Notinsella, FD herb. No. 1728 (SAFD); Gokama Forest: Marais

758. Port St. Johns: O. B. Miller D/88 = FD herb. No. 3864 (SAFD); Noxolweni ^

Forest, Mogg 13089 (PRE).

Natal. — Umzinto: Dumisa, Rudatis 964 (L). Pietermaritzburg: Killick 308 '

(PRE). Durban: Sanderson s.n. (= 657?) (PRE), Wood s.n. BOL, SAM = prob.

Inanda, Wood 732, GRA); Marriott PS 230 (PRE), Thorns s.n. (NH no. 23407),

Lavoipierre 94 (NU^ Eshowe: Lawn 57 (NH), Gerstner 1920 (NU, PRE), Fqrbes 681

(NH), Codd 1860 (PRE). Hlabisa: Hluhluwe Game Reserve, Codd 2050, Ward

1692 (PRE). “ Zululand ”: Gerstner 2618 (BOL).

Transvaal. — Pilgrims Rest: Mariepskop, Killick & Strey 2496 (PRE).

Swaziland. — Miller S. 263 (PRE).

Tanganyika. — Devern: Burtt s.n. (J); Zigigler (or Zigiglen): Burtt s.n. (J), ■

East Usambaras: Mlinge-Tongwe: Greenway 6064.

Two specimens from Uganda (Greenway 6064 and Greenway & Eggeling 7076 in

EA and PRE) are very similar if not conspecific. Two species described from E. Africa,

viz. Chrysophyllwn holtzii Engl, et Krause in Engl. Bot. Jahrb. 49: 390 (1913), and

C. tessmannii Engl, et Krause (op. cit., p. 389) may well be synonyms of Pouteria natalen-

sis.

As regards its taxonomical position, Engler placed P. natalensis in the subgenus

or section Gambeya (Pierre) Engl. (= Afrochrysophyllum Engl, p.p.) of Chrysophyllwn,

but this species has several features which distinguish it from Gambeya-. according to

Engler’s diagnosis of Gambeya (Engler, op. cit., p. 43), the stamens are inserted at or

below the middle of the corolla-tube (“ Staminum filamenta ad basin tubi vel medio

libera ”), the corolla lobes are ciliate (Corollae tubus lobis ciliatis aequalis vel longior ”)

and the nervation of the leaves is different: (“ . . . nervis lateralibus I pluribus

numerosis arcuatim patentibus prope marginem sursum versis, nervis lateralibus II

inter primarios obliquis ”). The fruit is also different (1-seeded in P. natalensis, usually

several-seeded in Gambeya, the seed-scar occupying the whole ventral side of the seed

in P. natalensis, only the greater part of the ventral side in Gambeya) and, finally the

ovules are basolaterally attached in Gambeya and distinctly laterally in P. natalensis.

I cannot refer P. natalensis to any other African genus but Pouteria on account of the

fruit characters (thin testa, large scar and lack of endosperm), in spite of the apparently

constant absence of alternipetalous staminodes (which would not, however, exclude it

from Pouteria sensu Van Royen).

Although most modern authors do not agree with Baehni’s very broad conception

of the genus Pouteria, several tropical species which fall into Pouteria sensu Van Royen

were either overlooked, or expressly excluded from the genus by Baehni. I cannot see

why these species were omitted or excluded, as Baehni did not give any reasons for his

actions. It seems necessary to make the recombinations in Pouteria, because it may

guide future workers on the flora of tropical Africa as regards my conception of the

genus. This list is not complete, for only those species are included which could be \

studied from authentic herbarium specimens or from good plates.

341

Pouteria adolfi-frederici (Engl.) A. Meeuse, comb. nov. Sideroxylon{l) Adolfi-frederici

Engl, in Mildbr., Wiss. Ergebn. dent. Zentral-Afr. Exp. 1907-1908, 2: 519,

t. 52 (1913). Aningeria adolfi-frederici (Engl.) Robyns & Gilbert in Robyns, FI.

Spermatophyt. Nat. Parc Albert 2: 43 (1947).

The plate and a duplicate of a syntype (named by Engler), viz., Mildbraed 2528

(PRE), were available for study. There is no doubt that it is a Pouteria.

Pouteria cerasifera (Welw.) A. Meeuse, comb. nov. Sapota cerasifera Welw., Apontam,

585, No. 17 (1859). Chrysophyllum cerasiferum (Welw.) Hiern, Cat. Afr. PI.

Welw. 3: 643 (1898). Sersalisia cerasifera (Welw.) Engl., Mon. Sapot Afr. 30

(1904).

Pouteria disaco (Hiern.) A. Meeuse, comb. nov. Chrysophyllum disaco Hiern, Cat.

Afr. PI. Welw. 3: 642 (1898). Sersalisia disaco (Hiern) Engl., Mon. Sapot. Afr.

30, t. lOA (1904).

Mr. B. de Winter compared a few specimens from Portuguese East Africa with

authentic material (Welwitsch 4812 in BM). Some of these are in fruit and can, there-

fore, be referred to Pouteria sensu Van Royen.

Pouteria msolo (Engl.) A. Meeuse, comb. nov. Chrysophyllum msolo Engl., Pflanzenw.

O. Afr. C., 306, t. 37 (1895). Pachvstela msolo (Engl.) Engl., Mon. Sapot. Afr.

38 (1904).

This plant is so closely related to “ Pachystela brevipes ” that Engler did not hesitate

to refer it to Pachystela and it is altogether incomprehensible to me how Baehni could

exclude Pachystela msolo (in Candollea 9: 428), while referring P. brevipes to Pouteria

(op. cit., 291). The same applies to the following species which is also very similar to

P. brevipes and yet excluded by Baehni (op. cit., 428) from Pouteria.

Pouteria zenkeri A. Meeuse, nom. nov. Pachystela robusta Engl, in Engl. Bot. Jahrb.

49: 386 (1913), non Pouteria robusta (Mart, et Eichl.) Eyma.

A true isotype, named by Engler, viz. Zenker 3697 (in PRE) was studied. The

specific epithet “ robusta ” being preoccupied on account of an earlier combination

for an American species, the name “ zenkeri ” was chosen to commemorate its first

collector, well-known for his extensive West-African collections. It is strange that no

reference is made to this Cameroons species in Hutchinson and Dalziel’s FI. W. Trop.

Afr. vol. 2 under Sapotaceae.

4. VINCENTELLA

Pierre, Not. botan. Sapot. 37 (1891). Bakerisideroxylon Engl, (as a section of Sider-

oxylon) in Engl. u. Prantl, Natiirl. Planzenfam. ed. 1, 4, 1: 144 (1890), and in

Nachtrage 276 (1897); (as a genus) in Mon. Sapot. Afr. 33 (1904).

Sideroxylon sensu Baker in Oliv., FI. Trop. Afr. 3: 502 (1877), pro parte.

Bakeriella Dubard in Lecomte, Not. Syst. 11: 89 (1911) and in Ann. Mus. Col.

Marseille 20: 26 (1912), pro parte.

Pouteria Aubl. sectio Bakerisideroxylon (Engl.) Baehni in Candollea 9: 382 (1942).

Type species: Baehni, in Candollea 7: 497 (1938), p. 497, mentions as the type

species Vincentella longistyla (Baker) Pierre, Not. Bot. Sapot. (1891), p. 37. However,

this is a synonym of Pachystela brevipes and as Pachystela was only validly published

in 1904, V. longistyla at the same time being removed from Vincentella and transferred

to Pachystela by Engler, the type species of Vincentella must be among the other two

species mentioned by Pierre and retained in Bakerisideroxylon by Engler, viz., V.

densifiora (Baker) Pierre and V. revoluta (Baker) Pierre. Priority of place would indicate

Sideroxylon densifiorum Baker = Vincentella densifiora (Baker) Pierre (from San

Tome Island) as the type species, as was rightly pointed out by Exell in Cat. Vase.

342

PI. S. Tome 235 (1944).

Shrubs or trees. Leaves generally oblong, with usually rather distant prominent

secondary nerves and fine tertiary reticulate nervation. Stipules early deciduous.

Flowers small, 5-merous, in many-flowered fascicles in the leaf axils and especially

in the axils of fallen leaves on the twigs; pedicels very slender to capillary often

rather long and more or less pendulous. Sepals small, free nearly to the base, later

patent or reflexed. Corolla-tube very short, the lobes many times longer, elongate-

oblong or linear-oblong, strongly reflexed. Staminodes and stamens inserted at the

throat of the corolla-tube. Alternipetalous staminodes narrowly linear, inserted between

and about as long as the corolla-lobes (but erect), entire or rarely dentate-serrate, acute

or acuminate. Filaments filiform, erect, several times longer than the oblong-sagittate,

minutely apiculate anthers. Ovary large, ovoid, villous, contracted in the usually

rather long and filiform-cylindric, glabrous style, 5-loculated; ovules with lateral

attachment, pendulous. Fruit oblong-ovoid, 1 -seeded with 4 sterile loculi. Seeds

oblong; testa crustaceous; scar long, linear, occupying the upper part of the ventral

side of the seed; cotyledons thick and fleshy, endosperm membranous or absent.

An African genus of four spec^, on species extending into Southern Africa.

As Baehni, l.c. pointed out, Vincent'ella Pierre (1891) was legitimately published

as a genus, although it corresponded with a section Bakerisideroxylon (1890) distin-

guished by Engler in the genus Sideroxylon which was later (1904), given the status of

a genus by Enger. Pierre was not compelled to take up Engler’s section name when

he established a new genus and Vincentella, therefore, stands.

The typical slender pedicels, reflexed narrow corolla-lobes and very short corolla-

tube, the fairly large ovary, erect long anthers and long staminodes make it possible

to recognise a species of Vincentella almost at once. Baehni retains Vincentella as a

section Bakerisideroxylon of his large genus Pouteria Aubl. sensu Baehni, but it is so

distinct that, in my opinion, it deserves generic rank.

V. sapinii {De Wild.) Brenan in Mem. New York. Bot 8: 498 (1954).

Bakerisideroxylon sapinii (“ Sapini De Wild, in Rev. Zool. Afr. 7, suppl. bot. B 16

(1919), and in PI. Bequart. 4; 116 (1926), type; Sapin s.n. BR, holo.; K, iso.

Pouteria tridentata Baehni in Candollea 9: 386 (1942); type: Stolz 1889 from Tan-

ganyika in G, holo; K and PRE, isos.

Bakerisideroxylon stolzii Mildbr. Ms.

Vincentella stolzii (Mildbr. Ms. ex) Hutch., Botan. South Afr. 506 (1948), nomen

nudum.

General Distribution. — Southern Tanganyika, Nyasaland and the Niassa Province

of Portuguese East Africa, Northern Rhodesia and Belgian Congo, seems to prefer

streambanks (teste Gomes e Sousa, Benson, Brass).

Nyasaland. — Kota-Kota distr. : Benson 257, 765 (PRE); Chia area. Brass 17510

(PRE).

Portuguese E. Africa. — Niassa: Nampula distr., Ribaue, Gomes e Sousa 2305

(PRE).

Tanganyika Territory. — Kyimbila; Stolz 1889 (isotype of Pouteria tridentata

Baehni, PRE).

“ A shrub or small tree, 4-6 m. high”. (Gomes e Sousa). Young branches terete,

densely rusty-tomentose-hirsutulous, later becoming shortly and more greyish-tomentose,

the older ones ultimately glabrescent. Leaves oblanceolate-oblong to obovate-oblong.

343

with obtuse, rounded or emarginate apex and attenuated-cuneate base, with reflexed

margin, 4-10 cm. long, 3-5 cm. wide (according to Baehni: 10-14 cm. long and 3-5

cm. wide), chartaceous or subcoriaceous, soon glabrous, rather shiny above, paler and

duller beneath; midrib impressed and keeled above, very prominent and when dry

longitudinally sulcate below, glabrescent but retaining the original tomentum some-

what longer than the rest of the blade; secondary nerves impressed above, prominent

below, 8-12 on either side, rather distant (4-10 mm. apart), at first rather straight,

ascending at an angle of 45°-70°, arcuate-ascending near the margin and reaching it

except those near the apex; tertiary nerves inconspicuous above, subimpressed and

rather distinct below, slender, mainly perpendicular to the secondary ones; ultimate

nervation very fine reticulate, tessellate, rather distinct at least below. Petioles stout,

densely rusty-tomentose-hirsutulous, 3-10 mm. long, flattened and canaliculate above.

Stipules linear-subulate 4-5 mm. long, deciduous. Flowers fragrant (teste Benson).

Pedicels 5-7-5 mm. long, hirsute, terete, slightly widening under the calyx. Sepals

ovate-triangular, hairy outside glabrous inside, about I f mm. long. Corolla white

(teste Gomes e Sousa and Benson) glabrous, lobes narrowly-oblong, obtuse, zb 3 mm.

long zb 1 mm. wide. Staminodes linear, acute, about as long as the corolla-lobes, with

or a few small or minute lateral teeth. Filaments slender, 2 • 5-3 mm. long, the anthers

:b0-5 mm. long. Ovary ovoid-conical, hirsute, zb 1+ mm. long and zb 1 mm. in

diameter, attenuated into the long-subulate-linear, sulcate (at least when dry), subacute

and glabrous 14—2 mm. long style. Fruit and seed not seen but fruit reported by Brass

to be yellow, soft and edible.

5. LECOMTEDOXA

{Engl.) Dubard in Ann. Mus. Col. Marseille 23: 31 (1915); Baehni in Candollea 7: 456

(1938); Lam in Blumea 4: 348,350(1941).

Mimusops subgenus Lecomtedoxa (Pierre ex) Engl., Mon. Sapot. Afr. 82, t. 24, Fig. A.

(1904).

Mimusops subgenus Quaternaria sectio Inhambanella Engl., op. cit., 80; Pilger in

Engler & Prantl, Naturl. Pflanzenfam., ed. 1, Nachtrage 1897-1904), exclus. descr. of

the flowers.

Inhambanella (Engl.), Dub., tom., cit., 42, as to type species.

Type species. — Mimusops klj^iniana Pierre ex Engl., Mon. Sapot. Afr. 82 (1904) =

Lecomtedoxa kleiniana (Pierre ex Engl.) Dub. in Ann. Mus. Col. Marseille 23: 31

(1915).

Small to very large trees. Leaves more or less distinctly crowded at the tips of the

branchlets, coriaceous, more or less shiny above, paler and dull beneath. Flowers in

few to many-flowered fascicles in the axils of the leaves or of leaf-scars below the leaves.

Calyx-lobes 4-6; sometimes 3, sometimes unequal and subbiseriate, more or less erect,

more or less concave. Coro//a isomerous with the calyx; the tube short to rather long

the lobes with each 2 lateral appendages (sometimes with only one appendage); the

appendages entire, either very broad and larger than the lobes, or small. Alterni-

petalous staminodes lanceolate to ovate-lanceolate or long-triangular (resembling those

of Sideroxylon), rather large and conspicuous, alternating with the corolla lobes.

Stamens inserted in the throat of the corolla-tube or slightly higher up; filaments

short or rather long; anters apiculate, included or slightly exserted. Ovary 5(-6)-

loculate, hairy; ovules with lateral attachment; style glabrous, rather short, either

capitate or tapering at the apex. Fruit 1 -seeded, rather large. Seed with long scar

occupying the ventral side of the seed; testa crustaceous; endosperm O or very thin;

cotyledons thick and fleshy.

344

General Distribution. — 2 or 3 species in tropical West Africa, one in Portuguese

East Africa and Zululand.

L. henriquesii {Engl, et Warb.) A. Meeuse, comb. nov.

Mimusops henriquesii Engl, et Warb. in Engl., Mon. Sapot. Afr. 80 (1904), (sphalm.

“ henriquezii cf. “ Corrigenda ”, op. cit., p. 88); type: Rolla Ferreira s.n. from

Portuguese East Africa in COI, holo! M. henriquesiana (Sphalm.?) Sim, For.

FI. Port. E. Afr. 80, t. 77, A (1909); Gerstner in J. S. Afr. Bot. 12: 54 (1946).

Inhambanella henriquesii (Engl, et Warb.) Dub. in Ann. Mus. Col. Marseille 23: 42

(1915).

As Engler’s and Sim’s descriptions are very incomplete and the flowers had hitherto

not been known, a very detailed description is given.

Smalt to medium-sized tree, up to 20 m. high, with spreading branches and abun-

dant heavy foliage. Branches rather stout, terete, longitudinally striate and sulcate,

soon quite glabrous, greyish-born, later turning light grey and usually marked with

scars. Innovations rusty-tomentose, but vegetative parts soon quite glabrous. Stipules

subulate-falcate, pubescent, 3-7 mm. long, early deciduous. Petioles rather strongly

canaliculate above and with a very narrow dorso-lateral wing or ridge on either side

(which is a continuation of the edges of the decurrent leaf-base), when dry longitudinally

sulcate, 12-35 (-52) mm. long. Young growths bright red to red-brown. Leaves

varying from oblanceolate-oblong to oblong or obovate-oblong, obovate or broadly

elliptic, coriaceous but rather thin, green (drying pale green or greyish green), 5-12

(-17) cm. long and 2|-5 (-8) cm. wide, with obtuse, rounded or bluntly acuminate,

often distinctly emarginate and more or less deflexed apex, acute or somewhat attenuate

and slightly decurrent at the base, with reflexed margin, the edges often more or less

undulate; almost invariable some leaves of each specimen with raised round flat or

semiglobose galls; midrib impressed and keeled above, very prominent and when dry

longitudinally sulcate below; secondary nerves 6-9 on either side, rather distant (5-10

mm., sometimes up to 25 mm. apart), not very conspicuous and often subimpressed

(but sometimes slightly raised) above, usually prominent and more conspicuous below,

slender, at flrst patent (ascending at angles between 45° and 90°), but mostly soon

arcuate-ascending, the majority very strongly curved, becoming more or less sinuous

and parallel with the edge of the leaf, ultimately reaching the nerve above it and thus

forming a more or less distinct intra-marginal vein close to the margin, but some bifur-

cate or archingly joining; tertiary nerves more or less perpendicular to the midrib and

joining the secondary ones, forming a coarse reticulation which is slightly prominent

below; ultimate nervation very flne, reticulate, usually conspicuous at least on the

lower surface. Flowers in few to many-flowered fascicles; bracts ovate, usually strongly

concave to almost keeled and acute, 1-3 mm. long, pubescent; pedicels 5-20 mm. but

usually 10-15 mm. long, terete or slightly angular, brownish-tomentose, rather ab-

ruptly widening into the calyx. Sepals 5 (or 4 to 6) erect, more or less concave, 4-5

mm. long, unequal to sub-biseriate ; the outer 2-3 ovate-triangular from a broad base,

3^ mm. wide, subacute, brownish-tomentose outside and inside near the margin and

towards the apex; the inner ones thinner in texture, ovate-oblong, elliptic or broadly

Fig. 8. — Lecomtedoxa henriquesii, 1: Flower, x 5; 2: outer sepal; 3: inner sepal;

4: corolla tube, opened and seen from inside, x 10; 5: part of corolla tube seen

from outside, showing the lateral appendages, x 10; 6: ovary, x 10,7: fruit and

8: seed (7 and 8 from the type specimen, Ferreira s.n. in COI).

345

346

oblong, with a narrower base, obtuse or rounded at the apex, pale fawnish-tomentose-

sericeous outside, glabrous inside, finely ciliate along the margin, usually only 2-3

mm. wide. Corolla glabrous, yellowish or white; the tube cylindric-campanulate,

about 3 mm. long, the inside with distinct thickened lines below the stamens and the

staminodes; the lobes elliptic or elliptic-oblong, 4-4-5 mm. long, and 2-3 mm. wide;

lateral appendages shorter than or as long as the corolla-lobes, attached near the middle

of the corolla-lobes or near the base, usually asymmetrical, ovate-lanceolate to lanceolate-

falcate, 1-4 mm. long. Alternipetalous staminodes ovate-lanceolate or lanceolate to

elongate-triangular, often somewhat undulate, entire, acute, or obtuse, glabrous, about

3 mm. long. Stamens inserted on the corolla-lobes slightly above the throat of the tube;

filaments linear, 1-2-5 mm. long, glabrous, more or less winged and often with broad

somewhat auriculate base; anthers oblong, minutely apiculate, 1-5-2- 5 mm. long.

Ovary ovoid-conical, about 1-5 mm. long and 1-5-2 mm. in diameter, more or less

distinctly lobed, densely greyish-tomentose, usually 5-loculated; style glabrous, thick,

columnar, terete or somewhat angular, about 1 -5 mm. long and 0-5 mm. thick; stigma

capitate, indistinctly 5-lobed and sometimes also bilobed. Fruit ellipsoid, about 4 cm.

long and 2-2-5 cm. in diameter. Seed oblong, somewhat compressed, about 3 cm.

long, 1-5 cm. wide and 1-2 cm. thick; testa crustaceous, shiny, scar oblong, about

28 mm. long and 6 mm. wide; cotyledons about 5 mm. thick.

ZuLULAND. — Hlabisa; False Bay (according to Gerstner, l.c. ; no specimen seen);

St. Lucia Estuary, Ward 441 (NU); Mdlozi Peninsula, Ward 3032 (PRE).

Portuguese E. Africa. — Sul do Save: Maputoland, Santaca; Gomes e Sousa

3816 (COI, PRE). Manica e Sofala: Inhambane, RoUa Ferreira s.n. (COI, type );

Beira, Vila Fontes, leg. N.N., Laboratorio Quimico Herbario No. 77 (SRGH); Chis-

sadze, Cheringoma, Simao 1559 (PRE); Dondo Junction, Honey 871 (BOL, PRE).

Zambezia: Quelimane, between Mopeia and Morzumbala, Barbosa & Carvalho 3960

(PRE). Locality not known to me (near Inhambane?): Maguiya da Costa, Sim 2091 1

(PRE).

The type is a specimen leg. A. J. Rolla Ferreira in 1903 (s.n.), “ Regioes de Gaza e

Inhambane ”, but most probably from Inhambane, Portuguese East Africa, because

Engler named it Inhambanella" and Sim (l.c.) mentions “Ferreira’s farm near

Inhambane” as the place of origin. Engler erroneously mentioned: “Herb. Univ.

Cordoba ” instead of “ Herb. Univ. Coimbra “ as the location of the type specimen.

Sim refers to this plant as “ Mimusops henriquesiana ” but as he mentions Ferreira’s

farm and the specific epithet is so similar, I am of the opinion that “ henriquesiana ”

is a lapsus calami for “ henriquesii

As regards the identity of Mimusops henriquesii and the other specimens cited

above, the only fruiting specimen that I could study was the type, but the vegetative

parts and the calyx agree in every respect with those of the flowering specimens cited

here so that I feel certain that the identification is correct.

The species under discussion differs from the West African representatives of

Lecomtedoxa, in that the lateral appendages are small and the fruit is only slightly

attenuate near the base (in the W. African species the lateral appendages are larger

than the corolla-lobes, and the fruits are markedly attenuate at the base). In all other

essential characters, the South African species agrees very well with the West African

ones, such as the structure of the calyx, the staminodes, the attachment of the ovules,

the one-seeded fruit, and the seeds with long ventral scar and without (or with very

scanty) endosperm. I see no reason to create a separate genus nor even a subgenus

or section for the South African species, because the differences are only relative ones

and, in my opinion, not at all important.

347

Lecomtedoxa is a remarkable genus in that it combines the possession of lateral

appendages with (usually) 5-merous flowers, a monoseriate calyx (which, however,

tends to be biseriate) and seeds with long scar (ovules with lateral attachment), and it

forms a veritable link between the subfamily Mimusopoidea H. J. Lam (with biseriate

calyces and 3-merous or 4-merous flowers) and the subfamily Sideroxyloideae H. J.

Lam (with usually 4-merous flowers, and monoseriate calyx); especially the tribe

Pouterieae H. J. Lam (which also has long seed-scars). The only other genera in which

5-merous flowers are found, combined with the presence of lateral appendages and

monoseriate calyx, are the American genera Bumelia and Dipholis, of the Sideroxy-

loideae-Bumelieae. These two genera show great affinities to the genus Sideroxylon

and certainly deserve their place in the Sideroxyloideae. Lecomtedoxa on the other hand,

shows distinct affinities with the Mimusopoideae (subbiseriate calyx) and should, in

my opinion be retained in this subfamily as was done by Dubard and by Lam.

In a note on Ward 3032 the collector mentions that the young growths (young

leaves) are bright-red to red-brown which makes the trees quite conspicuous in spring.

6. AUSTROMIMUSOPS

A. Meeuse, gen. nov. Inhambanella Dubard, in Ann. Mus. Col. Marseille 23: 42

(1915), pro parte, non Mimusops Subgenus Quaternaria sectio Inhambanella

Engl, in Mon. Sapot. Afr. 80 (1904). Mimusops Auct., pro parte.

Arbores vel frutices. Folia chartacea vel subcoriacea, conspicue ad apices ramu-

lorum conferta; foliorum nervi secundarii et tertiarii atque venae tenuae, dense sub-

tiliter reticulates. Flores 1-4 in axilis ad apices ramulorum (3-) 4 meri. Sepala (3

+ 3 vel) 4+4, rarissime 5 + 5, elongata-triangularia vel lineari-oblonga vel lineari-

lanceolata exterioribus et interioribus subaequilongis extus tomentosis, acutis vel

interioribus obtusis, interioribus pallidioribus. Corollae tubus brevis; segmenta

(6-) 8 appendiculis binis integris instructa. Stamina fertilia (6-) 8, staminus filamenta

antheris breviora; antherae oblongo-lanceolatae vel lanceolatae connectivo apiculato.

Staminodia elongata-triangularia vel lanceolata, Integra vel interdum ad apicem plus

minusve dentata, extus pilosa. Ovarium ovoideum vel subglobosum, hirsutum (6-) 8

loculare; ovula ad medium loculorum vel basi-laterale affixa; stylus exsertus, cylindra-

tus vel subulatus vel filiformis, glabrus. Bacca ovoidea vel ellipsoidea, apiculata,

monosperma raro 2-sperma. Semen ellipsoideum; area derasa (= cicatrix) magna,

lata, semen longitudine subaequalis; testa Crustacea vel plus minusve pergamacea;

albumen nullum; cotyledones plano-convexae, crassae.

Type species: Mimusops marginata N. E. Br. (= M. natalensis Schinz non Engl.)

= Austromimusops marginata (N.E. Br.) A. Meeuse.

Large shrubs or small to medium-sized trees with the leaves crowded at the very

tips of the, often thick, branches. Leaves exstipulate, firm, but usually not coriaceous,

not very shiny above, with a very fine tessellated reticular nervation which is always

conspicuous at least on one side. Petioles distinctly, and usually widely, canaliculate

above. Flowers in the axils of the leaves or of scaly bracts, 1-4 together, but as the

leaves are crowded at the tips of the branches, apparently forming dense umbels of up

to 20 flowers and over. Pedicels usually more or less pendulous. Calyx biseriate, the

lobes free nearly to the base, (3 + 3 or) 4 + 4, very rarely 5 + 5, 3- and 4-, or 4- and

5- merous calyces occuring in one specimen; outer calyx lobes thicker and usually

broader than the inner ones, which also differ in pubescence and are more or less

distinctly midribbed. Corolla 3-merous and 4-merous (also occasionally 5-merous?)

in one specimen, or 4- merous; the tube very short; the lobes and lateral appendages

subequal. Stamens (6 or) 8, resembling those of Mimusops, i.e., the anthers longer

348

than the filaments, apiculate. Staminodes as in Mimusops, i.e., not deeply incised,

lobed or fimbriate, usually quite entire, concave, hairy outside. Ovary 6- or 8- locu-

lated, usually subglobose; ovules with lateral or sometimes basi-lateral attachment;

style terete, subacute or subtruncate at the apex. Fruit one-seeded, rarely 2-seeded.

Seed with rather thin, crustaceous or tough almost pergamaceous testa and a large

broad scar occupying nearly the whole ventral side of the seed; endosperm absent;

cotyledons thick and fleshy.

The genus contains at least four species, three found in Southern Africa, and one

in tropical East Africa.

Engler, l.c., based his section Inhambanella of Mimusops on a specimen without

flowers, mainly on the characters of the seed. Dubard, l.c., noticed that the ovules of

the species Mimusops natalensis Schinz non Engl. (= Mimusops marginata N.E. Br.)

are laterally attached and concluded that these ovules would give rise to seeds with a

long lateral scar, so that this species could not be a true Mimusops. Although Dubard

did not study the type specimen of Mimusops henriquesii (the type of Engler’s section

Inhambanella), he referred Mimusops natalensis Schinz to Inhambanella, which he raised

to generic rank. This was a conjecture, because the flowers of M. henriquesii were

unknown and Dubard had not seen the seed of M. natalensis Schinz.

However, Mimusops henriquesii is, in my opinion, a species of Lecomtedoxa (see

p. 344) and therefore, the name Inhambanella either as a subgenus or as a genus being

typified by Mimusops henriquesii Engl, et Warb., becomes a synonym of the genus

Lecomtedoxa (Engl.) Dub. Dubard was, to my mind, quite right in excluding M.

natalensis Schinz from Mimusops, but the name Inhambanella Dub. cannot be used for

it and the congeneric forms, included here in Austromimusops. The affinities of Austro-

mimusops are most probably with Baillonella Pierre which it resembles in floral charac-

ters, which are also very similar to those of Mimusops, but from which it differs in the

absence of endosperm {Baillonella has a thin layer of endosperm), and the thin crusta-

ceous to pergamaceus testa (thick and bony in Baillonella), and with the more or less

dubious genera Dumoria Chev. and Tieghemella Pierre (= ‘IDumoria).

Although the floral characters are very similar to those of Mimusops, there are so

many differences that the generic distinction is not very difficult. The vegetative charac-

ters alone are almost sufficient to typify Austromimusops, apart from the characters of

the seed. The following table (see Table II) shows the differences between the genera

Mimusops S.S., Austromimusops, Baillonella and Manilkara. Dumoria and Tieghemella

are not included, because it is very likely that they are identical with Baillonella, at any

rate the characters of fruit and seed are apparently the same as those of Baillonella.

Leaves usually more than 6 cm. long, petiole usually over 10 mm. long. Pedicels 2-5 cm.

long 1. A. marginata.

Leaves (at least the majority) under 6 cm. long, petiole 3-8 mm.; rarely up to 10 mm. long.

Pedicels 0-9-2 cm. long:

Young parts, pedicels and calyces buffy-brown pubescent. Leaves f-2 (-2f) cm. wide,

soon quite glabrous. Petioles soon glabrous. Pedicels 9-16 mm. long. Calyx-

lobes 5-6 mm. long. Natal 2. A. dispar.

Young parts, pedicels and calyces rusty-pubescent. Leaves If-Bf (-4f) cm. wide, often

showing vestiges of the rusty brown pubescence. Petiole often rusty-tomentose.

Pedicels usually Ij— 2 cm. long. Calyx-lobes + 7 mm. long. E. Southern

Rhodesia 3. A. sylvestris.

1. A. marginata (N.E. Br.) A. Meeuse, comb. nov.

Mimusops marginata N.E. Br. in Kew Bull. 108 (1895); Engl., Mon. Sapot. Afr. 71

(1904); Wright in Dyer, El. Cap. 4, 1 : 441 (1906); Sim, For. FI. Cape Col. 254,

pi. 97, Fig. 1 (1907); Gerstner in J. S. Afr. Bot. 12; 54, Figs. 8, 9 (1946); type:

Flanagan 27, K, lecto., BOL, GRA, NBG, PRE. isos! M. natalensis Schinz in Bull.

349-350

Table 11.

6096259

351

Herb. Boiss. 4: 441 (1896); type; Schlechter 6220 in Z. “ M. transvaalensis

Schinz ” (sphalm.), Radik, in Zahlbr., PI. Penther. I, in Ann. K.k. Naturh. Mus.

Wien 15: 63 (1900). M. schinzii Engl. op. cit., 70, t. 29, Fig. A; Wright, op. cit.,

443, Gerstner, l.c., 54 (same type as M. natalensis Schinz).

Inhambanella natalensis (Schinz) Dubard in Ann. Mus. Col. Marseille 23; 42 (1915).

A tree found in rather moist forests. Stem straight, 6-20 m. high and 30-60 cm.

in diam. Branches terete, grey, more or less rough; ultimate branches short and rather

stout, usually over 3 mm. thick, often much thicker, glabrous. Innovations densely

rusty tomentose-villous, but all vegetative parts soon quite glabrous. Leaves obovate

or elliptic-obovate, sometimes elliptic-oblong or elliptic-(ob)lanceolate, thinly coria-

ceous with a dull shine (but not smooth) and drying a greyish green (rarely brown) above,

paler and duller below, 3-15 (usually 6-13) cm. by 2-9 cm., as a rule more or less acumi-

nate with obtuse apex, a narrowed or somewhat rounded base and subreflexed margin;

midrib distinct but not conspicuously keeled or channelled above, prominent below;

lateral nerves slender but prominent beneath; ultimate nervation very fine, reticulate,

usually conspicuous on at least one surface. Petioles (5-) 10-20 mm. long, rather stout,

semi-terete, strongly and widely canaliculate. Flowers 1-3 in axils of leaves and of

scales inside them, more or less pendulous on 2-5 cm. long pedicels, 3-merous or 4-

merous. Calyx biseriate, outer lobes rusty-pubescent, inner ones pale-pubescent,

all acuminate, acute, 7-5-12 mm. long. Corolla dull white; tube 1-2 mm. long, pubes-

cent outside, lobes and appendages subequal, 6-9 • 5 mm. long. Staminodes 6-8, densely

villous outside, 4 -5-5 -5 mm. long. Stamens 6-8 mm. long. Ovary 6- of 8- celled;

style 9-1 1 mm. long. Fruit rather large, ovoid or ellipsoid, apiculate or attenuate-

apiculate, pointed, ultimately glabrous, purplish red, up to 5 cm. long and 3^ cm. dia-

meter, on the slightly incrassate, but not lengthened pedicel; fruiting calyx spreading-

reflexed. Seed broadly ellipsoid, 20-25 mm. long, about 20 mm. wide and about 18 mm.

thick; testa crustaceous, dull, pale buff when dry; scar somewhat shorter than the

seed, occyping about half its surface area, elliptic or oblong in outline, emarginate at the

apex and about 20 mm. broad in the widest place.

N. E. Brown did not designate a type specimen but Mr. de Winter informed me

that the only specimen at Kew with fruits is Flanagan 27, so that most probably the

fruits were described from this specimen. Accordingly, I propose Flanagan 27 (in K)

as the lecto-type. Type locality: Komgha, E. Cape.

Distribution.-^From East London northwards into Natal, Zululand and Portu-

guese East Africa, just crossing the Transvaal border.

Cape Province. — East London: Sim 2182, 2183, 2190 (= 72194 collect. No.)

(NU), 2194 (BOL), 2602, s.n. (PRE); Acocks 10979, 12298 (PRE), Courtney Latimer

s.n. (PRE), Rattray 1371 (BOL). Komgha: Gwenkala, Flanagan 27 (PRE, GRA,

BOL, SAM, isotypes); Schlechter 6220 (GRA, isotype of Mimusops natalensis Schinz

= M. schinzii Engl.). Kentani; Pegler 692 (PRE, BOL, NBG).

Natal. — Port Shepstone: near Mehlomnyama, Marais 7871 (PRE). Umzinto:

Dumisa, Rudatis 450 (L). Pinetown: Dellville, Smuts s.n. (NH No. 17830); Warner

Beach, Ward 977 (NU, PRE); Amanzimtoti, Williams 65 (NU); Marianhill, Forbes

1041 (NH); Umlaas, Wood 5440 (NH); Kotze 436 (PRE) = FD Herb. No. 6858

(SAFD). Durban: Inanda, Wood 1661 (NH). Camperdown: Hammarsdale, Forbes

310 (NH). Empangeni: Utimona, Gerstner 2748 (BOL, NH, PRE). Nongoma:

Wendelane Kloof, Gerstner 4657 (NH, PRE, BOL, NBG). Hlabisa: Gerstner 3817

(NH); Hluhluwe Game Reserve, Ward 1599. Ngotshe: Ngome Bush. Gerstner

2591 (NH, BOL). Ingwavuma: Cecil Mack’s Pass, about 8 m. N. of Ingwavuma,

Acocks 13129 (PRE); Codd 2074 (PRE). Without precise locality: “Zululand”,

Gerstner 2820 (BOL).

352

Transvaal. — Nelspruit: Kruger National Park, Crocodile River Poort, van der

Sc/iijff 3960 (PRE).

2. A. dispar {N.E. Br.) A. Meeuse, comb. nov.

Mimusops dispar N.E. Br. in Kew Bull. 1895: 107; Engl., Mon. Sapot. Afr. 71 (1904);

Wright in Dyer, FI. Cap. 4, 1 : 443 (1906); type: Thresh in herb. Wood No. 5425

from Natal, K, lecto., GRA, NH, isos,!

A large shrub or small tree, up to about 10 m. high. Branches light grey, terete,

usually rather rough and the younger ones marked in stretches with the scars of fallen

leaves of a previous generation, but these stretches are not so sharply defined as in

A. sylvestris (see No. 3); ultimate branches short to very short (unbranched twigs

usually less than 10 cm. and often less than 5 cm. long). Innovations fulvo-sericeous,

but leaves, petioles and twigs soon glabrescent. Leaves oblanceolate-cuneate, or oblan-

ceolate-obovate to obovate-oblong, firm but not coriaceous, rather dull and drying

green or yellowish-green above, paler and duller beneath, 2-6 (-7) cm. long and 0-75-2

(-2-5) cm. wide, on a 3-8 (-10) mm. long, subterete, widely canaliculate petiole; blade

with a very slightly reflexed margin, obtuse, subacute or shortly and bluntly acuminate,

but not usually rounded at the top, gradually narrowing into the acute or acuminate-

decurrent base; midrib hardly prominent above, rather prominent below; secondary

nerves 6-9 on either side, not very distinct from the tertiary ones, ascending at an angle

of 45°-60°, rather straight, bifurcate or branched well within the margin and more or

less irregularly archingly joining; tertiary nerves mainly parallel to the secondary

ones, but branching and anastomosing to form a rather coarse reticulum which is filled

up by a very fine areolate nervation, the latter conspicuous on both sides of the leaf.

Flowers few to many (over 20) on one twig, bracts minute; pedicels rather slender,

more or less angular, especially in the slightly thickened part under the calyx, 9-16 mm.

long, bulfy-brown tom.entose. Sepals 3 + 3 or 4 + 4; the outer ones ovate-triangular,

acute, or subacute, buffy-brown-tomentose outside, with a whitish margin, greyish-

tomentose inside near the tip, 5-6 mm. long and about 3 mm. wide; the inner ones

about as long and as wide as the outer ones, thinner in texture, ovate-oblong, more

obtuse, greyish-pubescent outside with a slightly darker longitudinal streak in the

middle and inside at least in the upper half. Corolla yellowish, glabrous; the tube

very short, about 0-5 mm. long; the lobes and lateral appendages subequal, linear-

lanceolate, acute, orsubobtuse, 5-6 mm. long and 1-1-5 mm. wide. Stamens 6 or 8;

filaments 1 - 5-2 mm. long, subulate from a broad flattened base, the apical part capillary;

anthers ovate-lanceolate, acute, 2-3 mm. long, minutely apiculate. Staminodes ovate-

lanceolate, long-acuminate, concave, hairy at the back, about 3 mm. long. Ovary

subglobose, densely villous, about 2 mm. in diameter; style glabrous, rather short and

comparatively thick, terete, subulate-columnar, 2-3 mm. long, the apex subacute.

Fig. 9. — Austromimusops marginata, seed and fruit (from Pegler 692, Kentani, E.

Cape, in BOL).

Fig. 10. — Austromimusops dispar, seed and fruit (from Pentz et ah “ 1 A”, Weenen

Veld Reserve, Natal).

PiG_ ] 1. — Austromimusops sylvestris: (a) immature fruit (to show the dense pubescence;

from Chase 4328, Umtali, S. Rh.) (b) mature seed and fruit (from seeds sent by

Mr. N. C. Chase, coll, in Waranki Reserve, Umtali, S.R.)

JV.B. — The seed figured was from a large 2-seeded fruit; as a rule the seeds are

shorter and comparatively thicker.

353

354

Fruit green or yellowish-green, ellipsoid or ovoid, apiculate, more or less densely

covered with a short brown pubescence, 2-3 cm. long, 1-5-2 (-2-5) cm. in diam. ;

pericarp thin, almost leathery when fresh, 1- or rarely 2-seeded. Seed pale brown,

2-2-5 cm. long and 1-5-2 cm. in diam. with pale whitish scar.

Natal. — “ Thorns” nr. Greytown, Thresh in herb. Wood 5425 = NH No. 7175

(NH, GRA, isos! ). Estcourt: Mooi Rivier, “ Thorns ”, JTocxf 4472 (NH); 5//n 2187,

2188 (NU). “Upper Tugela River:” Gerrard (and McKen) 1482 (NH). Weenen:

Weenen Veld Reserve, Acocks 10148 (PRE, NH), Pentz & Acocks = Acocks 10721

(PRE, NH), Pentz 116, 596; “ 1 ” (Coll, in 1952), “2” (coll, in 1952). “ lA ” and

“ IB ” (coll, in 1953 from same trees as “ 1 ” and “ 2 ” respectively); all in PRE, the

last three fruiting specimens); West 1189, 1190 (PRE); Muden, Sim 19078, 19138

(PRE), Verdoorn 1727 (PRE). Msinga: Confluence of Mooi and Tugela Rivers,

Ngobevu, Edwards 881 (NU, PRE). Umvoti: Keats Drift, Edwards 916 (NU, PRE).

3. A. sylvestris (S. Moore) A. Meeuse, comb. nov.

jV///nM5’op5 5>7ve5/ra S. Moore in J. Linn. Soc. (Bot.) 40: 132 (1911); type: Swynnerton

570 from Southern Rhodesia in BM, holo.

A shrub or small tree, up to about 7 m. high, with smooth bole. Branches terete,

greyish, with longitudinal shallow grooves and with slightly thicker stretches marked

with rough transverse leaf-scars, especially towards the tips of the twigs below the leaves;

ultimate branches rather short (5-15 cm. long) and rather stout (more than 2 mm.

thick, sometimes much thicker and up to 6 mm. in diam.), usually with a thicker scarred

area just below the leaves. Buds, young leaves and young twigs, pedicels and bracts

densely rusty-tomentose. Leaves obovate or obovate-oblong, sometimes oblong or

oblong-cuneate, 2-5-6 (-7-5) cm. long and 1 -5-3-5 (-4-5) cm. wide, on a rather stout,

terete, rather widely canaliculate, rusty-pubescent but ultimately glabrescent, 2-6

mm. long petiole; blade with revolute margin, an obtuse, rounded, emarginate, or more

or less retuse, rarely very shortly and bluntly acuminate apex and a narrowed, usually

acute but sometimes rounded base, rather firm but not coriaceous in texture, ultimately

glabrous and green but not smooth or shiny above, glabrescent but not soon quite

glabrous, paler and duller below, midrib usually keeled, sometimes immersed above,

rather prominent beneath and keeping vestiges of the original rusty pubescence for a

long time; secondary nerves (5-) 7-10 (-12) on either side, as a rule very distinct below,

immersed above, rather straight, making an angle of about 60° with the midrib, usually

bifurcate well within the margin and more or irregularly joining, but mostly with one

distinct ascending branch; tertiary nerves mainly parallel to the secondary ones, but

usually sinuous, branched or more or less reticulate; ultimate reticulate nervation

very fine, areolate, in dried leaves usually conspicuous on both sides. Flowers 1-3

together; bracts densely rusty-tomentose, ovate, 2 mm. long, deciduous; pedicels

usually 1 -5-2 cm. long, rather stout, terete, but gradually widening near the apex into

the calyx, usually pendulous; in fruit not lengthened but slightly incrassate. Sepals

4 + 4 or 5 + 5; the outer ones ovate-lanceolate to oblong-lanceolate, about 7 mm.

long and 3-3 - 5 mm. wide, acuminate or narrowed towards the subobtuse apex, densely

rusty-tomentose with a distinct greyish-tomentose margin outside, greyish-tomentose

inside at least in the upper half ; the inner ones somewhat shorter and narrower, greyish

pubescent outside with a darker longitudinal streak and inside at least in the upper half.

Corolla glabrous; the tube about 1 mm. long; the lobes about 7 mm. long, oblong-

lanceolate; the appendages slightly shorter. Filaments more or less flattened, rather

slender, 1-5 mm. long; anthers oblong, obtuse, apiculate, 3 mm. long. Staminodes

long-triangular-lanceolate, concave, with acuminate, dissected, 2-3 fid or somewhat

fimbriate apex, villous with long sinuous hairs outside, the upper half strongly indexed.

Ovary sub-globose, about 2 mm. in diam, densely villous; style glabrous, 6-7 mm.

long, terete, tapering towards the apex, somewhat truncate-capitate at the tip. Fruit

355

ovoid-ellipsoid, attenuate-apiculate, 3-4-5 cm. long and 2-3 cm. in diam., when young

densely rusty-tomentose, glabrescent and dull brown when mature, 1 -seeded or 2-seeded;

pericarp not juicy, rather dry and leathery. Seed broadly ellipsoid to subglobose

(those of 2-seeded fruits ellipsoid, more slender), 20-32 mm. long, ± 20 mm. in diam.;

testa dull, very pale straw-coloured; the scar duller and somewhat lighter in colour

(see Fig. 13).

S. Rhodesia. — Umtali; Dora Farm, Chase 964, 965, 966, 774 (SRGH, Chase 965

also in PRE); Glenshiel Farm, Chase 1670 (SRGH); Zimunya Reserve, Chase 4238

(SRGH, PRE); Maranki Reserve: Chase s.n. (ripe fruits, PRE). Bikita: Wild Al>93>

(PRE, SRGH).

Mr. B. de Winter compared the type, Swynnerlon 570, with Chase 966 (Kew)

and informed me that there are some minor differences such as shorter and more hairy

petioles, smaller flowers, wrinkled leaves. However, the material 1 have seen is rather

variable and the differences mentioned by Mr. de Winter are well within the range of

variation, so that I have no doubt about the specific identity.

Apart from the three species dealt with, there is at least one species in tropical

Africa. This species, described as Mimusops ciineata Engl., has several characteristics

of an Austromimusops : leaves congested at the ends of the branchlets, with their largest

width above the middle, axillary flowers on pendulous pedicels and laterally attached

ovules. Judging by Engler’s figure and a specimen named Mimusops cuneata (Drum-

mond & Hemsley 4203 from Kenya, Kwale Distr., EA, K, PRE) this is very close to

Austromimusops dispar (but it is much more glabrous than the latter) and to A. marginata,

and although the fruits are apparently still unknown, I do not hesitate to refer it to the

same genus:

Austromimusops cuneata (Engl.) A. Meeuse, comb. nov.

Mimusops cuneata Engl., Pflanzenw. O. Afr., C, 307 (1895), and Mon. Sapot. Afr. 70,

t. 23, C (1904); Brenan and Greenway, Checklist Tang. Terr. 2: 565 (1949).

Recorded from Tanganjdka (West-Usambara) by Engler; also in Kenya.

7. MIMUSOPS

L., Gen. PI. ed. 5: 175 (1754), p.p.; Harvey, Gen. S. Afr. PI. 224 (1838), A. DC. in DC.,

Prodr. 8: 202 (1844), p.p.; Bentham et Hook., Gen. PI. 2: 661 ( 1876), p.p. ; Baker

in Oliv., FI. Trop. Afr. 3: 505 (1877), p.p.; Hartog in J. Bot. 17: 358 (1879),

p.p.; Engler in Engl. & Prantl., Naturl. Pflanzenfam., ed. 1, 2, 4: 150 (1899),

and in Nachtrage (1897) 278, pro parte; Engl., Mon Sapot. Afr. 50 (1904), p.p.;

Pilger in Engl, & Prantl, Nachtrage 1897-1904, 288 (1906) p.p.; Wright in Dyer,

FI. Cap. 4, 1: 439 (1906), p.p.; Dubard in Ann. Mus. Col. Marseille 23: 46

(1915), Baehni in Candollea 7: 465 (1938); Lam in Blumea 4: 345-347 (1941);

Phillips, Gen. S. Afr. FI. PI. ed. 2, 568 (1951), p.p.; Royen in Blumea 6: 594

(1952).

Type species: Mimusops elengi L., Sp. PI. ed. 1 : 349 (1953). Linnaeus mentioned

two species, M. elengi and M. kauki. Mimusops kauki L. is now generally assumed

to be the type species of Manilkara Adans. [see discussion by Van Royen in Blumea

7 : 406 (1953)], which makes M. elengi L. undoubtedly the type species of Mimusops L,

apart from priority of place.

Trees or shrubs. Leaves extipulate, not conspicuously crowded towards the tips

of the branches and without sclereids and with a rather lax, not conspicuously parallel

or “ striate ” secondary nervation. Flowers axillary, pedicellate, constantly 4-merous.,

Calyx biseriate (4 -f 4); the sepals free nearly to the base, often long and narrow.

356

Corolla 8-lobed, each lobe with 2 dorsal appendages; the latter entire or divided (in

S. Africa always entire). Stamens 8, inserted in the throat of the corolla tube; the

anthers usually (in S. Africa always) shorter than the long, more or less lanceolats and

apiculate anthers. Alternipetalous staminodes 8, entire, or somewhat dentate, lacerate

or fimbriate at the apex only, often pubescent at least at the back or along the edges

(in S. Africa never glabrous), incurved, and more or less covering the style. Ovary

8-loculated; ovules with basal attachment; style rather long and slender, cylindric-

subulate to filiform. Fruit a 1- to few-seeded berry. Seeds with a small, circular and

almost basal scar; testa shiny hard and thick; endosperm copious, cotyledons thin,

foliaceous.

The genus Mimusops as understood by Bentham and Hooker, Hartog, Engler and

several others is most heterogeneous and comprises forms with 3-merous and others

with 4-merous flowers, species with long and narrow ventral seed-scar and others with

circular basal scar, plants with staminodes and others without staminodes, forms with

endosperm and others without endosperm, etc. Dubard (1915) excluded many species

and referred them to several other genera, the majority to ManUkara Adans. The latter

genus is now almost universally adopted, cf. Van Royen in Blumea 7: 401 (1953),

but is should be borne in mind that the large genus Mimusops sensu Hartog and Engler

does not become more homogeneous if only ManUkara is taken out. If, however,

Muriea, Austromimusops, Baillonella, Lecomtedoxa and perhaps several other smaller

genera are also exluded, the remainder of Mimusops is a reasonably well defined homo-

geneous genus. In its limited sense, Mimusops L. sensu Dubard, Lam and Van Royen

comprises between 30 and 40 species, all occuring in Africa, Madagascar, and the

Mascarenes, with the exception of M. elengi L., a coastal form which occurs in tropical

Asia and the W. Pacific. Three species in Southern Africa:

Leaves obovate, obovate-cuneate or almost obcordate, always distinctly narrowing towards

the base, usually emarginate or retuse, with strongly revolute edges and at least when

young, with white sometimes fulvous adpressed, more or less silky pubescence below

(very old leaves glabrous). Coastal tree M. caffra.

Leaves different, glabrous when old or with “ powdery ” vestiges of a rusty brown pubescence:

Petioles under 1 cm. long; leaves usually rather small (mostly under 6 cm. long) and

usually drying dark brown; branches and leaves very soon glabrous; flowers

about 2 cm. in diameter when fully expanded, very rarely smaller, usually solitary

in the axles, sometimes 2 together, usually not numerous on a single branch-

let 2. M. ohovata.

Petioles over 1 cm. long; leaves often over 6 cm. long, innovations and young leaves, as

well as tips of young branches densely rusty-pubescent, the tips of the branches

often remaining pubescent for a considerable time, flowers up to 1 • 5 cm. in

diam, when fully expanded, often in clusters of more than 2 flowers, often

numerous on a single branchlet 3. M. zeyheri.

1. M. caflFra E. Mey. ex A. DC., Prodr. 8; 203 (March 1844); Wood, Natal PI. 1 : 36,

t. 43 (1898); Engl., Mon. Sapot. Afr. 72, t. 27, Fig. B (1904); Wright in Dyer,

FI. Cap. 4, 1: 441 (1906), Sim, For. FI. Cape Col. 255, fol. 97 (1907); and For.

FI. Port. E. Afr. 80, pi. 75 (1909); Gerstner in J. S. Afr. Bot. 12: 52, Fig. 7 (1946).

M. caffra E. Mey. ex Drege, Zwei Pflanzeng. Doc. 155 (1843), nomen tantum;

type: Drege s.n. from Pondoland in G ex herb. DC, holo, L, iso! M. revoluta

Hochst. apud Krauss in Flora 27: 825 (Dec. 1844); type: Krauss 76 from Durban

in ?, holo, K. iso .

A small tree or large shrub, forming a large propertion of the vegetation on the

sand dunes, growing down to the high-water mark and fully exposed to sea winds and

spray; exposed specimens usually dwarfed and gnarled, but in sheltered places growing

up to about 10 m. high and over 50 cm. stem diameter. Branches terete, usually rather

stout (about 3 mm. thick), densely leafy towards the tips. Innovations densely rusty-

tomentose. Leaves obovate, obovate-oblong or obovate-cuneate to almost obcordate.

357

3- 6 (-7) cm. long and 1 ■ 5-3 (-4) cm. wide; blade firmly coriaceous usually with strongly

reflexed margin, glaucous and glabrous above, paler and adpressed silky-pubescent

beneath (almost invariably white or silvery), with usually rounded emarginate to retuse

apex, tapering into the 5-10 (-15) mm. long petiole; midrib usually a little prominent

above in the lower half of the leaf, or flush, rarely somewhat immersed, prominent

beneath at least in the lower half; secondary nerves almost straight, slightly prominent

on both sides, forming an angle of 30°-50'’ with the midrib, joining a conspicuous,

sinuous intramarginal vein; tertiary nerves and finer nervation hardly more slender

than the secondary nerves, and mainly parallel to the latter. Petioles terete, slightly

thickened towards the base, narrowly but rather deeply channelled above, ultimately

glabrous, 5-10 (-15) mm. long. Flowers usually numerous on one branchlet, often

2-4 together in the leaf axils. Pedicels usually recurved, 2-3 cm. long, mostly distinctly

4- angled, shortly rusty-tomentose (this pubescence tending to become grey later),

more or less gradually widening into the calyx. Flower-buds about 12 mm. long just

before opening. Sepals lanceolate, acuminate, about 1 cm. long; the outer ones rusty-

tomentose outside, about 3 mm. wide, the inner ones with pale tomentum, about 2

mm. wide. Corolla about as long as the calyx, glabrous; the tube short, the lobes

about 10 mm. long, each with two about 7 mm. long appendages, all segments lanceolate,

usually in two rows consisting of an outer row of 16 appendages and an inner row of

8 corolla lobes. Filaments about 2-5 mm. long; anthers about 6 mm. with acute

apiculum. Staminodes triangular-ovate, slightly longer than the filaments, densely

pilose outside with long hairs. Ovary ovoid, densely pilose, about 2 mm. long; style

long-cylindrical, tapering into an acute point, about 11 mm. long. Fruit ovoid, 1-5-2

cm. long, 1-1 -5 cm. diam.. more or less rounded at the top but often contracted into

and crowned by the persistent style, red when ripe, edible, usually (always?) 1-seeded;

fruiting pedicels hardly lengthened, slightly incrassate, up to 3 cm. long and about

2 mm. thick; calyx-lobes persistent under fruit, adpressed to the fruit, greyish pubescent.

Seed oval, subcompressed, indistinctly keeled at the ventral side, not produced at the

base, 13-17 mm. long, 8-9 mm. wide and 5-7 mm. thick in the middle; testa shining

brown.

Selected Citations.

Cape Province. — Bathurst: Kowie and Port Alfred, Burchell 3805; Britten 2107,

Mor/ot/z 18097; Burt- Davy 1S56; Tyson s.n.; Story 2\63. East London: Galpin IS35,

9285; Smith 3%\1 . Komgha: Kei Mouth, West 2024. Kentani: Pegler \29%. Elliot-

dale: O. B. Miller ED herb. No. 5591 (SAED). Port St. Johns: Schonland 4038,

Mogg 770 (= photo of tree, PRE); Howlet 44.

Cape Province or Natal. — Bizana or Port Shepstone: “between Umtentu and

Umsamkulu Drege (L, isotype ).

Natal. — Port Shepstone: Margate, Bayer 1305 (NU); Port Shepstone, Mogg

13197. Umzinto: “South Coast, nr. Botha House”, Smuts 2325; Sezela, Smuts

s.n. Pine Town: Winkle Spruit, Rudatis 1608; Amanzimtoti, Kotze 453. Durban:

near Durban, Forbes & Obermeyer 72. Lower Tugela: Stanger, Pentz 386. Mtunzini:

Inyoni bush, Gerstner 1936. Hlabisa: St. Lucia Bay, Lansdell 31 (NH). Ubombo:

near Sordwana Bay, Gerstner 733; Ward 3013.

Portuguese E. Africa. — Sul do Save: Lourengo Marques, Schlechter 11986;

Rogers 2\314\ Junod 134; Borle 30; Pedro 59; Gomes e Sousa 3110; Maputo, Hornby

2619; Inhaka Island, Mogg s.n.; Chongoene, Pedro & Pedrogdo 1645. Manica e

Sofala: Inhambane, Earthy 172 (PRE); Gomes e Sousa 1701, 1739, 1908 (COI).

As regards the synonymy, Mimusops cajfra was the only name used for this species

in all recent publications, but Krauss in 1844 validly published Mimusops revoluta

Hochst., based on Krauss 76 from Durban. Strangely enough, this name was not

358

mentioned by Engler (op. cit.) as a synonymn of M. cajfra and is not cited in FI. Cap.,

although the number Krauss 76 is quoted in the latter publication. Mr. de Winter

kindly supplied the information that an isotype is present in the Kew herbarium and

that it is identical with M. cajfra. Fortunately, the publication of M. cajfra E. Mey

ex A. DC, antedates that of M. revoluta Hochst. apud Krauss by several months (mid

March 1844 against December 1844) so that the well-known and generally used name

M. cajfra need not be changed.

M. caffra is found on coastal sand dunes from Port Alfred (Bathurst) eastwards and

northwards to Portuguese E. Africa, also here and there along large rivers more inland

on sandy soil.

2. Mimusops obovata Sond. in Linnaea 23: 17 (1850); Harvey, Thes. Cap. 1: 28,

t. 44 (1859); Engler, Mon. Sapot. Afr. 72, t. 27, Fig. D (1904); Sim, For. FI.

Cape Col. 254, pi. 96 (1907); Wright in Dyer, FI. Cap. 4, 1 : 442 (1906); Marloth,

FI. S. Afr. 3: 36, t. 10 (1932); Gerstner in J. S. Afr. Bot. 12: 54, Fig. 10 (1946);

type: Ecklon & Zeyher “ Sideroxylon No. 16” from Alexandria, lecto! in S,

isos! in GRA, PRE.

Imbricaria obovata N. ab. E. ms. ex Sonder, l.c. in syn., Engler, op. cit., 72, in syn.

Mimusops oleifolia N.E. Br. in Kew Bull. 1895: 109 Engler, op. cit., 73, t. 34, Fig. B;

Wright, op. cit., 442; type: Gerrard 1642 in K, holo., NH, iso! from “ Tugela ”,

Natal.

M. woodii Engl., op. cit., p. 65, t. 26, Fig. A; Wright, op. p. 440; type: Wood 683 in

B, holof, BOL, NBG, isos! from Inanda, near Durban.

M. rudatisU Engl, et Krause in Engl. Bot. Jb. 49 (1913), p. 395; type: Rudatis 1136 in

B, holof, L, PRE, isos! from Dumisa, Natal.

A medium-sized tree, up to 20 m. high and 60 cm. stem diam., according to Sim

(in MS.) occuring mainly in rather open mountain forests, but also in coastal areas.

Branchlets terete, glabrous, rather slender, usually longitudinally wrinkled and light

grey, as a rule uniformly leafy. Innovations rusty-tomentose, very soon glabrescent,

older parts ultimately quite glabrous. Leaves variable in size and shape, but usually

obovate, obovate-oblong or obovate-cuneate, sometimes more oval or elliptic or

obovate-oblanceolate, 2-6 (-7) cm. long and 1-3 (-4) cm. wide, rarely narrowly

lanceolate, 2-5-6 cm. long and 4-9 mm. wide, more or less thinly coriaceous,

usually drying very dark brown and shiny above, pale brown and dull beneath,

sometimes drying grey above; the apex obtuse or rarely subacute, sometimes

rounded, but mainly in the obovate-cuneate type of leaf) not infrequently with

a short, blunt acumen, tapering at the base, which is always acute or subacute,

and with minutely reflexed margins; midrib almost flush above, slightly prominent

beneath; secondary nerves ascending at an angle of ± 45°, rather straight, but

very few reaching the leaf margin without brancing, joining other veins or slightly

deflexing in the points where the tertiary nervations join them, archingly joining

near the margin; tertiary veins forming a rather fine distinct reticular nervation

which is usually very conspicuous in dried leaves and slightly prominent, at least on the

lower surface. Petioles comparatively slender, terete, channelled above near the leaf-

base, 3-9 mm. long. Flowers white, fragrant, solitary or in twos in the leaf axils on 1-3

(usually 1 • 5-2) cm. long, brownish-tomentose, slender and suberect or patent, usually

not distinctly drooping pedicels; flower buds usually ± 1 cm. long just before opening.

Outer sepals (6-) 8-12 mm. long and 3-3-5 mm. wide, shortly rusty-tomentose with

a very narrow whitish or pale-grey edge, narrowly ovate-lanceolate, acuminate; inner

sepals slightly shorter and narrower, with a pale grey or whitish tomentum and minutely

ciliate, soon reflexed. Corolla-tube short, 1 mm. long; the 8 lobes linear-lanceolate or

359

narrowly oblong, ± 1 cm. long, rarely smaller (6-8) mm., the lateral appendages about

as long, usually more acute than the lobes. Filaments thick, 1-2-5 mm. long, much

shorter than the sagitate-oblong, or linear oblong, apiculate, 3-6 mm. long anthers.

Staminodes long-triangular or lanceolate-subulate from a broad base, villous outside,

shorter than the stamens but longer than the filaments. Ovary ovoid or oblong-ovoid,

± 2 mm. high, densely pilose; style glabrous, terete, slender, longer than the corolla

and up to i 12 mm. long, slightly tapering towards the apex which is truncate and often

subcapitellate. Fruiting pedicels somewhat incrassate; calyx under fruit persistent,

reflexed or ultimately deciduous. Fruit ovoid or ovoid-acuminate, 2-3-5 cm. long,

1-2 cm. in diam., often 1-seeded, ultimately glabrous, smooth, orange-red or yellow

when ripe. Seed, when single, 2-2-5 cm. long, 8-10 wide and 6-8 mm. thick in the

centre, oblong, with rounded apex and obliquely truncate-notched base, often distinctly

keeled on the side above the small, circular, ± 2 mm. wide scar, but if more seeds are

present in one fruit, often smaller and i irregularly shaped, flattened and less dis-

tinctly truncate-notched or keeled; testa brown, shiny.

General Distribution. — From the Eastern Cape eastward and northwards into

Natal, Swaziland, Zululand, Eastern Transvaal and Portuguese East Africa (possibly

extending into tropical East Africa, because several species described from that area

appear very similar from the descriptions, but no actual specimens seen) in evergreen

forests in frost-free areas, mainly at low to fairly low altitudes.

Cape Province. — Alexandria: Olifantshoek, Ecklon & Zeyher (S, lecto., PRE, GRA

isos); Zeyher (PRE), Ecklon 8c Zeyher or Zeyher (SAM). Bathurst: Acocks 11139

(PRE); Kariega Mouth, Acocks 18348 (PRE); Port Alfred, Britten 1681 (GRA, PRE).

Albany: near Grahamstown, MacOwan 258 (GRA, HN, BOL, SAM); Galpin 179

(GRA, PRE); Beggar’s Bush Forest Reserve, Archibald 5966 (PRE); Blaauwkrantz,

Britten 897 (GRA, PRE); Paradise Kloof, Story 3128 (PRE). Stutterheim: Sim 2273

(or 21922?; NU); Acocks 8939; Story 1242. King William's Town: Pirie, Sim 1333

(NU); Ross s.n. (SAFD No. 1903). East London: Galpin 3164 (GRA, PRE), 9518,

10432 (PRE); 5m/7/? 3783, 3818 (PRE); 5/m 2189 pp. (NU). Komgha: Flanagan 249

(PRE, BOL, SAM). Kentani: Pegler 765 (PRE, BOL, GRA, NBG); Qolora, Edward

in h. Moss 17544 (J). Engcobo: Manina, Zahn 22, v.d. Merwe FD No. 2200 (SAFD).

Ngqeleni: Qokama, Acocks 13426 (PRE). Port St. Johns: near Port St. Johns, Moss

4686 (J); BoshojfFD No. 5032 (SAFD). “ Pondoland ”: Egossa and Port St. Johns,

Sim 2418 (NU, PRE). Lusikisiki: Acocks 13426 (PRE), Miller FD No. 5771 (SAFD).

Flagstaff : R. C. Coloured School s.n. (GRA).

Natal. — Port Shepstone: “ South Coast Natal ”, Pole Evans 761 (PRE); South-

broom, Codd 9705 (PRE). Umzinto: Dumisa, Rudatis 1136 (L, PRE, isos of M.

rudatisii Engl, et Krause). Umlazi: Park Rhynie, "'Indian collector'" s.n. (NH No.

14819); Winkle Spruit, Van der Bijl s.n. (NH No. 16136). Durban: Isipingo, Ward

342 (NU), 503 (NU, PRE); near Durban, Wood 5797 (BOL), 9112 (PRE); Gerrard &

McKen 720, 869 (NH); Rogers 24502 (PRE); Stayner 20 (BOL); Wylie s.n. (= 23114,

also PRE); Inanda, Wood 6S3 (BOL, NBG, isos of M. woodii Engl.); Umgeni Dam,

Bayer 1387 (NU, NH, PRE. BOL). Pine Town: Marian Hill, Forbes 1039 (NH).

Verulam: Umhloti Beach, Codd 1499 (PRE). Kranskop: Jameson’s Drift, Bayer

540 (NU). Msinga: Nogbeva. Edwards 878 (PRE, NU). Nkandla: Middeldrift,

Edwards 1423. 1424 (PRE, NU). Eshowe: Lawn 666 (NH); Gerstner 1922 (PRE),

2262, 2440 (NH, PRE), 2819 (BOL), 2994 (NH); Entumeni Waterfall, Wylie s.n. p.p.

(NH No. 12940, the remainder is M. cajfra). “Tugela”: Gerrard 1642 ( = iso. of

oleifolia N. E. Br., NH). Mtunzini: near Mandeni, Edwards 1616 (PRE, NU). Lower

Umfolozi; Umfolozi Game Reserve, Ward 3159 (PRE); Kluge 19 (NH); Mtonjaneni:

Gerstner 3675 (NH, PRE); Empangeni. Gerstner 2767 (NH), 2730 (NH, BOL, PRE).

Hlabisa: Hluhluwe, 1700 (PRE); Masimba Hill, 2078 (NH, PRE); Hlabisa

360

c/5

TD

<U

<u

OT)

cq

^ T3

lO P

<U

-Si t3

«

0, flj

O u

t3 ^

C c/5

c3 "O

<U

c/3 aj

s-

s I

Q «

S >-

2 o

H

s *

S i"

d

£

£ cd

^ ^***

D- O

d aj

O D,

i-a^-g

^ ^ ^ C3

O ^ (D

^ C ^ N

"S n* •-

> 2 CU

"S _ =

c ^ c ^

2 '.:

H p:s.g

- c .V n

^ >

^ cd ^ <-■

c/5 ^ ^

§ p "2 aj

o o Q-> ■4— •

-S' o c o

O c« iS

nS

O

go.

!5 oo

i: C

O C3

.2^

s 2 C

2 o

2 w

Ji

■a*

c aj

3 ' — '

K 3 r,

:^S o

'5 o

> a

dii

aj o

ss

o ^

-o

aj 0/ c^ \

2^ m

c B ^

— '' c5 o S

^ -S''^ ^

a/ 2! w

-05 c/;

■>3 ao ST "S

- 2

cd

1^ O 2

S d

I ^ r^'

d

[li

361

or Umbombo: near northern end of St. Lucia Lake, Gerstner 4987 (PRE). Nongoma:

Wendelane Kloof, Gerstner 4659 (PRE, NBG). Ngwavuma: Bayer 760 (PRE). Natal

without precise locality: Gueinzuis (original syntype gathering, presumably from the

vicinity of Durban, in SAM); Cooper 1241 (BOL, NH, PRE).

Swaziland. — Stewart s.n. (TRV No. 8824 in GRA, PRE); Ubombo Mts. near

Stegi, Keith s.n. (PRE).

Transvaal. — Sebasa : Phephidi River (prob. near Phephidi Falls), Legat 1 (SAFD).

Letaba: Patatabos, Botha s.n. = FD. No. 7154 (SAFD. PRE); The Downs, Benny

DB 52 (PRE); De Hoek, Keet FD No. 7413 (SAFD).

Portuguese East Africa. — Sul do Save: Lourengo Marques, Gomes e Sousa

3859 (COI, PRE).

N.B. — The specimen mentioned in FI. Cap. 4, 1 : 442 from Warmbaths, Transvaal

{Burtt Davy 2625) is not M. obovata Sond., but M. zeyheri Sond.

The shape of the leaves of one specimen is usually fairly uniform but the various

forms described as M. oleifolia N.E. Br., M. woodii Engl, and M. rudatisii Engl. & Krause,

are, in my opinion, only ecotypes, as was already suggested by Gerstner. The forest

form has rather thin leaves which are large and obovate, and corresponds with the

descriptions of M. obovata and of M. woodii \ a depauperate form growing in rocky

places in the sun, has small coriaceous leaves, flowers more freely and corresponds

with the narrow-leaved M. oleifolia. Not only are all these forms linked up in the

herbarium by intermediate specimens, but the flowers, branches, pedicels, fruits and

seeds are also identical.

A type specimen was not defined, because Sonder mentions the numbers Gueinzius

583 and 101 ex parte from Durban, and Ecklon & Zeyher’s “ Sideroxylon No. 16”

from Olifantshoek. The Ecklon & Zeyher collection being the best known and better

represented in various herbaria, I propose the E. & Z. specimen in herb. Sonder (S)

as the lectotype.

3. M. zeyheri Sond. in Linnaea 23: 44 (1850); Engl., Mon. Sapot. Afr. 73, t. 27, Fig. C

(1904), inch var. laurifolia Engl.; Wright in Dyer, FI. Cap. 4, 1: 441 (1906);

Gerstner in J. S. Afr. Bot. 12: 54 (1946); Meeuse in R. A. Dyer, FI. PI. Afr. 30:

t. 1164 (1954), type: Zeyher 1130 in herb. Sonder, holo., BOL, NBG, isos.! M.

kirkii Baker in Oliv., FI. Trop. Afr. 3: 567 (1877); Engl., op. cit. 67; Gerstner,

op. cit., 55, type: Kirk s.n. from Lower Shire Valley and Kirk s.n. from Rovuma

in K, syns. M. monroi Sp. Moore in J. Bot. 49: 55 (1911); type: Monro 761 in

BM, holo; BOL, SRGH, isos!

A tree reaching a height of at least 10-15 m., with spreading branches and a leafy,

umbrageous canopy, but often appearing as a large shrub. Innovations densely ad-

pressed rusty-tomentose, the twigs often retaining the pubescence for some time, the

leaves sometimes in the form or irregular, powdery patches or along the midrib;

branches, petioles and leaves ultimately glabrous. Leaves coriaceous or thinly cori-

aceous, shiny above, paler beneath, usually drying a pale greyish-green (young ones

drying dark brown above, brownish-grey beneath), ovate-lanceolate to oblong-ovovate,

sometimes obovate or broadly elliptic, sometimes broadly lanceolate, usually obtusely

acuminate or gradually tapering into the obtuse or more rarely slightly emarginate,

occasionally rounded apex, tapering, acute or subacute at the base, with slightly

thickened and subreflexed margin, 4-9 (-11) cm. long (on coppice shoots occasionally

up to 15 cm. long) and 2-4 (-5) cm. wide, on a (10-) 15-35 mm. long, terete, petiole

which is rather firm, narrowly canaliculate at least near the leaf-base and not infrequently

shows vestiges of the original reddish-brown tomentum; midrib forming a fine keel

above, prominent below, very gradually tapering towards the apex of the leaf ; lateral

362

nerves ascending at an angle of 45°-60°, usually not reaching the margin without

branching joining or deflexing. archingly joining near the edge of the leaf, not or hardly

distinct from the tertiary nerves, the latter first more or less parallel to secondary ones,

then becoming irregular, forming a rather coarse reticulate meshwork filled up by a

finer reticulate nervation; secondary and tertiary nerves usually conspicuous, slightly

raised, finer nervation often inconspicuous. Pedicels often numerous on the twigs in

fascicles of 3 or more in the leaf axils, sometimes only 1-3 per axil, 10-30 mm. (but almost

invariably between 15 and 25 mm.) long, rusty-pubescent, recurved, not lengthening

but slightly incrassate (up to i 1 • 5 mm. thick) in fruit. Flower-buds ]u?,i before opening

5-7 mm. long and 3-4 mm. thick. Sepals long-triangular to ovate-lanceolate, acute;

the outer ones rusty-pubescent often with a narrow pale edge, 5-6 (-7) mm. long and

2-3 mm. wide, the inner ones pale greyish-white tomentose, slightly shorter and narrower.

Coro//a-lobes about equalling the calyx, about 6 mm. long, linear-lanceolate, acute,

their appendages about or long as shorter (4-6 mm.), linear-lanceolate and acute.

Anthers elongate-sagittate, apiculate, about 3-5 mm. long on subulate, 1‘5 mm. long

filaments. Staminodes long-triangular or triangular-lanceolate, either shorter than the

stamens and acute, or longer than the stamens and long-acuminate to nearly aristate,

densely pilose outside. Ovary globose or ovoid, about 2-5 mm. long, villous, attenuated

into a 5-10 mm. long style, the latter glabrous, terete, longsubulate, tapering towards

the apex. Fruiting calyx not accrescent, somewhat spreading but not quite reflexed

as a rule. Berry ovoid, ellipsoid or almost spherical, 2-3 cm. long, 1-2-5 cm. diam,

sometimes smaller, glabrous and yellow when ripe, usually crowned by the persistent

style (at least when young), with mealy, edible pulp, 1-4 seeded. Seeds obovate or

elliptic, compressed, usually 15-20 mm. long, 9-12 mm. wide and 4-5 mm. thick,

sometimes considerably smaller; often (especially when only 1 or 2 seeds are present)

more or less laterally produced or at any rate somewhat attenuated at the base; the

scar nearly basal, almost horizontal, in a hollow sinus; testa light brown, shiny.

General Distribution. — Bechuanaland, Transvaal, Swaziland, Portuguese East

Africa, Rhodesia and extending into tropical East Africa.

Bechuanaland. — Gaberones, Van Son s.n.; Kanye, O. B. Miller B/298; Lobatsi

Govt. Farm, O. B. Miller B/245 (PRE); Molepolole, O. B. Miller B/220 (PRE);

Chobe Riv. nr. Serondela, O. B. Miller B/337 (PRE).

Transvaal. — Marico: Louw 610; Zeerust, Thode A1438. Rustenburg: Rusten-

burg, Sutton 885, 885a, 1023; Phillips s.n.; Phillips & Schweickerdt 3550; Rose-Innes 1

(PRE); Pretoria, Brits and Magaliesberg Range: Zeyher 1130 (BOL, SAM, isos!);

Burke 72 (PRE, SAM, for practical purposes duplicates of Zeyher 1130); Leendertz

s.n. 949; Galpin 6975; Lam & Meeuse 4824 (L); Meeuse 9095, 9095a; Hutchinson &

Mogg 2905. Johannesburg: Modderfontein, Putterill 161. Middelburg: Olifants

Riv. Gorge, Mogg 22447. Warmbaths: Warmbaths, Burt t- Davy 2149, 2625; Galpin

8853. Waterberg: Nylstroom, Prosser 1753, Rietspruit, 40 miles NNW of Vaalwater,

Smuts 362. Potgietersrust: Galpin 8823, 9023; Smuts 2006, Pietersburg: Blaauw-

berg, Leipzig, Bremekamp & Schweickerdt 1118; Daviesville, Setali, Gerstner 5429;

Woodbush, Hojfmann 30; De Hoek Forest Reserve, Keet h. No. FD 7413. Zoutpans-

berg: near Louis Trichardt, Obermeyer, Schweickerdt & Verdoorn 361a; Hutchinson

2004; Rodin 3987; Rogers 2177; Hanglip, Gerstner 5994; 20 miles NW of Louis

Trichardt, Codd 3023; Happy Rest, Gerstner 6031; 10 miles E. of L. Trichardt,

Gerstner 5731; Tshakoma, Obermeyer 1092; ± 10 rn. W. of Wyllie’s Poort, Gerstner

363

5923 (PRE). Sebasa: Phepidi Falls, Smuts & Gillett 3308, 3286; Curson & Irvine 55,

109; Legal 4819; Kruger National Park, near Pafuri, v. d. Schijff 578, 3812; id.,

Punda Maria, Lang s.n.; Codd 6514; v. d. Schijff & Marais 3728. Letaba: Duiwels-

kloof, Gerstner 5881. Lydenburg: Sekukuniland, van Warmelo 94; Mogg 16916

(PRE); Barnard & Mogg 762; Barnard 13 (PRE). Pilgrims Rest: Mariepskop,

Scheffler ED herb. No. 9946 (SAFD); Lothian, ED herb. No. 890 (SAFD). Nel-

spruit : Kruger National Park, Numbi : van der Schijff 25 ; 'near Sabin, Codd 4409

van der Schijff 696, 1461 Barberton: Thorncroft 2177; Liebenberg 2630; Burtt-Daw

2810.

Swaziland. — Lebombo Mts., Hornby 2805; Stegi, Codd & Dyer 29\Q, O. B. Miller

S/76; Acocks 15350.

Portuguese E. Africa. — Sul do Save: Guija, Myre & Balsinhas 788 (PRE);

Pedrogao 262 (PRE); Pedro & Pedrogdo 291 A (PRE) Chegua, Chongoene, Gomes e

Sousa 3999 (PRE); Matola, Imputa River, Mogg s.n. (PRE); Macovane, Hornby 2714

(SRGH, PRE). Manica e Sofala: Serra da Gorongoza, Simdo 952 (PRE); “ a milha

25 da T.Z.R.”, Simao 818 (PRE); Lifumba (Mutarara), Simdo 1488 (LM). Zambezia:

Mocuba, Namagoa, Faulkner 56 (PRE, COI); betw. Regulo Ingive and Nante, Barbosa

& Carvalho 4203 (PRE). Niassa: Pto. Amelia, btw. Mocimboa da Praia and Palma,

Barbosa 2122 (PRE).

S. Rhodesia. — Bulalima-Mangwe: Plumtree, Davies 280 (SRGH); Mangwe,

Marula, Plowes 1309 (SRGH, PRE). Bulawayo: Matopos, O. B. Miller B/1257 (PRE),

Mundy 892, 904, 909 (SRGH); Plowes 1331 (SRGH, PRE); West 2941 (SRGH);

Orpen 049/50 (SRGH); Borle 45 (PRE); West 2494 (SRGH). Belingwe: Harvie

9/51 (SRGH); 9/51 (SRGH); West 2792 (SRGH); Nuanetsi: Matibi I Reserve,

Davies 1773 (SRGH, PRE); Ft. Victoria: Eyles (?) in SRGH No. 3863; N.N. 52/51

in SRGH No. 34663; McGregor 2/47 (SRGH); Acheson 13 (SRGH); Monro 690

BOL); 761 (SRGH, BOL isos of Mimusops monroi Sp. Moore); Zimbabwe, Hornby

2831 (PRE); Smuts s.n. (PRE); Gerstner 6987 (PRE); Seward 47/51 (SRGH);

Seymour-Hall 3/51 (SRGH) Chilimanzi: Wormold 21/51 (SRGH); Mylne 1/51

(SRGH); Kirkham 18/51 (SRGH); Gibson 5/51 (SRGH); Greenhow 24/51 (SRGH,

PRE). Sabi-Lundi: IL/W 3353, 3377 (SRGH, PRE). Ndanga: 2338 (SRGH,

PRE), 2416 (SRGH). Gwelo: Eyles 5555 (SRGH); leg. Eyles (?) (SRGH No. 3331);

Steedman 210 (SRGH). Hartley: Hornby 3230 (SRGH); Jack s.n. (SRGH No.

4082); Umniati Riv., Mills 5/47 (SRGH). Umtali: Obermeyer 2070 (PRE); Chase

317, 1314, 1531 (SRGH), 869 (SRGH, PRE). “Eastern Border”: Charley s.n.

(SRGH No. 3685). Vumba Mts.: Chase 1741 (SRGH). Salisbury: Concession,

Eylesl in SRGH No. 1124. Mazoe: Eyles 4773 (SRGH); Mundy s.n. (SRGH No.

2090); Ford s.n. (SRGH No. 5146). Lomagundi: Eyles 2689, 2715 (SRGH, PRE,

SAM), 3130 (SAM, PRE) 4971, 4973 (SRGH) Pardy s.n. (SRGH Nos. 5918, 5924).

“Zambesi”: West 2900 (SRGH). Urungwe: Lovemore 291, 361 (SRGH, PRE).

Victoria Falls: Flanagan 3302 (BOL); Galpin 7044 (GRA, PRE, BOL), 14999 (PRE);

Rogers 5310 (GRA, BOL, SAM); 7446 (J); Schwartz s.n. (BOL 25012); Rodin 4492

(PRE, SRGH); Wild 3130 (SRGH); Greenway 6250 (PRE), Poynton 12 = ED

herb. No. 10403 (SAFD).

Northern Rhodesia. — Nega-Nega Hills: Burn Davy 762 (PRE).

Nyasaland. — Likubula Gorge; Brass 16366 (= Vernay Nyasaland Exp. 1946)

(PRE).

364

As regards the identity of M. zeyheri Sond and M. kirkii Baker the following

“differences” can be found in the original descriptions (see also Engler, op. cit.):

These differences appear to be very slight, and the most important ones seem to be the

length of the lateral appendages, the length of the style and the shape of the fruit.

However, subequal corolla-lobes and appendages can be found occuring with short

staminodes and with longer ones, with short styles and with longer ones. The fruits

in M. zeyheri vary from globose to ovoid and the seeds of fruiting specimens referred

to M. kirkii (including the fruiting type specimen) in various herbaria show the charac-

teristic produced base reported for M. zeyheri. Mr. B. de Winter compared specimens

of Mimusops from the Transvaal with the type of M. kirkii and with authentic specimens

of M. zeyheri at Kew and some of them correspond with either form, but there are

intermediate specimens which link them up. The Rhodesian material I saw shows a

complete continuous transition from the narrow leaved forms (= M. zeyheri) to the

broad-leaved forms (= M. kirkii), and the rusty pubescence of the innovations occurs

in both forms. In my opinion the form which corresponds with M. zeyheri is the Bush-

veld ecotype and often shrubby, whereas the broad-leaved form is a Lowveld ecotype

(more tropical hence large trees), but there is no question about their specihc identity.

Mimusops monroi S. Moore is a narrow-leaved form of “ typical ” M. zeyheri.

Mr. de Winter kindly compared the type of M. monroi [Monro 761 in BM) and reported

that it is an almost perfect match of Burke 72 from the Magaliesberg (in herb. Kew),

which number is a “twin-type”, practically identical with Zeyher 1130, the type of

M. zeyheri, so that there is no doubt about the conspecifity of both species. Duplicates

of Monro 761 in BOL and SRGH which I have seen are indistinguishable from most

Transvaal specimens and certainly conspecific.

Excluded species of Mimusops.

Mimusops altissima Engl. = Muriea spec, [an Muriea discolor (Sond.) Hartog?],

see p. 379.

M. angolensis Engl. = ManUkara cuneifolia (Baker) Dubard (a duplicate of the

type, Welwitsch 4836 in COI, was studied).

M. buchananii Engl. = Muriea spec, (an M. discolorl, see p. 379).

M. concolor Harv. ex Wright = ManUkara concolor (Harv. ex Wright) Gerstner.

365

M. cuneata Engl. = Austromimusops cuneata (Engl.) A Meeuse, see p. 355.

M. cuneifoHa Baker = Manilkara cuneifolia (Baker) Dubard.

M. densiflora Baker = Manilkara multinervis (Baker) Dubard (t. Engler and Hutch*

& Dalz.)

M. densiflora Engl, non Baker = Manilkara mochisia (Baker) Dubard, see p. 369.

M. discolor (Sond) Hartog = Muriea discolor (Sond.) Hartog, see p. 344.

M. dispar N.E. Br. = Austromimusops dispar (N.E. Br.) A. Meeuse, see p. 352.

M. flscheri Engl. = Manilkara spec., prob. M. mochisia (Baker) Dubard.

M. hcnriquesiana Sim = Lecomtedoxa henriquesii (Engl, et Warb.) A. Meeuse,

see p. 344.

M. henriquesii Engl, et Warb. = Lecomtedoxa henriquesii.

M. klaineana (Pierre ex) Engl. = Lecomtedoxa klaineana (Engl.) Dubard.

M. macaulayae Hutch, et Corb. = Manilkara macaulayae (Hutch, et Corb.)

H. J. Lam, see p. 373.

M. marginata N.E. Br. = Austromimusops marginata (N.E. Br.) A. Meeuse, see

p. 348.

M. menyhartii Engl. = Manilkara mochisia (Bak.) Dubard, see p. 369.

M. mochisia Baker = Manilkara mochisia (Bak.) Dubard.

M. natalensis (Pierre) Engl, non M. natalensis Schinz = Muriea discolor, see p. 377.

M. natalensis Schinz = Austromimusops marginata (N.E. Br.) A. Meeuse.

M. schinzii Engl. = Austromimusops marginata (N.E. Br.) A. Meeuse.

M. spiculosa Hutch, et Corh. ^ Manilkara macaulayae (Hutch, et Corb.) H. J.

Lam.

M. silvestris S. Moore = Austromimusops silvestris (S. Moore) A. Meeuse, see

p. 354.

“ M. transvaalensis Schinz” (sphalm.) ex Radik, (recte: natalensis) = Austromi-

musops marginata (N.E. Br.) A. Meeuse.

M. umbraculigera Hutch, et Corb. = Manilkara macaulayae (Hutch, et Corb.)

H. J. Lam.

M. welwitschii Engl. = Manilkara cuneifolia (Baker) Dubard.

M. zanzibarensis Engl. — Manilkara zanzibarensis (Engl.) Dubard.

8. MANILKARA.

Adans. emend. Gilly in Trop. Woods 73: 1-22 (1943); Lam & Royen in Taxon.

2, 5: 112 (1953); Royen in Blumea 7: 401 (1953), nomen conserv. propos.

Manilkara Adans., Earn. PI. 2: 166 (1763); Dubard in Ann. Mus. Col. Marseille

23: 6(1915); Baehni in Candollea 7 : 462 (1938); Lam in Blumea 4: 323(1941).

Mimusops pro parte, A. DC. in DC., Prodr. 8: 203 (1844) (quoad sect. Ternaria ADC.);

Bentham & Hook., Gen. PI. 2: 661 (1876); Baker in Oliv., FI. Trop. Afr. 3: 505

(1877); Engl., Mon. Sapot. Afr. 55 (1904) (quoad section Euternaria Engl, exclus.

Muriea)', Wright in Dyer, FI. Cap. 4, 1: 439, 1906; Phillips, Gen. S. Afr. FI.

PI. ed. 2, 568 (1951).

366

Trees or large shrubs. Stipules caducous or none. Leaves usually crowded at

the tips of ihe branches; often more or less oblong-obovate with rounded (and emargi-

nate) apex; lower side often lighter than upper one, mesophyll with sclereids; the

nervation is usually parallel (tertiary nerves parallel to secondary ones), often causing

a striate appearance of the leaf. Flowers axillary, usually in the axils of the lower leaves

of the branches or of the scars of fallen leaves, 3-merous or occasionally 4-merous.

Sepals biseriate, 3 + 3 (or 4 + 4), often rather broad. Corolla 6- (or 8)- lobed, each

lobe (in the South African species) with two dorsal appendages. Stamens 6 (or 8),

inserted in the throat of the corolla tube; filaments usually longer than the rather

small anthers. Alternipetalous staminodes 6 (or 8), or sometimes fewer than the number

of stamens, but never absent, usually small, often more or less dentate, lobed, fim-

briate or divided, glabrous; not incurved and not covering the style as in Mimusops,

but erect with the stamens of reflexed with the corolla. Ovary 6- (or 8)-loculated,

sometimes up to 1 5-loculated ; ovules ventrally or basiventrally attached. Berry ov\t-

to several-seeded. Seeds more or less laterally compressed; testa crustaceous, brittle,

or hard and bony; scar ventral, long and narrow, or basiventral and rather large and

wide, rarely small, basal; endosperm copious, cotyledons thin, foliaceous.

Type species: Mimusops kauki L., Sp. PI. ed. 1, 349 (1753) = Manilkara kauki (L.)

Dub. in Ann Mus. Col. Marseille 23: 9 (1915), fide Van Royen in Blumea 7: 402.

Distribution. — Circumtropical; about 25 species in tropical America, about 30

in tropical Africa, South Africa and the Mascarenes and about 15 in the Far East and

Pacific islands.

Manilkara is now almost universally recognised as a distinct genus. For a detailed

discussion of the differences between this genus and Mimusops, cf. Lam., op. cit.,

345-347. For the area under discussion, the main differences are given in Table II.

Both Manilkara and Mimusops have alternipetalous staminodes (which distin-

guish. these two at once from Muriea, which has either 12 fertile stamens and no alterni-

petalous staminodes, or 12 sterile stamens), and both have seeds with endosperm,

which distinguishes the two from Austromimusops apart from the very large seed scar

in the latter. Gilly, in Trop. Woods 73 (1943), p. 1-22, extended the genus Manilkara

by including Achras L. This would mean that Manilkara Adans. (1763) has to be

replaced by Achras L. (1753) and all the old world species of Manilkara apart from the

American ones would have to be renamed. Gilly’s interpretation is accepted by Lam

and Van Royen who drew the consequences and proposed that it would be better to

conserve Manilkara against Achras (see Taxon 2, 5: 112). In anticipation of adoption

of this proposal for conservation, I retain the African representatives in Manilkara.

Twigs slender, not conspicuously zig-zag, divaricate or subverticillate; leaves not congested

at the very tips of the branches ;

Sepals 5-6 mm. long; staminodes about as long as the stamens, not fleshy, more or less

irregularly serrate, incised or lacerate; leaves often more than 6 cm. long and

over 2-5 cm. broad; petioles often more than 1 cm. long 1. M. zanzibaremis.

Sepals about 3-5 mm. long; staminodes usually shorter than the filaments, more or less

fleshy, occasionally with a long apical filamentous portion; leaves usually less

than 6 cm. long and 2 - 5 cm. broad; petioles 5-8 mm. long 2.M. concolor.

Twigs thick, conspicuously zig-zag, divaricate or subverticillate; leaves almost invariably

crowded in fan-like groups at the very tips of the branches or on short lateral side-

shoots; small trees resemblirg species of Terminalia:

Leaves quite glabrous, young oner, very soon losing their pubescence; pedicels 8-12 mm.

long, sparingly pubescent; sepals sparsely pubescent outside 3. M. mochisia.

Leaves usually retaining vestiges of the dense pubesce of the young leaves, at least near

the midrib or towards the base, rarely becoming quite glabrous; pedicels often

more than 12 mm. long; sepals tomentose outside A. M. macaidayae.

367

1. M. zanzibarensis (Engl.) Dub. in Ann. Mus. Col. Marseille 23: 26 (1915); Brenan &

Greenw., Tanganyika Terr. Check List. Mimusops zanzibarensis Engl, in Pflan-

zenw. O. Afr., C, 307 (1895), and in Mon. Sapot. Afr. 58, t. 21, Fig. B (1904),

type: Stuhlmann coll. No. 1, 1009 in B, holof, from Zanzibar.

A large shrub or small tree up to 15 mm. high. Branches terete, glabrous, faintly

longitudinally striate or sulcate, the ultimate ones rather uniformly leafy. Leaves

elliptic, or obovate-oblong to (elliptic-) oblong, 3-12-5 cm., but usually 6-10 cm.,

long, 2-5 cm. wide, coriaceous, rather opaque but sometimes shiny with a greyish or

silvery shine beneath, glabrous, obtuse, rounded or sometimes emarginate at the apex,

acute at the base, with distinctly reflexed margin; midrib immersed but often more

or less keeled above, very prominent and when dry longitudinally finely striate or

sulcate below; secondary nerves numerous (16-25 on either side), slender, impressed

hence conspicuous above, far less conspicuous beneath, straight or ascending; tertiary

nerves parallel to secondary ones, but somewhat thinner and often shorter; ultimate

nervation fine, reticulate, immersed and conspicuous above, less conspicuous beneath.

Petioles stout, terete, canaliculate above, 0-5-3 -5 but usually 1-2 cm. long, glabrous,

in dried specimens more or less sulcate or striate below. Flowers trimerous, in fascicles

in the axils of the leaves or of leaf-scars, few or many together; pedicels 5-8 mm. long,

more or less angular, densely and shortly rusty-pubescent. Sepals ovate-oblong to

oblong, acute, 5-6 mm. long; the outer ones about 3 mm. wide, shortly rusty-pubescent

outside and whitish-pubescent inside near the margins and near the apex, with a broad

base; the inner ones thinner in texture, 2-3 mm. wide, whitish-tomentose outside and

glabrous inside except near the very apex, ciliate along the margins, with a narrower

base. Corolla white, glabrous; the tube 2 mm. long, the lobes linear-oblong, 4-5 mm.

at the apex, sometimes bifid and the lobes variously incised, glabrous. Filaments long-

subulate, dz 2 mm. long; anthers biapiculate, ovate-elliptic, about as long as the fila-

ments. Ovary depressed semi-globose, about 10-loculated and more or less distinctly

lobed, densely covered with greyish hairs, about 0-75 mm. high and about 1-5 mm. in

diameter, very abruptly narrowing into the terete, glabrous, di 5 mm. long, truncate

style. Fruit (according to Engler) obovoid, 10 mm. long and ±7-5 mm. in diameter,

2-4 seeded; seeds oblong, compressed dr 8 mm. long, with a short cicatrix above the

base.

N.B. — The figure A of tabula 21, B in Engler's monograph (1904) is not correct

in that the flowers are drawn far too small although the legend mentions natural size.

The other figures are properly drawn to scale.

General Distribution. — Coast of East Africa, from Zanzibar southwards and ex-

tending into the province of Niassa, Portuguese East Africa.

Zanzibar. — Imp. For. Herb. 834 leg. Vaughan (PRE).

Tanganyika. — Mafia Island, Greenway 5013, 5356 (PRE).

Portuguese E. Aerica. — Niassa: between Macomia and Mipande near Pto.

Amelia, Barbosa & Lemos 2302 (LM); between Mahate and Metuge near Pto. Amelia,

Barbosa 2346 (LM).

2. M. concolor (Harv. ex Wright) Gerstner [sphalm. (“ E. Mey) Gerstner ”] in J. S. Afr.

Bot. 14: 171 (1948).

Mimusops concolor Harv. ex Wright in Dyer, FI. Cap. 4, 1 : 443 (1906); type: Gerrard

& Macken 1662 in TCD, holo, teste C. H. Wright, K. iso, from Zululand.

Although the combination “ Manilkara concolor (Harvey) Gerstn. ” was erroneously

published by Gerstner by putting “ E. Mey ” as the original author, it is evident from

some of his notes and names on labels in the National Herbarium, Pretoria, that he

actually meant the combination “ Manilkara concolor (Harv.) Gerstn. ”, and he was

credited as the author of this combination in the Index Kewensis.

368

A much branched large shrub or small tree, up to 15 m. but usually well under 10

m. high. Branches terete, greyish, more or less distinctly longitudinally fissured and

with very prominent and large scars, the ultimate ones rather slender and almost in-

variably under 3 mm. in diameter, soon quite glabrous. Innovations glabrous or at

least very soon quite glabrous. Leaves crowded at the ends of the twigs but not strictly

terminal as in the next two species, varying from obovate-oblong or narrowly spathulate-

oblong to oblanceolate-oblong, sometimes elliptic or elliptic-oblong, but usually narrow-

ly spathulate-oblong or almost oblong (but with the greatest width just above the

middle), 2-5 cm., rarely up to 8-f cm. long and 1-2 (-3 ■ 5) cm. wide, coriaceous, glabrous,

with subreflexed margin an emarginate or retuse, rarely rounded apex, and narrowed,

but never very acute, base; midrib slightly raised or subimmersed, flush or minutely

keeled, but always narrow above, rather prominent below; secondary nerves very

slender, inconspicuous or more or less conspicuous because they are impressed on

both sides. 10-15 on either side, parallel, straight or rather straight, and ascending at an

angle of 60°-90°; tertiary nerves rather few, parallel to but shorter than the secondary

ones; ultimate reticulate tessellate nervation fine, immersed on both sides and as. a

rule distinct or conspicuous. Petiole semi-terete, flattened and channelled above,

rather stout, 3-8 mm. long, glabrous. Flowers in the axils of the leaves and of scars

of fallen leaves, fasciculate, often very numerous, 3-merous or sometimes 3-merous

and 4-merous on one specimen; pedicels comparatively stout, terete or sub-glabrous,

abruptly widening into the calyx. Sepals ovate or ovate-elliptic, obtuse, about 3-5

mm. long; the outer ones tomentose outside and inside near the apex, 2-2-5 mm.

wide; the inner ones thinner, petaloid, slightly narrower, about 2 mm. wide, vaguely

midribbed, whitish tomentose outside, glabrous inside. Corolla yellow or yellowish,

glabrous; the tube 0-75-1 mm. long, the lobes oblong-linear or oblanceolate-spathulate,

obtuse or rounded, with a very narrow, base, 3-3-5 mm. long and about 0-75 mm.

wide; the appendages about as long and wide, but from a broad base lanceolate-linear,

acute or acuminate, often with a few coarse serrations near the apex. Staminodes

sometimes fewer than the number of calyx-lobes, glabrous, fleshy, usually

ovate-suborbicular or subquadrate, and much shorter than the filaments, usually

more or less trilobed or tridentate, occasionally some produced in a thin filiform apical

portion equalling the filaments and if so. equalling or longer than the stamens. Fila-

ments 1-5-2 mm. long, glabrous; anthers somewhat sagittate, apiculate, 1-1-5 mm.

long. Ovary 6- (or 8-) loculated, semiglobose-conical, faintly lobed, hairy, nearly 1 mm,

long and about 1 -25 mm. in diameter, more or less gradually passing into the rather thick,

subulate, glabrous and truncate, about 2-5 mm. long style. Fruiting pedicels not or

hardly lengthened, somewhat incrassate; calyx persistent. Berry 1- or 2-seeded, edible,

ellipsoid or subglobose, 10-15 mm. long and 8-15 mm. in diameter. Seeds obovoid,

hardly compressed when single, often with one flattened lateral side when from a 2-

seeded fruit, obliquely truncate at the base, 9-11 mm. long 6^8 mm. wide, those from

1-seeded fruits 5-7 mm. thick, those from 2-seeded fruits 3-5 mm. thick; testa brown,

smooth and shiny, hard; scar ovate, oblong or elliptic, occupying the truncate basi-

lateral side of the seed. 6-8 mm. long and i 3 mm. wide.

Natal. — Kranskop: Tugela near Jameson’s Drift, 5<7>’c/- 648 (NU). Mtonjaneni:

Mtonjaneni, Gerstner 3710 (NH, PRE); Fule Drift near junction with Umhlatuzi,

Gerstner 2735 (BOL). Lower Umfolozi: Umsindusi Bridge, West 1873, 1911 (NH);

Empangeni, Nagana Res. St., Kluge 30 (NH); Umfolozi Game Reserve, Ward 1467

(NH, PRE). Hlabisa: Duku-Duku forest, Forest Dept. FD h. No. 8603 (Pre, SAFD);

near Matubatuba, Gerstner 8974 (NH, PRE); Hluhluwe Game Reserve, Ward 1593

(NH, PRE); False Bay, Gerstner 4791, 5061 (NBG, PRE), 5235 (PRE). Nongoma:

about 13 m. N. of Nongoma on road to Magut, Acocks 13015, Codd 1943 (PRE);

Wendelane Kloof. Gerstner 2522 (BOL), 4683 (NBG, PRE). Ubombo: Mkusi, Galpin

13319 (PRE).

369

Portuguese East Africa. — Sul do Save; Louren^o Marques distr., Maputo,

Santaca, Gomes e Sousa 3799, 3819 (COI, PRE), 3861 (PRE).

3. M. mochisia (Baker) Dub, in Ann. Mus. Col. Marseille 23: 26 (1915); Gerstner in

J. S. Afr. Bot. 14: 171 (1948), sphalm. “(Baker) Gerstner”. Mimusops mochisia

Baker in Oliv., FI. Trop Afr. 3: 506 (1877); Engl., Mon. Sapot. Afr. 63, t. 22,

Fig. B (1904); type: A'/rk 304 from Tete, K, lecto. ; fragment in PRE!.

Mimusops densiflora Engl., Pflanzenw. O.-Afr., C, 307 (1895), and in Mon. Sapot.

Afr. 63, t. 22, Fig. C (1904), non Baker in Kew Bull. 1895: 148; type: Stuhlmann

coll. No. 581 in herb. Hamburg, lecto.

Mimusops menyhartii Engl., op. cit. (1904), 63, t. 23, Fig. D; type: Menyhart 111 from

Boruma in Z, holo., PRE, photo!.

“ Manilkara densiflora Engl.”, Dale, in Imp. Forestry Inst. Paper 18: 25 (1939); M.

densiflora Dale, q.e. nom. nov., ex Brenan and Greenw., Check list Tanganyika

Terr. 2: 563 (1949). M. densiflora (Engl.) H. J. Lam in Blumea 4, 2: 355 (1941),

nomen illeg. M. menyhartii (Engl.) H. J. Lam, op. cit., 356..^ Sideroxylon flscheri

Engl, in Pflanzenw. O. Afr., C, 306(1895). == Mimusops (Engl.) Engl., Mon.

Sapot. Afr. 64 (1904). = Manilkara flscheri (Engl.) Lam in Blumea 4: 355 (1941).

A much branched deciduous large shrub or small to medium-sized tree, up to 10 m.,

rarely up to 15 m., high, with divaricate, subverticillate or zig-zag wise arranged, often

crooked branches, resembling several African species of Terminalia in habit. Branches

dimorphous; main shoots rather long (internodes usually 5-10 cm. long); “lateral”

short branches usually under 20 mm. long; internodes of main shoots terete, rather

smooth, finely longitudinally sulcate and, if not very young, glabrous and over 3 mm.

thick, only the very young growing ones bearing a few leaves that are not strictly ter-

minal. Short shoots and apices of the branches frequently somewhat thicker than the

adjoining internodes, very rough with the scars of fallen leaves; youngest shoots at

first pale yellowish- or whitish-tomentose, glabrescent. (New shoots are formed

laterally under the terminal thickened and leaf-bearing portions of the branches and

and this causes the peculiar divaricate or subverticillate branching, the original terminal

portion becoming a short “ lateral ” branch). Leaves in rather dense, more or less

fan-wise spreading groups on the short lateral branches and terminal apices (apparently

both lateral and terminal thickened portions of the stems being able to produce young

leaves during several season in succession); young leaves almost completely glabrous

as soon as they appear, the older ones quite glabrous; blade narrowly elliptic-obovate,

narrowly spathulate-oblong, cuneate-oblong or more or less oblanceolate-oblong,

sometimes obovate-oblong, 1 • 5-4 (-6) cm. long and 0 • 75-2 (-2 -75) cm wide, coriaceous,

or subcoriaceous, with a usually distinctly emarginate or retuse, sometimes rounded,

apex, and narrowing towards the base, with minutely reflexed margins, drying greyish-

green, paler beneath; midrib usually not very prominent on either side, but as a rule

flush or slightly immersed above and slightly prominent beneath, distinct and discolorous

in dried specimens; lateral nerves slender, immersed, ascending at an angle of 45°-60°;

secondary nerves rather numerous (10-15 on either side), tertiary nerves not or hardly

discernible from the ultimate, fine, impressed, reticulate nervation. Petioles 2-4 mm.,

rarely up to 7 mm., long, flattened above, on either side minutely winged by a con-

tinuation of the leaf margins. Flowers trimerous (at least no 4-merous ones seen),

often appearing with the young leaves or just before the leaves, solitary or in fascicles

in the axils of fallen leaves below the young growths; pedicels and calyx greenish or

green tinged with brown. Pedicels 8-12 mm. long, rather slender, more or less angular,

sparingly whitish pubescent, abruptly widening into the calyx. Sepals 3 + 3; the

370

outer ones ovate, sometimes oblong, 4-5 mm. long, 2-5-3 mm. wide, sparingly pube-

scent outside, pubescent inside near the apex and margins; the inner ones more petaloid,

oblong, 4-4-5 mm, long and 2-2-5 mm. wide, acute, tomentose outside, glabrous

inside, vaguely midribbed. Corolla glabrous; the tube about 1 mm. long; the lobes

elliptic-lanceolate from a narrow base, obtuse or rounded; about 4 mm. long and

0-75-1 mm. wide, the appendages about as long, linear, obtuse or rounded, 0-5-0-75

mm. wide. Staminodes variably in size but usually much shorter than the filaments,

rather fleshy, subquadrate, ovate or spathulate, trilobed, tridentate, bifid or incised,

occasionally produced at the apex in a long subulate or filamentous appendage equalling

the filaments, rarely even longer. Filaments long-subulate, about 2 - 5 mm. long, anthers

oblong, apiculate, about 1 - 5 mm. long. Ovary 6-loculated, semiglobose-conical, about

1 mm. high and about 1 - 5 mm. in diameter, densely covered with rather long hairs

which also cover the base of the about 3 mm. long, subulate, terete, glabrous, truncate

style. Fruiting pedicels hardly lengthened, somewhat incrassate, fruiting calyx persistent

Berry ellipsoid or ellipsoid-obovoid, yellow when ripe, highly appreciated by natives

Fig. 15. — Manilkara concolor, seed and fruit (from Gerstner 3759, “ Nangusi Forest”,

Zululand, in NH).

Fig. 16. — Manilkara mochisia, seed and fruit (from Wild 2748, Ndanga, S. Rh., in

SRGH).

Fig. 17. — Manilkara macaulayae, seed and fruit {{vova Lovemore 250, Sebungwe, S. Rh.,

in SRGH).

Fig. 19. — Muriea discolor, seed and fruit (from Wood 1349, Inanda, Natal, in NH).

371

and baboons for the flavour, 10-12 mm. long and 8-10 mm. in diameter, those I have

seen 1-3 seeded. Seed brown, obovate-oblong or oblong, compressed, 8-11 mm.

long, 5-7 mm. wide and 3-4 mm. thick, sub-produced at the base; testa brown, rather

smooth, not very shiny, hard; scar basilateral, linear-elliptic or linear, 3-5 mm. long

and 0-5-1 -5 mm. wide, surrounded by a rather wide, pale brown, callus-like zone which

includes the whole subproduced base of the seed.

A type specimen was originally not designated, as Baker, l.c., mentions two speci-

mens, via., a specimen from Batoka country, N. Rhodesia and one collected near Tete

(Zambezia, P.E.A.), both leg. Kirk. As Engler only mentions the specimen from Tete

in his monograph, I propose to make this one the lecto-type (in herb. Kew.)

Natal. — Umbombo: Otobotini, near Ndumo Game Reserve, Gerstner 3438

(NH).

Swaziland. — Probably near Stegi, Pole Evans 3488 (= 66).

Transvaal. — Nelspruit: Kruger National Park, Skukuza, Codd & deWinter 4992,

van der Schijff \262. Pretoriuskop, v. d. Schijff 2014; Mbeyamide, v. d. Schijff 2044,

1136. Pilgrims Rest: Kruger National Park, near Skukuza Codd 4388; Nwanetsi,

V. d. Schijff 320. Letaba: Kruger National Park, near Shingwedsi, Codd & Dyer 4660;

near Letaba, Lamont 33. Sebasa: Punda Maria Codd 4238, Codd & Dyer 4537, v. d.

Schijff 56^, 570; Nuanetsi, Gerstner 6061 (all PRE).

Portuguese E. Africa. — Sul do Save: Maputoland, Pto. Henrique, Gerstner

6670 (PRE); Lourengo Marques, Borle 144 (PRE); nr. Macia, Pedro & Pedrogao

1442 (PRE); nr. Chibuto, Pedro & Pedrogao 1535 (PRE); Guiza, Pedrogao 348 (PRE);

nr. confluence of Limpopo and Nuanetzi Rivers (nr. Transvaal border); Smuts P.

332 (PRE); Maringua, Sabi Riv., Chase 2454 (SRGH). Zambesia: Tete, Kirk 304

(fragment of type in PRE, ex herb. Kew); id., Baroma, Sisitso Riv., Chase 2753 (SRGH);

Menyhart 771 (Z, photo PRE!, holotype of M. menyhartii Engl.).

Southern Rhodesia. — Ndanga: Wild 2748 (SRGH); Chase 2367 (SRGH)-

Chipinga: Makosa Hills, Phelps 200 (PRE, SRGH). Melsetter: Chase 1767 (SRGH)-

West Nicholson: Plowes 1524 (PRE, SRGH).

Nyasaland. — Chiromo: Hornby 2902 (PRE).

Tanganyika. — Lindi: Lake Lutamba, Schlieben 5500 (PRE). In addition, the

following material was kindly sent on loan from the E. African Herbarium, Nairobi,

which I refer to this species:

Tanganyika Territory. — Moa-Mwakijembe: Kermode AY19930. Tabora: Forest

Guard Tabora 1517.

Kenya. — Bura, Tana Riv.: Bally B2805. Kitui Distr., Nzui; Edwards E79.

Coast, Kiunga : Elliott OX SSS; N. Giriama: Kn/e K 3663. Kibwezi: Gibbons K2571.

Engler describes Mimusops mochisia in his monograph as having staminodes often

bearing a long filamentous portion at the apex (Engler, op. cit., 1904, t. 22, Eig. B, d,

and in the description, op. cit., 63: “ staminodiis e parte basali subquadrati in appen-

dicem filiformem filamenta aequantem saepe productis”). Baker (l.c.) does not mention

372

the filiform appendages, but describes “ 6 minute toothed petaloid cuneate truncate

glabrous staminodes Miss Kies, our officer stationed at Kew in 1950, examined

one of the original specimens (leg. Kirk at Tete), and reported that this specimen does

not possess the filamentous appendages (at least not in the flowers she dissected). Upon

examination of a fragment of the specimen Kirk 304 from Tete, kindly sent through

the courtesy of the Keeper of the Herbarium at Kew, I came to the same conclusion.

In my opinion, the shape of the staminodes in this group of Manilkara (Section Iso-

gyne of Engler, op. cit., 1904) is not a constant character. Both in Manilkara concolor

and in M. macaulayae a great variation in the shape of the staminodes is found, from

small and subsquamiform to long-subulate, or some having filamentous appendages

(see Fig. 17), and. in the few flowering specimens of M. mochisia I could examine, several

different forms of staminodes were found: trilobed, spathulate, dentate, bifid, etc.

Apart from the differences in the shape of the staminodes, the differences mentioned by

Engler between M. mochisia, M. densiflora and M. menyhartii are negligible. The

leaf-shape reported for M. menyhartii is frequently found on specimens also having

some longer and narrower leaves. The three type specimens are evidently only forms

of one rather variable species, and Mimusops mochisia Baker being the oldest name,

the name Manilkara mochisia (Baker) Dubard. applies to all three forms under discus-

sion. A syntype, which I propose as the lecto-type, of Mimusops densiflora Engl, non

Baker is still extant (in herb. Hamburg) and was compared by Mr. de Winter with

specimens of M. mochisia. He reported that they are identical. An isotype of M.

menyhartii in herb. Zurich also proved to be a good match.

Mimusops flscheri (Engl.) Engl., the type of which was destroyed during the last

war and of which no isotypes could be traced, is most probably closely related to M.

mochisia, but I hesitate to refer it to this species, because in the description of M. flscheri,

the innovations are reported to be densely tomentose, whereas the older leaves are said

to be glabrous. The young parts of M. mochisia are never densely tomentose in the

specimens I have seen and on the other hand the older leaves of M. macaulayae, which

species also comes into consideration, are very rarely quite glabrous, but may by an

oversight have been described as glabrous in Engler’s description. The old leaves

are described as coriaceous which applies to M. mochisia rather than to M. macaulayae.

The pedicels of the type specimen, which was in bud, were described as 3-6 mm. long,

which points to M. mochisia rather than to M. macaulayae . Upon my request. Dr. B.

Yerdcourt of the East African Herbarium, Nairobi, sent me their material labelled

M. densiflora. The specimens in question, including those from the type area of M.

flscheri, proved to be M. mochisia and did not include specimens referable to M. macau-

layae. For these reasons the identity of Mimusops flscheri Engl, remains somewhat

uncertain, but the evidence is mostly in favour of M. mochisia, and I therefore regard

Mimusops flscheri as a probable synonym of M. mochisia.

The history of the name of Manilkara densiflora is interesting from a nomenclatural

point of view, although the name falls into synonomy. Engler and Baker published

the binomial Mimusops densiflora almost simultaneously, but for different plants.

Baker’s name (in Kew Bull. 1895, p. 148) antedates Engler’s [Pflanzenw. Ost-Afr., C.

(1895), p. 307]. Strictly speaking, all combinations based on Engler’s name are invalid,

but of course the name densiflora, not being preoccupied in Manilkara, can still be

used for this species. Brenan and Greenway, therefore, took up the combination made

by Dale as a “ new name ”, Manilkara densiflora Dale. Mimusops densiflora Baker was

reduced to Mimusops multinervis Baker by Engler in his Sapotaceae monograph, and

373

Mimusops densiflora Engler, as was pointed out above, becomes a synonym of Manil-

kara mochisia. M. mochisia is closely related to M. concolor and M. macaulayae,

and resembles especially the latter very much in habit. There is no doubt that they

represent three distinct species which can be distinguished as follows:

Although M. mochisia and M. macaulayae are distinct species, it is difficult to

distinguish specimens of the latter with only glabrous old leaves from M. mochisia.

In this case the additional characters used in the key, viz., the length of the pedicels

and the pubescence of the calyx, ca.n be used to distinguish them.

4. M. macaulayae {Hutch, et Corb.) H. J. Lam in Blumea 4: 356 (1941).

Mimusops macaulayae Hutch, et Corb. in Kew Bull. 1924: 329,330, fig. A (1924);

type: MacAulay 1002 from Northern Rhodesia in K, holo; fragment in PRE.

M. spiculosa Hutch, et Corb., op. cit., 330, 331, Fig. B., type: Allen 185 from

Victoria Falls in K, holo; BOL! and PRE!, isos. M. umbraculigera Hutch, et

Corb. l.c., and Fig. C (p. 330); type: N.N. from Southern Rhodesia in K, holo,

= SRGH No. 2639, iso!. ? Mimusops fischeri (Engl.) Engl., Mon. Sapot. Afr. 64

(1904).

Manilkara spiculosa (Hutch, et Corb.) H. J. Lam, l.c., and M. umbraculigera (Hutch,

et Corb.) H, J. Lam, l.c.

A small to medium-sized deciduous tree up to about 15 m. high, with divaricate,

subverticillate or zig-zag wise arranged, often crooked branches. Branches dimor-

phous: main shoots rather long (internodes usually 5-10 cm. long), lateral short

branches usually ±5 - ± 15 mm. long, rarely longer; internodes of main shoots

terete, rather smooth with faint longitudinal fissures, striations or wrinkles and (if

not very young) glabrous and at least 3-6 mm. thick, and only the very young ones

bearing a few leaves that are not strictly terminal; short shoots and apices of branches

frequently thicker than the adjoining internodes, very rough with the close scars of

fallen leaves; youngest shoots at first densely yellowish- or pale rusty-tomentose.

6096259—6

374

glabrescent. (New branches are formed under the terminal thickened portions of the

branches and this causes the peculiar divaricate or subverticillate appearance, the

original terminal portion becoming a short lateral branch). Innovations, young leaves

(mainly below), pedicels and outside of calyx-lobes with a pale buff pubescence, the

latter often, and on the leaves always, turning white. Leaves in rather dense, more or

less fan-wise spreading groups on the short lateral branches and terminal apices (ap-

parently both terminal and lateral thickened portions of the stems being able to produce

young leaves in several seasons in succession); blade narrowly elliptic-obovate to

narrowly spathulate-oblong or more or less cuneate-oblong, usually rather narrow,

3-5 (-7-5) cm. long and 1-2 (-3) cm. wide, with a usually distinctly emarginate or

retuse, sometimes rounded apex, narrowing to the base, with minutely reflexed margins,

rather firm but not coriaceous (M. mochisia has more coriaceous leaves as a rule),

drying an opaque green or brown, slightly paler beneath, densely tomentose or pube-

scent when young, at least on the lower surface, very rarely becoming quite glabrous,

but usually at least retaining vestiges of an adpressed whitish pubescence beneath,

especially near the midrib and towards the base; midrib either slightly raised or slightly

immersed, flush or minutely keeled and not very wide above, prominent beneath,

usually discolorous in dried specimens; secondary nerves slender, often inconspicuous,

rather numerous (± 10 to ± 15 on either side), parallel and straight, ascending at an

angle of about 60", faintly raised above, subimmersed below, usually bifurcate well

inside the margin and their ramifications soon merging with the finer reticulate ner-

vation; tertiary nerves not or hardly descernible from the very fine, tessellate, reticulate

nervation which is always distinct in older leaves. Petioles flattened and canaliculate

3-7 (-10) mm. long, glabrescent but as a rule never becoming quite glabrous. Flowers

often appearing with or just before the young leaves, solitary or in small fascicles in the

axils of fallen leaves below the young growths; pedicels and calyx of a greenish-fawn

or buff colour. Pedicels often more than 12 mm. long, rather slender and faintly angular,

abruptly widening into the calyx. Sepals 3 + 3 or sometimes 4 + 4; those of the outer

and inner rows subequal, ovate-elliptic (3-) 4 mm. long and (2-) 3 mm. wide, but the

outer ones with a broad base, the inner ones slightly thinner in texture, with a narrower

base and the greatest width just above the middle, all obtuse, tomentose outside,

glabrous inside except near the apex. Corolla glabrous, probably yellow or yellowish;

the tube less than 1 mm. long, the lobes and lateral appendages subequal, oblong-

lanceolate, 3-4 long and + 1 mm. wide, rounded or obtuse at the apex. Staminodes

sometimes fewer than the number of calyx-lobes, usually much shorter than the filaments

but sometimes nearly as long, glabrous, very variable in shape, varying (often in one

specimen or even in one flower) from triangular, trilobed, tridentate or subquadrate to

oblong, irregularly dentate or serrate to linear-subulate or subulate from a sub-quadrate

or semi-orbicular basal portion. Filaments subulate or linear-terete, 1-5-2 mm. long;

anthers ovate-oblong or ovate-lanceolate, 1 • 5-2 • 5 mm. long. Ovary 6- (or 8-)-loculated,

ovoid-globose, more or less angular, l^-lf mm. long and 1-1^ mm. in diameter, pubes-

cent, more or less gradually tapering into the glabrous, long-subulate, angular to sub-

terete and minutely truncate, 3-4 mm. style. Fruiting pedicels lengthening and up to

25 mm. long, slightly incrassate and ultimately about 1 mm. thick, glabrescent; calyx

under fruit persistent. Berry ellipsoid or obovoid-oblong, with a narrow base, about-

18 mm. long, 8-10 mm. in diameter, edible, the few seen all one-seeded. Seed obovate

oblong, compressed, ± 16 mm. long, ± 7 mm. wide and 3^-4 mm. thick, narrowed

and more or less unequal at the base, laterally produced at the side of the scar; testa

a dull greyish brown, more or less rugose, or tuberculate, hard; scar linear, + 8 mm.

long and 0-5-1 mm. wide, surrounded by a light brown and rather shiny, slightly

thickened zone which is about 12 mm. long and about 3 mm. wide and includes the

whole produced base of the seed.

Angola. — Caposolo: Faulkner (A) 64 (PRE).

375

Bechuanaland. — Chobe: Serondela, O. B. Miller B/1128, B/1203, (PRE).

Kachikau on road to Kasane, Erens 380 (PRE, SRGH). Small island in marches

near Ghanzi, 22' 30^ E, I90 5' S: Story 4789 (PRE).

Southern Rhodesia. — “ Matabeleland ” : Mrs. Pardy s.n. (SRGH No. 5027).

Urungwe: Wild 4163 (PRE, SRGH). Sebungwe: Lovemore 250 (SRGH). Hartley:

Umswewe River, Hodgson H. 35/48 (SRGH); Gatooma, Golding s.n. (SRGH No.

31015, 31267); Eyles in QVM. Herb. No. 7264, 7507 (SRGH); Hartley: Eyles in

QVM. Herb. No. 7507 (SRGH). Wankie: Wankie Camp, Pole Evans 2751 (PRE,

SRGH); Victoria Falls. Allen 185 (isotype of Mimusops spiculosa. Hutch et Corb.,

(PRE, BOL); Galpin 7052 (PRE); Wild 3088, 3106 (SRGH). Bulawayo: Hodgson

s.n. (PRE, SRGH). Without precise locality: N.N. in SRGH 2639 (isotype of Mimu-

sops umbraculigera Hutch, et Corb.); specimen sent by Dept, of Munitions, S. Rh.,

to Bolus Herb (BOL. 25013).

Northern Rhodesia. — Victoria Falls, N. bank, Pole Evans 2751 (8) (PRE, SRGH)

Lusaka: Mumbura. MacAulay 1002 (fragment of type of Mimusops macaulayae,

PRE).

A study of the few flowering specimens I have seen (including a fragment of the

type specimen and, in addition, isotypes of Mimusops spiculosa and M. umbraculigera)

showed that the staminodes are very variable in shape and length and that in a single

flower triangular, serrate, bifid, dentate and subulate ones may occur (Fig. 18). In the

isotype of M. umbraculigera, I found bifid and triangular staminodes apart from the

irregularly dentate ones described in the species diagnosis by Hutchinson and Corbishley;

in the isotype of M. spiculosa bifid ones occur apart from the subulate ones mentioned

in the original species diagnosis. In Eaulkner (A) 64 such a great variation is found

(see Fig. 18), that I am convinced that the three species described by Hutchinson and

Corbishley, are conspecific. Apart from the character of the staminodes, the differences

they mentioned are very slight and as far as I have seen, not constant (such as the shape

of the filament, relative lengths of filaments and anthers, pubescence) so that there is no

reason to distinguish more than one species. As regards the possible identity with

Mimusops fischeri Engl., see the discussion under Manilkara mochisia.

d A M

c d

F:c. 18. — Shapes of staminodes in Manilkara macaulayae: [a) 3 staminodes found in

one flower of Faulkner (A) 64 from Angola; (b)5 staminodes of another flower of

the same specimen; (c) 2 staminodes from isotype of Mimusops umbraculigera

(SRGH No. 2639 from Rhodesia); {d) 2 staminodes out of one flower of O. B. Miller

B/1128 from Bechuanaland.

376

9. MURIEA

Hartog in J. Bot. 16: 145 (1878) (diagnosis on p. 72 sub. Eichleria Hartog); Dubard

in Ann. Mus. Col. Marseille 23: 28 (1915); H. J. Lam in Blumea 4: 348 (1941)

and Table 1 (p. 350). Eichleria Hartog in J. Bot. 16: 72 (1878), non Eichleria

Progel in Mart., FI. Bras. 12, 2: 518 (1877).

Labourdonnaisia Boj. ex parte, Benth. & Hook, Gen. PI. 2: 659 (1876); Gerstner in

J. S. Afr. Bot. 12: 40 (1946).

Mahea Pierre, Not. bot. Sapot. 8 (1890).

Mimusops L. sect. Muriea Hartog in J. Bot. 17: 358 (1879).

Mimusops L., subgenus Ternaria (A. DC) Engl., sect. Euternaria Engl., subsect. Muriea

(Hartog) Engl, in Mon. Sapot. Afr. 55 (1904), and sect. Mahea (Pierre) Engl.,

op. cit., 65.

Mimusops L., ex parte, C. H. Wright in Dyer, FI. Cap. 4, 1 : 439 (1909), p. 439; Phillips

in Genera S. Afr. Flow. PL, ed. 2: 568 (1951).

Type species: Labourdonnaisia discolor Sond. in Linnaea 23: 73 (1850) = Muriea

discolor (Sond.) Hartog.

Trees with exstipulate, parallel-nerved leaves. Flowers trimerous, axillary, solitary

or in fascicles. Sepals 3 -f 3; corolla with short tube and 6 lobes, these lobes either

bearing two lateral appendages (in this case 12 fertile stamens, sub-bisei iate : 6 epipeta-

lous and 6 alternipetalous) or the lateral appendages reduced, sometimes 0 (in this case

usually stamens more or less sterile and transformed into subulate staminodes).

Stamens inserted in the throat of the corolla tube. Ovary 6-loculated, hairy; style

long-cylindric-subulate ; ovules with basal or basilateral attachment. Berry (always?)

1-seeded. Seed with basilateral, oblong or elliptic scar; testa rather thin and brittle;

endosperm copious; cotyledons thin, foliaceous.

Number of species: At least one in the coastal regions of Southern East Africa,

possibly one or two more in tropical East Africa (if not conspecific with the first) and,

according to Hartog (1878) and others, another species in the West Indies.

The nomenclature of the type species Muriea discolor (Sond.) Hartog is rather

complicated. Sonder described it as a Labourdonnaisia and it was included in this

genus by Bentham and Hooker in Gen. PI. 2: 659. Hartog (1887) pointed out that

the petals in Labourdonnaisia are equal and placed in one whorl, their number varying

from 12-17, with an equal number of stamens, whereas the species described by Sonder

has 6 corolla lobes with each 2 lateral appendages. On account of this difference,

among other things, Hartog decided that the plant belongs to a different genus, which

he described as Eichleria. This name being illegitimate on account of Eichleria Progel,

Hartog changed the name to Muriea. A year later, he regretted this decision and in-

cluded his genus Muriea as a section in his large genus Mimusops which comprised

practically all the Sapotaceae — Mimusopoideae. Engler (1904), whose conception of

Mimusops agrees with that of Hartog, included Muriea as a subsection of his section

Euternaria (the greater part of which is now generally recognised as belonging to the

genus Manilkara Adans.)

Dubard in 1915 resuscitated Muriea, which genus he placed near his section Mahea

of Manilkara, on account of the absence of staminodes.

Baehni in Candollea 7: 467 (1938), reduces Muriea to Mimusops, which is alto-

gether incomprehensible to me, because Baehni, op. cit., 465-466 excludes all forms

with 3-merous flowers from Mimusops (and includes most of them in Manilkara).

Mahea Pierre remained a more or less problematic genus for a long time. Engler

(1904) reduced it to a section Mahea of a large genus Mimusops. Dubard (op. cit., 27)

reduced Mahea to a section of Manilkara. Baehni (op. cit., 461) follows Dubard.

377

Gerstner in J. S. A. Bot. 12: 49 (1946), finally, resuscitated the name Labour-

donnaisia discolor Sond. In a subsequent publication in J. S. Afr. Bot. 14: 173 (1948),

Gerstner pointed out that the anthers and the corolla-segments of this plant are reduced

when the flowers develop in dry weather, or if the tree is not growing on fertile soil.

He distinguished four/ormae, a forma perfecta and three progressively more depauperate

forms. The most depauperate form corresponds with Mahea natalensis Pierre, so that

Mahea becomes a synonym of Muriea*. Incidentally, a specimen of Wood 1134 (type

number of Mahea natalensis Pierre) in BOL bears fruits and its fruits and seeds are

identical with those of Muriea discolor.

The synonymy of the plant described as Mahea natalensis Pierre is also rather

complicated. Engler, when he reduced Mahea natalensis to Mimusops, made the com-

bination Mimusops natalensis (Pierre) Engl. (1904). However, Schinz had validly

published as specific name Mimusops natalensis Schinz in Bull. Herb. Boiss. 4: 441

(1896) and Engler renamed this species Mimusops schinzii Engl. Under the present

rules, M. natalensis (Pierre) Engl, is illegitimate, being antedated by M. natalensis

Schinz. The first name becomes a synonym of Labourdonnaisia discolor Sond., the

second of Mimusops marginata N.E. Br. [= Austromimusops marginata (N.E. Br.)

A. Meeuse, quod vide, see p. 350].

My reasons for resuscitating Muriea Hartog, at least for the species Labourdonnaisia

discolor Sond., are the following:

1. Labourdonnaisia was described as having 12-18 equal corolla-lobes (and not

6 lobes each with 2 lateral appendages) and 12-17 stamens, sometimes with a few

staminodes, whereas Muriea has 6 corolla-lobes, each having, in perfect flowers, 2

lateral appendages, and always 12 stamens (or rarely 12 staminodes).

2. Labourdonnaisia is reported to have a large and basal seed scar, whereas Muriea

has a basiventral, rather narrow seed scar. Muriea has its affinities clearly with Manil-

kara; it is, in fact, a Manilkara with normally 6 + 6 stamens instead of 6 stamens + 6

alternipetalous staminodes. Its habit is also very much that of a Manilkara (parallel

substriate nervation of the leaves, etc.). However, the absence of alternipetalous

staminodes distinguishes it clearly from Manilkara.

The synonymy of the only South African species, therefore, becomes as follows: —

M. discolor (Sond.) Hartog in J. Bot. 16: 145 (1878); Dubard in Ann. Mus. Col.

Marseille 23: 28 (1915).

Labourdonnaisia discolor Sond. in Linnaea 23: 73 (1850); Gerstner in Jl. S. Afr. Bot.

12:49 (1946), and 14: 173(1948); Syntypes: Gueinzius \2% and Gueinzius 541 from

Durban, Natal, in herb. Sonder nunc S. L. sericea Benth. et Hook, f., Gen. PI.

2: 660, (1876), nomen.

Eichleria discolor (Sond.) Hartog in J. Bot. 16: 72 (1878).

Mimusops discolor (Sond.) Hartog in J. Bot. 17: 358 (1879); Engl., Mon. Afr. Sapot.

55, t. 34, Fig. A (1904); C. H. Wright in Dyer, FI. Cap. 4, 1 : 440 (1906).

Mahea natalensis Pierre, Notes Bot. Sapot. 10 (1890); type: probably Wood 1134,

holo in P?, also in BOL, GRA.

Mimusops natalensis (Pierre) Engl, Mon. Afr. Sapot. 65, t. 25, Fig. B (1904) non

Mimusops natalensis Schinz (1896); C. H. Wright, l.c.

A medium-sized tree, 10-20 m. high with a stem diameter up to about 60 cm.

with grey, longitudinally fissured bark; on young branches the bark is smoother and

often full of leaf-scars. Ultimate branches rather stout (at least 1 -5 mm. thick), terete.

* This reduction of the lateral appendages in one species could possibly occur in the related genus

Manilkara and that is why, to my mind, the species of Gilly’s subgenus Manilkariopsis (Trop. Woods

73: 9) should be re-examined in the light of Gerstner’s observations on Muriea, i.e., in relation to

prevailing ecological conditions at the time of flowering.

378

Innovations shortly puberulous-tomentose, the pubescence cinnamon-coloured; older

parts except the lower surface of the leaf very soon glabrous. Leaves more or less

distinctly crowded at the tips of the branches, rather uniform in shape, almost invariably

obovate-oblong, sometimes oblong, or more or less spathulate- or lanceolate-oblong,

narrowed but usually not distinctly cuneate at the base, with a rounded, obtuse, subacute

or very shortly acuminate and usually recurved, almost invariably emarginate apex,

green above (often drying brown or greyish-brown), densely adpressed-silvery pubescent

beneath, except on the midrib, with slightly recurved margins, 3-5-7 (-10) cm. long,

1-5-3 -5 (-4-5) cm. wide; midrib immersed and distinctly channelled above, very

prominent beneath, at first finely rusty-pubescent, ultimately glabrous and of a different

colour than the silvery lower surface of the leaf (yellowish brown or dark brown when

dry); secondary nerves 10-15 or more on either side, straight, ascending at an angle of

70^-90e, archingly joining near the margin, immersed but rather conspicuous above,

immersed and partly hidden by the silvery tomentum below; tertiary nerves parallel

to the secondary ones, usually inconspicuous below, but appearing as a fine striation

above; fine ultimate nervation reticulate, usually invisible below except in the oldest

leaves, but distinct and impressed above and giving the upper surface of the leaf a

minutely tessellate, dull appearance. Petioles terete, hardly thickened towards the

base, narrowly canaliculate above, glabrous, 6-15 mm. long. Flowers in few-flowered

fascicles (often about 3 together). Perfect flowers', pedicels 5-10 mm. long, rusty-

pubescent; flower-buds rounded at the top, broadly obovoid, i 5 mm. long and 3-4

mm. in diameter; sepals almost completely free; the three outer ones rather coriaceous,

broadly ovate, narrowed towards the tip, but not acute, ± 5 mm. long and ± 4 mm.

wide, rusty-pubescent outside, subciliate, glabrous inside except near the apex, the inner

ones narrower, 5-5-5 mm. long and ±3 mm. wide, yellowish, whitish adpressed-

pubescent outside and near the apex inside; corolla white or yellowish; tube 1 - 5-2 mm.

long, cylindric; the lobes long-spathulate with a long-tapering, narrow base and rounded

top, the appendages broader, lanceolate-oblong or linear, acuminate, all segments

3-3-5 mm. long; filaments 2-2-5 mm. long, filiform and slender; anthers 1 -5-2 mm.

long, ovate-cordate, apiculate; ovary depressed-conical, it 1 mm. high and i 1 mm.

in diameter, densely sericeous-pilose; style glabrous cylindric, 3-3-5 mm. long, rather

gradually tapering into the flat, subcapitate and it 6-lobed stigma. Imperfect flowers

on short pedicels (sometimes under 5 mm. long), often considerably smaller (e.g.,

calyx and corolla lobes only 2 mm. long); corolla lobes with reduced lateral appendages,

trifid, tridentate or entire; anthers very small or wanting or sometimes developed,

ovary and style as above but sometimes ovary more subglobose and style only it 2 mm.

long. Fruit spherical, edible and red when ripe, according to Gerstner (1946), but

when dried ellipsoid, 8-12 mm. long and 5-10 mm. wide, crowned by the peristent

style. Fruiting pedicels not appreciably lengthened or incrassate; fruiting calyx adpressed

to the fruit. Seed elliptic in outline, compressed, it 8 mm. long, i 5 mm. wide, and

+ 3-5 mm. thick in the middle; scar basilateral, oval or elliptic, 3-4 mm. long, and

1-2 mm. wide in the widest place; testa pale fawn-coloured when dry, thin and brittle.

Distribution. — Natal, from Durban northwards, Zululand, Swaziland, P.E. Africa

and extending into East Tropical Africa.

Natal. — Durban: Thorns s.n. (NH No. 40612, PRE); Durban, Albert Park,

Van der Byl s.n. (HN No. 16245 and 16246 in NH, GRA); Gerstner 4703 (PRE),

Butcher (= Gerstner 4712) (PRE, SRGH); Gerstner 6820 (NBG, PRE). Pinetown:

Umzinyati, Wood 11440 (PRE, NU, J). Ndwedwe: Inanda, Wood 1134 [GRA, BOL,

one of the numbers cited by Engler sub. Mimusops natalensis (Pierre) Engl.], 1349

(NH, BOL, SAM). Palmiet, Wood 8378 (NH, NU, PRE). Lower Tugela: Darnal,

Schmidt 39, 40 (NH). Hlabisa: near Hlabisa, Gerstner 4117 (NH, PRE) 4527 (PRE);

Emgangado, Gerstner 5077 (RE, BOL, NBG); Hluhluwe Game Reserve, Ward 2639

379

(NH, PRE), “ Zululand Gerstner 2539 (BOL). Ubombo: Bangazi Lake, Boocock

FD herb. No. 5720 (PRE, SAFD). Swaziland (?): Mangusi Forest (prob. near Hlati-

kulu) FD. Herb. No. 5314 (PRE, SAFD).

Portuguese East Africa. — Sul do Save: Maputo, Sim 20992 (NU, PRE);

Lourengo Marques: Oliveira 74 (LM); Inhaca Island, Mrs. Moss (J. No. 27637, also

PRE) between Chongoene and Chidinguel: Pedro 247 (PRE); Chibuto e Vila Gomez

da Costa, Pedro & Pedrogao 1559 (PRE). “ Zavala Distr.,” Afr. Music Res. Libr. Ill

(PRE); Inhambane, Gomes e Sousa 1657 (PRE, COI), 1949 (COI); Inhambane,

Maxixe, Amostra 1 (LM). Manica e Sofala: Macuacua, Simao 1251 (PRE); Gaza-

land. Earthy 27 (PRE); Masiyeni, Earthy 169 (PRE). Niassa: Porto Amelia, Barbosa

1865 (LM), Barbosa & Lemos 2031 (PRE); entre Pundankas e Nanga, Barbosa 2190

(PRE).

Southern Rhodesia. — Umtali: Chase 4664, 4559, 5122, Ball 1 (PRE, SRGH).

As this species is most probably found in other parts of tropical East Africa, it is

not unlikely that it was described from this area under a different name. It is quite

likely that Engler’s species Mimusops buchananii in Pflanzenw. O. Afr., C, 307 (1895)

and in Mon. Sapot. Afr. 56, t. 19, Fig. B (1904) and Mimusops altissima Engl., op. cit.

(1904), 55, are referable to this species. These two have 12 stamens and the descriptions

(and Figure of M. buchananii) agree very well indeed. This would extend the range of

Muriea discolor to Nyasaland and Tanganyika.

381

An Enumeration of the Maytenus Species

of Southern Africa.

By

W. Marais.

The revision of the South African Celastraceae by Davison in Bothalia 2: 289-346

(1927), was, for many years, the latest work available in South Africa. Loesener’s

treatment of the family in Engl. & Prantl, Nat. Pflanzenfam. ed. 2, 20b: 87-197 (1942),

did not become available until several years after the war. He transferred the unarmed

African species of Gymnosporia to Maytenus, a genus which had, up to that time, been

regarded as being restricted to tropical and subtropical America.

Exell in Bol. Soc. Brot., Ser. 2, 26: 222 (1952) regards Gymnosporia, as defined by

Loesener, as too artificial to be maintained, and transfers several of the armed species

to Maytenus. Brenan in Mem. N.Y. Bot. Card. 8: 238 (1953) and Blakelock in his

series of notes on African Celastraceae [the first in Kew Bull. 1956: 237 (1956)] also

hold the view that the separation of the two genera on a basis of the presence or absence

of spines and short shoots is unsatisfactory when the supposed difference of a two-

loculed versus a three-loculed ovary breaks down.

Ding Hou in Ann. Miss. Bot. Card. 42: 215-302 (1955) expresses the view that

Gymnosporia and Maytenus are distinct, though very closely related, and provides a

tabular key using a large number of characters. After a careful comparison, however,

there seems to be no character or combination of characters constant enough to justify

the retention of Gymnosporia.

The following key and enumeration is tentative, but it has become imperative to

make the combinations under Maytenus and to try and clarify some of the specific-

level problems.

See Taxon 3: 196 (1954) for the proposal to conserve the generic name Maytenus

H.B.K.

My thanks are extended to the Directors of all the herbaria who kindly sent material

on loan to Kew.

Flowers in axillary fascicles, common peduncle none or less than 2 mm. long; plants unarmed

and without short shoots; internodes sometimes much abbreviated, but leaves never

in true fascicles;

Leaves with resinous threads on being broken M. acuminatus.

Leaves without resinous threads ;

Leaves entire, suborbicular, obovate to obovate-elliptic M. Incic/us.

Leaves variously dentate, serrate or angled ;

Capsules yellow, orange to orange-brown; leaves green or yellow-green; margin

strongly revolute with 3-5 (rarely 6) teeth or angles on either side;

apex rounded or obtuse M. procumbens.

Capsules greenish or whitish, never brightly coloured; leaves green, grey-green or

whitish on lower surface; margin flat or slightly revolute, obscurely

angled to sharply and spinously dentate or serrate, usually with more

than 6 teeth on either side; apex often acute or acuminate, but some

rounded or obtuse M. undatus.

382

Flowers in cymes; cymes axillary or appearing fascicled on short shoots; peduncle some-

times reduced, but then plants armed and with short shoots; leaves often fascicled;

Plants unarmed and without short shoots ; leaves never fascicled :

Leaves with resinous threads on being broken M. acuminatus.

Leaves without resinous threads:

Plants, or at least the young parts pubescent M. peduncularis.

Plants glabrous:

Leaf-margin closely serrulate; blade narrowly lanceolate; veins prominent

on lower surface M. bachmannii.

Leaf-margin entire; blade broadly ovate to narrowly lanceolate, thick and

leathery with veins more or less immersed M. oleoides.

Plants armed and with short shoots; leaves often fascicled:

Young parts puberulous or pubescent:

Leaves lanceolate, oblanceolate or narrowly elliptic, entire or scarcely dentate.

Capsule papery. Flowers greenish- white to cream M. tenuispimis.

Leaves ovate to ovate-elliptic, closely serrate or crenate-serrate; capsule leathery

to almost woody; flowers pink to red M. mossambicensis var. ruber.

Plants glabrous :

Ovary two-celled:

Leaves oblanceolate to obovate, flnely and regularly serrate M. senegalensis.

Leaves linear to lanceolate or lanceolate-oblong, entire, repandly dentate or

coarsely serrate M. linearis.

Ovary three-celled:

Leaves entire or nearly so ; bark developing very early on young twigs which

are brown or whitish, sometimes longitudinally wrinkled but not

angular-striate (see also M. cymosa):

Leaves obovate-cuneate, thick, the veins usually completely

immersed ; capsule thick, woody M. capitatus.

Leaves oblanceolate or oblanceolate-oblong, often discolorous,

the veins raised, at least on lower surface; capsule

thin-leathery M. poly acanthus.

Leaves serrate, dentate or crenate-serrate, rarely almost entire, but then

young twigs usually green and angular-striate:

Mature leaves thin, ovate or ovate-lanceolate, often rounded at the

base or broadest below the middle; margin often sharply

serrate; twigs slender, often reddish; inflorescence very lax,

often few-flowered, with long, slender main branches

M. mossambicensis

var. mossambicensis.

Mature leaves leathery to very thick leathery, variable in shape, often

cuneate and broadest above the middle, except for leaves on

young growth, or else very thickly leathery if rounded at the

base; margin crenate-serrate to subentire or entire; twigs

brown, grey or green; inflorescence lax to very dense, with

rather stout or short main branches:

Leaf margin revolute; blade more or less elliptic above the distinct

petiole, almost invariably drying brownish, discolorous;

spines absent or present, the whole or at least their tips

often shiny dark-brown; twigs sometimes lon^tudinally

wrinkled but often striate with definite raised lines

M. nemorosus.

Leaf-margin usually flat; blade very variable in shape and texture,

but not often as above; petiole sometimes distinct but

when blade cuneate it is often decurrent on the petiole;

usually drying green or grey-green, only rarely discolorous;

spines absent or up to 20 cm. long, but never shiny dark-

brown; twigs often strongly angled or striate M. cymosus.

383

1. M. lucidus (L.) Loes. in Engl. & Prantl, Nat. Pflanzenfam. ed. 2, 20b: 140 (1942).

Celastrus lucidus Linn., Mant. 49 (1767); Sond. in Harv. & Sond., FI. Cap. 1 : 456

(1860).

Gymnosporia lucida Loes. in Engl & Prantl, Nat. Pflanzenfam. 3, 5: 208 (1892);

Davison in Bothalia 2 : 294 ( 1 927).

2. M. procumbens (L./.) Loes. in Engl. & Prantl, Nat. Pflanzenfam. ed. 2, 20b: 140

(1942).

Celastrus procumbens Linn, fil., Suppl. : 153 (1781); Sond., l.c. 457.

Gymnosporia procumbens Loes. in Engl., Bot. Jahrb. 39: 169 (1906); Davison,

l.c. 297.

3. M. undatus {Thunb.) Blakean Kew Bull. 1956: 237 (1956).

Celastrus undatus Thunb., Prodr. PI. Cap.: 42 (1794); Sond., l.c.

C. ilicinus Burch., Trav. Int. S. Afr. 1: 340 (1822); DC., Prodr. 2: 7 (1825).

Catha fasciculata Tul. in Ann. Sci. Nat., Ser. 4, 8: 98 (1857).

Celastrus zeyheri Sond., l.c. 456.

Gymnosporia undata SzyszyL, Polypet. Discifl. Rehmann. : 34 (1888); Loes. in

Engl. & Prantl, Nat. Pflanzenfam. 3, 5: 208 (1892); Davison, l.c. 296.

G. rehmannii SzyszyL, l.c.; Loes., l.c.

G. zevheri SzyszyL, l.c. 33; Loes. in Engl., Bot. Jahrb. 17: 548 (1893); Davison,

l.c. 294.

G. fasciculata Loes. in Engl. & Prantl, Nat. Pflanzenfam. 3, 5: 208 (1892); Davison,

l.c. 298.

Celastrus albatus N.E. Br. in Kew Bull. 1906: 16 (1906).

Gymnosporia deflexa Sprague in Kew Bull. 1906: 246 (1906); Davison, l.c. 299.

G. albata Sim., For. & For. FI. Col. Cape Good Hope: 186 (1907): Davison, l.c.

G. ilicina Loes. in Engl., Pflanzenw. Afr. 3, 2: 225 (1921); Davison, l.c. 296.

G. peglerae Davison, l.c. 298.

Maytenus ilicinus Loes. in Engl. & Prantl, Nat. Pflanzenfam. ed. 2, 20b: 140 (1942).

M. fasciculatus Loes., l.c.

M. zeyheri Loes., l.c. 138.

4. M. oleoides {Lam.) Loes., l.c. 137.

Celastrus laurinus Thunb., l.c. non Maytenus laurinus Briq. 1919.

C. oleoides Lam., Tabl. Encycl. & Meth. 2: 293, No. 2696 (1797).

Scytophyllum laurinum E. & Z., Enum. PL Afr. Austr. : 124, No. 966 (1834-5);

Sond., l.c. 471.

Scytophyllum angustifolium Sond., l.c. 472.

Gymnosporia laurina SzyszyL, l.c. 35; Loes. in Engl. & Prantl, Nat. Pflanzenfam.

3, 5: 208 (1892); Bolus & Wolley-Dod in Trans. S. Afr. Phil. Soc. 14: 247 (1903);

Davison, l.c. 310.

G. angustifolia Loes., l.c.; Bolus & Wolley-Dod, l.c.; Davison, l.c. 300.

G. monococca Davison, l.c.

Maytenus monococcus Loes. in Engl. & Prantl, Nat. Pflanzenfam. ed. 2, 20b:

138 (1942).

M. angustifolius Loes., l.c.

5. M. acuminatus {L.f.) Loes., l.c.

Celastrus acuminatus Linn, fil., l.c. 154 (1781); Sond., l.c. 454.

C. cordatus E. Mey. ex Sond., l.c.

Gymnosporia acuminata SzyszyL, l.c. 33; Loes. in Engl. & Prantl, Nat. Pflanzenfam.

3, 5: 208 (1892); Davison, l.c. 311.

384

C. cordata Sim, l.c. 184; Davison, l.c. 300.

Maytenus cordatus Loes. in Engl. & Prantl, Nat. Pflanzenfam. ed. 2, 20b: 138 (1942).

Gymnosporia filiformis Davison in Bothalia 2: 311 (1927), may belong here. All

the flowers examined showed a 5-celled ovary with more than 2 ovules in each cell.

More material is required before this can be cleared up.

6. M. peduncularis (Sond.) Loes., l.c. 136.

Celastrus peduncularis Sond., l.c.

Gymnosporia peduncularis Loes. in Engl. & Prantl, Nat. Pflanzenfam. 3, 5: 208

(1892); Davison, l.c. 308.

7. M. bachmannii {Loes) Marais, comb. nov.

Gymnosporia bachmannii Loes. in Engl., Bot. Jahrb. 19: 232 (1894); Davison l.c.

8. M. capitatus {E. Mey. ex Sond.) Marais, comb. nov.

Celastrus capitatus E. Mey. ex Sond., l.c. 458.

Gymnosporia capitata Loes. in Engl. & Prantl, Nat. Pflanzenfam. 3, 5: 207 (1892);

Davison, l.c. 293; Loes. in Engl. & Prantl, Nat. Pflanzenfam. ed. 2, 20 b: 150

(1942).

9. M. polyacanthus {Sond.) Marais, comb. nov.

Celastrus polyacanthus Sond., l.c. 455.

Gymnosporia polyacantha SzyszyL, l.c. 34; Loes. in Engl. & Prantl, Nat. Pflanzen-

fam. 3, 5: 207 (1892); Davison, l.c. 302; Loes in Engl. & Prantl, Nat. Pflanzen-

fam. ed. 2, 20b: 150(1942).

G. vaccinifolia Conrath in Kew Bull. 1908: 211 (1908); Davison, l.c. 303.

10. M. tenuispinus {Sond.) Marais, comb. nov.

Celastrus tenuispinus Sond., l.c. 456.

Gymnosporia tenuispina Szyszyl., l.c. 33; Davison, l.c. 306; Loes., loc. cit.

G. botsabelensis Loes. in Bull. Herb. Boiss. 4: 429 (1896).

11. M. linearis {L.f.) Marais, comb. nov.

Celastrus linearis Linn, fil., l.c. 153; Sond., l.c. 455.

C. lanceolatus E. Mey. ex Sond., l.c. 456.

Gymnosporia linearis Loes. in Engl. & Prantl, Nat. Pflanzenfam. 3, 5: 207 (1892)

Davison, l.c. 305; Loes. in Engl. & Prantl, Nat. Pflanzenfam. ed. 2, 20b: 149

(1942).

G. lanceolata Loes. in Engl. & Prantl, Nat. Pflanzenfam. 3, 5: 207 (1892); Davison,

l.c.; Loes in Engl. & Prantl, Nat. Pflanzenfam. ed. 2, 20b: 149 (1942).

12. M. senegalensis {Lam.) Exell in Bol. Soc. Brot., Ser. 2A, 26: 223 (1952).

Celastrus senegalensis Lam., Encycl. Meth. Bot. 1 : 661 (1785).

Gymnosporia senegalensis Loes. in Engl. & Prantl, Nat. Pflanzenfam. 3, 5: 207

(1892); Davison, l.c. 320; Loes. in Engl. &. Prantl, Nat. Pflanzenfam. ed. 2, 20b:

147, 149 (1942).

G. dinteri Loes. in Bull. Herb. Boiss., Ser. 2, 3: 823 (1903).

385

13. M. mossambicensis (Klotzsch) Blakelock in Kew Bull. 1957: 37 (1957) var. raossam-

bicensis.

Celastrus mossambicensis Klotzsch in Peter’s Reise Mossamb. Bot. 112 (1861).

Gymnosporia mossambicensis Loes. in Engl., Bot. Jahrb. 17; 547 (1893); in Engl.

& Prantl, Nat. Pflanzenfam. ed. 2, 20b: 152^(1942).

Celastrus concinnus N.E. Br. Kew Bull. 1906: 16 (1906). Cr

var. ruber {Harv.) Blakelock, l.c.

Celastrus ruber Harv. in Harv. & Sond., El. Cap. 2: 592 (1862).

Gymnosporia rubra Loes., l.c.; Davison, l.c. 309; Loes. in Engl. & Prantl, Nat.

Pflanzenfam. ed. 2, 20b; 150(1942).

14. M. nemorosus (E. & Z.) Marais, comb. nov.

Celastrus nemorosus E. & Z., l.c. 129, No. 938; Sond., l.c. 460.

Gymnosporia nemorosa Szyszyl., tom. cit. ; 35; Davison, tom. cit.; 316.

15. M. cymosus {Soland.) Exell, l.c. 222.

Celastrus buxifolius Linn., Sp. PL: 197 (1753) pro parte, excl. Tab. Plukenet;

Sond., l.c. 459; non Maytenus buxifolius Griseb.

C. ellipticus Thunb. in Hoffm., Phytogr. Blaett. 1; 22 (1803); Thunb., FI. Cap.

119 (1818); Sond., l.c. 458; non Maytenus ellipticus Krug. & Urb. ex Duss. (1897).

C. cymosus Soland. in Sims, Bot. Mag. t. 2070 (1819).

C. heterophyllus E. & Z., l.c. 120, No. 943 ; Sond. l.c.

C. angularis Sond., l.c. 460.

Gymnosporia buxifolia Szyszyl., l.c. 34; Loes. in Engl. & Prantl, Nat. Pflanzenfam.

3,5: 207(1892); Davison, tom. cit. : 317; Loes. in Engl. & Prantl, Nat. Pflanzen-

fam. ed. 2, 20b: 152 (1942).

G. woodii Szyszyl., l.c. 35; Loes., l.c.

G. heterophylla Loes. in Engl. & Prantl, Nat. Pflanzenfam. 3,5: 207 (1892); Davison,

l.c.; Loes. in Engl. & Prantl, Nat. Pflanzenfam. ed. 2, 20b: 152 (1942).

G. condensata Sprague in Kew Bull. 1906; 246 (1906); Davison l.c. 305.

G. angularis Sim, l.c. 186; Davison, l.c. 314.

G. elliptica Schonl. in Bot. Surv. Mem. S. Afr. 1: 73 (1919); Davison, l.c. 306.

G. angularis var. orbiculata Davison, l.c. 316.

G. angularis var. grandifolia Davison, l.c.

G. cratoegiflora Davison, l.c. 314; Loes., l.c. in syn.

G. uniflora Davison, l.c. 294.

Solander, in his manuscript in the library of the British Museum of Natural History,

gives the following reference; “ H.S. 261. 34. Habitat ad Prom. b. spei (Desmaretz.)”

Vol. 261 in the Sloane Herbarium contains plants collected by Desmarest at the Cape.

The specimen on p. 34 must be the type of Celastrus cymosus Soland., and not the

specimen figured on Bot. Mag. t. 2070. I can trace no specimen of the figured plant,

which seems to be nearer to the form described as “ Gymnosporia angularis ”.

It has not been possible to ascertain the identity of all the species described from

South African plants. Those not identified are as follows: —

G. eremoecusa Loes. in Engl., Bot. Jahrb. 41 : 299 (1908).

The type of this has been destroyed. Judging from the description I have no doubt

that it is the same as M. senegalensis (Lam.) Exell.

386

G. crenulata Engl., Bot. Jahrb. 10: 38 (1888).

Type destroyed. I have examined several sheets that have been called by this name,

but they all belong to either M. cymosus (Soland.) Exell or M. senegalemis (Lam.)

Exell.

G. acanthophora Loes. in Engl., Bot. Jahrb. 41: 299 (1908).

The type has been destroyed. All material that I have seen under this name belongs

to M. cymosus (Soland.) Exell.

Maytenus rudatisii (Loes.) Loes. in Engl. & Prantl, Nat. Pflanzenfam. ed. 2, 20b: 138

(1942). G. /Wat/m Loes. in Engl., Bot. Jahrb. 41 : 307(1908).

I have not been able to locate the type specimen; it has probably been destroyed.

From the description it seems to be related to Cassine eucleaeformis (Sond.) O. Ktze.

G. schlechteri Loes. in Bull. Herb. Boiss. 2: 193 (1894).

Schlechter 241 differs from M. cymosus (Soland). Exell in the very lax and slender

inflorescences, and its leaves are somewhat narrower than those of other specimens

from the Cape Peninsula. It seems remarkable that no other material like it is available

from such a well-collected area, and 1 am inclined to regard it as just another local

form of the extremely variable M. cymosus (Soland.) Exell.

G. integrifolia (L.f.) Glover in Ann. S. Afr. Mus. 9 : 206 (1915).

I have not yet been able to examine any authentic material. All the specimens

quoted by Davison were misidentified ; most of them belong either to Maytenus

cymosus (Soland.) Exell or to Putterlickia pyracantha (L.) Endl.

G. saxatilis (Burch.) Davison in Bothalia 2: 303 (1927). Celastrus saxatilis Burch.,

Trav. Int. S. Afr. 2: 264 (1824).

Burchell cites only his No. 1671, which has 6 ovules per locule and therefore belongs

to Putterlickia. In making the new combination, Davison added 3 more Burchell

specimens. One of these is a sterile scrap, and both the others are, like the type, Putter-

lickia pyracantha (L.) Endl.

387

A New Genus of Gramineae.

By

B. de Winter.

Diandrochloa de Winter, genus nov. (Eragrosteae), affinis Eragrosti ciliari (L.) R. Br.

et al., sed ita differt, ligula membranacea, palearum carinae leves vel scabrae,

flores staminibus binis.

Spikelets laterally compressed, awnless, 2-14-flowered solitary, pedicelled, laterally

placed in relation to the rhachis, not exceeding 3-5 mm. in length; rhachilla disarti-

culating between the florets and above the glumes, glabrous or scabrid, not produced

beyond the uppermost floret. Florets hermaphrodite, falling entire with a segment of

the rhachilla attached; callus very short, rather swollen, truncate, glabrous. Glumes

persistent, unequal to subequal, membranous, often sub-hyaline, ovate to lanceolate,

keeled, apex rounded or acute, one-nerved, nerve prominent and often green. Lemmas

0 ■ 5-1 • 5 mm. long, strongly keeled, often depressed between the keels, usually translucent

or thinly coriaceous, pallid, greenish or flushed with purple, rounded, emarginate or

acute or sometimes somewhat erose, lanceolate to broadly ovate-oblong when flattened,

3-nerved ; nerves distinctly raised, very often dark green, glabrous or scabrid, midnerve

percurrent, lateral nerves evanescent. Palea subequal to the lemma, membranous,

broadly-oblong when flattened, 2-keeled nerves strongly developed in the lower part

of the keels, evanescent upwards, smooth or scabrid, apex truncate or rounded or

3-lobed. Lodicules 2, truncate, tapering downwards. Stamens 2, anthers broadly

oblong to rotund, 0-2-0 -4 mm. long, anther thecae slightly divergent, attached more

or less centrally. Ovary glabrous, styles terminal, free, stigma plumose. Caryopsis

obovate-oblong to broadly spindle-shaped, brown, smooth and semitransparent when

mature; hilum basal, punctiform; embryo ^ to nearly \ the length of the grain, disc

shaped, elliptic-oblong; starch grains simple.

Hygrophyllous caespitose annuals or perennials, culms rather soft, hollow, pallid,

greenish or purplish, geniculate or erect, few to many- noded, branched or simple;

basal sheaths rather loose, pallid-green or pinkish; leaf blades linear, flat, occasionally

flaccid, tapering to a setaceous point; panicles usually rigid, always much longer than

broad, contracted and dense or much branched and divaricate, branches in pseudo-

whorls on the strongly developed central axis.

Type species; Diandrochloa namaquensis (Nees) de [Winter. Species about 7.

Natives of the Americas, Australia, Asia and Africa.

It is surprising that this genus has not before been recognised as distinct. Possibly

this is because it is represented in most countries by only one or two species. It was

not even placed in a distinct section of its own within the genus Eragrostis but was

always associated with species such as E. tenella and E. ciliaris which it resembles

superficially. E. tenella and related species have ciliate ligules and the keels of the

palea are fimbriate. These species should not therefore be confused with any of the

species of Diandrochloa.

According to Dr. J. M. J. de Wet of the Division of Botany the chromosomes of

E. namaquensis are smaller than those of the other species of Eragrostis which he has

investigated. The basic number of 10 with the genome of 40 chromosomes is typical

of Eragrostis and its allies.

388

The following species of Eragrostis should be transferred to the genus Diandrochloa:

E. glomerata (Watt) Dewey (S. & N. America); E. confertiflora Black (Australia);

E. japonica (Thunb.) Trin. (Asia); E. diarrhena Steud. (India); E. diplachnoides

Steud. (Africa and Asia). The synonymy of most of the species listed above is very

complicated and in some the status of the taxa is uncertain. For this reason I have

refrained from making the combinations for species outside South Africa.

There are only two South African representatives of the genus Diandrochloa

namely; —

(a) Diandrochloa namaquensis (Nees) de Winter, comb. nov. {Eragrostis nama-

quensis Nees).

{b) D. pusilla (Hack.) dc Winter, comb. nov. {Eragrostis pusilla Hack.)

They can be distinguished as follows: —

Spikelets 1-1-5 mm. long, 2^-flowered ; lemmas -5- -6 mm. long; inflorescence

branches with spikelets patently spreading when fully developed. ... D. pusilla.

Spikelets 2-3 mm. long, 4-8-flowered; lemmas about 1.0 mm.; inflorescence

branches somewhat contracted D. namaquensis.

1. D. namaquensis (Nees) de Winter

Eragrostis namaquensis Nees in Linnaea 12, 542 (1838) and FI. Afr. Austr. 1.408

(1841); Chiov in Ann. Inst. Bot. Roma 8, 63 (1903), do. l.c. 363 (1908); do. in

Nuovo Giorn Bot. Ital. n.s. 19, 423 (1912); de Winter in Grasses and Pastures

S. Afr. 182 (1955). E. namaquensis var. robusta Stapf. in FI. Cap. 7, 630 (1900).

Catabrosa micrantha Hochst. ex A. Rich. Tent. FI. Abyss. 2, 426 (1851) in synonymy.

Annual or subperennial, caespitose, culms erect 15-130 cm. high, hollow, rather

soft, glabrous, pallid or greenish, finely striate, 1-6-noded, simple or branched upwards,

upper internode exceeding the others and usually long exserted; sheaths striate, lax

and slipping from the culms, pallid or flushed with purple, especially towards the base;

leaf blades linear and tapering to a fine point, flat or somewhat rolled, 5-25 cm. long;

up to 5 mm. broad, nerves fine, slightly raised, scabrid; ligule membranous, about

0-5 mm. long. Panicle tightly contracted and linear or open and lax, 10-60 cm. long,

erect, branches solitary or approximate and in pseudowhorls, erect and appressed or

obliquely spreading, repeatedly divided from the base, subdivisions subcapillary,

glabrous or scabrid. Spikelets pedicelled, more or less crowded on the branchlets,

elliptic to oblong-elliptic, obtuse, 2-3 mm. long, 4-8-flowered, purplish or greenish-

brown, rhachilla disarticulating above the glumes and between the florets, smooth,

glabrous. Florets falling entire, hermaphodite; callus glabrous, slightly obliquely

truncate. Glumes unequal to subequal, broadly oblong, obtuse and emarginate, about

0-75 mm. long, 1-nerved, persistent. Lemmas broadly ovate-oblong when flattened,

obtuse, emarginate, about 1-0 mm. long, strongly keeled, 3-nerved, slightly depressed

between the nerves, nerves smooth. Palea subequal to the lemma, falling with the lemma,

keels smooth. Lodicules truncate, cuneate, very small. Stamens 2, anthers 0-3-0 -4

mm. long, purple, broadly oblong. Ovary glabrous, styles free, stigmas plumose.

Grain obovate-oblong; hilum basal, punctiform; embryo 2/5 of the length of the

grain, rotund.

Very widely distributed in Africa, West Tropical Africa, British East Africa and

South Tropical Africa. Also found in all the provinces of South Africa including S.

West Africa.

This grass prefers moist localities and inhabits river banks and pools. Usually

an annual, it is apparently sometimes a weak perennial. It varies tremendously in

size, and can be from 15 cm. to over 130 cm. high. The shape of the inflorescence is

also variable.

Holotype: Namaqualand, banks of Orange River, Drege 2569 (B; PRE, fragment).

389

Cape Province. — Port St. Johns: Umzimvubu River, Sidey 567; Mt. Frere:

Acock s \2>565-, Mafeking: Brueckner 367 ; Hay: Langebergen, /Jcoc/rj' 8531 ; fountain

at Rietkloof, Acocks 8534.

Natal. — Vryheid: White Umfolosi, Curson 148; mountain top, sides of stream,

Hlohane, Johnstone 423] Hlabisa: Hluhluwe, river banks, 2581.

Transvaal. — Potchefstroom : Vaal River, Nooitgedacht, Louw 1702; Pretoria:

Petronella, Acocks 11713; 19 m. N.E. of Premier Mine, Codd 2764; Rustenburg:

Elandskraal, de Winter 232 B; Ermelo: Morgenson, Bose Innes s.n.; Nylstroom:

Warmbad, Schweickerdt 1786; Potgietersrust: Mosdene, Galpin M. 539; Barberton:

Bolus 9794; Kruger National Park: Numbi, v. d. Schyff2651] Shangani, v. d. Schyff,

2820; Punda Maria, Codd 5363; Soutpansberg: Tshokoma, Obermeyer s.n. Tvl.

Mus. No. 31657.

Swaziland. — Mbabane: Mbeluzi Falls, near stream, Compton 25086.

Bechuanaland. — Makarikari Basin: Nata River, Van Son, Tvl. Mus. No. 28614.

South West Africa. — Tsoachaubsand Salem, Dinter 140; Otjiwarongo: Water-

berg, Vlok 2240; Stream near Great Waterberg, Rodin 2568; Kaokoveld: banks of

Kunene, Story 5859.

Angola. — Boss s.n., Tvl. Mus. No. 36763.

Southern Rhodesia. — Odzani Valley, Eyles 1685; Zambesi River, Hitchcock

24190,24345; Cleveland Dam, GiV/zVontf 46; Sabi-Lundi Junction, Wild 3339.

Northern Rhodesia. — Matonchi Farm, Milne Redhead 2758; Kashitu, Stohr

2845,2844,2834; River Mekeba, Milne Redhead 7 59] between Livingstone and Kaloma.

Pole Evans 2797 (4); Broken Hill, Rogers 26093.

Portuguese East Africa. — Tumbini slopes, Hornby 3372.

Nyasaland. — Mlanje, Palombe Plain, Laurence 376; Benga, Lake Nyasa, Brass

17497; Mperere Mission, Chankalamu Dambo, Jackson 242.

Tanganyika — Ardai Plain, Greemvay 7014; Ngudu, Mwanza, Lewys Lloyd 16;

Lindi, Mikindani, Schlieben 6555.

Belgian Congo. — Elizabethville, Rogers 26271; Quarre 4570, 5530.

Abyssinia. — Scholoda Mnt., Schimper 406 (fragment).

2. D. pusilla (Hack.) de Winter.

Eragrostis pusilla Hack, in Bull, Herb. Boiss. 4, app. 3, 27 (1896); de Winter in

Grasses and Pastures of South Africa, 184 (1955).

Caespitose annual up to 30 cm. high, culms erect or slightly geniculate, 1-3 (usually

l)-noded, soft and compressible, pallid, finely striate, simple or branched; sheaths

chartaceous, rather loose, pallid, greenish or brown, margins membranous; ligule

a membranous rim; leafblades linear, thin, flat, tapering to a fine setaceous point,

4-13 cm. long and up to 0 - 5 mm. broad, primary nerves 5-7, nerves glabrous or scabrid.

Panicle densely and divaricately branched, 8-15 cm. long, narrowly elliptic to narrowly

oblong in outline, central axis strongly developed and rigid, branches clustered together

in pseudowhorls especially lower down or solitary upwards, spreading obliquely or

nearly at right angles, repeatedly and finely branched from the base upwards, divisions

glabrous or sparsely scabrid. Spikelets awnless, more or less broadly oblong 1-1-5

mm. in length, 2-4-flowered, brownish to purplish, pedicelled, pedicels 0-5-1 -5 mm.

390

long; rhachilla disarticulating above the glumes and between the florets, glabrous.

Florets hermaphrodite, falling with an internode of the rhachilla attached, callus trun-

cate. Glumes unequal, lower narrower and shorter than the upper, oblong, one-nerved,

obtuse or slightly emarginate, persistent. Lemmas broadly oblong, obtuse, emarginate,

0-5-0 -6 mm. long, 3-nerved, keeled, depressed between the nerves and membranous,

nerves smooth, strongly raised. Palea subequal to and falling with the lemma, strongly

keeled, keels smooth. Lodicules truncate, oblong-cuneate. Stamens 2, anthers 0.3-

0-4 mm. long, broadly oblong. Ovary glabrous, styles free, stigmas plumose. Caryop-

sis obovate-oblong, hilum basal, punctiform, embryo more or less 2/5 the length of

the grain.

This is a small delicate and rather decorative hygrophyte, inhabiting the margins

of seasonal pools in the subtropical drier areas of the Transvaal, Cape, South West

Africa and Bechuanaland and probably also occurring in S. Rhodesia and Angola.

Holotype: Kalahari: in stagnis exsiccatis prope Uugua, May, 1891’ Fleck 321,

(Herb. Hackel, Vienna; fragment in PRE).

Cape Province. — Kimberley: Riet Pan, near Riverton, Reinhardt 3228; Vryburg:

farm Welgelegen, Pentz s.n., N.H. No. 14876.

Transvaal. — Bloemhof: Louw 1823; Soutpansberg: Farm Hamilton, de Winter

& Codd 313.

South West Africa. — Maltahohe: Farm Urusis, Kinges 2548; Gibeon: Fish

River Gorge, F. r/'ot/za s.n.; Rehoboth: Kalkrand, z/e 3521 ; Karibib: Farm

Otiimbojo East, Kinges 3392; Okahandja: Waterberg, Bradfield 270; de Winter 2807;

Okahandja River, Dinter 117; Grootfontein : Nosib, Schoenfelder S. 878; Nosib

road, Dinter 7433; Tsumeb, Dinter 7432. Okavango: Omuramba Omatako, Seiner

685.

Most of the specimens named E. namaquensis Nees var. uninodis Hack, are in fact

E. pusilla. I have not been able to trace a valid publication of the var. uninodis.

391

The Opening of the Pretoria National

Botanic Garden.

By

R. A. Dyer.

After my return in 1934 from the Royal Botanic Gardens, Kew, I cherished the

idea of a botanic garden attached to the National Herbarium, Pretoria. Kirstenbosch

served the needs of the winter rainfall region of the Union, and a similar institution

to serve the summer rainfall area was an obvious necessity. That it should be com-

plementary to the National Herbarium also seemed obvious.

In 1944, towards the end of the war, the idea was given its first airing, in connection

with the training of student gardeners, but the time was not propitious. In 1945 the

botanical survey section of the Division of Botany was resuscitated with Dr. L. E. Codd

in charge. The question of a new botanic garden was then brought into the open and

the environs of Pretoria were explored for a suitable area.

At this juncture contact was made with Professor H. B. Davel, then director of the

Agricultural Research Institute of the University of Pretoria. Together we inspected

the piece of the University farm cut off from the main block by the Silverton and Brum-

meria roads, which had been found unsuitable for farm experimental purposes owing to

its isolation and to the presence of an abundance of gifblaar, Dichapetalum cymosum.

Other areas were considered including a portion of Rietondale Experimental Station,

but the University ground was favoured above all others and this project gained the

support of the then Minister of Agriculture Mr. J. G. N. Strauss.

The University Council gave formal approval of the development of the area as

a botanic garden in June, 1946. Soon afterwards Mr. Jan Erens, head gardener at

the Division of Botany, dre>v up a rough plan of the area, planted the first avenue of

experimental trees, supeivised the sinking of boreholes and saw to the building of 2

small reservoirs.

The decision in 1947 to purchase the western corner of portion F of Koedoespoort

farm 299 proved abortive, owing to a ridiculously high price being placed on the property

by the owner.

Mr. Erens was promoted into the Public Works Dept, in 1949 and Mr. A. van der

Ende later took over his duties at the Division of Botany at Vredehuis, while Mr. J.

Admiraal was appointed to the gardening staff and placed in charge of the new botanic

garden.

In 1951 the University Council donated 200 acres on the opposite (south) side of

the Brummeria road to the C.S.I.R. for its head office and laboratories. The Division

hoped to be treated in similar fashion, but the University Council decided to retain the

botanic garden ground under the Agricultural Research Institute.

In 1954, the Bureau of Standards negotiated with the University for the purchase

of the botanic garden ground, and interested parties suggested that an area in the

Fountains Valley or the Wonderboom Reserve would serve equally well as a botanic

392

garden. A Commission consisting of Professors W. J. Lutjeharms, A. W. Bayer, R. S.

Adamson and N. P. Badenhuizen, reported strongly in favour of the existing project and

their recommendation was accepted by the Minister of Agriculture at the time, Mr.

S. P. le Roux and the University Council.

The building of a residence on the botanic garden for the gardener in charge was

sanctioned in 1955, and was soon completed under the guidance of the P.W.D.

With the support of the Minister, Mr. le Roux, the Secretary for Agriculture,

Dr. C. H. Neveling and the Director of Technical Services Dr. M. S. du Toit, we suc-

ceeded, in a matter of 13 days, in obtaining Treasury authority to purchase, on 15/10/55,

portion G of farm Koedoespoort No. 299, nearly 8 morgen in extent, with northern

aspect, to add to the botanic garden.

Progress has been most encouraging and the present Minister of Agriculture

(Technical Services) Mr. P. K. le Roux, accompanied by Mrs. le Roux, opened the

garden officially on 23rd October, 1958.

By the greatest good fortune, considering the unsettled weather at the time, the

morning of the 23rd was perfect for the outdoor ceremony. A Cabinet Meeting called

for 11 a.m. on the 23rd, threatened to wreck the opening ceremony, but all obstacles

faded away at the last moment.

Mr. le Roux addressed the gathering of about 150 guests, speaking for the first half

in Afrikaans and then in English. The text of his address, with a translation, was as

follows : —

,, Die grond waarop ons vanmore staan is ’n deel van die Pretoria Universiteits-

plaas — ’n deel van die plaas wat vir die Fakulteit van Landbou opsygesit is. Effens

oor die tien jaar gelede het die Universiteitsraad toegestem dat ’n botaniese tuin op

hierdie terrein tot stand gebring kon word. Plantkunde, soos u sal begryp, is een van

die basiese kursusse in die eerstejaars studiekursus van alle studente wat Landbou loop,

dus is daar alle rede om noue samewerking tussen die Fakulteit en die Afdeling Plant-

kande te he.

Dit doen my groot genoee om op hierdie geskiedkundige geleentheid, naamlik die

opening van hierdie botaniese tuin, die Rektor — Prof. Rautenbach — en die Universi-

teitsraad vir hulle versiende beleid om hierdie ontwikkeling aan te moedig, te bedank.

Die grond is ongeveer 68 morg groot. Deur slegs rond te kyk sal u besef wat ’n uit-

stekende figging dit as botaniese tuin het. Dit is maklik bereikbaar daar dit op die

nasionale pad van Pretoria na die Oostelike Transvaal en die Kruger Wildtuin gelee is.

A1 hoe meer oorsese besoekers sal langs hierdie roete reis en hierdie vertoonvenster van

ons Suid-Afrikaanse flora sien. Dit sal gedurig in die oog van die publiek wees.

Daar is noordelike en suidelike hange; daar is ’n groot golwende gebied wat uit

diep sanderige grond bestaan, en daar is hierdie mooi koppie agter my wat die hele

terrein oorheers en vanwaar ’n groot deel van Pretoria sigbaar is. ’n Mens kan ook na

die imposante reeks geboue wat deur die Wetenskaplike en Nywerheids Navorsingsraad

oorkant die pad opgerig is, tuur, en peins omtrent wat alles in hulle gedoen word.

Ja, hierdie is waarlik ’n stimulerende omgewing en ons kan verwag dat hierdie tuin

in die jare wat voorle iets om op trots te wees sal word.

Plantkunde in die bree sin van die woord ken geen grens tussen sulke onderwerpe

soos Tuinbou, Plantsiektekunde, Weidingsnavorsing, Veldbeheer, Plantfisiologie en

Planteteelt nie. Plantkunde dien hulle almal. Wanneer ’n nuwe gebied ontgin word

is een van ons eerste pligte om ’n opname van die inheemse plantegroei te maak. Die

samestelling van die natuurlike onbeskadigde veld toon beste aan hoe die land gebruik

moet word om die maksimum landboukundige opbrengs daaruit te verkry.

393

My vriende wat boer is sterk bewus van die aanhoudende pogings van die Departe-

ment om die land meer bekwaam te gebruik. Hulle is bewus van die verslag oor hierdie

onderwerp wat deur die Afdeling Ekonomie en Markte uitgegee is. Veldwerkers van

die Afdeling Plantkunde het ’n belangrike bydrae tot hierdie werk gelewer deur veld-

tipes en hulle verspreiding af te baken.

Die boer self moet sy veld ken. Hy moet soveel as moontlik van die afsonderlike

plante, wat sy veld vorm ken. Hy moet die goeie voerplante, wat aangemoedig en

teen oorbeweiding beskerm moet word, ken. Hy moet ook die giftige plante, wat

gedurig ’n bedreiging vir sy vee is, ken. Die Afdeling Plantkunde het bale inligting

omtrent hierdie onderwerpe versamel, maar sal in hierdie ontsaglik studieveld nooit

alles kan weet nie. Ons kennis is nooit volledig nie. Plante uit die veld sal in hierdie

tuin vir verdere studie gekweek word en die inligting aldus verkry sal soos in die verlede,

vir die boeregemeenskap en algemene publiek, beide in die Unie en oorsee, beskikbaar

gestel word.

Die geskeidenis van Landbou in die Unie gaan terug na die aankoms van Jan

van Riebeeck aan die Kaap in 1652. Een van sy eerste pligte was om ’n tuin vir die

kweek van groente, om handelsskepe van die Hollands-Oos-Indiese Kompanjie daar-

mee te voorsien, tot stand te bring. Die geskiedenis van die Landbou in Transvaal

dek egter ’n bale korter tydperk en begin kort voor 1900. Sommige van die vroee

gebeurtenisse is van algemene belang en toepaslik op hierdie geleentheid. In 1896

het die Volksraad ’n veearts, Arnold Theiler, aangestel om die Runderpes, wat toe

gewoed het, te bestry. ’n Plantkundige, Burtt Davy, is in 1903 aangestel, dus het die

Departement van Landbou in die Transvaal met die aanstelling van ’n veearts en ’n

plantkundige begin. Die plig van die plantkundige was om in die veld uit te gaan,

boere te ontmoet, en die landboubehoeftes van die verskillende streke te bespreek,

om die voerplante van die natuurlike veld te bestudeer, sowel as die giftige plante,

skadelike onkruid, ens. Daarby moes hy verteenwoordigende versameling van hierdie

plante wat die kern van ’n herbarium sou vorm, bewaar.

’n Plantsiektekundige is in 1905 tot die span gevoeg om graansiektes te bestudeer.

Na Unie was die ontwikkeling vinniger en is die staf aansienlik uitgebrei. By die

funksies van die Afdeling Plantkunde en Plantsiektekunde is Tuinbou, Akkerbou,

Weiding, Insektekunde, Grondbewaring en Uitbreiding gevoeg, en die hele organisasie

het die Afdeling van Plantnywerheid geword. By die hoofkwartier te Vrede Huis was

slegs ’n klein botaniese tuin en bykomstige proefstasies om inheemse en ingevoerde

plante te ontvang en te toets ten opsigte van hulle landboukundige moontlikhede, is

opgerig.

Die skepper van die Afdeling van Plantnywerheid, Dr. I. B. Pole Evans het in

1939 op pensioen afgetree, en dit is toe besluit dat die leiding van die verskillende lyne

van navorsing wat hy tot stand gebring het onder onafhanklike Hoofde van Afdelings

voortgeset wou word. Op die manier het Tuinbou, Akkerbou en Weiding, Insekte-

kunde, Grondbewaring en Uitbreiding afgeskei en die Plantkunde en Plantsiektekunde

seksies was weer op hulle eie. ’n Tydjie later het Plantsiektekunde ook mondig geword

en ’n Afdeling op sy eie geword.

Nadat hy hierdie satelliete met welslae afgevuur het, kon die Afdeling Plantkunde

sy aandag skenk aan ’n meer eng botaniese navorsingsveld, en so is dit teweeggebring

dat plantkundiges in die Nasionale Herbarium, wat grootliks besiggehou word met die

opname en klassifikasie van ons inheemse flora, gevoel het dat ’n botaniese tuin, wat

die naam waardig sou wees, waarin lewende plante gekweek en waargeneem kan word —

’n tuin waarin die moontlikhede van elke soort afsonderlik bestudeer kan word —

noodsaaklik geword het. Tegelykertyd sou dit ’n opvoedingssentrum en skoonheids-

oord word.

394

Op hierdie tydstip is die verbond tussen die Afdeling Plantkunde en die Universiteit

van Pretoria, waarna ek in die begin verwys het, gemaak.

Sedert die vroegste tye is plante in tuine versorg. Die vroegste tuine het hoof-

saaklik te doen gehad met die kweek van geneeskragtige kruie en was werklik die eerste

wetenskaplike plantkundige inrigtings. Selfs vandag, hoewel daar so ’n menigte van

die nuwe sintetiese medisynes is, is daar nog ’n diep belangstelling in navorsing van

die wereld se plantegroei met die oog op die ontdekking van nuwe medisynes of nuwe

bronne van bekende samestellings. Suid-Afrika dra reeds ’n belangrike deel by deur

die uitvoer van aalwyn, boegoeolie en diosgenin, wat uit Dioscoria verkry word vir die

vervaardiging van cortisone, om slegs drie voorbeelde te noem. Op die huidige tydstip

is daar ’n skema wat deur handelsbelange in Amerika ondersteun word, om ’n pos vir

’n Afrikanerskeikundige wat die plantegroei van die Unie meer intensief as ooit tevore

sal bestudeer, te skep. Die persoon aan wie die pos geskenk word sal van beide die

W.N.N.R. en die Afdeling Plantkunde leiding ontvang, en in hierdie tuin waarin ons

vanmore saamgekom het, sal plante deur horn versamel en gekweek word.

Wanneer ons die taamlik onlangse geskiedenis van botaniese tuine beskou vind

ons dat baie van hulle dwarsdeur die wereld ’n belangrike rol in die toets en verspreiding

van plante wat van ekonomiese belang is, gespeel het. Daar is die beroemde geval

van die plantkundige wat toevallig ’n paar sade van die rubberboom in sy natuurlike

omgewing in Suid-Amerika versamel het, en hulle na Kew Botaniese Tuin in Engeland

gestuur het. Die jong rubberbome wat daaruit gespruit het is toe na die Singapoer

Botaniese Tuin gestuur, en uit hierdie klein begin is ’n industrie met ’n geweldige waarde

vir die mens, en bepaald baie miljoen pond werd, opgerig. Botaniese Tuine van die

wereld het 'n ewe belangrike rol in die ontwikkeling en uitbreiding van die katoen,

kiena en tee-industriee gehad, en uit Australie het ons die voorbeeld van die uitgebreide

koringnywerheid wat ontstaan het vanuit ’n paar sade wat na die Sydney Botaniese

Tuin ingevoer is. Ek kan u ook as ’n voorbeeld noem een van ons eie veldgewasse,

wat ’n belangrike plek in die landbou buite ons grense ingeneem het, deur die toedoen

van wyle Dr. Maria Wilman wie op die flora van Griekwaland-Wes gewerk het. Dr.

Wilman het saad van een van ons gewone veldgrasse — een wat destyds geen groot naam

in die Unie gehad het nie — na Amerika gestuur. Hierdie gras, Eragrostis curvula,

het spoedig in die Amerikaanse landbou in die guns geraak, met die gevolg dat dit

vandag as weiding en grondverbetering oor groot landstreke gesaai word, en in die

algemeen die “ Wilman Love Grass ” genoem word. Die Amerikaners het reeds

verbeterde tipes daarvan by hulle navorsingstasies voortgebring. Die waarde van

hierdie gras word nou ook in die Unie meer besef en in die Griekwaland-Oos distrik

alleen is reeds meer as 10,000 akker binne die afgelope 8 jaar met Love Grass beplant.

Die plantkundige in sy Botaniese tuin was bepaald al in menige ekonomiese belangrike

landboukundige sake betrokke.

Ek het met opset u aandag op hierdie paar voorbeelde gevestig, om u te herinner

hoe nou die landbou en ons alledaagse voorspoed met plantkunde verbonde is. Sedert

die tydperk toe dit slegs van mediese belang was, het die funksies van ’n botaniese

tuin egter geleidelik verander. Dit is nog ’n setel van opvoeding en geleerdheid, maar

geleidelik het die estetiese aspek na vore getree. Laasgenoemde is ook van groot belang

wanneer ons sy invloed op die opbou van ons nasionale kultuur oorweeg. ’n Op-

gevoede gemeenskap moet besef dat dit sy plig is om op intelligente wyse in die flora

van sy land belang te stel. En wat is daar in ons flora om ’n gevoel van nasionale trots

aan te wakker?

In die eerste plek kan ons daarop aanspraak maak dat die Unie van Suid-Afrika,

in verhouding tot sy grootte een van die rykste floras in die wereld het. Hier word

oor die 20,000 soorte plante gevind. Om ’n paar statistieke te noem het ons ongeveer

500 soorte Erica, 300 soorte Protea en sy verwante, oor 880 soorte grasse, en heelwat

meer as 1 ,500 soorte vygies en ander vetplante.

395

Die versamel en kweek van vetplante, wat vanaf die vroegste dae toe ontdekkings-

reise in die Kaap gedoen is, populer was, is vandag nog ’n stokperdjie waarin baie

spesialiste verdiep is. Daar is ’n aanhoudende belangstelling en handel in hierdie

groep, waarin die eindelose verskeidenheid van vorm en kleur altyd ’n bron van be-

wondering is. Daar is die Aalwyne, die vensterplante wat half begrawe is in harde

grond, die klipagtige plante wat op die oppervlakte groei en so harmonies met hulle

omgewing saamsmelt dat hulle besonder moeilik is om raak te sien, en ’n menigte

ander eienaardige vorms. Nerens in die wereld is daar ’n vetplante-flora so ryk soos

ons s’n nie.

In sommige droe dele van Amerika is daar ’n groot verskeidenheid cacti, ’n groep

wat in die Unie afwesig is; maar die vetplant-flora van die Unie is nie alleen geweldig

ryk in soorte nie maar het ook verteenwoordigers in ’n aantal heeltemal onverwante

families waarin parallelle ewolusie plaasgevind het.

Maar dit is nie al nie. Die Unie het ’n aansienlike bydrae tot botaniese tuine en

tuinbou dwarsdeur die wereld gelewer, vanuit die seldsame en pragtige flora waarmee

die natuur ons so mildelik bedeel het.

Selfs voor die eerste blanke nedersetting aan die Kaap ongeveer 300 jaar gelede,

het daar reeds ’n klein handel in Kaapse plante met ’n aantal koninklike tuine in Europa

ontstaan. Matrose wat in die tuinbou belanggestel het, het saad en bolle versamel

wanneer hulle tydelik geland het. Hoewel van Riebeeck die eerste nedersetters aan die

Kaap streng beveel het om nie verder as die Skiereiland te reis nie, is hierdie perke

spoedig oorskry. Later het die twee Goewerneurs, Simon en Adriaan van der Stel,

vader en seun, spesiale ontdekkingsreise na die binneland georganiseer om onder andere

plante en saad te versamel om die beroemde tuine van Holland te verryk.

Gedurende die agtiende eeu was daar ’n standvastige vooruitgang in die invoer van

Suid-Afrikaanse plante in Europa totdat, aan die einde van die eeu, dit gese kon word

dat ons Ericas, Proteas, Stapelias en baie bolplante byna die kweekhuise van nie alleen

Holland, maar ook van Engeland, Oostenryk, Swede, Italie en Frankryk oorheers

het, Tussen die besitters van tuine was daar skerp mededinging om die rykste versame-

ling uitheemse plante te mag besit. Die Kaapse flora was ’n ontsaglike ryk bron om

uit te put.

Gedurende die neentiende eeu het die ambisie om nuwighede te kweek bedaar,

en ons Flora het gedeeltelik sy aantrekkingskrag verloor, egter nie voor sulke plante

soos Agapanthus, Nerine, Gladiolus, Pelargonium en vuurpyle, varkblomme en seifs

sommige van die Namakwalandse madeliefies, blywende plek in die tuine van Europa

en die Nuwe Wereld ingeneem het nie. Terwyl hierdie Suid-Afrikaanse plante oorsee

makgemaak is, het hulle die aandag van plantekwekers getrek, met verreikende tuin-

boukundige gevolge. Onder die kruisings wat eerste in guns geraak het, was die Pelar-

goniums— gewoonlik onder die naam Geranium bekend — wat tot vandag so uitgebreid

in tuine gekweek word. U sal u herinner dat die Geranium Vereniging van Britanje

onlangs aan die Unieregering ’n versameling van die beste moderne kruisings geskenk

het. as ’n blyk van waardering aan die land van hulle herkoms. Hierdie versameling

is nou in bewaring van die Afdeling Tuinbou.

Dan is daar die rooi Barbertonse Madeliefie, wat deur seleksie en kruising ’n

byna eindelose verskeidenheid van kleure en dubbelvorms voortgebring het, en wat

die besonder goeie hoedanigheid het dat dit so lank in water kan hou. Om die waarheid

te se is daar maar min van ons gekweekte blomme wat hulle oorspronklike vorm

behou het. Besoekers aan openbare tuine oorsee dink nie eers om navraag te doen

oor die land van oorsprong van die vele tuinboukundige wondere nie.

Is dit dus nie paslik dat ons ons eie flora behoort te eer nie; dit in sy natuurlike

toestand moet bestudeer, dit in hierdie botaniese tuin moet kweek en bewaar sodat

almal dit kan sien en waardeer nie? Hier sal geen plek vir uitheemse plante wees nie.

396

Sender twyfel is daar in die teenwoordige eeu dwarsdeur die Unie ’n al gemene

oplewing van belangstelling in die kweek van ons inheemse flora. Groot stappe vooruit

is gemaak in die openbare tuine van sommige van die groter en heelparty van die kleiner

dorpe. Uitgebreide gebiede is opsygesit vir die bewaring van aantreklike en interessante

soorte. Dit werk alles ten goede en die neiging behoort soveel moontlik aang jmoedig

te word, maar uit die aard van die saak is die wetenskaplike toenadering afwesij of slegs

tot ’n beperkte mate in ag geneem. Slegs inrigtings soos die Afdeling Plantki nde met

sy staf bestaande uit opgeleide plantkundiges en tuiniers kan wetenskaplike studie met

die kweek van plante verenig.

Die Nasionale Botaniese Tuin te Kirstenbosch, wat in 1913 tot stand gekom bet,

het ’n soortgelyke organisasie — ’n tuin en ’n opgeleide plantkundige staf. Om die

waarheid te se word dwarsdeur die wereld die kombinasie van ’n botaniese tuin met ’n

herbarium en biblioteek erken as die ideale organisasie om plantkunde as wetenskap

te bevorder.

Die Nasionale Botaniese Tuin te Kirstenbosch het tans ’n wereldwye reputasie

deur sy pragtige omgewing en groot natuurlike skoonheid. Sy reputasie is onlangs

verder bevorder deur die toekenning van ’n gone medalje op ’n internasionale blomme-

skou in New York vir ’n versameling van Suid-Afrikaanse blomme wat hoofsaaklik

uit Proteas bestaan het. Kirstenbosch kan egter nie ’n beeld gee van die flora van die

hele Unie nie. Geen enkele inrigting sou dit kon doen nie.

Soos u almal weet word die Unie volgens klimaat in ’n oorwegende winterreen-

valstreek en ’n oorwegende somerreenvalstreek verdeel, met ’n betreklik klein streek

met reen dwarsdeur die jaar in die George-Knysna gebied. Soos te begryp is die flora

van die twee hoofgebiede baie verskillend. Kirstenbosch, wat ’n hoe winterreenval

geniet, is dus die mees gepaste plek om die flora van daardie gebied te bestudeer. Die

plantegroei van die somerreenvalstreek is nie daar tuis nie, en dit is hierdie afdeling wat

die meeste aandag in hierdie nasionale Botaniese Tuin sal geniet. Die Afdeling

Plantkunde wil egter nie sy ambisie verswyg om ook sukses met sommige van die Kaapse

plante te behaal nie, en aan my is meegedeel dat hierdie plan bevorder word deur Mev.

Vogts, wat self vanmore hier aanwesig is, en self in Pretoria sukses daarmee behaal het

na vroeere ondervinding aan die Kaap.

Die gedagte het seker al by meeste van u verrys dat die ontwikkeling van ’n botaniese

tuin met die bou van ’n nasie vergelyk kan word. Die stigters het hulle ideale en strewe;

hulle le die fondament, maar sien selde hulle arbeid tot voile bloei kom. Vandag le

ons dus spreekwoordelik die hoeksteen van ’n gebou wat baie jare sal neem om sy voile

grootte te bereik. As simbool van die strewe van die Afdeling Plantkunde is ek gevra

om ’n inheemse boom as aandenking aan hierdie geleentheid te plant. Die groei en

ontwikkeling van die boom wat jaar na jaar in grootte en skoonheid sal toeneem. sal ’n

inspi’rasie wees vir latere geslagte van tuiniers en plantkundiges. En met hierdie voor-

uitsig waarin ons ten voile vertroue het doen dit my genoee om hierdie Nasionale

Botaniese Tuin oop te verklaar. Ons sal nou voortgaan om die boom te plant.”

“The ground we are standing on this morning is part of the Pretoria University

farm — part of the farm set aside for the Faculty of Agriculture. It is now a little over

ten years ago that the University Council agreed to the establishment of a botanic

garden on this site. Botany, it will be appreciated, is one of the basic courses in the

first year curriculum of all students taking Agriculture, so there is every reason why

there should be close collaboration between the Faculty and the Division of Botany.

It gives me great pleasure on this historic occasion of the official opening of this

botanic garden to thank the Rector — Professor Rautenbach — and the Council of the

University for their far sighted policy in fostering this development. The area is approxi-

mately 68 morgen in extent. You have only to glance round to appreciate what an

397

excellent site it is for a botanic garden. It is readily accessible. It is on the national

road from Pretoria to the eastern Transvaal and the Kruger National Park. Along this

route more and more overseas visitors will travel and see this display window of our

indigenous flora. It will be constantly in the public eye.

In this garden there are north and south facing slopes; there is a large undulating

area of deep sandy soil, and there is this fine koppie behind me which dominates the

whole site and from which one can survey a wide expanse of Pretoria. Or one can

gaze and ponder about what goes on across the road in that imposing array of buildings

erected by the Council for Scientific and Industrial Research. Yes, this is a truly stimu-

lating environment and we can expect this garden to become a place of pride in years

to come.

Botany in its broad sense knows no boundary between such subjects as Horti-

culture, Plant Pathology, Pasture Research, Veld Management, Plant Physiology and

Plant Breeding. Botany serves them all. When a new territory is opened up one of the

first duties is to take stock of the wealth of the indigenous vegetation. It is the com-

position of the natural undamaged veld which affords the best indication of how the

land should be utilised for the maximum agricultural output.

My farming friends are well aware of the continuous efforts of the Department

for more efficient land utilisation. They are aware of the report on this subject by the

Division of Economics and Markets. The field officers of the Division of Botany

played an important part in this project in demarcating veld types and their distribution.

The farmer himself must know his veld. He must know as many as possible of

the individual plants which go to make up his veld. He must know the good fodder

plants, those to be encouraged and protected against overgrazing. He must know

also the poisonous plants which are a constant menace to his stock. The Division of

Botany has a good store of information on these subjects but can never find out all

there is to know in such a vast field of study. Our knowledge is never complete. Plants

from the veld will be introduced into this garden for further study and the information

obtained, will, as in the past, be made available to the farming community and general

public both in the Union and oversea.

The history of Agriculture in the Union dates back to the arrival of Jan van Riebeeck

at the Cape in 1652. One of his first duties was to establish a garden for the cultivation

of vegetables to supply trading vessels of the Dutch East India Company on their

way to and from the East Indies. The History of Agriculture in the Transvaal, how-

ever, covers a very much shorter period and begins shortly before 1900. Some of

the early events are of general interest and relevant to the present occasion. In 1896

the Volksraad appointed a veterinarian, Arnold Theiler, to combat the scourge of

Rinderpest which was then raging. A botanist, Burtt Davy, was appointed in 1903.

Thus it was that the Department of Agriculture in the Transvaal had its beginning

in the appointment of a veterinarian and a botanist. The duty of the botanist was to

go out into the veld, to meet farmers and discuss the agricultural needs of the different

areas, to study the fodder plants of the natural veld, the poisonous plants, noxious

weeds and any plant which might have some economic value. In addition he had to

preserve for permanent record representative collections of these plants as the nucleus

of a herbarium.

A Plant Pathologist was added to the team in 1905 to study cereal diseases.

After Union there was more rapid development and considerable staff expansion.

To the functions of the Division of Botany and Plant Pathology were added Horticulture,

Crops, Pastures, Entomology, Soil Conservation and Extension, and the whole organi-

sation became the Division of Plant Industry. There was only a small botanical garden

398

at the headquarters at Vrede Huis and additional experimental stations were established

to receive indigenous and introduced plants where they could be tested for their agri-

cultural potentialities.

The Architect of the Division of Plant Industry, Dr. I. B. Pole Evans, retired on

pension in 1939, when it was decided that the direction of the several lines of research

which he had initiated should be placed under independent Chiefs of Divisions. Thus

Horticulture, Crops and Pastures, Entomology, Soil Conservation and Extension

hived off leaving the Sections of Botany and Plant Pathology on their own again. Some-

time later Plant Pathology also attained its majority and became a fully fledged Division.

Having successfully launched these important satellites the Division of Botany

was set to concentrate on a more strictly botanical field of research, and so it was that

botanists in the National Herbarium, who are largely occupied with the stocktaking and

classification of our indigenous flora, felt the need of a botanical garden, worthy of the

name, in which to grow and observe plants in their living state — a garden in which to

study the potentialities of each individual species. In would be at the same time an

educational centre and a place of beauty.

It was at this stage that the alliance was made between the Division of Botany

and the University of Pretoria to which I referred at the outset.

From time immemorial plants have been cared for in gardens. The earliest gardens

were concerned mainly with the cultivation of herbs with medicinal properties and were

in fact the first scientific botanical institutions. Yet even today, with the spate of new

synthetic drugs there is still an intense interest in the exploration of the vegetation of

the world for the discovery of new drugs or new sources of known compounds. South

Africa is already making valuable contributions by the export of “ aloes ”, “ buchu

oil ”, diosgenin, from Dioscorea for the manufacture of cortisone, to mention only three

examples. There is now on foot a scheme sponsored by commercial interests in America

to create a Fellowship for a South African to explore the chemical properties of the

flora of the Union more intensively than ever before. The holder of the Fellowship

will receive direction from both the C.S.I.R. and the Division of Botany, and this

garden, in which we are gathered this morning, will receive any plant material which

shows promise of beneficial properties.

If we look into the fairly recent history of botanical gardens we find that many

of them throughout the world have played an important part in the trial and distribution

of plants of economic importance. There is the famous case of the botanist casually

collecting a few seeds of the rubber tree in its wild habitat in South America and sending

them to the Kew Botanic Gardens in England. The resulting young rubber trees were

then forwarded to the Singapore Botanic Gardens, from which small beginning an

industry of immense value to mankind, and certainly one worth many millions of pounds,

has been established. Botanical Gardens of the world played an equally important part

in the development and expansion of the cotton, quinine and tea industries, and from

Australia we have the example of the extensive wheat industry being founded on a few

seeds introduced into the Sydney Botanic Gardens.

I can give you an example also of one of our veld plants gaining a place of irnpor-

tance in agriculture beyond our borders through the action of the late Dr. Maria Wilman

who worked on the flora of Griqualand West. Dr. Wilman sent seed of one of our

ordinary veld grasses — one with no great reputation in the Union at the time — to

America. This grass, Eragrostis curvula, soon found favour in American agriculture,

with the result that today it is sown as a pasture and soil improver over vast areas,

and is commonly referred to as the ‘ Wilman Love Grass ’. The Americans have, in

fact, already produced improved strains at their experimental stations. The value of

399

the grass is now appreciated more in the Union too, and in the Griqualand East district

alone over 10,000 acres have been planted with Love Grass within the last 8 years.

The botanist in his botanical garden has certainly had a finger in many a rich agricultural

pie.

I have purposely drawn your attention to these few examples to remind you how

intimately Agriculture and our every day welfare have been linked with the subject

of botany. The functions of botanical gardens have, however, undergone a gradual

change since the days of the purely medicinal concept. It is still a seat of learning and

education and gradually the aesthetic aspect has gained prominence. This is also of

great importance when we weigh its influence in the building up of a national culture.

An educated community must recognise that it is its duty to take an intelligent interest

in the flora of the country in which it lives.

And what have we in this flora of ours to inspire a national pride?

For a start we can claim that, for its size the Union of South Africa has one of the

richest floras in the world. It supports nearly 20,000 species. To mention a few statistics

we have about 500 species of Erica, 300 species of Protea and its allies, over 800 species

of grass and well over 1,500 species of vygies and other succulents.

The collection and cultivation of succulents, which was popular from the earliest

days of exploration at the Cape, is still an absorbing hobby for many specialists. There

persists a steady interest and trade in this group in which the infinite variety of form

and colour is a continual source of wonder. — There are the Aloes, the window plants

half buried in hard ground, the stone-like plants growing on the surface and blending so

harmoniously with their surroundings that they are extremely difficult to detect, and

a host of other curious forms. Nowhere in the world can our succulent flora be matched

for its richness.

Some arid areas of the Americas boast a great variety of Cacti which are absent

from the Union, but the succulent flora of the Union is not only immensely rich in

species but is represented in a number of unrelated families in which there has been

parallel evolution.

But this is by no means all. The Union has contributed handsomely to botanical

gardens and horticulture throughout the world from the rare and beautiful flora with

which nature has so lavishly endowed us.

Even before the first European settlement at the Cape about 300 years ago, there

had developed a small traffic in Cape plants to several royal gardens of Europe. Sailors

with an interest in horticulture had gathered seed and bulbs during temporary landings.

And although the first settlers at the Cape were given strict instructions by van Riebeeck

to confine their waderings to the Peninsula, it was not long before these bounds were

broken. Later the two Governors, Simon and Adriaan van der Stel, father and son,

organised special expeditions into the interior in order, among other things, to collect

plants and seed to enrich the famous gardens of Holland.

Steady progress with the introduction of South African plants to Europe was made

during the 18th century, until towards the end of the century, our Ericas, Proteas,

Stapelias and many bulbous plants, may be said almost to have dominated the green-

houses, not only of Holland but of England, Austria, Sweden, Italy and France. There

was great rivalry among the owners of gardens to claim the richest collection of rare

plants from foreign lands. The Cape Flora was a tremendously rich reservoir from

which to draw.

400

During the 19th Century the ambition for novelty subsided and our Flora lost

some of its attraction; not, however, before such subjects as Agapanthus, Nerine,

Gladiolus, Pelargonium, the red hot pokers. Arum lilies, and even some of the Nama-

qualand daisies, had earned for themselves a permanent place in the gardens of Europe

and the New World. During their domestication overseas these South African plants

attracted the attention of plant breeders with far reaching horticultural results. Among

the earliest hybrids to gain favour were the Pelargoniums — popularly referred to as

Geraniums — which are so extensively used to this day. It will be recalled that the

Geranium Society of Great Britain quite recently presented the Union Government with

a collection of the best modern hybrids as a token of gratitude to the Country of their

origin. This collection is now under the care of the Division of Horticulture.

Then there is the red Barberton daisy, which by selection and hybridisation has

yielded an almost endless variety of shades and double forms, whose lasting quality

as a cut flower is a special asset. There are in fact very few of our cultivated flowers

which have retained their original purity. Visitors to public gardens overseas take

so many of these wonders for granted that few pause to enquire about the country of

origin.

Is it not fitting therefore that we should honour our own flora, study it in its natural

condition, cultivate and preserve it in this botanical garden for all to see and appreciate?

There will be no place here for exotics.

There is no doubt that the present century has seen a general awakening of interest

in the cultivation of our indigenous flora throughout the Union. Rapid strides have

been made in the public gardens of some of the larger and several of the smaller towns.

Extensive areas have been set aside for the preservation of attractive and interesting

species. This is all to the good and the trend should be encouraged as much as possible;

but in the nature of these undertakings the scientific aspects are absent or, at best,

observed only to a limited degree. It is only institutions such as the Division of Botany

with a staff of trained botanists and gardeners that can combine scientific study with the

art of cultivation.

The National Botanic Garden of Kirstenbosch established in 1913 has a similar

organisation — a garden and a trained botanical staff. In fact thoughout the world

the combination of a botanical garden with a herbarium and library is recognised as

the ideal organisation for the advancement of botanical science.

The National Botanic Garden at Kirstenbosch has earned for itself a world-wide

reputation for its magnificent setting and its great natural beauty. Its reputation was

enhanced still further recently by the winning of a gold medal award at an international

wild flower show in New York for a collection of South African flowers consisting

mostly of Proteas. But Kirstenbosch cannot feature the whole of the Union’s flora.

No single institution could do so.

As you all know the Union is divided climatically into a predominantly winter

rainfall region and a predominantly summer rainfall region, with a relatively small

region of equal proportions of winter and summer rainfall in the George-Knysna

area. As one could guess, the flora associated with the 2 main regions is distinctive.

Thus Kirstenbosch, which enjoys a high winter rainfall, is suited best for the study of the

flora of that region. The summer rainfall flora is not at home there and it is this section

which will receive most attention in this Botanic Garden in Pretoria. The Division of

Botany does not disguise its ambition, however, to achieve success with some of the

Cape specialities, and I have been told that this project is fostered by Mrs. Vogts, who

is present this morning, and has herself had success in Pretoria after earlier experience

at the Cape.

401

It must have occurred to most of you that the development of a botanical garden

has a likeness to the building of a nation. The founders have their ideals and aspirations ;

they set the foundation, but rarely see the full harvest of their labours. Thus today,

we are, so to speak, laying the foundation stone of an edifice which will take many

years to reach maturity. As a symbol of the ambition of the Division of Botany I have

been asked to plant an indigenous tree to commemorate this occasion. The growth

and development of the tree — year by year gaining in stature and beauty — is to be an

inspiration to future generations of gardeners and botanists. And on this note of

confident expectation, I have pleasure in declaring this national Botanic Garden open.

We shall now proceed to the planting of the tree.”

One of the staff said later: While we entertained our guests to tea some phrases of

the Minister’s address seemed to linger on the air- a botanical garden worthy of the

name, in which to grow and observe our plants in their living state — in which to study

the potentialities of each individual species — a garden that would be at the same time

an educational centre and a place of beauty.

Now we look forward to the building of a new herbarium on a site within the new

botanic garden. It is sorely needed.

The title Pretoria National Botanic Garden — Pretoria Nasionale Botaniese Tuin —

was approved by the Council of the Pretoria University and by the Department of

Agriculture (Technical Services) in 1959.

403

Notes on some South African species

of the Genus Euclea

By

B. de Winter.

Euclea crispa, E. lanceolata, E. ovata and some hybrids.

The fact that Celastrus crispus Thunb. represents a species of Euclea was realised

by Sonder who states in a note in Harvey & Sonder FI. Cap. 1 : 461 (1859-60) “ Celastrus

crispus Thunb. = Eucleae (Ebenac) ”. Later this was taken up by Guerke in Pflan-

zenfamilien 4, 1 : 158 (1891) who credited the combination to Sonder.

Under the modern Rules of Botanical Nomenclature a new combination is only

valid if the combination is actually made, hence in the present case, Guerke not Sonder

is the effective author.

Guerke regarded Celastrus crispus as representing the same species as E. ovata

Burchell, Pflanzenfam. 4, 1: 158 (1891). When the type specimen of C. crispus was

examined by me, it was found to agree more closely with the type of E. lanceolata

E. Mey. ex DC. However, Hiern, in the FI. Cap., and later Sim, in his Forest Flora

of Cape Col. p. 262, noted that E. lanceolata and. E. ovata cannot easily be distinguished.

I have failed to find sufficient difference to justify keeping them as distinct species.

I regard E. lanceolata to be a synonym of E. crispa and E. ovata a variety of E.

crispa.

Euclea crispa {Thunb.) Sond. ex Guerke, aggregate species.

(a) var. crispa.

E. crispa (Thunb.) Sond. ex Guerke in Pflanzenfam. 4, 1: 158 (1891). Type;

Cape, Thunberg s.n. E. lanceolata E. Mey ex DC. Prodr. 8; 217 (1844); Hiern in F.C.

4, 1 : 467 (1906). Type; Klipplaat River, Drege s.n.

Celastrus crispus Thunb. in Hoffm. Phytogr. Blatt 1 ; 23 (1803).

{b) var. ovata {Burch) de Winter comb, et stat. nov.

E. ovata Burch. Trav. S. Afr. 1 ; 387 (1822); Hiern in F.C. 4, 1 ; 468 (1906). Type:

Hay, Sluipslang, Burchell 1706.

When studying the var. ovata it was found that the leaves of some specimens are

indistinguishable from the leaves of E. coriocea, a species very different in its floral

characteristics. The undulate margin, indumentum and acute apex of the leaves of

var. ovata are reminiscent of E. coriacea, and the area of distribution is so situated

that the idea that var. ovata originated due to hybridization between E. coriacea

and E. crispa presents itself. The floral structure, however, offers little evidence

for the suggestion since the flowers of the var. ovata have deeply cleft corollas, quite

unlike those of E. coriacea, in which the corollas are only very shallowly lobed. One

could assume that the gene for a deeply cleft corolla is dominant. In some forms

of Euclea natalensis, which, on the evidence available, can only be regarded as the result

of hybridization between E. natalensis and E. tomentosa, the same obstacle presents

404

itself. Here again the acceptance of the deeply lobed corolla as being controlled by

a dominant gene will give a reasonable explanation of the abscence of plants with

intermediate floral characteristics. Should the presence of this dominant gene be

accepted, it is clear that the var. ovata could have acquired some of the leaf characteris-

tics of E. coriacea by introgressive hybridization, while the flowers have remained more

or less unaltered and resemble those of E. crispa. E. coriacea has on the other hand

apparently remained unchanged, a fact which could be regarded as evidence against

hybridization having taken place. It is significant, however, that the studies of Stebbins

and others on hybridization and its effects (Variation and Evolution in Plants 1950)

offer examples which agree almost in detail with the case discussed here (Chap. 7 : 262-

265).

On the importance of the environment for the success of hybridization I quote

Stebbins : “ Hybridization between well-established and well-adapted species in a

stable environment will have no significant outcome or will be detrimental to the species

populations. But if the crossing occurs under rapidly changing conditions or in a region

which offers new habitats to the segregating offspring, many of these segregates may

survive and contribute to a greater or lesser degree to the evolutionary progress of the

group concerned ”.

According to Acocks in Veld Types of S. Afr., Bot. Survey Mem. No. 28, the Karroo

vegetation has advanced considerably eastward since the advent of man in the southern

tip of Africa. When the maps drawn up by Acocks depicting conditions in 1400 A.D.

and 1950 A.D. are superimposed, the supposed original boundary between Karroo

and tropical grassveld and the actual boundary in 1950, enclose an area which agrees

in remarkable detail with the area of distribution of var. ovata. It can be inferred

that conditions in this area, which can be regarded as transitional, are unstable

and offer a range of new habitats. On the other hand it seems unlikely that the var.

ovata is of very recent origin, which suggests that the unstable conditions of this area

go back to much earlier times than estimated by Acocks.

Another possibility is that the var. ovata has become better adapted to arid con-

ditions than either of its parents, due to a recombination of characters. In this case

the distribution may be mainly determined by rainfall. The distribution of rainfall,

however, does not agree nearly as closely as the former with the area of distribution of

the var. ovata, nor does geological data combined with rainfall offer a better solution.

Euclea schimperi, E. daphnoides and E. macrophylla.

An examination of the Abysinnian species E. schimperi, shows that the two S.A.

species E. macrophylla and E. daphnoides are not specifically distinct from it. E. macro-

phylla agrees in all essentials, but E. daphnoides differs somewhat in the shape and texture

of the leaves; these are narrower and often become thicker and more coriaceous than

those of typical E. schimperi. E. macrophylla is, therefore, placed in synonymy under

E. schimperi and E. daphnoides is reduced to varietal rank in the same species: —

E. schimperi {A. DC.) Dandy, aggregate species.

(a) var. schimperi.

E. schimperi (A. DC.) Dandy in Andrews Flow. PI. Anglo-Egypt. Sudan 2: 370

(1952). Type: Abyssinia, Schimper s.n. E. Kellaua Hochst. in Flora 26: 83 (1843).

E. macrophylla E. Mey. ex A. DC. Prodr. 8: 218 (1844); Hiern. in F.C. 4, 1: 472

(1906). Type: Cape, between Kei and Bashee Riv., Drege s.n. E. racemosa Murr.

var. burchellii Hiern. in F. C. 4, 1 : 474 (1906).

405

Kellaua schimperi A. DC. in Ann. Sc. Nat. 2, 18 : 20 (1842).

{b) var. daphnoides (Hiern.) de Winter stat. nov. E. daphnoides Hiern in Trans. Cambr.

Phil. Soc. 12: 102 (1872): and in F.C. 4, 1: 42 (1906). Lectotype: Humansdorp,

Burchell 4909. (I nominate the sheet at Kew, SwrcAe// 7/4909/, „western bank of the

Wagenbooms River on the north side of Lange Kloof”, a female specimen, as the type

of E. daphnoides Hiern).

6096259—7

407

Notes and New Records of African

Flowering Plants.

By

Various Authors.

ACANTHACEAE.

Justicia minima A. Meeuse, nom. nov. — J. spergulaefolia sensu C. B. Clarke in Dyer,

FI. Cap. 5, 1 : 60 (1901), non T. Anders. Type; Schlechter 4267 (K, holo.!, PRE,

iso.!).

A number of specimens of a Justicia had been collected in the Transvaal, which

run in the key in Flora Capensis to “ J. spergulaefolia ” and indeed match Schlechter

4267 (PRE), the only gathering cited by Clarke. All these specimens, including the

type gathering, came from a fairly restricted area in the Warmbaths, Waterberg and

Potgietersrust districts of the Transvaal and it was felt that they might not be conspecific

with the type of J. spergulifolia which was collected in Damaraland. Mr. W. Marais

borrowed the holotype of the latter from T.C.D. (Dublin) and reported that it is un-

doubtedly a species of Monechma, probably M. namaense C. B. Clarke. The Transvaal

specimens referred to above appeared to be without a valid name and the name “ mini-

ma ” was chosen because this reflects the very slender habit of this plant and the small

flowers.

Characteristic of this species is the varying degree of pubescence. Specimens

which are almost or completely glabrous are growing side by side with specimens which

are rather densely pubescent with short stiff hairs. Very glabrous specimens also have

glabrous inflorescences, but very hairy ones have a pubescent rhachis, pubescent bracts,

bracteoles and calyx-segments.

The species can be best characterised as a miniature of the tropical African J.

linearispica C. B. Clarke, which it resembles in habit, in the shape of the leaves and in

the inflorescence, but all parts are considerably smaller. J. minima grows almost in-

variably in rock crevices on quartzite.

Transvaal. — Locality not quite certain, but probably Potgietersrust district :

Magalakwin River, Schlechter 4267 (K. holo! , PRE, iso! ). Potgietersrust: between

Bokpoort and Palala Heights, Meeuse 9349. Waterberg: Vyeboom near Nylstroom,

Meeuse 9334, 9334a; near Naboomspruit, Meeuse 9736. Warmbaths: Hills North

of Warmbaths on road to Nylstroom, Smuts and Gillett 3314, Codd 3448, Repton 793,

Sidey 1329. (All cited specimens in PRE).

Justicia montis-salinarum A. Meeuse spec. nov. ex affinitate J. orchioides L.f., J. odorae

Vahl, J. philipseae Rendle et J. lorteae Rendle, sed inter alia habitu, indumento

puberulo, foliis linearibus, corolla extus pubescent!, capsula minute denseque

hirtiuscula differt.

Suffrutex perennis densus multicaulis ramosissimus habitu semigloboso 30-60

cm. altus. Caules plurimi lignosi subteretes glabrescentes cortice griseo sublevi vel

ubrugoso obtecti, ramulis ulterioribus flavo-viridibus suffruticosis subquadrangulatis

408

densissime minuteque puberulis vel farinosis, internodiis 1-2 cm. longis. Folia lineari-

lanceolata subrigida Integra sessilia subacuta subpungentia minute puberula in siccitate

flavo-viridia 10-25 mm. longa et 1-2-5 mm. lata, nervo medio subtus prominulo.

Flores solitarii in axillis foliorum superiorum sessiles. Bracteolae 2 lineari-lanceolatae

attenuato-acutae minute hirtiusculae uninerves 5-6 mm. longae c. 0-75 mm. latae.

Calyx profunde 5-fidus, firmiter herbaceus, dense minuteque hirtiusculus, tubo c. 1 mm.

longo, lobis erectis subaequalibus lineari-lanceolatis attenuato-acuminatis uninervibus

c. 4 mm. longis 0-5-0 -75 mm. latis. Corolla alba intus in faucibus purpureo-maculata,

extus dense minuteque strigoso-pubescentis intus in faucibus sparse pilosula, 8-9 mm.

longa, labio antico 3-lobato in faucibus transverse ruguloso cum lobo mediano sub-

orbicularibus lobis lateralibus oblongo-rotundis, labio postico concavo leviter emargi-

nato. Stamina in faucibus aflfixa, filamentis ad basin pilosulis, loculis antherarum

inferioribus distincte caudatis, granulis pollinis ut in Justicia, poris 3 instructis. Ovarium

ad apicem hirtiusculum, stylo infra medium sparse pilosulo. Capsula dilute ochracea,

dense minuteque strigoso-hirsutula, stipitata, 8-9 mm. longa, parte stipitata complanata

c. 3-5 mm. longa. Semina rufo-brunnea tumida papilloso-verruculosa, c. 2-5 mm.

longa, 1-5 mm. lata, 1-25 mm. crassa.

Transvaal.^ — Soutpansberg: Vivo, schoolgrounds, Mogg 24448 (PRE, J); slopes

of Soutpansberg near Saltpan, Obermeyer, Schweikerdt & Verdoorn 168 (PRE); southern

entrance of Sandrivierspoort, about 4 miles north of main road bridge, Meeuse 10213

(PRE, holo.!).

Erect perennial suffrutex, greyish to subcanescent in appearance when fresh,

forming semi-globose densely tufted bushes 30-60 cm. tall. Stems numerous, much-

branched from the base up, the older parts woody, subterete, grey, with a smooth

or faintly longitudinally wrinkled bark; ultimate ramifications wiry, subquadrangular,

yellowish-green when dry, finely puberulous to farinose; internodes 1-2 cm. long.

Leaves linear-lanceolate, stiff, entire, sessile, subacute to subpungent, thinly puberulous

or farinose, drying yellowish-green, 10-25 mm. long and 1-1-5 mm. broad; midrib

prominent on lower surface, remaining venation consisting of a sinuous marginal vein

connected by a few (usually 3-6) short transverse lateral veins with the midrib, pro-

minent on lower surface, indistinct on upper surface. Flowers solitary, sessile in upper

leaf-axils, almost running into a sparse leafy terminal spike; bracteoles 2, linear-lanceo-

late, resembling the leaves in shape and texture, attenuate-acute, 1 -nerved, 5-6 mm.

long and about 0-75 mm. broad. Calyx deeply 5-fid, the same colour as the leaves,

densely covered outside with very short stiff bulbous-based hairs and inside mainly

in the upper part; tube about 1 mm. long; lobes erect, equal, linear-lanceolate, attenu-

ate-acuminate, about 4 mm. long and 0-5-0-75 mm. broad at the base. Corolla white

with purple markings in the throat, outside densely pubescent with stiff short hairs,

inside with a few stiff hairs in the throat, 8-9 mm. long; anticous (lower) lobe 3-lobed

with the middle lobe subrotundate and the lateral lobes oblong, rounded at the apex;

the throat on this side transversely rugulose; posticous (upper) lip concave, with a

rugula containing the style, slightly emarginate. Stamens with a few stiff hairs near the

base of the filaments; lower anther-cells with a long nearly straight tail; pollen of the

Justicia-iyg)&, ellipsoid, 3-colporate. Ovary glabrous except near the top; style shortly

and thinly pilose in lower half. Capsule pale ochre, densely strigose-hirsutulous with

short stiff usually retrorse hairs, stipitate, 8-9 mm. long of which 3-5 mm. taken up

by the “ stalk ”. Seeds rather thick, elliptic-subquadrate in outline, of a somewhat

reddish brown colour, finely and rather evenly papillose-verrucose, about 2-5 mm.

long, 1 -5 mm. broad and 1 -25 mm. thick.

This rather insignificant Justicia belongs to a group of species which have axillary

flowers not forming a distinct terminal inflorescence but sometimes running into a

leafy sparse spike which is not clearly separable from the remainder of the stem (section

409

Calophanoides of FI. Cap. and FI. Trop. Afr.). In this group it must be placed in the

affinity of the shrubby perennial species such as J. orchioides L.f., J. odora Vahl, J.

phillipseae Rendle and J. lorteae Rendle, but it differs from the first two in the smaller

flowers and the narrow leaves, from the other two among other things in the pubescence,

hairy corolla and nairy capsule. The pollen is of the Justicia-iypQ (Kndtchenpollen of

Lindau), with three germination pores.

The first gathering was apparently by Obermeyer c.s. twenty years ago, but the

material was rather scanty and nobody ventured a description. Recent gatherings

provided ample material and field-notes. The specimens O.S.V. 198 and Meeuse 10213

were compared at Kew and reported by Mr. W. Marais to be unmatched.

This plant is easily recognised in the field by its very dense low bushy habit which

is not found in other species of the genus occurring in Southern Africa. It seems to

prefer warm, dry and rocky situations and has so far only been found in a fairly small

area comprising some drier localities in the foothills of the western part of the Soutpans-

berg range. It is not gregarious, the specimens gathered being apparently taken from

single individual plants, and it is probably rare, but a careful search for this not very

conspicuous species in suitable localities may prove that it is more wide-spread than the

present records indicate.

Petalidium (Sect. Petalidium = Sect. Haplanthus Lindau) luteo-album A. Meeuse, spec,

nov. ex affinitate P. linifolii T. Anders, et P. lucentis Oberm. sed inter alia foliis

ellipticis vel oblongis (nec linearibus nec lineari-lanceolatis), floribus baud lila-

cinis, bracteis nec atri- nec purpurei-venosis differt.

Frutex glaber densus multiramosus habitu semigloboso, 0-50-1 m. altus. Caules

lignosi cortice griseo vel atri-purpureo obtecti, ramulis ulterioribus subdense foliosis

viridibus vel albescentibus. Folia firmiter herbacea in siccitate pergamacea, flavo-

viridia, elliptica vel oblonga, Integra, obtusa vel subacuta interdum minute apiculata,

2-3-5 cm. longa, 0-5-1 -5 cm. lata, cystolithis indistinctis raro infra conspicuis, petiolis

brevibus ad c. 6 mm. longis. Flores axillares, solitarii, pedicellis gracilibus curvatis

vel sinuatis 4-7 mm. (post anthesin ad 12 mm.) longis, bracteolis 2 planis late cordatis

asymmetricis submembranaceis vel papyraceis dilute albo-viridibus interdum plus

minusve purpurascentibus abrupte apiculatis subglabris integris distincte denseque

reticulato-venosis 2-5-3 cm. (post anthesin ad 4-5 cm.) longae et latae. Calyx viridis,

5-fidus, tubo 2-3 mm. longo, lobis erectis inaequalibus linearibus acutis minute sparseque

glanduloso-puberulis ciliatisque 6-10 mm. longis. Corolla dilute flava labio antico

flavo, in faucibus fuscovariegata, extus praecipue infra medium et ad nervos sparse

adpresse pubescens, intus ad basin loborum pilis albidis rigidis retrorsis paucis sparsim

strigosa, tubo (parte cylindrata) c. 10 mm. longo 3 mm. lato, faucibus (parte infundi-

buliformi) c. 15 mm. longis ad apicem c. 15 mm. latis, lobis 4 posticis subrotundis

c. 8 mm. longis et latis, lobo antico late orbiculari-rhomboideo truncato emarginato

c. 11 mm. longo 8 mm. lato. Capsula fusca glabra subnitida apiculata, 11-12 mm.

longa, c. 6 mm. lata. Semina 1-2, 6x4 mm.

South West Africa. — Kaokoveld Reserve: Okonjombo, Watt O.P. No. 2132fG

(PRE); near Orupembe (Anabib), Story 5730 (PRE, K); 20 m. W. of Otjiku on road to

Orumpembe, de Winter & Leistner 5683 (PRE, holo. ! ; isotypes to be distributed to

K, M, Windhoek, B, SRGH and elsewhere).

A much branched rounded shrublet 0-50-1 m. tall. Older stems woody, terete,

somewhat swollen at the nodes, grey to rather dark purplish grey, nearly smooth but

here and there with a few small and slightly raised circular white lenticels, ultimate

branchlets mostly appearing as short lateral shoots, green or ash-grey to white, glabrous

or puberulous when still very young, rather densely leafy and bearing the solitary flowers

in the leaf-axils. Leaves firm, yellowish-green when dry, more pale greyish green when

410

fresh, elliptic or oblong, entire, obtuse with rounded or minutely apiculate apex,

gradually narrowing at the base, very soon quite glabrous, 2-3-5 cm. long and 0-5-1 -5

cm. broad; cystoliths in dried leaves indistinct or occasionally discernible on lower

surface; midrib slender, prominent on lower surface, lateral veins usually only 2-3

on either side, finer nervation almost invisible except in the youngest leaves; petioles

short, some attaining 6 mm. Pedicels slender, curved (usually upwards) or sinuous,

4-7 mm. in fruit up to 12 mm. long. Bracteoles very large, flat, greenish-cream some-

times more or less suffused with purple, orbicular-cordate, abruptly and sharply api-

culate with asymmetrical cordate base and more or less triangular rather narrow basal

sinus, entire, papyraceous and rather brittle when dry, nearly glabrous except for a

rim of a sparse very short pubescence near the margin inside, with a distinct midrib

and a fine reticulate nervation, 2-5-3 cm. long and broad under open flowers, in fruit

enlarging to 4-5 x 4-5 cm. Calyx green, 5-fid; the tube 2-3 mm. long; lobes linear,

erect, acute, minutely and sparsely glandular-puberulous and ciliate, unequal in length,

6-10 mm. long, the tube in fruit elongating to 4 mm., the longest lobes to 12 mm.

Corolla pale-cream with brown markings inside in lower side of throat and the lower

(anticous) lip yellow; outside, mainly below the middle and on the veins, shortly and

sparsely pubescent, inside with a few white stiff retrorse hairs at the bases of the lobes;

the cylindrical part of the tube about 10 mm. long and 3 mm. in diam., the funnel-

shaped portion (throat) about 15 mm. long and as much in diam. at the mouth; four

posticous lobes suborbicular, about 8x8 mm.; the anticous (lower) lobe broadly-

orbicular-rhomboid, truncate and emarginate at the apex, about 11 mm. long and

14 mm. broad. Capsule brown, rather shiny, apiculate, 11-12 mm. long and 6 mm.

broad. Seeds 1 or 2, much compressed, about 6x4 mm.

This very distinct species was, as far as can be ascertained, collected for the first

time by Dr. J. Watt, now Director of Agriculture, Windhoek, about twenty years ago

when he was a Government Veterinary Officer. The two more recent gatherings pro-

vided ample material for the description.

It is a typical representative of the genus even in such minute details as the pubes-

cent style. The solitary axillary ebracteate flowers with deeply 5-fid calyx indicate its

place in the section Petalidium (= Haplanthus Lindau). Its nearest affinities appear

to be with P. linifolium T. Anders, and P. lucens Oberm. which also possess large, flat

and reticulate-veined bracteoles, but it differs in several respects such as broader leaves,

a different colour of the corolla (which is mauve or bluish in the other species) and the

concolorous (not blackish or purple, discolorous) venation of the bracteoles. The

specimen Story 5730 was compared at Kew and the British Museum herbarium by our

officer at Kew, Mr. W. Marais who reported that this plant is not represented in either

herbarium and confirmed my conelusion that there is no other described species whose

diagnosis could possibly apply to the Kaokoveld plant.

The specific epithet chosen refers to the colouring of the corolla, although it should

be mentioned that cream or yellow colours are found in the corollas of several species

of Petalidium; however as far as all available records show, the combination of cream

and yellow in a single flower has not been observed in previously described species.

Rhinacanthus xerophilus A. Meeuse, spec, nov., R. rotundifolio C. B. Clarke arete affinis,

sed caulibus subteretibus (baud acute sexangulatis), indumento caulium et foliorum

densiore praecipue differt.

Suffrutex perennis e basi ramosus 20-70 cm. altus. Caules adscendentes vel sub-

erecti, basi lignosi glabrescentes fuscentes, supra virides firmiter herbacei rigidi sub-

teretes indistincte longitudinaliter sulcati, pilis brevibus curvatis dense pubescentes.

Folia late elliptica vel ovata vel obovata vel triangulari-ovata interdum suborbicularia,

subsessilia vel petiolis 1-6 mm. longis instructa, lamina acuta vel obtusa interdum

411

emarginata vel minute apiculata basi rotundata vel subcordata raro decurrenti vel

subcuneata Integra vel subcrenata supra intense viridi in siccitate olivacea vel nigricanti

minute subsparseque strigoso-pubescenti subtus pallidiore et praecipue ad nervos pilis

plus minusve curvatis nunc subsparse nunc dense pubescent! ad subvelutina demum

glabrescenti 1-4 • 5 (-6) cm. longa et 0 • 5-4 (-5) cm. lata. Flores sessiles vel subsessiles

in paniculam terminalem 8-18 cm. longam 2-10 cm. latam more Rhinacanthi dispositi.

Pedunculi teretes longitudinaliter subsulcati dense minuteque pubescentes vel pilosuli.

Bracteae bracteolaeque minutae, pilosulae. Calyx profunde 5-fidus, pilis brevibus

patentibus et pilis glandulosis pubescens, 5-6 mm. longus, tubo subnullo, segmentis

anguste lineari-lanceolatis erectis acutis. Corolla alba intus in faucibus maculis dilute

purpureis variegata, extus villosula et sparse glanduloso-pilosa, intus (labio postico

excepto) subglabra, tubo tubuloso (17-) 22-25 mm. longo 1-1 -5 mm. lato, lobis anticis

obovato-ellipticis obtusis intus sparsissime praecipue ad basin glanduloso-pilosis

7-10 mm. longis 3 -5-4 -5 mm. latis, lobo postico lineari-oblongo acuto intus sparse

glanduloso-pubescenti 6-8 mm. longo. Genitalia et granula pollinis ut in typo, stylo

sparse pilosulo, ovario pilis brevibus subpatentibus et pilis glandulosis paucis dense

pubescent!. Capsula apiculata, pilosula, 15-18 mm. longa, parte stipitata 7-9 mm.

longa. Semina nigra, c. 3 mm. longa et lata.

Transvaal. — Soutpansberg: Messina, Rogers 20759, 22599; Msekwa’s Poort,

Meeuse 9188; near Mara, Meeuse 10198. Sebasa; near Lake Fundusi, Bremekamp

& Schweikerdt 373. Pietersburg: Soekmekaar, Meeuse 9221 (PRE, holo.!, K, iso.!);

Mokeetsi, Breyer h. No. 21450; 3 miles W. of Maliepsdrift, Meeuse 10644; Nelspruit:

Kruger National Park, near Skukuza, Codd 5215, Van der SchijJfMOl.

Portuguese East Africa. — Sul do Save: Lourengo Marques distr., Ressano

Garcia, Schlechter 11881.

Southern Rhodesia. — Ndanga: Mtilikwe Riv., Bangara Falls, Wild 43>6\ (SRGH,

PRE).

(All cited numbers in PRE, Codd 5215, Meeuse 9188, 10198, Van der Schijff 3401

also in K).

Straggly to suberect subshrub, branched mainly near the base, 20-70 cm. tall.

Rootstoek woody. Stems ascending to suberect, woody at the base and there becoming

covered with a thin pale brown bark; upper parts green, firmly herbaceous to wiry,

rigid, subterete, indistinctly longitudinally sulcate and densely covered with a pubescence

of short, patent to more or less curved hairs. Leaves firmly herbaceous, varying from

broadly elliptic to (ob)ovate or broadly deltoid-ovate, sometimes suborbicular, sub-

sessile or on 1-6 mm. long shortly pubescent petioles, acute or obtuse, often minutely

apiculate, occasionally emarginate; rounded, truncate to subcordate, rarely decurrent

or subcuneate at the base, entire or faintly crenate, dark green drying blackish, and with

a rather sparse short strigose pubescence on upper surface, paler and usually much

more densely shortly pubescent to velutinous on lower surface (more densely so on the

veins), rarely noticeably glabrescent, 1-4-5 (-6) cm. long and 0-5-4 (-5) cm. broad.

Flowers sessile or subsessile in small clusters or solitary borne on a typical paniculate

branched Rhinacanthus inflorescence 8-18 cm. long and 2-10 cm. broad. Peduncles and

branches of the panicles very similar to the stems in appearance but pubescence more and

more interspersed with toadstool-shaped glandular hairs as the ramifications become

finer. Bracts and bracteoles minute, thinly covered with short stiff hairs. Corolla white

marked with mauve in the throat, pilose outside, nearly glabrous inside except on the

upper lip; the tube (18-) 22-25 mm. long and 1-1 -5 mm. in diam.; anticous lip of 3

obovate-elliptic obtuse 7-9 mm. long lobes, which are very sparsely glandular-pubescent

mainly near the base inside; posticous lip linear-oblong, acute, sparsely glandular-

pubescent inside, 6-8 mm. long. Pollen-grains as in the type species. Ovary densely

412

pubescent with short stiff hairs interspersed with a few glandular hairs; style very

sparsely pilose. Capsule apiculate, stipitatc, hairy, 15-18 mm. long of which 7-9 mm.

taken up by the “ stalk ”. Seeds black, orbicular in outline with a notch near the hilum,

about 3 mm. in diam.

This species is very close to R. rotundifolius C. B. Clarke and Mr. D. J. B. Killick

who compared several Transvaal specimens such as Meeuse 9221 (an isotype), van der

Schijff 3407 and Codd 5265 with a syntype of the latter {Thomas 7, K) reported the

following differences, which were confirmed when a specimen of R. rotundifolius from

Kismayu, Somalia {Bally 9508) was received on loan from the East African Herbarium,

Nairobi.

Thomas 1.

1 . Leaves strictly ovate (in Bally 9508 elliptic to

oblong).

2. Leaves sparsely hairy with veins prominent

on lower surface.

3. Stems sharply 6-angled in cross-section with

2-3 grooves between the angles, with very

small white adpressed retrorse hairs, mainly

on the angles.

R. xerophilus.

1. Leaves ovate-cordate and sometimes cordate.

2. Leaves densely pubescent on lower surface,

veins not so prominent.

3. Stems subterete, more undulate in cross-

section without angles or intermediate grooves,

abundantly hairy all round with suberect

hairs.

These differences may seem to be small, but they are apparently constant and in

this genus the differences between the species are always small, so much so that the

wide-spread African species R. gracilis Klotzsch has for a long time been treated as

identical with (and a taxonomical synonym of) the Indian R. nasutus (L.) Kurz (= R.

communis Nees).

To the differences mentioned by Mr. Killick may be added that the corolla-tube

in R. rotundifolius was described as 1^ inch (about 30 mm.) long (in Bally 9508 it is

30-50 mm.) long, whereas in the plant described here the length of the corolla-tube,

even in vigorous cultivated specimens, does not exceed 25 mm.; the width also differs:

1-1 -5 mm. in R. xerophilus, 0-5-1 mm. in Bally 9508. Another difference between

R. xerophilus and some related species may be found in the ecology. R. xerophilus

seems to be restricted to open, gravelly or rocky open places in dry bushveld vegetations

in areas with a low annual rainfall, classified by Acocks (Veld types of South Africa,

1953) as Arid Lowveld, Arid Sweet Bushveld and Mopani Veld, hence the specific

epithet. R. gracilis is found in more humid situations, often in tropical lowland

vegetation. The ecology of R. rotundifolius (recorded from Kenya, Tana River, and

Somalia) is not known, but it is likely to grow in similar arid conditions as the species

from Southern Africa and this may account for the great similarity in habit between these

two geographically separated forms. It is possible that the two are derived from the

same common ancestral form and developed independently into the present specifically

different taxa, because they became geographically separated.

AMARYLLIDACEAE.

CyrtanthuserubescensK/7//c/:, sp. novadistinctissima,nullis e specibus notibus propinqua.

Bulbi ad 10 cm. alti 4-6 cm. diam. Folia 2-4-lorata 20-60 cm. longa 2-6 cm.

lata basi rubro-marginata. Scapi 20-50 cm. alti 1-1 -5 cm. diam. Bracteae 2, 4-7 cm.

longae 1 -5-2 cm. latae. Umbellae 9-15-florae, pedicellis ad 5 cm. longis (fructiferis ad

7 cm.) erectis vel suberectis. Perianthium rubicundum 3-4 cm. 3-4 longum, tubo 8-10

mm. longo campanulato, lobis curvatis patentibus 2-3 cm. longis oblongo-lanceolatis,

exterioribus 6-5 mm. latis, interioribus 6 mm. latis. Stamina 6, 3 in parte superiore

faucium et 3 basi loborum interiorum inserta, antheris versatilibus. Ovarium 6 mm.

longum 4-5 mm. diam. Loculi multiovulati. Stigma trifida, lobis recurvatis et margini-

bus leviter revolutis. Capsulae 2-5 cm. longae 1-7 cm. latae, seminibus compressis

nitidis nigris.

413

Natal. — Bergville District: locally frequent on grassy banks of stream, 9400 feet,

Cathedral Peak Forest Station, Killick 1840; locally frequent in subalpine grassveld

in moist gully below Windy Gap, c. 9300 feet, Cathedral Peak Forest Station, Killick

and Marais 2187 (PRE, holotype); near upper contour path near minor stream below

Inner Buttress, 6500 feet. Cathedral Peak Forest Station, Ndnni s.n. (Nat. Herb. No.

28587).

Bulbs up to 10 cm. tall and 4-6 cm. diam., more or less gradually tapered from

near base to neck, with membranous tunic. Leaves 2-4, developing at same time or

slightly later than the inflorescence, strap-shaped, 20-60 cm. long, 2-6 cm. wide, central

midrib prominent on under surface, margin red in lower third of blade. Scape 20-50

cm. tall, 1-1-5 cm. diam., hollow. Bracts 2, subtending umbel, 4-7 cm. long, 1-5-2

cm. wide, the outer appreciably larger than the inner. Umbel 9-15-flowered; pedicels

varying in length, but rarely more that 5 cm. long (elongating to 7 cm. in fruit), erect

or suberect. Perianth pink, 3-4 cm. long, with one of the inner perianth segments

uppermost very slightly less spreading than all the others and with the style arching

somewhat under it. Tube 8-10 mm. long, campanulate, lobes curved, spreading, 2-3

cm. long, oblong-lanceolate; outer lobes 6-5 mm. broad with small tuft of hairs within

apex; inner lobes 6 mm. broad with few hairs at apex. Stamens 6, 3 attached to the

throat of the perianth opposite the outer segments and 3 on the inner segments just

below their base; anthers versatile. Ovary 6 mm. long, 4-5 mm. diam. Loculi many-

ovuled, in 2 ranks. Stigma trifid, the lobes recurved with the margins slightly revolute.

Capsule 2 - 5 cm. long, 1 - 7 cm. broad ; seeds compressed, shiny, black.

This very attractive plant with umbels of blush-pink flowers has been found in

only two places in the Drakensberg, both in the Cathedral Peak area. It was first

collected in December 1952 by the author in a moist gully at 9300-9400 feet some dis-

tance below Windy Gap (Organ Pipes Pass). There it is fairly common in subalpine

grassveld where it is often associated with boulders. The second station is a few miles

further north below the Inner Buttress at 6500 feet. Climbers in the Drakensberg have

occasionally come across the plant and have called it the “ pink Agapanthus ”. Its

potential value as a cultivated plant has apparently been recognised, because climbers

have at times made special expeditions to look for the plant.

When one uses Phillips’s key to Amaryllidaceae in his Gen. S. Afr. PI. (2nd Edition,

1951), this plant runs down to Anoiganthus. It agrees with Baker’s original description

of that genus in Journ. Bot. 76 (1878) except in two respects. Firstly, the stamens are

not all inserted “ at the throat of the tube ”; three are situated on the inner perianth

segments about 1 mm. above their base. Secondly, the flowers are pink instead of

whitish or pale yellow. Neither of these characters ought to preclude this plant from

Anoiganthus, consequently its inclusion in Cyrtanthus requires some explanation.

Baker in his Handb. Amaryll. ix (1888) and in FI. Cap. 6, 171-172 (1896) dis-

tinguishes Cyrtanthus and Anoiganthus on the presence of versatile or basifixed anthers

respectively. None of the specimens of Anoiganthus examined by the author had

basifixed anthers; they were versatile. Baker himself in his original description of

Anoiganthus describes the anthers as versatile. It is surprising, therefore, that he should

use this character at all.

Phillips (l.c. p. 199) distinguished Anoiganthus from Cyrtanthus on the relative

length of perianth tube to segments. In Anoiganthus the perianth tube is shorter than

the segments and in Cyrtanthus the tube is much longer than the segments. The Drakens-

berg plant has segments which are 2-3 times longer than the tube, which would put it in

Anoiganthus. In most species of Cyrtanthus the tube is longer that the segments, but in

several species e.g. C. affinis R. A. Dyer, C. clavatus (L’Herit.) R. A. Dyer, C. bicolor,

R. A. Dyer and C. brachyscyphus Bak. the tube is often equal in length to the segments.

6096259—8

414

In C. ihorncroftii C. H. Wright, the segments are as long as or longer than the tube.

It seems reasonable then to enlarge our concept of Cyrtanthus to include plants with

perianth tube shorter, equal or longer than the segments. The general facies of the

Drakensberg plant is that of a Cyrtanthus.

Anoiganthus and Cyrtanthus have both been investigated cytologically. Gouws in

Plant Life 5 (Herbertia Edition), 66-67 (1949) found that they have the same chromosome

number, namely 2n = 16. Wilsenach (ms.) investigated Cyrtanthus (among other

genera of Amaryllidaceae) and arrived at the same number. In a summary of his work

seen by the author Wilsenach suggests that our conception of Cyrtanthus should be

modified to include Anoiganthus. However, the fate of Anoiganthus is best left to the

considered judgement of some future monographer.

COMPOSITAE.

Helichrysum milfordiae Killick, sp. nov. affine H. albo N.E. Br., sed foliis capitulisque

multo minoribus, involucri bracteis apice sanguineis differt.

Herba nana pulvinata 1-3 cm. alta omnino incana. Folia basalia dense rosulata

obovata — spathulata 0 -6-1 -4 cm. longa 0-49-0 -9 cm. lata apice subacuta ad rotunda,

folia caulina lanceolata elliptica vel spathulata apice acuminata ad rotundata saepe

arista membranacea subamplexicaulia. Scapi 3-12 cm. alti. Capitula 2-5 -2-9 cm.

diam. solitaria terminalia homogama. Bracteae involucri 4-5 seriatae lanceolatae ad

ovatae acutae 0 -6-1 -2 cm. longae 1-4-5 mm. latae niveae saepe apice et basi sanguineae.

Receptaculwn leviter convexum et breviter alveolatum. Styli rami truncati. Pappi

setae graciles apice plumosae. Achaenia glabra.

Basutoland. — Drakensberg Area: on rocks, summit of Mount aux Sources,

9000 feet, Thode s.n. (Nat. Herb. No. 2749); summit of Mont aux Sources, 9500 feet,

Flanagan 1972; rocky top of mountain in crevices between boulders, Thabana Ntlenyana,

11400 feet, Guillarmod 2332; in fissures of rocks on top of Thabana Ntlenyana, 11475

feet, Coetzee 597 ; fairly common on cliff faces, Castle Buttress Area, 9800 feet, Killick

2333.

Natal. — Bergville District: locally abundant in cave above Organ Pipes Pass,

Cathedral Peak Forest Station, 9600 feet, Killick 1990; forms extensive carpets in

overhang at top of Organ Pipes Pass, Cathedral Peak Forest Station, 9700 feet, Killick

2322 (PRE, holotype); between rocks, in mats, rare. Cleft Peak, 9500 feet, Esterhuysen

23023.

A dwarf, cushion-forming herb, 1-3 cm. high, grey-lanate all over. Basal leaves

densely rosulate, obovate to spathulate, 6-14 mm., 4-9 mm., apex subacute to round;

cauline leaves lanceolate, elliptic or spathulate, apex acuminate to round often with

a prominent membranous arista, subamplexicaul. Scapes 3-12 cm. tall. Heads 2-5-

2 - 9 cm. diam., solitary, terminal, homogamous. Involucral bracts 4-5-seriate, lanceolate

to ovate, acute, 6-12 mm. long, 1-4-5 mm. wide (progressively longer and narrower

towards centre except for innermost row which is shorter than penultimate row), white,

shiny, often crimson (occasionally brown) at apex and base. Receptacle convex,

shallowly honeycombed, corners of pits slightly produced. Corolla 3-5 mm. long,

filiform, glabrous except on outside of lobes. Style branches truncate. Pappus of fine

bristles, plumose at apex. Achenes glabrous.

This new Helichrysum is an attractive carpet or cushion-forming plant consisting

of scores of tightly congested grey rosettes connected by horizontal rhizomes. The

cushions are found on rocks on the summit of the Drakensberg above 9000 feet. In

415

the Organ Pipes Pass area H. milfordiae usually occurs on cliffs with a southern aspect

or immediately below them. Occasionally it is associated with Crassula sp. (Killick

2321).

H. milfordiae is closely related to H. album N.E. Br., but differs in the very much

smaller leaves and heads, and the crimson- (sometimes brown-) tipped involucral

bracts. It should be noted that the gatherings of H. milfordiae from the Cathedral

Peak area {Killick 1990, 2322 and 2333, and Esterhuysen 23023) are consistent in having

an indumentum which is lanate-floccose i.e. the hairs intertwine in a random manner,

whereas in the remaining specimens the hairs are more or less straight and adpressed.

At first it was thought that this might be a specific difference, but the plants agree so

closely in other respects that they have here been treated as one species.

The first gatherings of this species were made in the 1890’s. They were Thode s.n.

in February 1893, Flanagan 1972 in January 1894 and Thode 29 in January 1896 —

all from Mont aux Sources. It seems that fifty years elapsed before it was collected

again as a herbarium specimen; the author found it growing at the top of the Organ

Pipes Pass in May, 1953. Since then the plant has been collected several times as

indicated by the other specimens cited.

This plant has been cultivated in gardens in Britain for many years under the name

of Helichrysum marginatum DC. It differs from typical H. marginatum in its dwarf

habit (rosette plus scape up to 12 cm. high as against 25 cm.), shorter leaves (up to 1 - 4

cm. instead of 5 cm. long), in being lanate all over, and in having crimson-tipped involu-

cral bracts. Interest in the plant was probably stimulated by an article by Dr. R. Selig-

man in Bull. Alpine. Card. Soc. 2, 168-178 (1933) entitled “ Wild Flowers of South

Africa”. The part dealing with H. milfordiae reads: “The summit (Machache in

Basutoland) and its approaches brought me three new treasures, a white Dimorpho-

theca nestling among the stones — a true scree plant if there ever was one — a tiny

Helichrysum wedged firmly into the living stone, and in a sheltered spot a dwarf

Geranium.’’’ Dr. Seligman included a photograph of the plant taken on the summit of

Mount Machache (9300 feet).

How this species came to be grown in Britain can be gleaned from letters attached

to a specimen of the plant in Kew Herbarium. The plant was sent to Kew for naming

in June, 1947 by Dr. J. S. L. Gilmour, then Director of Wisley Royal Horticultural

Society Gardens. Mr. A. A. Bullock dealt with the plant and decided that it was new

to science, but that he would have to know more about its origin before describing it.

Dr. Gilmour subsequently found out from Mr. W. E. T. Ingwersen that the plant was

collected by the late Mrs. Helen A. Milford at Mont aux Sources in Basutoland and

brought to England where it found its way to the gardens of alpine plant enthusiasts.

The author has adopted Mr. Bullock’s manuscript name of H. milfordiae honouring

Mrs. Milford.

CONVOLVULACEAE.

Merremia guerichii A. Meeuse, spec, nov., — M. guerichiana (Engl, ex) Hallier f. in Engl.

Bot. Jb. 18: 69 (1893), nomen tantum. Ab M. quercifoliae Hallier f. foliis profun-

diter palmatilobis praecipue differt, ab M. ampelophyllae Hallier f. inter alia

indumento setoso-piloso baud molliter pubescenti, ab M. bipinnipartitae (Engl.)

Hallier f. indumento distinguenda.

Suffrutex. Radix lignosa perennis. Caules pauci vel plurimi, prostrati vel sub-

erecti, 20-60 cm., interdum ad 90 cm., longi, 1-5-2 -5 mm. diam., firmiter herbacei

ad basin plus minusve lignosi, subtereti, sublevi vel longitudinaliter striati, pilis basi

incrassatis subsparse vel subdense piloso-setosi demum glabrescentes et scabridi.

416

Folia firmiter herbacea in siccitate pergamacea, lamina ambitu ovato-orbicularia pro-

funde 3-5 palmatifida glabra vel ad nervos sparse setosa 1-5 cm. longa et lata, lobis

lyratis vel pinnatifidis vel crasse acuteque dentatis, petiolis setosis 0-5-3 cm. longis.

Tnjiorescentiae 1-2-florae, pedunculis satis gracilibus subteretibus sparse setosis 0-5-4

cm. longis, bracteolae minutae, pedicellis 0-5-2 cm. longis. Calyx viridis in parte

inferiora subdense pilis basi incrassatis setosus, lobis erectis oblongis vel anguste

ellipticis vel cuneato-oblongis acutis vel subobtusis frequenter minute apiculatis subae-

qualibus 12-18 mm. longis. Corolla infundibuliformis, alba vel flavescens, vinoso-

oculata, glabra, 5-lobata, 2 -5-3 -5 cm. longa lataque, lobis obtusis interdum minute

apiculatis. Stamina subaequalia ad basin dilata et pilis minutis rectis papilloso-puberula,

pollen Merremiae. Ovarium subglobosum, glabrum. Capsula subglobosa, straminea,

apiculata, glabra, c. 12 mm. diam. Semina atrogrisea vel fusca, puberula, c. 8 mm.

longa.

South West Africa. — Kaokoveld: between Orupembe Waterhole and Kunene

River, de Winter & Leistner 5741 (PRE, holo.! EA, K, M, SRGH, isos.!); Anabib

(Orupembe), 5698; without precise locality: Von Koehnen s.n. {M). Omaruru:

Brandberg, Tsisab Gorge: R. & F. von Wettstein s.n. (M), Rhodes University Student

s.n. (PRE), Merxmiiller & Giess 1654 (M, PRE).

Perennial suffrutex. Stems several from a woody rootstock, rather firm to wiry,

young stems often erect, older ones often prostrate; young parts rather densely setose-

pilose with patent bulbous-based stiff hairs; older parts glabrescent and becoming

scabrid, usually more or less woody, ultimate length about 60 cm., occasionally up to

90 cm., width near the base 1-5-2 -5 mm.; internodes on prostrate stems up to about

7 cm. long but usually shorter (2-5 cm.). Leaves firmly herbaceous drying pergama-

ceous; the lamina in outline ovate-orbicular, deeply 3-5-palmatifid, usually nearly

glabrous with only a. few setose hairs on midrib and main veins towards the base, 1-5

cm. long and broad; the lobes lyrate to pinnatifid or coarsely and acutely, usually

more or less coarsely, dentate; the petiole setose, 0-5-3 cm. long. Flowers solitary

or in 2-flowered cymes ; peduncles rather slender but firm, nearly terete, sparsely setose,

0-5-4 cm. long; bracteoles minute; pedicels 0-5-2 cm. long. Calyx green, in basal

portion rather densely setose or even bristly with stiff bulbous-based patent hairs;

sepals subequal, erect, oblong, narrowly elliptic or cuneate-oblong, acute or subobtuse

and usually minutely apiculate, 12-18 mm. long; in fruit turning brown, slightly en-

larged (mostly in width) and glabrescent becoming somewhat scabrid in lower portion.

Corolla funnel-shaped, white or pale yellow with wine-coloured or maroon “ eye ”,

glabrous, 2 -5-3 -5 cm. long and as much in diam., with a 5-lobed spreading limb,

lobes of limb obtuse, occasionally minutely apiculate. Stamens subequal, at the laterally

broadened base with papillose short stiff straight hairs. Pollen grains typical of the genus.

Ovary subglobose, glabrous. Capsule subglobose, apiculate, straw-coloured, glabrous,

about 12 mm. in diam. Seeds dark grey or brownish grey, puberulous, about 8 mm.

long.

This plant has been collected on several occasions but seems to be restricted to a

comparatively small area including a part of the Kaokoveld and the area of the Brand-

berg (Omaruru district) in South West Africa. It was most probably collected by

Guerich for the first time about 70 years ago, but his specimen (in Berlin) was destroyed

and it was never properly described, so that the evidence that “ Merremia guerichiana ”

is the same plant as the one described here as “ guerichii ” is circumstantial: (1) Guerich

69 was collected at Soris-Soris, which is not far from the Brandberg, and (2) a second

specimen was cited under M. guerichiana by Hallier, viz., Fischer 34 from East Africa.

This caused some confusion when Dr. B. Verdcourt of Nairobi studied Fischer 34 a

few years ago and decided it is M. ampelophylla. As he did not see the Guerich gathering

he took it for granted that the two specimens cited by Hallier were indeed identical.

417

so that in a remark in Kew Bull. 1957: 348 Dr. Verdcourt stated that M. ampelophylla

occurs both in East Africa and in South West Africa. The South West African plant is

undoubtedly specifically different from M. ampelophylla, but is closely related to it,

so that it is not surprising that one could confuse the species described here with the

East African one. It follows that Guerich 69 must have been conspecific with M. gueri-

chii, otherwise Hallier could not have cited it together with Fischer 34. Dr. Verdcourt

has seen a specimen of M. guerichii and agrees that it is distinct from M. ampelophylla.

The identity of M. guerichiana seems, therefore, to be definitely established. (^) How-

ever, the specific epithet “ guerichii ” was chosen which conveys the same but is not

nomenclaturally synonymous, so as to avoid the dangerous action of validating a name

without access to the original type {Guerich 69). In this way a new gathering could be

chosen as the type of M. guerichii, a solution which is much more satisfactory and not

at variance with certain recommendations in the Code of Botanical Nomenclature.

GRAMINEAE.

Andropogon ravus J. G. Anderson, sp. nov., A. sylvatico C. E. Hubbard affinis, sed ita

differt: habitu brevior, cum rhizomatibus, culmi 2-3-nodosi glabri, vaginae quam

internodia breviores, spiculae pedicellatae longiores, glumae spicularum pedicella-

tarum et glumae superiores spicularum sessilium sine aristis.

Gramen perenne ad 90 cm. altum (plerumque 45-60 cm.) rhizomatibus ramosis.

Culmi erecti simplices 2-3-nodosi glabri obscure striati vel leves teretes, nodis glabris,

internodiis exsertis glabris levibus. Folia ad 23 cm. longa 3-6 cm. lata plana linearia.

Vaginae glabrae internodiis breviores, lamina summa saepe ad aristam brevem redacta,

ligula membranacea ad 2 cm. longa. Racemi 7-12 cm. longi, spiculis pallidi-viridibus

purpura suffusis. Spiculae sessiles lineari-oblongae 6-9 mm. longae callo obtuso

barbato. Glumae subaequales, inferior canale profunde per mediam fundum, tenuiter

coriacea apicem versus admodum carinata carina anguste alata in dimidio superiore

scabrida vel ciliata, superior naviformis triente superiore carina scabrida vel ciliata

marginibus leviter ciliatis subacuta 1-3-nervata. Lemma floris inferioris 6-8 mm.

longum 2-carinatum obtusum 3-nervatum marginibus acute inflexis hyalinis et in dimi-

dio superiore ciliatis. Palea non nervosa hyalina 2-3 mm. longa Lemma floris superioris

5-6 mm longum inter lobos acutos aristatum arista ad 2 cm. longa infra genu contorta

Palea nulla. Spicula pedicellata dorsaliter compressa. Glumae inaequales, inferior

7-12 mm longa plana carinata carinis et marginibus ciliis hyalinis rigidis apice mucro-

nato vel obtuso, superior breviter naviformis 6-9 mm. longa 3-nervata vel obscure

5-nervata marginibus cilatis apice acuto. Flos inferior sterilis lemmate 7-8 mm. longo

marginibus inflexis in dimidio superiore obscure ciliato hyalino obtuso 3-nervato

palea non nervosa hyalina. Flos superior masculinus, lemmate 6 mm. longo apice

obtuso marginibus inflexis palea nulla. Antherae lineares 4-5 mm. longae.

Cape Province. — Maclear District: near lower Pitsing, sandy banks of Lusio

River, locally frequent, c. 4,700 feet, Acocks 12191.

Natal. — Bergville District: Cathedral Peak, towards Indumeni, Cleft Peak,

c. 7,000 feet, J. M. Gomell; occasional in Themeda triandra Grassveld c. 6,700 feel,

Killick 1261 (PRE, holotype). (*)

(*) Note added in proof: Dr. S. J. van Ooststroom of the Rijksherbarium, Leiden, kindly sent me

the notes on M. guerichiana which he found among the unpublished manuscripts left by Hallier. These

notes include a detailed description which agrees perfectly with the plants cited above (bulbous-based

hairs on vegetative parts and calyx, and similar details) and Hallier mentions only Guerich 69 (herb.

Berol.). This is the decisive evidence for the identity of “ M. guerichiana ” with the plants described

here as M. guerichii.

418

Orange Free State. — Bethlehem District: Golden Gate, c. 6,000-7,500 feet,

Sto)-y 1957.

Swaziland. — Mbabane: Ukutula, c. 4,300 feet, Compton 25593.

Perennial, up to 90 cm. tall, usually 45-60 cm., with a branched knotty system of

rhizomes. Culms erect, simple, 2-3-noded, glabrous, faintly striate or smooth, terete;

nodes glabrous; internodes exserted, glabrous and smooth. Leaves up to 23 cm.

long, 3-6 mm. wide, flat, linear, tapering to a setaceous point, narrowed towards the

base or base slightly rounded, glabrous, glaucous, margins often scabrid; sheaths

glabrous, striate, pallid or upper often suffused with purple, shorter than the internodes,

leaf blade of uppermost sheath often reduced to a short awn, mouth often with a few

long hairs; ligule membranous, scarious, up to 2 mm. long. Inflorescence of 2 (very

occasionally 3) spikelike racemes, shortly pedicelled or lowermost sessile. Racemes

7-12 cm. long; spikelets light green suffused with dull purple, paired, lower sessile,

upper pedicelled. Sessile spikelets linear-oblong, 6-9 mm. long; callus obtuse, bearded.

Glumes sub-equal, lower with a deep median groove, thinly coriaceous, strongly keeled

upwards, mucronate, keel narrowly winged, scabrid to ciliate in the upper half; upper

boat-shaped, with a scabrid or ciliate keel in the upper third, finely ciliate on the margins,

subacute, 1-3-nerved. Lemma of the lower floret 6-8 mm. long, 2-keeled with sharply

inflexed margins; margins ciliate in the upper half, hyaline, obtuse, 3-nerved. Palea

a nerveless hyaline scale 2-3 mm. long. Lemma of upper floret 5-6 mm. long, awned

from between the acute lobes; awn up to 2 cm. long, twisted below the knee. Palea

absent. Pedicelled spikelet dorsally compressed. Glumes unequal; lower 7-12 mm.

long, keeled, keels and margins with stiff hyaline cilia, apex mucronate or obtuse;

upper shallowly boat-shaped, 6-9 mm. long, 3- or faintly 5-nerved, margins ciliate,

apex acute. Lower floret sterile. Lemma 7-8 mm. long, margins inflexed, sparsely

ciliate in upper half, hyaline, obtuse, 3-nerved. Palea a nerveless hyaline scale. Upper

floret male. Lemma 6 mm. long, apex obtuse, margins inflexed. Palea absent. Anthers

linear, 4-5 mm. long.

Andropogon ravus is occasional in Themeda triandra grassveld on mountain slopes

near Mbabane in Swaziland and in the foothills of the Drakensberg from Bethlehem in

the Orange Free State to Maclear in the Cape Province. In the Bergville and Estcourt

districts it occurs on the terrace immediately below the main Drakensberg escarpment

between 6,000 and 7,000 feet. According to Mr. D. J. B. Killick, who has analysed the

sourveld in the Cathedral Peak area, A. ravus provides 0 ■ 75 per cent to 0 -95 per cent of

the basal cover in a total grass cover of between 40 per cent and 47 per cent. He also

states that it is readily recognised in the field, even when not in flower, because of its

glaucous-grey colour.

This grass is closely allied to A. sylvaticus C. E. Hubbard described from Nyasaland.

Our material was compared with this at Kew. The main differences are as follows:

A. sylvaticus is a tufted perennial 1-1 -3 m. high with the leaf-sheaths longer than

the internodes and the upper sheaths pilose with lax soft hairs, whereas A. ravus has a

system of branched knotty rhizomes, is generally shorter and rarely up to 90 cm. tall

and the glabrous leaf-sheaths are shorter than the internodes.

In A. ravus the pedicelled spikelets are longer and up to 12 mm. long compared with

those of A. sylvaticus which are 7-8 mm. long. The apex of both the upper and lower

glumes of the pedicelled spikelet of A. sylvaticus is aristate whereas the upper glume in

A. ravus is acute and the lower mucronate or obtuse. The apex of the upper glume of

the sessile spikelet in A. sylvaticus is aristate at the apex and subacute in A. ravus.

The awn of the lemma of the upper floret of the sessile spikelet is up to 2-0 cm.

long in A. ravus and only up to 1-5 cm. long in A. sylvaticus.

419

Danthonia stereophylla J. G. Anderson, sp. nov., D. drakensbergensi affinis, sed ita

differt: folia rigidissima inflorescentiae aequalia vel inflorescentiam superantia,

appendices laterales lemmatis angustiores, aristae loborum lemmatis longiores.

Gramen perenne caespitosum ad 80 cm. altum. Culmi erecti simplices compressi

obscure nodosi glabri leves. Folia rigidissima subpungentia glabra levia inflorescentiae

aequalia vel inflorescentiam superantia, vagina rigida pallida striata intus pubescente,

inferiore indurata, ligula ad seriem densam ciliarum brevium redacta, laminis filiformibus

rigidis ad 36 cm. longis conduplicatis supra striatis minute scaberulis, supra ligulam

sericeo-pubescentibus marginibus integris. Paniculae erectae laxe contractae angustae

lanceolatae ad 18 cm. longae, rhachide subangulata striata glabra levi, ramulis sub-

teretibus leviter patentibus solitaribus in axillaribus sericeis, pedicellis 3-20 mm. longis

angulatis vel subteretibus glabris nonnunquam scaberulis. Spiculae ad 2 cm. longae

pallidi-flavae nitidae. Glumae lanceolatae, 1-1-1 -8 cm. longae subaequales pallidae

nitidae glabrae leves, superiores marginibus basin versus saepe pubescentes, 1-nervosae

carinatae acutae. Rhachillae internodii glabri apicem versus ampliati complanati.

Lemmata lanceolata ad 16 mm. longa bilobata, supra glabra, intus sparse pilosa, margini-

bus media sub parte pilosa, apicem versus sparse pilosa pilis non caespitosis, lobis ad

9 mm. longis margines versus tenuiter scabris acuminatis in aristam gracilem scabram

productis. Callus truncatus dense barbatus. Arista media geniculata ad 2 cm. longa

scaberula, apicem versus angustior basin versus complanata et spiraliter contorta

non lobis adnata. Paleae ad 7 cm. longae intus pubescentes apice bilobatae distincte

carinatae carinis et lobis breviter ciliatis lobis ad 2 cm. longis. Antherae circiter 3 mm.

longae. Ovarium glabrum stylis distinctis.

Basutoland. — Drakensberg Area; often dominant on rock outcrops on summit of

Drakensberg between Indumeni Dome and Cleft Peak, 9,800 feet, Killick 2349; common

on rock outcrops on summit of Drakensberg in Cleft Peak Area, c. 9,800 feet, Killick &

Marais 2183.

Natal. — Estcourt District: edge of south krantzes, frequent, Tabamhlope, c.

.6,400, Acocks 1 1472; summit and elsewhere common, Bushman’s Pass, 7,100-9,000 feet.

West 1690; occasional on slopes of Drakensberg near Champagne Castle, 10,000 feet,

West 799. Bergville District: Tutumi Valley, 9,000 feet, 2339; Cathedral Peak

Forestry Station, 6,300 feet, Killick 1317; 7,400 feet, Killick 1184 (PRE, holotype).

A rigid, wiry, tufted, perennial. Culms erect, simple, up to 80 cm. high, compressed,

obscurely noded, glabrous and smooth. Leaves very rigid, subpungent, glabrous,

smooth, reaching up to or overtopping the inflorescence; sheaths rigid, pale, striate

above and minutely scaberulous, sericeous above the ligule, margins entire. Panicle

erect, loosely contracted, narrowly lanceolate, up to 18 cm. long; rhachis sub-angular,

striate, glabrous and smooth; branches sub-terete, slightly spreading, solitary or

apparently binate, lower up to 7 cm. long, upper shorter, sericeous in the axis; pedicels

3-20 mm. long, angular or subterete, glabrous occasionally scaberulous. Spikelets

up to 2-0 cm. long, pale yellow-green, shiny. Glumes lanceolate, 1-1-1 -8 cm. long,

sub-equal, pale, shiny, glabrous and smooth, upper glume often with a few long hairs

on the margin towards the base, 1 -nerved, carinate, acute. Rhachilla internodes

glabrous, broadened towards the apex and flattened. Lemmas lanceolate, bilobed, up

to 16 mm. long (including the lobes) glabrous dorsally. sparsely hairy on the inside,

pilose near the margin from about the middle downwards, sparsely pubescent on the

margin upwards, hairs usually not arranged in distinct tufts; lobes up to 9 mm. long,

finely scaberulous towards the margin, acuminate and gradually tapering into a scaberu-

lous, slender awn. Callus truncate, densely barbate. Central awn geniculate, up to

2 cm. long, scaberulous, tapering towards the apex, flattened towards the base and

spirally contorted, not adnate to the lobes. Palea up to 7 mm. long, pubescent on the

inside, apex 2-lobed, distinctly keeled; keels and lobes shortly ciliate, lobes up to 2 mm.

long. about 3 mm. long. Orary glabrous; styles distinct.

420

Danthonia stereophylla is closely related to D. drakensbergemis Schweick. The

distribution of these two species in the Drakensberg overlaps, but they can be dis-

tinguished on both ecological and morphological grounds.

D. stereophylla occurs on the Little Berg and its outliers (e.g. Tabamhlope), and

on the main Drakensberg escarpment from 6,000 feet to the summit at c. 11,000 feet.

So far it has only been collected between Bushman’s Pass (Langalibalele’s Pass) and

Cleft Peak, but it has been recorded further north at Mont Aux Sources by Mr. D. J. B.

Killick. At its lower altitudinal limits it grows on outcrops at the edge of koppies,

and in the alpine zone it is often dominant on dry basalt cliffs and horizontal pavements.

Its habitat is essentially xeric. D. drakensbergensis is restricted to the sub-alpine and alpine

zones above 8,000 feet and, in contrast to D. stereophylla it occupies mesic situations on

streambanks and seepage areas.

Vegetatively, the plants can be distinguished quite easily in the field. D. stereophylla

is a wiry plant up to 80 cm. tall with rigid, erect or suberect, grey-green leaves, whereas

D. drakensbergensis is a taller plant (up to 100 cm. high) with softer, olive-green leaves.

A peculiar feature of D. drakensbergensis (also of D. macowanii Stapf) which is not

evident in D. stereophylla concerns the behaviour of old leaves; the leaves break off

a little distance above the ligule and the portion remaining splits along the middle nerve

and the resultant halves recurve outwards at the apex.

The main differences in the spikelet are :

The hairs of the lemmas of D. drakensbergensis are arranged in 3 distinct tufts

on each side, arranged in an oblique row at about the middle while the margin of the

lemma is glabrous downwards. In D. stereophylla the hairs are arranged in a more or

less continuous band along the margin, with long soft hairs from the base to the middle,

and short rather sparse hairs from the middle to the apex. There is usually no definite

formation of tufts, but the hairs at the middle occasionally aggregate to form a single

tuft.

In D. drakensbergensis the lobes of the lemmas are broad, and taper rather abruptly

into the awns. In D. stereophylla, however, the lobes are narrow and taper very gradually

into the long slender awns, so that it is impossible to say where the lobes end and the

awns begin.

The lemmas (lobes plus awn) of D. stereophylla are 10-16 mm. long and up to 10

mm. in D. drakensbergensis. Schweickerdt in his description of D. drakensbergensis

in Fedde Rep. XLIII, 88 (1938) gives the length of the lemmas as up to 14 mm., but

it is possible that his range of material included specimens of D. stereophylla.

Panicum volutans J. G. Anderson, sp. nov., affine P. obscuranti (Woodrow) Stapf, sed

nodis hirsutis, spiculis glumis et lemmatibus floris inferioris longioribus multo

acuminatioribus, gluma inferiore manifesto 3-nervata differt.

Gramen annuum. Cultni erecti basi geniculati nodis hirsutis. Folia hispida pilis e

tuberculis minutis ortis praesertim superne; vaginae externe pilis e tuberculis minutis

ortis. Paniculae 40 cm. longae 25 cm. latae effuse ramosae. Spiculae 6 0-6 -5 mm.

(raro 5 -5 vel 7 0 mm.) longae. Gluma inferior 4- 5-5 0 mm. longa, longo-acuminata

manifesto 3-nervata, nervo uno secundaris divergente ab nervorum exteriorum, nervis

primariis scaberulis; superior 6 mm. longa longo-acuminata manifesto 7-9 nervata

nervis scaberulis. Lemma floris inferioris 4 -0-4 -5 mm. longum longo-acuminatum

7-nervatum. Antherae 1 ■ 5 mm. longae.

Cape Province. — Vryburg District: Armoedsvlakte, c. 3,960 feet, Mogg 8108.

Natal. — Estcourt District: Moordspruit, 8 miles north of Estcourt, c. 3,500 feet,

Codd 2466; Moordspruit near Estcourt c. 3,500 feet. Acocks 9947.

421

Transvaal. — Rustenburg District: 35 m. north of Swartruggens, c. 3,900 feet,

Acocks 19179. Pretoria District: Onderstepoort, Theiler 21131 \ Leendertz 11335;

Wonderboom Poort, Mogg 9937; Boschkop, 20 m. east of Pretoria, along banks of

small river, Kinges 1721; Koedoespoort, Pole-Evans 321; Middelkop Farm near

Pienaars River, Smith 2169; Rissik, c. 4,500 feet, Robinson sn. ; Onderstepoort, c. 4,300

feet, (7w Toit 14; Franks \Q6\6. Ermelo District : Between Bethal and Ermelo, Co<7i/ and

Muller 268. Naboomspruit District: Vogelstruispan, Galpin M. 548. Standerton

District: Teakworth, i'rcw? sn.; New Denmark, A/orow 16. Waterberg District: Bingley,

Kopje Alleen, Lyall Soutter sn. Warmbaths District: Crecy, Barenbrug 1. Venters-

dorp District; abundant in cultivated lands in vlei, Louw 1713 (PRE, holotype).

A loosely tufted annual. Culms up to 75 cm. tall, often sparingly branched, erect

from a geniculate base, often rooting at the lower nodes; internodes terete, glabrous,

smooth or faintly striate, easily compressible, lower often tinged with purple; nodes

hirsute. Leaves up to 23 cm. long and 1 mm. broad, linear, tapering to a sub-setaceous

apex, hispid with bulbous-based hairs on both surfaces, the upper more hairy than the

lower, margins scabrid; ligule a membranous long-ciliate rim; sheath striate, outer

surface hispid with bulbous-based hairs, inner surface glabrous, usually tinged with

purple. Panicle large, up to 40 cm. long and 25 cm. broad, profusely branched, branches

with 1-3 spikelets at their apex, naked below, lowest primary branches whorled, base

often enclosed in the uppermost sheath; rhachis angular, grooved, scaberulous on the

angles, often tinged with purple. Spikelets 6 0-6 -4 mm. (rarely 5- 5-7-0 mm.) long,

borne singly at the tips of long pedicels, lanceolate, acuminate, green, usually tinged

with purple. Glumes unequal; lower 4- 5-5 .0 mm. long, long acuminate, membranous,

prominently 3-nerved with 1 secondary nerve diverging from each of the outer primary

nerves, primary nerves scaberulous; upper 6-0 mm. long, long acuminate, membranous,

prominently 7-9 nerved with the nerves scaberulous. Lower floret barren. Lemma

4 -0-4 -5 mm. long, long acuminate, membranous, 7-nerved. Palea 2-0 mm. long,

elliptic-oblong, hyaline. Upper floret bisexual. Lemma 2-75 mm. long, crustaceous,

hard, smooth, narrowly oblong, apex obtuse, faintly 5-nerved, yellowish-grey with the

nerves showing up as light-coloured lines. Palea 2-75 mm. long, narrowly oblong,

apex obtuse. Anthers 1 • 5 mm. long.

Panicum volutans is found mainly in black turf soil in cultivated and other dis-

turbed areas. It is one of the grasses known as “ Rolling grass ” or “ Tumble Weed ”,

an allusion to the panicle which breaks off as a unit when mature and is carried away

by the wind.

It is closely related to the Indian grass P. obscurans to which it was formerly erro-

neously referred. However, Mr. B. de Winter, found at Kew in 1952, that the South

African plant is specifically distinct. Another close relative is P. hippothrix K. Schum.

a tropical African species which is doubtfully distinct from P. obscurans (Woodrow)

Stapf. Stapf in Flora of Tropical Africa 9, 699 (1902) says that P. hippothrix is “ very

similar and possibly identical with P. obscurans from the Deccan, but the blades of

P. obscurans are much wider (up to 7 lin) and the panicle is perhaps on the whole more

open with slightly larger spikelets ”. Mr. de Winter supports the view that these species

are probably synonymous since he could find no significant difference when dissecting

spikelets of both at Kew.

Chippindall in the “ Grasses and Pastures of South Africa 328 (1955) refers to

it as Panicum sp. aff. P. hippothrix K. Schum. because a valid name was not available at

the time.

P. volutans differs from P. obscurans as follows : In P. obscurans the spikelets are

about 4 mm. long with the lower glume approximately 3 mm. and the upper approxi-

mately 4 mm. long. In P. volutans the spikelets are about 6 mm. (rarely 5-5-7 .0 mm.)

422

long with the lower glume 4- 5-5 0 mm. and the upper glume 6 0 mm. long. The

glumes and the lemma of the lower floret are much more acuminate than in P. obscurans.

The nervation of the lower glume and the fertile floret is also different. In P. obscurans

the lower glume is prominently 5-nerved, whereas in P. volutans it is prominently

3-nerved with a faint secondary nerve diverging on each side of the outer primary

nerve. The fertile floret tends to turn brown in colour with the nerves showing up as

light coloured lines. This is not the case in P. obscurans. Vegetatively there is very

little difference between the two species except that the nodes are glabrous in P. ob-

scurans and loosely to densely hirsute in P. volutans. The panicle is perhaps more

rigidly branched than that of P. obscurans.

TILIACEAE.

Corchorus confusus Wild, sp. nov. ; afflnis C. triloculari L. sed habitu perenni, a C.

asplenifolio Burch, sed pedunculo fructuoso recto folds anguste lanceolatis vel

ovatis, ramulis omnibus partibus patenti-pilosis, a C. angolensi sed petiolis distincte

brevioribus et indumento differt.

C. serraefolius var. lancifolius Szyszyl., Polyp. Thalam. Rehmann.: 61 (1887)

pro parte quoad specim. Rehmann 4200 (BM; K) nomen nudum. C. trilocularis sensu

Burtt Davy, Man. FI. PI. Ferns Transv. & Swazil. 1 : 257 (1926) pro parte excl. specim.

Thorncroft 2058 et Nelson 381. Non C. trilocularis L.

Herba perennis, radice lignosa, ramulis annuis ad 0-6 m. longis subprostratis vel

erectis omnibus partibus saltern juvenilibus patenti-pilosis. Folia petiolata, petiolo

ad 8 mm. longo omnibus partibus patenti-pilosis; lamina ad 7 X 2-6 cm., anguste

lanceolata vel ovata, utrinque praesertim nervis pilosa, apice acuta vel obtusa, basi

rotundata vel leviter cordata ecaudata vel aliquando caudata c. 2 mm. longa, margine

crenata vel crenato-serrata, nervis basalibus 3-5 valde arcuato-adscendentibus, nervis

lateralibus utrinsecus 6-12, nervis omnibus supra leviter prominulis, subtus promin-

antitibus; stipulae ad 6 mm. longae, setaceae, sparse pilosae. Inflorescentia axillaris;

cymae pro nodo unicae, oppositifoliae, (1) 2-3-florae; pedunculi 0-4-2-5 cm. longi,

omnibus partibus pilosis; pedicelli pedunculis similes, ad 0-8 cm. longi; bracteae

ad 6 mm. longae, setaceae, sparse pilosae. Alabastra ovoidea demum oblongo-ovoidea,

plerumque apiculata. Sepalaadl X 0- 15 cm., lineari-lanceolata vel anguste lanceolata,

apice caudato-acuminata, extus pilosa. Petala flava, quam sepalis paullo breviores.

oblanceolata vel obovata, basi breviter unguiculata ciliata. Androgynopliorum c. 0-5

mm. altum ad apicem annulo glabro leviter undulato instructum. Stamina c. 50,

filamentosa. Ovarium 3-loculare, trigono-cylindricum, dense pubescens; stylus c.

2-5 mm. longus, tenuis, glaber. Capsula c. 3 x 0-2 cm., subcylindrica, pedunculo

fructuoso recto, apice obtusa, sparsissime scabroso-pilosa demum glabrescens; semini-

bus numerosis c. 2 x 1-2 mm. intense brunneo-griseis.

Cape Province. — Griqualand East, Umzimkulu R., fl. & fr. xii. 1884, Tyson 1413

(K; PRE).

Natal. — Zululand, Egoa Farm, fl. & fr. i. 1922, Curson s.n. (PRE); Hlabisa,

fl. 10. iv. 1954, Ward 2306 (PRE). Fort Yolland, fl. & fr. 19. Hi. 1903, Medley Wood

8976 (PRE). Estcourt, fl. & fr. 15.xii.l937, West 527 (PRE).

Transvaal. — Barberton, fl. & fr. ix-xii.l889, Galpin 577 (K; PRE). Johannesburg,

fl. xi.l902, Rand 1061 (BM). Shiluvane, fl. iii-v. 1905, Junod in Herb. Transv. Mus.

4895 (PRE). Lydenburg, Waterval Boven, fl. & fr. 20.xii.l914, Rogers 10933 (PRE).

Pretoria, Aapies Poort, fl. & fr. 1875-1880, Rehmann 4200 (BM; K); Pretoria, fl.

6.xii.l919, Verdoorn 69 (PRE). Rietvlei Reserve, fl. & fr. xi.l946, Repton 3170 (PRE).

Kruger National Park, 5 m. N. of Pretorius Kop, fl. & fr. 4.ii.l949, CoddScde Winter

4960 (K, holo.; PRE).

423

Swaziland. — Without precise locality, fl. vi.l910, Stewart in Herb. Transv.

Mus. 8999 (PRE).

Mozambique. — Lourengo Marques, Incanhini, fl. & fr. 13.i.l898, Schlechter

12024 (K). Umbeluzi,fl. &fr. 27.x. 1952, Cor VO///0 1283 (LM; K).

Southern Rhodesia. — Gwelo District: 15.iv.l905, Gardner 40 (K). Gwanda

District, fl. 17.xii.l956, Davies 23%6 (K; SRGH).

This species has been frequently confused in the past with C. asplenifolius; super-

ficially there is a resemblance but the straight fruiting peduncles and stems hairy on

all sides render it easy of recognition as a rule. In addition there is on the whole a

difference in leaf-shape and the fruits are always quite straight; in C. asplenifolius

they are typically slightly falcate. Burtt Davy (loc. cit.) included this taxon under

C. trilocularis but our plant is an indigenous perennial and cannot be conspecific with

the annual weed C. trilocularis, which is distributed throughout the tropics and sub-

tropics of the Old World. Our plant is also very near C. angolensis but the latter species

has markedly longer, more slender petioles and acuminate, ovate-lanceolate leaves:

it is also more nearly glabrous and although quite closely related it has a noticeably

different facies.

Corchorus merxmuelleri Wild, sp. nov. ; C. angolensi Exell & Mendonga affinis sed

folds ovatis utrinque dense stellato-tomentellis.

Suffrutex c. 2 m. altus ramis numerosis late patulis griseo-tomentellis demum

glabrescentibus, cortice brunneo. Folia petiolata, petiolo ad 6 mm. longo, utrinque

griseo-tomentello; lamina ad 2-2 x 1-6 cm., ovata, apice acuta, basi leviter cordata,

margine valde serrato-dentata, utrinque griseo-tomentella, nervis lateralibus utrinsecus

4, supra inconspicuis, subtus prominentibus; stipulae ad 2 mm. longae, subulatae,

griseo-tomentellae. Inflorescentia axillaris, cymae pro nodo unicae, oppositifoliae,

1-2-florae; pedunculi ad 2 mm. longi, griseo-tomentelli; pedicelli ad 5 mm. longi,

griseo-tomentelli; bracteae 1-5 mm. longae, subulatae, griseo-tomentellae. Sepala

5-6 X 0-75-1 0 mm., anguste oblanceolata, apice longe acuminata, dorso leviter

carinata, extus griseo-tomentella, intus glabra. Petala flava, 9x4-5 mm., anguste

obovata, basi breviter unguiculata margine minute ciliata. Androgynophorum c. 0-5

mm. ahum. Stamina c. 60, filamentosa. Ovarium 3-loculare, trigono-cylindricum,

brevissime tomentellum; stylus 4 mm. longus, glaber. Capsula 3-4-5 x 0-15 cm.,

subcylindrica, leviter torulosa, obtuse 2 mm. rostrata, griseo-tomentella demum sparse

stellato-puberula, pedunculis maturitate rectibus; seminibus numerosis brunneis

c. 2 X 1 mm. subcylindricis.

South West Africa. — Karibib, Erongogebirge, fr. 17.ii.l953, H. Kinges 3257

(M): bush. Omaruru, Otjihorongo Reserve, N. sides of porphyritic koppies on the

Ugab E. of Rooipoort, fl. & fr. 15.ii.l958, H. Merxmuller & W. Giess 1620 (M, type):

bush 2 m. tall, 3 m. wide.

This species is similar in habit to and is related to C. angolensis Exell & Mendonga

but its small ovate tomentellous leaves give it a very distinctive appearance. It is an

unusually large plant for a Corchorus species.

Corchorus pinnatipartitus Wild, sp. nov.; C. asplenifolio Burch, affinis sed foliis profunde

pinnatipartitis.

Herba perennis, radice lignosa, ramulis annuis usque ad 20 cm. longis suberectis

vel prostratis glabris sed lineis pilis crispis munitis. Folia petiolata, petiolo ad 7 mm.

longo, parte adaxiali pilosa, parte abaxiali glabra; lamina ad 2-5 x 1-2 cm., ambitu

anguste oblonga sed profunde pinnatipartita usque ad | lobata, aliquando ad basin

424

leviter palmatipartita, lobis versus folii basin majoribus obtusis vel subacutis simplicibus

rare laterale 1-dentatis, apice obtusa vel subacuta, basi truncata vel leviter cordata

ecaudata, utrinque minute punctulata, utrinque glabra vel costa subtus minute sparse

puberula; stipulae 2 mm. longae, lanceolatae, apicibus subulatis. Inflorescentia axilla-

ris; cymae pro nodo unicae, oppositifoliae, 1 -3-florae; pedunculi c. 1 mm. longi vel

nulli; pedicelli ad 1 cm. longi, tenues, puberuli; bracteae c. 1 mm. longae, setaceae.

Sepala plerumque purpurina, c. 7 mm. longa, oblanceolata, apice subacuta vel acuta,

glabra. Petala flava vel plerumque roseo-purpurea suffusa, c. 7 mm. longa, anguste

obovata, basi breviter unguiculata minute ciliata. Androgynophorum c. 0-25 mm.

ahum. Stamina numerosa, filamentosa. Ovarium 3-loculare, ellipsoideum, aliquantum

trigonum, minutissime glanduloso-papillosum ; stylus 3 mm. longus, glaber. Capsula

c. 2 X 0-2 cm., subcylindrica, apice obtusa, sparse glandulosa, pedunculis plerumque

contortis; seminibus numerosis fuscis 1-5x1 mm. subcylindricis angulatis.

Cape Province. — Hay Division, Wilde-als Put, 16 miles East of Griquatown,

frequent on red sand over limestone, fl. & fr. ii.l937, Wilman 4075 (K; KMG). Barkly

West Division, Silverstreams, fl., 15.iii.l939, Esterhuysen 776 (BOL; K). Kuruman

Division, Kuruman, limestone, Edmonstone-Sammons 54 (KMG); Cardington,

common on surface limestone, fl. iv.i.l940, Ferrar s.n. (K; KMG); Cotton End,

fl. & fr., iv.l940, Ferrar s.n. (K); Cardington, fl. & fr. iv.l940, Esterhuysen 2188 (K,

type; BOL; PRE) Cotton End, limestone, fl. iv.l940, Esterhuysen 2251 (BOL; K;

PRE): flowers opening in afternoon.

This species is close to C. asplenifolius Burch, but is very distinct because of its

pinnatipartite leaves. It appears to be confined to limestone soils and to have a very

restricted range of distribution.

VERBENACEAE.

Priva auricoccea A. Meeuse spec, nov., P. curtisiae Kobuski arete affinis sed calycibus

fructiferis majoribus sparse minuteque tuberculatis, fructu aurantiaco spinis

longioribus horrido praecipue differt.

Suffrutex perennis c. 50 cm. altus e basi ramosus. Caules erecti quadrangulati

pilis rectis vel subuncinatis puberuli vel hirtelli, demum glabrescentes. Folia ovato-

oblonga vel subhastato-oblonga, obtusa, crenato-dentata, ad basin abrupte et anguste

attenuata, pilis minutis setulosis subscabra, petiolis applanatis hirtellis subnullis vel

interdum ad 7 mm. longis. Racemi 15-25 cm. longi, subhirtelli, pedicellis infra calycem

dilatatis articulatisque post anthesin 1-2 mm. longis, bracteis lanceolatis acuminatis

acutissimis 3-4 mm. longis. Corolla ignota. Calyx fructifer compresso-subglobosus

pilis uncinatis subvelutinus, sparse minuteque tuberculatus, 7-9 mm. longus et latus.

Mericarpia spinis rectis vel leviter curvatis 1-2 mm. longis horrida, fusco-aurea, hirsuto-

puberula, c. 6 mm. longa, 4 mm. lata, 2-5 mm. crassa.

South West Africa. — Kaokoveld: 1-6 m. N.E. of Kaoko-Otavi on road to

Ohopoho, de Winter & Leistner 5532 (PRE, holo. ! ).

A perennial herb with tough base about 50 cm. tall, producing several erect and

sparingly branched stems from the woody top of the rootstock. Stems more or less

distinctly quadrangular becoming more terete when passing into the terminal spike-

like racemes, rather densely leafy, shortly pubescent to hirsutulous with straight or

sometimes more or less distinctly uncinate hairs, the nodes between the petioles of the

opposite-decussate leaves marked with a narrow band of longer stiff hairs. Leaves

ovate or ovate-oblong to somewhat hastate-oblong, obtuse to rounded at the apex,

broadly cuneate to subtruncate at the base, abruptly narrowed at the base into narrow

lateral decurrent wings of the up to 7 mm. long, setulose-hispidulous petioles, with a

425

distinctly but usually not very deeply incised crenate or crenate-serrate margin, appa-

rently rather dark green above, paler beneath, on both surfaces subscabrid through

fine setulose-strigose hairs with slightly raised broad base; blade 2-4-5 cm. long and

1-2-3 cm. broad, with a venation (only conspicuous on lower surface, indistinct on

upper surface) consisting of a rather slender midrib from which on either side at the

base of the blade a usually unequally forked rarely single secundary vein branches

off and higher up 3-5 usually unforked ones branch off, connected by rather distant

more or less parallel tertiary veins oriented perpendicular to the secondary ones.

Racemes 15-25 cm. long; the axis a direct continuation of the stem and with the same

type of pubescence; pedicels ultimately 1-5-2 mm. long, dilated at the apex into a

flat disc-shaped articulation with the calyx; bracts lanceolate-acuminate, setulose,

3-4 mm. long. Corolla unknown. Fruiting calyx inflated, nearly closed at the orifice,

orbicular in outline, laterally compressed, subvelutinous with short stiff greyish-yellow

straight to uncinate setose hairs and sparsely tuberculate with blunt or pungent small

protuberances, 7-9 mm. long and broad, 3-4 mm. thick. Fruit of 2 dry 2-celled cocci

(mericarps); cocci spinescent with 1-2 mm. long straight to distinctly curved spines,

of a bright golden-brown colour caused by a dense pubescence of minute stiff hairs,

with the spines measuring about 6 mm. by 4 mm. by 2 - 5 mm.

Although flowers are lacking, the plant is clearly a species of Priva, a genus whose

species are mostly characterised by fruit characters, the flowers being, apart from dif-

ferences in size and pubescence, rather uniform in structure. In Moldenke’s monograph

of the genus in Fedde, Repert. 41 : 1-76 (1936) it would key out as P. curtisiae Kobuski,

an East African species, to which it is indeed closely related on account of the cocci

being spiny throughout on back and sides and which it also resembles very much in

habit, shape and texture of leaves, type of pubescence, “ bearded ” stem-nodes, etc.

but from which it differs in several respects. The petioles are longer, the fruiting calyx

and cocci are larger and the spines on the cocci longer and more curved than in P.

curtisiae, the calyx bears short protuberances (not found in P. curtisiae) and the fruit-

cocci, which are flat on the commissural surface (not excavated as in P. curtisiae),

are of a striking golden-brown colour, hence the proposed specific epithet. It is possible

that more differences will be found once the flowers of P. auricoccea have been collected.

Although the differences mentioned may appear to be rather small, they are not smaller

than those between some of the other species of long standing in this genus. In view

of the urgency in connection with the preparation of the proposed Flora of South West

Africa by the workers of the Botanische Staatssammlung at Munich, the description

is not delayed for lack of flowers which, in this genus, are not so important for diagnostic

purposes as in most other genera.

I

V*

1'

427

Homonyms in the Prodromi of Thunberg

and Burman

by

J. E. Dandy

I was asked by the Division of Botany, Pretoria, to give an opinion concerning

the status of the name Cestrum venenatum used by Thunberg in his Prodromus (1794),

the question being whether this was intended to be the name of a new species, or whether

Thunberg was simply adopting the already existing name C. venenatum Burm. f. (1768).

The two works concerned are Thunberg’s Prodromus Plantarum Capensium (1794, 1800)

and N. L. Burman’s Florae Capensis Prodromus appended to his Flora Indica (1768).

Investigation of these showed that there are a number of cases similar to that of Cestrum

venenatum, and I therefore thought it advisable, before forming an opinion, to carry

out a comparative analysis of all the specific names published in the two Prodromi.

Following are the results.

Thunberg dedicated his Prodromus to N. L. Burman but this is apparently a mark

of appreciation of and gratitude for help and encouragement given by Burman, not

a reference to Burman’s own published work on the Cape flora. In fact Thunberg

nowhere in his Prodromus makes any direct reference to Burman’s Prodromus, and the

only mention of N. L. Burman in the text is on p. 108 where he is cited under Heliophila

in connexion with a description of that genus published in a Swedish journal. The

other references to “ Burm.” given by Thunberg (e.g. on pp. 97, 98) are to the pre-

Linnaean work of J. Burman, the elder. On the other hand, Thunberg throughout his

Prodromus cites various Linnaean works, with which he was obviously familiar, and

also makes occasional references to post-Linnaean publications by Bergius, L’Heritier,

Sparrman, Cavanilles, Smith, Jacquin, Delaroche, Houttuyn and Acharius (lichens).

N. L. Burman in his Prodromus published about 260 legitimately named new species

from the Cape, usually providing a description of his own but sometimes citing an

earlier one. It is noteworthy that only 22 of the names listed by Thunberg correspond

(in form, at least) with new specific names published by Burman in his Prodromus.

They can conveniently be listed as follows, each with the Thunberg page-number and

then the Burman page-number in parentheses;

Gladiolus junceus 8 (2)

Echium spicatum 33 (4)

Erica hispida 70 (11)

Ononis prostrata 129 (21)

Othonna linifolia 167 (29)

Satyr ium cornutum 5 (30)

Polygala stipulacea 121 (20)

Lobelia volubilis 39 (29)

Aloe aracimoides 61 (10, “ arachnoidea ”)

Schoenus spicatus 16 (3)

Echium hispidum 33 (5)

Cestrum venenatum 36 (5)

Vitis capensis 44 (7)

Limeum aethiopicum 68 (11)

Reseda capensis 85 (13)

Orobanche capensis 97 (17)

Geranium ovale 113 (19)

Ononis strigosa 130 (21)

Phaseolus capensis 130 (21)

Psoralea linearis 135 (22)

Gnaphalium spathulatum 151 (25,

“ spatulatum ”)

Osteospermum incanum 166 (29)

428

Of these, Gladiolus junceus is G. junceus L. f. (1781), non Burm. f. ; Echium spicatum

is E. spicatum L. f. (1781), non Burm. f. ; Erica hispida is E. hispida Thunb. (1785),

non Burm. f. ; Ononis prostrata is O. prostrata (L.) L. (1771), non Burm. f. ; Othonna

linifolia is O. Unifolia L. f. (1781), non Burm. f. Satyrium cornutum is 5”. cornutum

(L.) Thunb., a new combination based on Orchis cornuta L., which has no connexion

with 5. cornutum Burm. f. Polygala stipulacea is evidently taken up from Linnaeus,

Mant. PI. Alt. 260 (1771), where Linnaeus makes an interpretation of P. stipulacea

Burm. f. Similarly volubilis is taken up from Linnaeus fil., Suppl, PI. 396 (1781),

where an interpretation is made of L. volubilis Burm. f. Aloe “ arachnoides ” is A.

arachnoidea (L.) Burm. f., this being Linnaeus’s A. pumila var. arachnoidea which was

raised to specific rank in 1768 both by Burman and (probably later) by Miller, Card.

Diet., ed.8; the spelling A. “ arachnoides ” was used by Thunberg in one of his earlier

works (Diss. Bot.-med. Aloe, 7) in 1785.

Thus only the 13 names in the right-hand column above have to be considered,

and it is at once notable that they all have common-place epithets which might occur

to any author coining names for new species. Furthermore, Thunberg’s descriptions

of these species all differ more or less from those given (or referred to) by Burman for

his species, and while Thunberg was in the habit of remodelling descriptions (even for

Linnaean species) the differences in some cases are so striking that it is difficult to believe

that he is referring to Burman’s species. For example;

Psoralea linearis Burm. f. : “ folds simplicibus . . . fforibus terminalibus ternis

Psoralea linearis Thunb. : “ folds ternatis . . . fforibus lateralibus solitariis ”.

Ononis strigosa Burm. f. ; “ fforibus axillaribus sessilibus ”.

Ononis strigosa Thunb. : “ umbellis terminalibus ”.

Limeum aethiopicum Burm. f. : “ foliis ovato oblongis ”.

Limeum aethiopicum Thunb. ; “ foliis lineari-lanceolatis ”.

From the above we have the following points:

(1) Thunberg nowhere refers directly to Burman’s Prodromus.

(2) Only very few of Thunberg’s names coincide with names proposed by Burman

and not used by any other author.

(3) Of these few, all have common-place epithets and all have descriptions

differing (sometimes fundamentally) from Burman’s.

My conclusion is that none of Thunberg’s names (except Polygala stipulacea, and

Lobelia volubilis which he adopted through Linnaean works) can be accepted as an

application of a new Burman name; and that all the names listed in the right-hand

column above, including Cestrum venenatum, are to be regarded as illegitimate later

homonyms published as new by Thunberg.

429

South African Labiatae

by

L. E. Codd

THORNCROFTIA

The genus Thomcroftia was described by N. E. Brown in Kew Bull. 1912:281 and

was based on T. longiflora N.E.Br., which he described at the same time. Only one

specimen was cited, namely: “ Transvaal, among rocks near Barberton, 1220 m,

Thorncroft 795”. Two specimens in the Transvaal Museum Herbarium (now incor-

porated in the National Herbarium, Pretoria), collected by George Thorncroft, supply

additional information. On No. 795 the label states: “ Herb, 2 ft., flowers pink, found

among sandstone rocks, Joe’s Luck near Barberton, alt. 4,000 ft., April 1911 and

on a second gathering without number: “Joe’s Luck footpath near Barberton, on

sandstone rocks, alt. 4,000 ft.. May 1913 ”.

The only other material of the species known until recently is a plant cultivated

by the Cambridge Botanic Garden from seed sent by Thorncroft, and figured for the

Botanical Magazine, t. 8824 (1919). In order to evaluate the plant as to generic status,

it was felt that more material was required and the assistance was enlisted of Barberton

residents, including George Thorncroft’s son, Mr. J. N. Thorncroft, and grandson,

Mr. N. G. Thorncroft.

In George Thorncroft’s time a flourishing settlement existed in the mountains

some miles to the north-east of Barberton, called Eureka City, which boasted of

amenities attractive to inhabitants of the neighbourhood. Joe’s Luck is the name of

a mine nearby and a railway siding in the low country at the foot of the mountains,

and it is reported that a well-used footpath connected the latter with Eureka City some

2,000 feet higher in altitude. At present, however, very little is left of Eureka City

and the footpath cannot be traced with certainty. Thus more than a year passed before

Mr. and Mrs. P. F. Clarke and Mr. N. G. Thorncroft found the species in the mountains

above Joe’s Luck Siding, probably not far from the type locality.

During their searches in the environs of Barberton, Mr. and Mrs. Clarke collected

two species which were clearly closely allied to Thomcroftia longiflora, and one of these

was again collected near Barberton by the late Prof. Werdermann of Berlin, during

his recent visit to South Africa. Further study revealed that these two species had been

described in the genus Plectranthus, as P. thorncroftii S. Moore and P. succulentus Dyer

and Bruce. The opportunity is now taken to transfer them to Thomcroftia.

Although the three species now in Thomcroftia superficially resemble Plectranthus

or Orthosiphon, they possess characters which clearly place them in a distinct genus.

These characters are: {a) the flowers are borne singly in the axils of the bracts, not in

several-flowered verticils; {b) the bracts are not clearly differentiated from the leaves,

but grade into the upper leaves on the flowering shoot; and (c) the corolla, although

obscurely bilabiate is, in fact, 4 lobed, comprising a lower, entire, boat-shaped lip,

an erect upper emarginate lobe and two lateral, narrowly oblong lobes which are deflexed

in the open flower and spread on each side of the lower lip. This may represent a more

primitive state than in Plectranthus and Orthosiphon, where the lateral lobes are

apparently fused with the upper, producing a more or less obscurely 4-lobed upper

lip, while the lower boat-shaped lip remains much the same.

In some minor respects the generic description of Thomcroftia must be modified

to accommodate the two additional species. For instance, the upper calyx lobe may

or may not be decurrent on the tube, and the corolla tube may be short or long.

430

The basic chromosome number in Thorncroftia appears to be n = 7, as in Plectmn-

lhus and Orthosiphon, judging by the number 2n = 28 recorded for T. succulentus by

de Wet in S.A. Journ. Sci. 54:153 (1958). Another point of general interest is that the

three species of Thorncroftia are almost invariably parisitised by a weevil, Apion sp.,

which causes thickened swellings in the succulent stems. It is not known how specific

this weevil is, but very rarely are Plectranthus specimens seen with these swellings and

Orthosiphon not at all.

Plants up to 25 cm tall; corolla tube less than 1 cm long I. T. thorncroftii.

Plants 30-120 cm tall; corolla tube T5-3-8 cm long:

Inflorescence relatively lax; corolla tube 3 -4-3 -8 cm long; pubescence consisting of

simple or multicellular, not dendroid hairs 2. T. longiflora.

Inflorescence compact; corolla tube T5-2 cm long; dendroid hairs present, mixed

with simple and multicellular straight hairs 3. T. succulentus.

1. T. thorncroftii (5. Moore) L. E. Codd, comb. nov.

Plectranthus thorncroftii S. Moore in Journ. Bot. 56:39 (1918). Type: Baiberton.

Thorncroft in Rogers 16987 (BM, holo., PRE, iso.). On the PRE specimen the number

has been altered to 14987.

Succulent herb 10-25 cm high; stem erect, sparingly branched, about 8 mm in

diameter at the base, tapering to 4 mm in diameter above, pubescent with glandular,

simple and scattered multicellular hairs. Leaves opposite, shortly petioled, obovate

to oblong-obovate, 1 • 5-2 cm long and 6-8 mm broad, fleshy, drying subcoriaceous,

thinly pubescent and glandular on both surfaces, cuneate at the base, apex rounded:

margin sparingly crenate-dentate in the upper third ; pubescence mainly of multicellular,

unbranched hairs. Inflorescence a raceme or sparingly branched panicle 5-8 cm long;

bracts opposite, leaflike, up to 6 mm long below, becoming progressively smaller and

narrower above to about 3 mm long; flowers solitary in the axil of each bract, pedicels

3- 6 mm long, glandular puberulous. Calyx 5-toothed, glandular puberulous, obscurely

2-lipped, 3-3-5 mm long in flowering stage increasing to 7 mm long in fruit, tube

campanulate; the upper tooth the largest, ovate-deltoid, decurrent on the tube, the

lower 4 subequal, lanceolate- to lineai -triangular. Corolla whitish with purplish spots

on the upper lobe, obscurely bilabiate, sparingly pubescent without; tube campanulate,

expanding abruptly from the calyx, 4-5-5 mm long and 2-5-3 mm broad; upper lip

erect, sub-orbicular, emarginate, 6 mm long and equally broad; two lateral lobes

oblong, 3-4 mm long, deflexed; lower lip boat-shaped horizontal, 6-7 mm long.

Stamens 4 exserted; filaments free, up to 6 mm long. Ovary 4-lobed, glabrous; style

slender, exserted; stigma shortly bifid.

Found among rocks in mountain grassland. Known so far only from the mountain

massif between Barberton and Havelock Mine.

Transvaal. — Barberton: Thorncroft in Hb. Rogers 16987 (or 14987); Havelock

road, growing among rocks and grass on koppie near Angle Station, alt. 5,000 ft..

24th March, 1956, Clarke 41; between Barberton and Piggs Peak, Werdermann 2197.

This species has a short corolla tube and the corolla bears a superficial resemblance

to certain species of Plectranthus, e.g. P. thunbergii, as remarked by Spencer Moore.

However, it has the essential features of Thorncroftia, namely the four-lobed corolla,

solitary flowers in the axils of the bracts and the rather leaflike bracts. The upper

calyx tooth is decurrent on the tube as in Orthosiphon.

2. T. longiflora N.E.Br. in Kew Bull. 281 (1912); Prain in Bot. Mag. t.8824 (1919).

Type: Transvaal, Barberton, Thorncroft 795 (K, holo., PRE, iso.).

Succulent herb or soft shrublet 30-60 cm high; stems several, arising from a

thickened rootstock about 4 cm in diameter, 1 cm in diameter at the base, tapering to

4- 5 mm in diameter above, sparingly branched, densely and shortly grey tomentose;

431

pubescence of simple and multicellular unbranched hairs. Leaves opposite, shortly

petiolate, elliptic to obovate, 1-2 cm long and 4-10 mm broad, cuneate at the base,

apex rounded, entire or faintly crenate-dentate in the upper third, densely short tomentose

and pitted with sessile glands on both surfaces. Inflorescence a panicle; racemes

3- 9 cm long, densely glandular puberulous; bracts opposite, leaflike, up to 1-2 cm

long below, becoming progressively smaller and narrower above to about 2-5 mm

long; flowers solitary in the axil of each bract; pedicels 2-5-4 mm long, glandular

puberulous. Calyx 5-toothed, glandular puberulous, obscurely 2-lipped, the upper

tooth the largest, ovate-deltoid, acute, decurrent on the tube, the lower 4 subequal,

lanceolate-triangular, acuminate. Corolla pink to mauve pink with deeper flecks on

the lateral lobes, obscurely bilabiate, glandular and sparingly pubescent without;

tube narrowly cylindrical, straight, 3 -4-3 -8 cm long and 1-1 -5 mm broad; upper lip

erect, oblong, emarginate, 7-8 mm long and 4-5 mm broad; two lateral lobes narrowly

oblong 5-6 mm long and 1-5 mm broad, deflexed; lower lip narrowly boat-shaped,

horizontal, 6-7 mm long. Stamens 4, exserted; filaments free, 6-7 mm long. Ovary

4- lobed, glabrous, style slender, exserted; stigma shortly bifid.

Found in crevices on relatively bare rock outcrops. Evidently rare and localised

in mountains north-east of Barberton.

Transvaal. — Barberton: Joe’s Luck footpath, April, 1911, Thorncroft 795;

May, 1913, Thorncroft in TM 12775; March, 1957, Thorncroft and Clarke s.n.

(lu

3. T. succulent|^ {Dyer and Bruce) L. E. Codd, comb. nov.

Plectranthus succulentus Dyer and Bruce in Flow. PI. Afr. 27:t.l073 (1949). Type:

Soutpansberg, Entabeni, Loock in PRE 27461 (PRE, holo.).

Succulent herb or shrub 60-120 cm high; stems several arising from a thickened

rootstock 3-6 cm in diameter, 0-7-1 -5 cm in diameter at the base and tapering to

4- 5 mm in diameter above, sparingly branched, densely and shortly grey tomentose.

Leaves opposite, shortly petiolate, ovate-elliptic or obovate 1 • 6-3 cm long and 1 - 5-2 cm

broad, cuneate or obtuse at the base, apex rounded, crenate in the upper two-thirds,

thinly to densely tomentose and glandular on both surfaces; pubescence consisting

of branched and unbranched multicellular or simple white hairs. Inflorescence a

congested racemose panicle 8-14 cm long, glandular and hispid with dendroid hairs;

racemes 2-5-4 cm long; bracts opposite, somewhat leaflike, ovate and 1-2 cm long

below, becoming progressively smaller and narrower above to about 5 mm long and

2-3 mm broad; flowers solitary in the axil of each bract; pedicels 0-5-2 mm long,

pubescent. Calyx 5-toothed, with scattered glands and dendroid hairs, obscurely

2-lipped, the upper tooth somewhat larger than the rest, ovate-deltoid, acuminate,

not obviously decurrent on the tube, the lower 4 subequal, lanceolate-triangular,

acuminate. Corolla pale lilac with darker spots on the upper and lateral lobes, obscurely

bilabiate, glandular and sparingly pubescent without; tube narrowly cylindrical,

straight, 1-5-2 cm long and 1-5-2 mm in diameter; upper lip erect, obovate-oblong;

emarginate, 6-8 mm long and 5-7 mm broad; two lateral lobes narrowly oblong,

5- 6 mm long and 1 mm broad, deflexed; lower lip narrowly boat-shaped, 5-6 mm long,

at first horizontal and later reflexed. Stamens 4, exserted; filaments free, 3-4 mm

long. Ovary 4-lobed, glabrous; style slender, exserted; stigma shortly bifid.

Occurs among rocks or in crevices of bare rock outcrops in mountains of eastern

and northern Transvaal.

Transvaal. — Soutpansberg: Soutpansberg Mountains, Entabeni, Loock in

PRE 27461 ; Bruce and Kies 1 ; Codd 4194; Summit of Franzhoek Peak, Galpin 14881 ;

Hangklip, Gerstner 5903; McKay s.n.; Meeuse 10163; 20 miles W. of Mountain Inn,

Meeuse 91%1 ; Wylliespoort, Greyer in TM 19451. Pilgrimsrest : Drakensberg excarp-

ment opposite Mariepskop, Codd 7904. Barberton: Hyslop’s Creek, Clarke 213.

432

THE COLEUS CANINUS COMPLEX

Coleus neochiius (Schltr.) L. E. Codd, comb. nov.

Plectranthus neochiius Schltr. in J. Bot. Lond. 34: 394 (1896); Cooke in FI. Cap. 5, 1:

285 (1910). Type: Transvaal, near Barberton, Galpin 968 (GRA, NH, iso.).

Coleus schinzii Guerke in Bull. Herb. Boiss. 6: 555 (1898); Bak. in El. Trop. Afr. 5:

430 (1900). Type: S.W. Africa, Ovamboland, Tsumeb, Schinz 56 (Z, holo.,

PRE, photo.). C. pentheri Guerke in Ann. Naturhist. Hofmus. Wien, 20: 48

(1905); Cooke in FI. Cap. 5, 1: 289 (1910); Bruce in Hook. Ic. PI. 34: t.3375

(1938). Type: Cape Province, Peddie District, Breakfast Vlei, Kwok in Hb.

Penther 1716 (W, holo., PRE, iso.). C. carnosus Dinter, ined. ; Eliovson, S.A,

Flow, for the Garden, 165 (1955) nom. nud.; non Hassk. nec A. Chev.

Found in relatively dry bushveld, or occasionally among rocks in moister grassland,

in the eastern Cape Province, Natal, Transvaal, S.W. Africa and Rhodesia.

Cape. — Peddie: Breakfast Vlei, Kwok in Hb. Penther 1716. East London:

Kintza River Mouth, Galpin 6554. Stutterheim : Kabaku Hills, Acocks 9547. Komgha :

Flanagan 557.

Natal. — Pietermaritzburg: Thornville Junction, Chapman 1368. Weenen:

1 mile W. of Muden, Codd 8602.

Transvaal. — Carolina: Waterval Boven, Rogers 14485. Barberton: near

Barberton, Galpin 968 (GRA, NH); Codd 9531; on Lebombo Range, near Kobinja,

Codd 119^. Pilgrim’s Rest: near Graskop, Galpin 14433; near Vaalhoek, Meeuse

10013. Lydenburg: Sekukuniland, Barnard and Mogg 704. Pietersburg: Munro in

TM 15282. Potgietersrus : Rogers 4817; Comins 917. Waterberg: Towoomba

Research Station, Sidey 1401. Pretoria: Rietvlei Research Station, Acocks 11262;

13 miles S.E. of Pretoria, Codd 2570; Quaggapoort, Verdoorn s.n. ; near Saltpan,

Pole Evans 4775; 10^ miles N. of Hammanskraal, Codd 9385. Rustenburg: 19 miles

S. of Northam, Cor/fl' 8632. Marico: near Zeerust, in TM 13219; Leistner 349.

Lichtenburg: Jenkins in TM 13220; Sutton 365.

S.W. Africa. — Ovamboland: Tsumeb, 5'c/2/«r 56 (Z). Grootfontein: Schoenfelder

S643; near Otavi, Dinter 5634; de Winter 2853.

S. Rhodesia. — Intabazinduna Reserve, Davies 19. Inyanga Village, Wild 3853.

Matobo: Bulawayo, Rogers 5933; 13657; Bima Kobila, Miller 2054; Matopos

Research Station, Plowes 1388. Mrewa: near Shawanoe, Leach 8084.

I am indebted to the Director of the Natural History Museum, Vienna, for presenting

the National Herbarium, Pretoria, with a portion of the type of C. pentheri Guerke,

and to the Director of the Zurich University Herbarium for sending us the type of

C. schinzii Guerke on loan. These were compared with material of the type gathering of

Plectranthus neochiius Schltr. {Galpin 968 in NH and GRA) and are considered to be

conspecific. The differences between the three types are mainly in habit and pubescence,

characters which vary a good deal in this group. In the eastern Cape Province, the

stems are semi-prostrate to decumbent, about 10-20 cm long, with numerous glands

and short, dense, adpressed pubescence on leaves and stems and scattered long, multi-

cellular hairs on the stems. In the Transvaal, S.W. Africa and Rhodesia, the stems are

more ascending, 30-45 cm high, and usually more conspicuously villous, though some-

times possessing only the glandular and adpressed pubescence. There is some indication

that the plants may behave as annuals in South West Africa, though this requires

investigation. Elsewhere the plants are perennial, often with somewhat tuberous roots

in the young stage. The leaves are usually obovate to elliptic-ovate, faintly crenate

at the rounded or obtuse apex and narrowly cuneate at the base. The inflorescence

characters are relatively constant. In the bud stage, the inflorescence is a 4-angled

spike 3-4 cm long, composed of 4 rows of densely imbricate, ovate, acuminate bracts.

433

The bracts are shed as each verticil of flowers starts to open. As flowering proceeds,

the rhachis elongates, with the result that the verticils become separated by intervals

of 5-15 mm, producing an interrupted spike of up to 15 cm long. Depending on the

length of the inflorescence, 5-12 spaced verticils may be seen below the coma of bracts,

the uppermost 3 or 4 still flowering and the lower ones in fruit, with the rhachis easily

visible between the verticils. The corolla is usually about 1 • 5 cm long, varying from

1 -2-1 -8 cm. The lower lip is large and boat-shaped, purple to violet, while the upper

lip is erect, small, whitish, mottled with bluish-purple.

Two closely related species, both of which bear epithets older than C. neochilus,

must be taken into consideration, namely, C. caninus (Roth) Vatke (syn. C. spicatus

Benth.) and C. comosus Hochst. ex Guerke. There seem to be adequate reasons at

present for maintaining the three as distinct, though the tropical material seen is

admittedly scanty. It is possible that when more good material becomes available

from tropical territories the position should be reviewed again. In any case, the following

notes will act as a guide to the characters to which collectors should give attention.

Both C. caninus and C. comosus can usually be separated from C. neochilus by

their shorter, denser inflorescences. As the inflorescences start to flower they elongate

slightly, but not to the extent of C. neochilus, so that the verticils do not become separate,

except perhaps the lowest one or two. Even when flowering is well advanced, the

spike is usually 3-5 cm long, rarely up to 9 cm long. The corolla in C. caninus is shorter

than in C. neochilus, rarely exceeding 1 cm long, while in C. comosus it is usually longer

than in C. neochilus, being 2-2-5 cm long. Usually in drying specimens the corolla

becomes shrivelled or distorted, making measurements difficult.

C. caninus was described from India, and herbarium specimens were kindly sent

on loan to us by the Forest Research Station, Dehra Dun, Pakhistan. Specimens

matching the Indian material have been seen from Kenya, Tanganyika, Rhodesia and

South West Africa. Seed was extracted from one specimen received from the latter

territory {Sachs 2) and the resulting plants in cultivation behaved as annuals. It is

also fairly definite that Wild 4544 from Rhodesia and Greenway 9190 from Tanganyika

are annuals. Whether all material referrable to C. caninus is of annual habit is not

known; if this proves to be the case, it would be an additional useful character for

separating it from the predominantly perennial species, C. comosus and C. neochilus.

Although no material has been seen of the type gathering of C. omahekensis Dinter,

it seems clear from the somewhat scanty description that it is conspecific with the

Sachs specimen, and so C. omahekensis is included in synonymy under C. caninus.

The leaves of C. caninus are relatively long in relation to their width, being ovate-

lanceolate to long-elliptic or oblanceolate, sparingly crenate-dentate in the upper half,

acute to obtuse at the apex and long-cuneate at the base. As may be expected, there is

considerable variation in stature and leaf size according to growing conditions.

C. comosus is essentially a plant of north-east Africa, chiefly Kenya and Abyssinia,

and specimens have not been seen from territories south of Kenya. The possibility

exists that it extends to Arabia. It is a perennial and, like C. neochilus and C. caninus,

the leaves are aromatic and fleshy. In general the leaves are broadly obovate, faintly

to distinctly crenate at the rounded apex and cuneate at the base. The flower spike is

short, rarely becoming lax at the base and is borne usually on a long, slender, leafless

rhachis.

The more important citations for C. caninus and C. comosus are set out for con-

venience below.

C. caninus {Roth) Vatke in Linnaea 37; 318 (1871), partly, excl. specimen cited; Guerke

in Bot. Jahrb. 19: 212 (1895). Type: India, Heyne s.n.

Plectranthus caninus Roth, Nov. PI. Sp. 279 (1821).

434

Coleus spicatus Benth. in Wall. PI. As. Rar. 2: 15 (1831); Wall. Cat. 2729 (1831);

Benth. in DC. Prodr. 12: 71 (1848); Wight, Ic. t. 1431 (1849). Type: India,

Wight s.n. in Hb. Benth. C. omahekensis Dinter in Fedde Rep. Beih. 53: 123

(1928), ex descr. Syntypes: S.W. Africa, Grootfontein, Etemba, Dinter 3265;

Otjikuara, Dinter 3265.

Found in India, east tropical Africa to Rhodesia and South West Africa.

S.W. Africa. — Grootfontein: Farm Kakuse, east of Etosha Pan, and plants

cultivated in Pretoria from seed obtained from this gathering, Sachs 2.

S. Rhodesia. — Hartley: Poole Farm, Wild 4544.

Tanganyika. — Kilimanjaro, Greenway 6897.

Kenya. — Nairobi, Verdcourt and Greenway 399.

C. comosus Hochst. ex Guerke in Bot. Jahrb. 19: 212 (1894); Bak. in FI. Trop. Afr. 5:

426 (1900); Bruce in Hook. Ic. PI. 34: t.3374 (1938). C. comosus Hochst. in A.

Rich., Tent. FI. Abyss. 2: 183 (1851), in synonymy. C. spicatus [non Benth.],

A. Rich., l.c. C. caninus [non (Roth) Vatke], Vatke in Linnaea 37: 318 (1872).

Type: Abyssinia, Schimper 1328.

Specimens seen include the following: a plant cultivated in Pretoria, collected by

Dr. P. J. Greenway near Nairobi; Codd 8238, a plant of unknown origin, cultivated in

gardens in South Africa; Gillett 14080 and 14312 collected with the Kenya-Ethiopia

Boundary Commission; and probably Williams 637 from Nairobi, though the flowers

of this specimen appear smaller than usual.

The main characters which may be used in separating the three species are

summarised below: —

C. neochilus. Plants perennial; leaves obovate to elliptic-ovate, faintly crenate

at the rounded or obtuse apex; flower spike 7-15 cm long, lax below with 5-12 spaced

verticils below the coma of bracts; corolla 1 -2-1 -8 cm long.

C. comosus. Plants perennial; leaves usually broadly obovate, faintly to distinctly

crenate at the rounded apex; flower spike 3-5, rarely up to 9 cm long with one or two,

rarely more, spaced verticils at the base ; corolla 2-2 • 5 cm long.

C. caninus. Plants annual or, possibly, perennial; leaves ovate-lanceolate, long

elliptic or oblanceolate, sparingly crenate-dentate in the upper half, apex acute to obtuse;

flower spike 2-5, rarely up to 9 cm long, dense at the base; corolla 0-8-1 cm long.

435

South African Species of Satureia

by

D. J. B. Killick.

SATUREIA

Linn., Sp. PI. ed. 1: 567 (1753), Gen. PI. ed. 5; 247 (1754); Briquet in Engl, and Prantl.

Nat. Pflanzenfam. 4, 3a: 296 (1896). Micromeria Benth. in Bot. Reg. 15: sub t. 1282

(1829).

The genus Micromeria was described by Bentham (1829), who distinguished it

from Satureia on the grounds that the calyx was 13-15 instead of 10-nerved. Baker in

Flora of Tropical Africa 5, 453 (1900) and Skan in Flora Capensis 5, 1: 306 (1912)

in their treatments of Labiatae both upheld Micromeria. Briquet, however, reduced

Micromeria to a synonym of Satureia explaining that the calyx nerves very often divide

and that intermediate species made delimitation of the two genera impossible. Later

taxonomists, for example Thonner, Flowering Plants of Africa, 479 (1915), Brenan in

Mem. N. Y. Bot. Gard. 9, 1: 45-53 (1954) and now Hedberg, Afroalpine Vascular

Plants, 160-164 and 317-318 (1957) have followed Briquet’s classification. Until a

critical revision of Satureia and related genera is undertaken, it seems desirable for the

sake of uniformity in Africa, to follow recent workers and place our species under

Satureia. The necessary name changes are made in this paper.

The generic name was originally spelled Satureja by Linnaeus, but the spelling

Satureia is more correct philogically and is used here in accordance with Art. 74 (2)

of the International Code of Botanical Nomenclature (1956).

Bracts foliose 1. S. grandibracteata.

Bracts linear:

Calyx tubular 2. S. bifiora.

Calyx campanulate:

Lower lip of corolla longer than upper 3. S. reptans.

Lower lip equal or shorter than upper 4. S. compacta.

1. S. grandibracteata Killick, nom. nov.

Micromeria grandiflora Killick in Bothalia 6, 2: 439-440 (1954). Type: Cathedral

Peak Forest Influences Research Station, Killick 1684 (PRE, holo.).

The epithet grandiflora cannot be used, because it is pre-occupied in S. grandiflora

Scheele (1843).

For a description of S. grandibracteata see Killick (l.c.). No additional specimens

of this species have been received since it was first described.

Natal. — Bergville: Cathedral Peak Forest Influences Research Station Codd

and Dyer 6241; Killick 1102, 1600, 1684.

2. S. biflora {Buck. Hamilt. ex D. Don) Briq. in Engl, and Prantl., Nat. Pflanzenfam.

4, 3a: 299 (1896). Type: Upper Nepal, Suemba, Hamilton s.n. (not located).

Thymus bflorus Buch. Hamilt. ex D. Don, Prodr. FI. Nepal, 112 (1825). Micromeria

b flora (Buch. Hamilt. ex D. Don) Benth., Lab. Gen. et Sp. 378 (1834).

436

A shrubby herb with numerous ascending branches, 20-60 cm high, pubescent,

glandular-punctate; internodes 0-5-3 cm long. Leaves subsessile, narrowly elliptic

to broadly ovate, 5-15 mm long, 1 -5-10 mm wide, apex acute to obtuse, base broadly

cuneate to round, entire. Cymes 1-9 flowered; peduncles 2-7 mm long, bibracteate;

bracts linear, 1-5-2 mm long. Calyx tubular, 15-nerved; tube 2-3 mm long; teeth

subequal, elongate-triangular, acuminate, somewhat spinous and reflexed. Corolla

white or mauve; tube 5-6 mm long; upper lip emarginate, 1 -5 mm long, 2 mm wide;

lower lip 3-lobed, 4 mm long, 4-5 mm wide, lateral lobes rounded, middle lobe emargi-

nate. Stamens didynamous, arcuate, upper 1 mm long, lower 2 mm long; anthers

2-celled, divaricate. Disc slightly lobed. Style included, 5 mm long; stigma bifid,

lobes subequal. Nutlets oblong, 1 mm long, 0-5 mm wide.

Basutoland. — Quthing: Moyeni Mountain, Lelvaleng, Dieterlen 1346.

Cape. — Engcobo: Emgwali River, 2809. Herschel: Sterkspruit,

5. Umtata: Baziya, Baur 220; Nqadu woods, Pegler 1575.

Transvaal. — Barberton: Lomati, Thorncroft 2115. Heidelberg: Mogg 24170.

Krugersdorp: Mogg 23161. Lydenburg: Kantoorbos, Codd 9787. Pietersburg:

Duivelskloof, Galpin 10101, 11393. Pretoria: Leikenhoutskloof, Mogg 12433. Roode-

poort: Roodepoort Ridge, Mogg 20254. Riistenburg: Farm Morgenzon, Rose-Innes

122. Soutpansberg: Louis Trichardt, Letty 247. District Unknown: Valdesia,

Schlechter 4534.

S. biflora is an extremely variable species as regards leaf-shape, number of flowers

and structure of calyx. It is here treated in its widest sense, following Baker (l.c. 452)

who regarded the Tropical African species S. punctata (Benth.) Briq., Micromeria

ovata Benth., M. schimperi Vatke and M. purtschelleri Giirke as synonyms. Subsequent

workers have variously upheld these species. For example, Brenan (l.c. 45) considered

S. punctata as distinct, likewise Hedberg (l.c. 161). Recently E. and K. Walther in

Mitt. Thiir. Bot. Gesell. 1,4: 1-12 (1957) in a very detailed study of S. bflora and its-

allies upheld all the species sunk by Baker. According to these two workers typical

S. bflora is restricted to the Himalayas and Eastern India and does not occur in Africa,

but they allow for the presence in Africa of two of its varieties, namely var. rhodesiaca

Walther and var. villosa Walther.

The problem was submitted to Mr. W. Marais our liaison officer at Kew who,

together with Mr. P. Taylor of that institution, examined type material and specimens

at Kew and the British Museum. Taking the whole range of material into consideration

they decided that there was no justification whatever for upholding the species and

varieties regarded by E. and K. Walther as distinct.

3. S. reptans Killick, nom. nov.

Micromeria pilosa Benth. in Hook. Icon. Plant. 15: t. 1522 (1886). Types: Faku’s

Territory, Sutherland s.n.; Natal, Medley Wood 3712 (both K).

Prostrate aromatic herb with slender branches up to 2 ft. 6 in. long, covered with

numerous silvery multicellular hairs and a few unicellular gland-tipped hairs, glandular-

punctate; internodes 1-5 cm long. Leaves ovate, 0-7-2 -5 cm long, 0- 6-2-0 cm wide,

apex acute to rounded, base shallowly cordate to broadly cuneate, margins crenate

with few teeth; petioles 1-5 mm long. Flowers 1-2 in axils of uppermost leaves (larely

3, Galpin 11745). Pedicels threadlike, 0-7-3 cm long, bibracteate about the middle;

bracts linear, 1-2 mm long. Calyx campanulate, 15-nerved; tube 2-5-4 mm long,

teeth subequal, deltoid, 1-2 mm long. Corolla “ white ” (Galpin 11745 and 11925) to

pale cobalt violet, often creamy yellow on lower side; tube 0-5-1 -2 cm long, 3-7 mm

wide at mouth, pilose and yellow inside on lower side; upper lip emarginate, 1-2 mm

long, 3-6 mm wide; lower lip 3-lobed, middle lobe emarginate, 4-7 mm long, 3-7 mm

wide, lateral lobes round, 2-3 mm diam. Stamens 4, didynamous, arcuate, lower 3 mm

long, upper 1-5 mm long. Disc slightly lobed. Style 1 cm long, the posticous lobe

the shorter. Nutlets oblong, 2 mm long, 1 mm wide.

437

Natal. — Bergville: Cathedral Peak Forest Station, Killick 1272, 1429. Estcourt;

Cathkin Peak, Galpin 11745; Bushmans River, Medley Wood 10894. Lions River:

Lidgetton, Mogg 6890. Richmond: Byrne, Galpin 11925. Underberg: Bamboo

Mountain, McLean 709.

A new name is necessary in Satureia, because of the existence of S. pilosa Velen.

(1899) described from Bulgaria.

4. S. compacta Killick, sp. nov., S', reptanti Killick et S. grandibracteatae YJiWickafhnxs,

ab ambabus habitu multo compactiore corollae labio antico quam posticum

aequilongo vel breviore facile distinguitur.

Herba prostrata compacta multo ramosa glanduloso-punctata pilis multicelluraribus

induta, internodiis 5-15 mm longis. Folia late ovata vel rotunda 0-5-1 -1 cm longa,

0- 4-1 -0 cm lata, apice et basi rotunda, marginibus breviter crenatis, petiolis 2-4 mm

longis. Flores in axillis foliorum solitarii. Pedicelli 3 mm longi infra medium bibracteati ;

bracteae lineari-lanceolatae 1-5 mm longae carinatae. Calyx campanulatus 15-nervis,

tubo ad 2 mm longo, lobis aequalibus triangulari-lanceolatis c.2 mm longis. Corolla

cobalto-violacea intus glabra; tubus anguste campanulatus 5 mm longus, labio antico

3-lobato lobis lateralibus rotundis 2 mm diam. lobo medio emarginato 2 mm longo

2-5 mm lato, labio postico emarginato 2 mm longo 3 mm lato. Stamina didynama,

antica 3 mm longa, postica 2 mm longa, antheris 2-locularibus divaricatis. Discus

leviter lobatus. Stylus inclusus 6 mm longus, stigmate bifido. Nucellae oblongae

1- 8 mm longae 1-2 mm latae minute pubescentes.

Natal. — Bergville: below the Amphletts, Cathkin Peak Area, Killick 1866 (PRE,

holo.).

Prostrate, mat-forming, much-branched herb, covered with multicellular hairs

(especially dense on stems), gland-dotted; internodes 0-5-1 -5 cm long. Leaves broadly

ovate to round, 0-5-1 -1 cm long, 0-4-1 -0 cm wide, round at base and apex, margins

shallowly crenate; petioles 2-4 mm long. Flowers solitary in axils of leaves. Pedicels

3 mm long, bibracteate below middle; bracts linear-lanceolate, 1-5 cm long, keeled.

Calyx campanulate, 15-nerved; tube just under 2 mm long; lobes equal, triangular-

lanceolate, 2 mm long. Corolla cobalt-violet, glabrous inside except for few hairs at

base of lower lip; tube narrowly campanulate, 5 mm long; lower lip 3-lobed, lateral

lobes round, 2 mm diam., middle lobe emarginate 2 mm long, 2-5 mm wide; upper

lip emarginate, 2 mm long, 3 mm wide. Stamens didynamous, somewhat arcuate,

lower 3 mm long, upper 2 mm long; anthers 2-celled, cells divaricate. Disc slightly

lobed. included, 6 mm long; stigma bifid, the posticous lobe the shorter. Nutlets

oblong, 1 -8 mm long, 1 -2 mm wide, minutely pubescent.

This species was found growing at the side of a bridle path in Festuca costata

Grassveld at 6,800 feet just below the Amphletts in the Cathkin Peak Area of the Natal

Drakensberg. Although the author has spent nearly three years in the Drakensberg

doing botanical survey work, he has seen S. compacta growing in only the one locality.

S. compacta is related to S. rep tans and S. grandibracteata, but can be readily

distinguished from these species. Vegetatively it dilTers in being denser and much

more compact in habit; it forms mats about 45 cm in diameter. Florally it differs

mainly in that the flowers are smaller and the lower lip of the corolla is equal or shorter

than the upper lip.

438

439

X Ruttyruspolia, a Natural Inter generic Hybrid

in Acanthaceae

by

A. D. J. Meeuse and J. M. J. de Wet

A plant recently collected near Wyllie’s Poort, Transvaal {Meeuse 9793) exhibited

characteristics of both the genera Ruttya Harv. and Ruspolia Lindau and could not be

matched by any herbarium specimens from tropical or South Africa. This, together

with the fact that this plant, of which only one single specimen was seen, did not produce

seeds, led to the assumption that it could be a hybrid between the only representatives

of these genera occurring in the Soutpansberg area, viz., Ruttya ovata Harv. and RuspoHa

hypocrateriformis (Vahl) Milne-Redh. var. australis Milne-Redh. As natural inter-

generic hybrids are rare and had not before been reported in Acanthaceae, additional

evidence was required to support the assumption.

Material and Methods

Flowers of the supposed Ruspolia parent were emasculated when the buds just

began to show colour and pollinated two days later with Ruttya pollen. Root-tips

for a study of the chromosome numbers were fixed in Randalph’s (1935) “ CRAF ”

fluid, dehydrated using n-butyl alcohol and embedded in a mixture of paraffin, beeswax

and rubber. Sections were cut 12 |U, thick and stained in Stockwell’s (1934) solution.

For a study of meiosis, anthers were fixed in 3 : 1 alcohol and propionic acid, squashed

and stained in a propionic acid-carmine solution as described by Swaminathan, Magoon

and Mehra (1954).

Results and Discussion

From the numerous pollinated flowers a few viable seeds were obtained. Two

plants were raised which started flowering six months after planting. These plants

are morphologically very similar to the clone Meeuse 9793 and differ from it only slightly

in the colour of the corolla. This provides a strong indication of the hybrid origin

of the plant discovered in Wyllie’s Poort.

In the artificial hybrid the corolla has a peculiar mauvish-pink colour which is

nearest “ Roeellin Purple” (Ridgway, 1912, XXXVIII, 71” V. RR. b)', in the wild

plant the colour of the corolla is of a deeper hue. In the Ruspolia parent the corolla

is a clear scarlet red (Bruce 1954) and in Ruttya ovata it is white with minute mauve-

purple dots in the throat on the side of the lower lip.

The generic characters of the two parent genera involved are practically identical

as regards genitalia, pollen morphology, fruit and seeds. The only essential difference

between them is in the shape of the corolla which is more or less distinctly bilabiate

with a short tube gradually widening into the throat in Ruttya, and sub-equally 5-lobed

with a long narrow cylindric tube in Ruspolia. Cytologically they differ in basic chromo-

some number, but not in chromosome size. Both genera are characterised by extremely

small chromosomes in multiples of « = 9 {2n = 36) in Ruttya, and n 1 {In = 42)

in Ruspolia. During pollen development the chromosomes associate into 18 pairs in

Ruttya ovata and into 21 pairs in Ruspolia hypocrateriformis var. australis.

440

The hybrids under discussion have most of the morphological characters of the

Ruspolia parent, but a markedly shorter and somewhat wider corolla-tube. The

inflorescence is elongated and pseudo-spicate and the corolla-limb subregularly 5-fid

as in Ruspolia. The colour of the corolla in the hybrids is found in neither parent, but

is characteristic of both the natural and the artificial hybrids. The corolla-lobes are

marked on the inside (upper surface in open flowers) near the base with fine dots of a

deep, somewhat purplish red to deep amaranth red, a character which is also derived

from the Ruspolia parent.

The natural and artificially produced hybrids are completely sterile. Back-crossing

with either of the parents proved unsuccessful. The natural hybrid has not been studied

cytologically. The artificial one has 39 small somatic chromosomes. During pollen

development pairing is almost completely absent between the two parental sets of

chromosomes. In the pollen mother cells 39 univalents are usually present and these

chromosomes are distributed at random over the two poles with an appreciable number

of laggards which are excluded from the two daughter nuclei. Only very occasionally

a maximum number of two bivalents was observed.

In our opinion this hybrid has excellent potentialities as an ornamental. Upon

introduction into horticulture a convenient name is desirable, but as the hybrid is a

sterile Fj it is not deemed advisable to give it a scientific name at the specific level.

Apart from the formal description of the hybrid genus, only a horticultural name is

proposed. It is thought that the artificial hybrid and the natural one do not differ

sufficiently to warrant different horticultural names for each of them.

X Rutty ruspolia A. Meeuse et de Wet, gen. hybr. nov. {Ruttya Yia.n.x Ruspolia

Lindau), a Ruttya tubo corollae subcylindrato longiore, limbo subregulare 5-fido

differt, a Ruspolia tubo corollae breviore recedit.

Ruttyruspolia ‘Phyllis van Heerdsn" = Ruttya ovata Haw. <^x Ruspolia hypo-

crateriformis (Vahl) Milne-Redh. var. australis Milne-Redh. $.

In order to distinguish this hybrid from other theoretically possible Ruttya x Ruspolia

hybrids, this name is intended only to include all Fj-hybrids (hybrid clones and sub-

clones, respectively) which are derived from the parent species indicated above and

resemble Ruspolia hypocrateriformis var. australis in morphological characters, except

in the much shorter corolla tube. The colour of the corolla is a shade of “ Rocellin

Purple ” (Ridgway XXXVIII, 71”. V. RR. b), or approximately so. The corolla lobes

are in addition marked near the base inside with fine dots of a deep red to amaranth

red colour.

The horticultural name proposed commemorates the fact that it was Mrs. P. van

Heerden of Louis Trichardt, Northern Transvaal, who first discovered the natural

hybrid and subsequently introduced it into horticulture.

It is feasible that other intergeneric hybrids in Acanthaceae can be artificially

produced, because although the basic chromosome number varies rather widely in

this family, the same number (or a near number) is sometimes found in different genera

(Grant, 1956, Raghavan, 1957, Takizawa, 1957, Mangenot and Mangenot, 1957).

In this way horticultural novelties could be developed. Most Acanthaceae can easily

be propagated by means of cuttings and the sterility in intergeneric crosses need, there-

fore, not be a hindrance. It may even be an asset because the flowers do not become

fertilized and, as we have observed in x Ruttyruspolia, remain fresh for several days

instead of wilting soon after having been pollinated (as is the rule in this family). Judging

by our example, experimental cross-breeding work in this family for horticultural

purposes seems promising and is to be strongly recommended.

441

Literature Cited.

Bruce, E. A. (1954)

Grant, Wm. F. (1956)

Mangenot, S. and G. Mangenot

(1957)

Ruspolia hypocrateriformis var. australis in: R. A. Dyer,

Flow. PI. Afr. 30 (1): PI. 1166.

A cytogenetic study in the Acanthaceae. BrittoniaS: 121-149.

Nombres chromosomiques nouveaux chez diverses dicotyle-

dones d’Afrique occidentale. Bull. Jard. Bot. (Bruxelles)

27: 639-654.

Raghavan, R. S. (1957)

Chromosome numbers in Indian medicinal plants. Proc.

Ind. Acad. Sect. B 45: 294-298.

Randolph, L. F. (1935)

Ridgway, R. (1912)

Stockwell, P. (1934)

SWAMINATHAN, M. S., M. L. MaGOON

AND K. L. Mehra (1954)

Takizawa, Y. (1957)

A new fixing fluid and a revised schedule for the paraffin

method in plant cytology. Stain Techn. 10: 95-96.

Color standards and color nomenclature. Washington.

A stain for difficult plant material. Science 80: 121-122.

A simple propionic acid carmine smear method for plants

with small chromosomes. Ind. Jour. Genet. Plant Breed.

14: 87-88.

Die Struktur der Pachytenchromosomen einiger Acanthaceen,

sowie eine Reihe neu bestimmter Chromosomenzahlen..

Cytologia 22: 118-126.

442

443

Notes and New Records of African Plants

hy

Various Authors

ACANTHACEAE

Barleria (Sect. Frionitis) ameliae A. Meeuse, spec, nov., B. delagoensi Oberm. arete

affinis, sed bracteolis brevioribus, sepalis eglandulosis, sepalo postico sepalo

antico excedens, praecipue differt.

Suffrutex erectus perennis pauce ramosus 30-60 cm altus. Caules subquadrangulati

plus minusve puberuli vel sparsissime strigosi demum tereti, glabrescentes. Folia

elliptica vel plus minusve ovato-elliptica vel subobovata firmiter herbacea in siccitate

papyracea vel subpergamacea, breviter acuminata obtusa vel subacuta, pungentia,

basi attenuata vel longe decurrentia, subglabra, 3-9 cm longa 2-5 cm lata, petiolis

apice vel totus alatis ad 3 cm longis, spinis axillaribus paucis vel interdum nullis

albescentibus 2-4 mm interdum ad 7 mm longis. Flores axillares subsessiles, in parte

superiore ramorum in inflorescentias spiciformes aggregati, bracteolis lineari-subulatis

pungentibus minutissime aculeato-hirtellis 4-6 mm longis. Sepala in apices pungentes

attenuata extus sparse intus subdense strigosa, eglandulosa, sepalo antico 11-13 mm

longo 5-6 mm lato, sepalo postico 14-17 mm longo 4-5 mm lato, sepalis interioribus

angustioribus 11-13 mm longis 1-5-2 mm latis. Corolla flava, extus minute molliter

pubescens, tubo 12-14 mm longo, labio superiore 18-20 mm longo 4-lobato lobis

subaequalibus obovato-ellipticis c. 13 mm longis, labio inferiore integro 10-11 mm longo.

Pistillum glabrum. Capsula 14 mm longa 4 mm lata attenuato-rostrata plus minusve

puberula.

South West Africa. — Caprivi Zipfel: Eastern area, Mpilila Island, under

trees on rock outcrop near Chobe River bank, Killick and Leistner 3391 (PRE, holo.!,

K,L,M, SRGH, Windhoek, isos.!).

Southern Rhodesia. — Wankie: Wankie, Le\y 150 (PRE), 1113 (PRE, SRGH);

Wankie Game Reserve, vlei edge. Wild 4751 (PRE, SRGH). Nuanetsi: Lundi River,

near rapids above big bend, in patches of woodland, Davies 2051 (SRGH).

Suffruticose erect perennial 30-60 cm high. Stems usually not much branched,

when young green (drying brown) and quadrangular, somewhat puberulous and/or

with very few strigose hairs, soon quite glabrous, ultimately terete and covered with a

thin greyish-yellow to pale brown bark. Leaves elliptic or somewhat ovate or obovate-

elliptic, firmly herbaceous drying papyraceous to somewhat pergamaceous, subentire,

cuneate-attenuate to long-decurrent at the base, shortly acuminate into a blunt or

subacute pungently mucronate apex, glabrous except a sparse strigose pubescence on

the basal portions of midrib and main veins (mainly on lower surface) and along the

subciliate margin, 3-9 cm long and 2-5 cm broad; cystoliths minute, inconspicuous;

petioles winged by the decurrent leaf-base in upper part or throughout, up to 3 cm long,

in the axils usually with straight to slightly curved whitish slender 2-4 mm, occasionally

up to 7 mm long spines. Flowers subsessile in the axils of normal leaves and the upper-

most ones in the axils of smaller floral leaves running into a terminal pseudo-spike;

bracteoles linear-subulate, pungent, minutely scabrid-hirtellous, 4-6 mm long, pedicels

very short, usually somewhat strigose. Sepals sparsely strigose outside, more densely

so inside, subciliate, eglandular, attenuate into the pungent tips; of the outer oblong

444

lanceolate concave ones the anticous one is 11-13 mm long, 5-6 mm broad and usually

with two adjacent small spines rarely a single spine at the apex, and the posticous one

14-17 mm long and 4-5 mm broad; inner sepals linear-lanceolate, 11-13 mm long and

1 • 5-2 mm broad. Corolla orange-yellow or creamy-yellow, on the outside finely and

softly pubescent, more densely so on the 12-14 mm long tube; the upper lip 18-20 mm

long, 4-lobed with subequal obovate-elliptic about 13 mm long lobes; the lower lip

10-11 mm long. Pistil glabrous. Capsule 14 mm long and 4 mm broad, greyish-brown,

attenuate-beaked at the apex, somewhat puberulous.

This species is in habit exceedingly like B. delagoensis Oberm., a coastal species

from Portuguese East Africa, but it differs constantly in the non-glandular sepals

(with stipitate glands in B. delagoensis), of which the posticous one is the longest (it is

the shortest in B. delagoensis), in the shorter bracteoles, in the usually shorter axillary

spines; in the somewhat puberulous capsules (pilose in B. delagoensis) and in several

other points. The leaves are more decurrent on the petiole and are relatively somewhat

broader in B. ameliae, the cystohths are inconspicuous (distinct in B. delagoensis),

the sepals are shorter than in B. delagoensis (the three shorter ones up to 13 mm long,

as against the longest three 17 mm long in B. delagoensis) and the relative lengths of

the parts of the corolla are not the same: tube 12-14 mm long against 12 mm in B.

delagoensis, the upper lip 18 mm long as against 21 mm; its lobes 13 mm against 15 mm

and the lower lip 10-11 mm against 16 mm, i.e. a relatively longer tube and relatively

(and absolutely) shorter limb in B. ameliae as compared with B. delagoensis. There are

also differences in the degree and type of pubescence of stems, leaves and sepals. All

these differences together indicate that in spite of the close resemblance the two forms

are not of varietal status but represent two distinct species. This view is supported

by the difference in ecology, B. delagoensis being a coastal plant found on alluvial

soils near the sea-shore in an area with very mild winters, a high rainfall and a high

humidity of the air almost throughout the year, whereas B. ameliae is a plant of the

interior growing at altitudes between 1,000 and 3,000 ft. in areas with a lower annual

rainfall and a pronounced dry winter season with low humidity and often low

temperatures.

This species is named after Mrs. A. A. Mauve (nee Obermeyer) who contributed

so much to the knowledge of the genus Barleria in Southern Africa and kindly studied

the cited material to confirm my views that the plant under discussion represents a

hitherto undescribed taxon.

A. Meeuse

Barleria argilHcoIa Oberm. sp. nov., B. bolusii Oberm. affinis, sed floribus solitariis

planta subglabra bracteis absentibus corolla 3 • 5 cm longa differ!.

Fruticulus parvus. Rami e rhizomate orti. Folia coriacea glabra nitida margine

alba undulata Integra vel parce spinosa. Flores solitarii. Bracteae absentes. Corolla

3 • 5 cm longa. Stigma capitata.

Small plants up to 20 cm high, deep rooted. Stem perennial, woody, rhizomatous

giving off short annual shoots which are unbranched or form 1-2 short basal branches,

minutely pubescent in the grooves, nodes bristly. Leaves lanceolate, ca 3 x 1 cm,

coriaceous, glabrous, shiny, apex tipped with a short spine, tapered at the base into a

short petiole, or sessile, margin white, wavy, entire or sparsely spinous. Flowers solitary,

axillary, bractless. Bracteoles about as long as the leaves but narrower and with a

prominent lateral nerve on each side of the midrib. Calyx with the posticous sepal

broadly ovate, reticulate, spine tipped, about as long as the bracteoles; anticous similar

to posticous but bispinous at the apex; lateral segments small, linear, 1 cm long. Corolla

pale mauve, with a narrow tube 2 cm long, limb 5-lobed, regular, somewhat shorter

than the tube. Stamens dissimilar, two protruding from tube, two smaller included,

the fifth represented by a short filament only. Ovary glabrous, style filiform, stigma

capitate. Capsule typical, 2 cm long.

445

Natal. — Estcourt: Rensburg Spruit near Estcourt, eroded thornveld slopes,

frequent in patches, in flower October 1944, Acocks 10701 (PRE, holo. NH iso); near

Estcourt in Bushman’s River Valley, eroded clay banks of Rensburg Spruit, with fruit,

December 1943, Acocks 9968 (PRE, NH); Edwards 2458 (PRE).

The species comes under the section Eiibarleria, Pungent es. The stigmas are

confluent, capitate. Its nearest ally seems to be B. bolusii Oberm. (B. mosdenensis

Oberm. is a synonym of B. bolusii Oberm.) but it is very different from this species.

The flowers are solitary, the plant is glabrous except for a slight short pubescence on

the stem and the corolla is 3 • 5 cm long. The narrow white leaf margin is a conspicuous

character. Mr. D. Edwards who visited Acocks’ locality in November 1959, found it

there exclusively on the eroded, clayey neutral to alkaline soils. Excellent photographs

and colour slides of the flowering plants were made by him.

Barleria saxatilis Oberm. sp. nov., B. eleganti S. Moore affinis, sed floribus minoribus

dilute purpureis cymis parvifloris differt.

Suffruticosa. Rami strigosi et pilis brevibus patentibus dense pubescentes. Eolia

oblonga herbacea lamina inferiore pallide viridia. Inflorescentiae cymae unilaterales

1-4 florae; bracteoli lineares spinoso-dentati; sepalum posticum ovatum, spinoso-

dentatum papyraceum reticulatum; sepalum anticum postico simile sed paulo minoiius;

corolla dilute purpurea, tubo angusto faucem versus gradatim paulo expanso 15 mm

longo, limbo 5-lobato sub-regulari; stamina perfecta 2, exserta; stamina imperfecta

brevia inclusa. Capsula glabra 1 cm longa 4 mm lata apiculata.

A straggling or erect shrub 30-150 cm tall. Branches woody white puberulous

and strigose. Innovations densely strigose. Leaves variable in size, oblong to lanceolate,

up to 3 cm long and up to 2 cm wide, usually smaller, apex obtuse, in 1-4 flowered

sessile cymes. Bracteoles unequal, firm, spinoso-dentate, the lower smaller, linear,

recurved from the base; the upper erect, lanceolate-acuminate. Outer sepals ovate,

acuminate, up to 2 cm long including spine, 9 mm wide, with ca 16 marginal spines

1-3 mm long, reticulate, sparsely strigose, green at first, stramineous in fruit but appa-

rently not enlarged. Inner sepals linear, 1 cm long. Corolla puberulous, bluish mauve;

tube narrow, 15 mm long, shghtly widened at the apex; limb 5-lobed ca 11 mm long.

Stamens inserted in lower half of tube, the fertile ones well exserted, filaments linear;

3 sterile short, included, one represented by a minute filament only. Ovary glabrous,

style filiform 2 cm long, with a ring of short hairs at the base; stigma short, filiform.

Capsule glabrous, just exserted from the outer sepals, 1 cm long, 4 mm broad, apiculate.

Elowering. Period: March onwards.

Distribution: In the drier parts of the northern and eastern Transvaal, usually

on rocky hillsides.

Transvaal. — Soutpansberg: Sandrivier Poort, southern end, on a stretch 1-8

miles from main road bridge, Meeuse 10203 (PRE, holo,) Farm Soutpan, lower northern

slopes of the Soutpansberge, Obermeyer, Schweickerdt and Verdoorn 130. About

two miles south of Wyliespoort, Meeuse 9790. Pietersburg: Rogers 14151; Mokeetsi,

Obermeyer TM 31977; two miles S.E. of Chuniespoort Hotel, Obermeyer and Verdoorn

13; Blaauwberg near Leipzig Mission Station, Codd 8713. Waterberg: Nylstroom

Mountains north of Warmbad, Obermeyer TM 31978. Pilgrims Rest: Kruger National

Park, Gorge, van der SchijffYJAP 2294. Lydenburg: Rustplaats, Taylor 1943.

The species is closely related to B. elegans S. Moore but can be distinguished by

its few flowered, mauve cymes, usually more dense pubescence and its thinner less

rigid branches. The capsule of B. elegans usually reaches a length of 15 mm whilst

that of B. saxatilis is only about 10 mm long. The calyx and bracteoles are also

correspondingly smaller in B. saxatilis. It inhabits dry hot areas of the Transvaal

446

bushveld, where it has established itself very firmly, being dominant for miles in some

parts. B. elegans occurs along the coast but is also found in more humid places in the

Low Veld near rivers.

At Leipzig Mission Station Dr. Codd noticed that bunches of dried stems were

tied round poles of grain stores to repel rats.

A. Amelia Obermeyer

Sclerochiton triacanthus A. Meeuse, spec, nov., 5. scissisepalo C. B. Clarke affinis,

sed inter alia ramulis pilosis, foliis minoribus angustioribus, bracteis fertilibus

apice breviter trispinosis diflfert.

Frutex suberectus e basi ramosus 0-50-1 -00 m altus. Caules subteretes lignosi

breviter albo-pilosi, demum glabri, cinnamomei vel cinerascentes. Folia ad apicem

ramulorum plus minusve aggregata, subcoriacea, subsessilia, nitida, oblongi-lanceolata

vel lanceolato-linearia, Integra, convexa vel subplana margine deflexa vel revoluta,

basin versus interdum ciliata, apice pungentia, subglabra, 2-5 cm longa, 4-9 mm,

interdum ad 15 mm, lata, costa media subtus prominenti interdum subtus basin versus

sparse pilosa, cystolithis inconspicuis. Inflorescentiae terminates vel laterales, subsessiles

pauciflorae densae, ad 6 cm longae, bracteis bracteolisque firmiter herbaceis plus

minusve concavis carinatis vel subcarinatis puberulis vel subglabris plus minusve

ciliatis. Bracteae oblongae vel lineari-oblongae, inferiorae steriles parvae innocuae

vel pungentes, fertiles majores 14 mm longae 4 mm latae spinulis 3 apicalibus pungentes.

PediceUi satis robusti, 3-5 mm longi, bibracteolati, bracteolis lineari-lanceolatis sparse

puberulis 16-18 mm longis 3-4 mm latis spinulis 1-3 pungentibus. Sepala 5, acuta,

pungentia, subglabra, plus minusve ciliata; sepalum posticum late lineare acutum

infra medium ciliatum, apice unispinulosum vel interdum 2-3-spinulosum, 23-24 mm

longum, 4-5 mm latum, sepala 2 antica lineari-lanceolata, 16-18 mm longa, 3 mm lata,

sepala lateralia lineari-lanceolata, 15-17 mm longa, 2-2-5 mm lata. Corolla dilute

azurea, coeruleo-venosa, tubo ca 7 mm longo, limbo ca 19 mm longo extus parce sericeo

intus lineis 2 pilorum subpatentorum cincto, lobis subaequalibus obovato-oblongis

apice rotundatis 4-5 mm longis. Stamina basin versus retrorse sericea, filamentis ca

1 1 mm longis, antheris hirsuto-pubescentibus ca 2 mm longis. Ovarium puberulo-

velutinum, apice pilis erectis comosum, 3-4 mm longum, stylo basin versus pilis setaceis

patentibus albido-fulvidis barbato 14-15 mm longo. Capsula anguste oblonga, glabra,

ca 14 mm longa.

Transvaal. — Barberton: Barberton, Rogers 24047 (PRE); about five miles

from Barberton on road to Florence Mine, Mr. and Mrs. P. F. Clarke 12 (PRE, holo. !).

A somewhat straggly to erect low shrub, usually 40-60 cm tall but occasionally

attaining about 1 m, branched from the base upwards, usually slender in habit, but in

some cases forming a “bush” or thicket 1-2-5 m across. Stems subterete, woody,

dark brown and shortly pilose with whitish patent hairs, usually rather densely so,

later glabrescent and turning cinnamon-coloured or ashy-grey, forming a thin, nearly

smooth or somewhat (longitudinally) wrinkled bark. Leaves more or less distinctly

aggregated at the tips of the branches or on short side-shoots, subcoriaceous, subsessile,

shiny, oblong-lanceolate to linear-lanceolate, entire, usually somewhat convex through

the deflexed margins, glabrous or ciliate near the base, sometimes sparsely pilose on

the midrib, narrowing at the base into the very short pilose petiole, subobtuse to acute

at the spine-tipped apex, distinctly veined, 2-5 cm long and 4-9 mm (on young shoots

occasionally up to 15 mm) broad; midrib prominent on lower surface; cystoliths

indistinct or invisible. Inflorescences terminating the branches and short side-shoots,

dense, few-flowered, up to about 6 cm long. Bracts oblong to linear-oblong, firm,

green; lower sterile ones short, innocuous or pungent, fertile ones larger, about 14 mm

long and 4 mm broad, more or less concave to boat-shaped and somewhat carinate,

as are the bracteoles, puberulous or glabrous usually ciliate along the margins, especially

about the middle, at the apex with three small spines. Pedicels rather stout, 3-5 mm

447

long, bibracteolate in the middle; bracteoles linear-lanceolate, with 1-3 small apical

spines, sparsely puberulous, often somewhat ciliate, 16-18 mm long and 3-4 mm broad.

Sepals 5, acute, pungent, subglabrous often somewhat ciliate; posticous sepal broadly

linear, acute with 1-3 apical spines, 23-24 mm long and 4-5 mm broad; 2 anticous

sepals linear-lanceolate, 16-18 mm long and 3 mm broad; 2 lateral sepals linear-

lanceolate, 15-17 mm long and 2-2-5 mm broad. Corolla a light but bright blue with

darker blue reticulate veins; tube about 7 mm long, glabrous; limb about 19 mm

long, on the outside (lower side) silky pubescent, on the inside with two broad lines of

hairs, its lobes subequal obovate-oblong rounded at the apex, 4-5 mm long. Stamens

towards the base retrorsely silky pubescent; filaments about 11 mm long, anthers

shortly hirsute-pubescent, 3-4 mm long. Ovary densely puberulous to velutinous and

with a dense apical tuft of erect stiff, almost bristly hairs, 3-4 mm long; style 14-15 mm

long towards the base with setaceous hairs which form an abaxial line becoming denser

and broader towards the base, these hairs and those on the ovary yellowish and fawnish.

Capsule narrowly oblong, glabrous, about 14 mm long.

This species was discovered by F. A. Rogers as early as 1921, but it was apparently

not collected again till Mr. and Mrs. P. F. Clarke found it in the summer 1955/56.

It is probably one of those local endemics in which the Barberton area is so rich. I am

indebted to the collectors for the following extensive notes:

“ The plant is a straggly, low-growing shrub, averaging about 2 ft. in height and

possibly attaining a maximum of 3 ft. It is branched, usually slender, but in some cases

forming a bush or thicket 3 or 4 ft. across. It is fairly common in the locality in which

it was found, but we have not yet observed it elsewhere. It grows in shallow shaly soil,

sloping steeply, on shale outcrops — a very well drained position, facing S.W. It occurs

in fairly open bush consisting mainly of Kirkia, Bauhinia galpinii, Peltophorum,

Bowkeria, Acacia ataxacantha. Heteropyxis, Ziziphus, Acacia karroo, Dalbergia,

and in association with Royena, grasses, aloes. Orthosiphon, Crossandra and many

other unidentified plants. The Schlerochiton is found in partial shade — where the bush

is thick it does not occur. The flowers are clear blue in colour (see accompanying sketch).

Unripe fruit are plentiful and a few ripe fruits were found.”

Characteristic of the species are the narrow linear-oblong leaves with deflexed

margins and the 3-spined fertile bracts, two characters which I have not found in any

other species I have studied. The seeds in the capsules had all been eaten by insects

and could not be described.

A. Meeuse

APOCYNACEAE

Acokanthera

When G. Don described the genus in his Gen. Syst. 4: 485 (1838), he explained

that the name is derived from the Greek “ acoce ”, a mucrone, referring to the mucronate

anthers characteristic of Acokanthera spp. In writing the Greek word he translated

the Greek kappa in both cases as a roman c, but he spelt the generic name Acokanthera.

Subsequent authors have considered that, to be consistent, the genus should be spelt

either Acocanthera, as was done by Endlicher, Gen. PI. Suppl. 1 : 1404 (1841), Pfeifer,

Nom. Bot. 1: 29 (1873) and K. Schumann in Pflanzenfam. 4, 2: 126 (1895), or

Akokanthera, as proposed by Walpers, Rep. 3: 122 (1845). Stapf retained the spelling

Acokanthera in FI. Trop. Afr. 4, 1 : 92 (1902) and in FI. Cap. 4, 1: 499 (1907) but, in

Kew Bull. 29 (1922), he stated a preference for Acocanthera “ in the place of the absurd

and barbarous form Acokanthera, found in G. Don’s Generum Systema, which is

evidently due to a printer’s error There seems, however, equal reason for concluding

that Don deliberately spelt the name Acokanthera and, unless there is clearer evidence

to the contrary, this spelling should be retained.

448

Acokanthera is closely related to Carissa, under which genus it is placed in synonymy

by Pichon in Mem. Mus. Nat. Hist. Paris, n.ser. 24: 132 (1948) and Bull. Jard. Bot.

Brux. 22: 109 (1952). It is considered, however, that there are adequate reasons for

maintaining it as distinct. In support of this view may be cited the essentially practical

reason that species of Acokanthera have a characteristic facies, which permits the

assignment of even sterile specimens without difficulty, even though the characteristics

are not easy to define in words. Florally, the two genera are closely allied but, in

Acokanthera, the inflorescence is axillary, not terminal or pseudo-axillary as in Carissa,

and the stamens are situated at the apex of the corolla tube, with the tips of the anthers

often exserted. In Carissa, the stamens are situated from about the middle of the tube

to near the apex, with the mouth of the tube slightly constricted, so that the tips of the

anthers are not visible. There are also small differences in the shape of anthers and

stigma.

The genus Carissa (in the sense of Stapf in Flora Capensis and Flora of Tropical

Africa) may be divided into two sections: Section Carissa, in which the corolla lobes

overlap to the right; and Section Arduina, in which the corolla lobes overlap to the

left (as in Acokanthera). Section Carissa includes armed and unarmed species, occurring

in Africa, Madagascar, Asia and Australia. In Section Arduina, all species are armed

(though occasional herbarium specimens may lack spines) and the species are limited

to eastern and southern Africa. Thus, although the presence or absence of spines

would not alone constitute a generic difference, the character may be used in conjunction

with the direction of overlap of the corolla lobes to separate Acokanthera and Carissa.

A further practical reason for keeping the two genera separate is that Acokanthera

species all have a highly toxic substance in the sap, while this does not appear to be

the case in species of Carissa.

Typification of the genus Acokanthera

As indicated by Phillips, Gen. S. Afr. Flow. PI. ed. 2: 583 (1951), the type of the

genus Acokanthera has been regarded as A. venenata G. Don, generally written A.

venenata (Thunb.) G. Don. It is clear from what follows that this view can no longer

be held without altering the circumscription of the genus. In order to retain the genus

Acokanthera in its present sense, therefore, it is proposed that the type of the genus

should be accepted as A. lamarckii G. Don, a superfluous name for Cestrwn

oppositifolium Lam., which is now combined in Acokanthera.

Acokanthera oppositifolia (Lam.) L. E. Codd, comb. nov. — Cestrwn oppositifolium

Lam. in Tab. Encycl. Bot. 2: 5, t. 112, fig. 2 (1797). Type: Africa, Sonnerat (P).

C. venenatum Burm. f., FI. Cap. Prodr. 5 (1768), non Acokanthera venenata G. Don

(1838). Type: South Africa, Banks (G). C. venenatum Thunb., Prodr. 1: 36

(1794); FI. Cap. ed. Schult. 193 (1823), non C. venenatum Burm. f. (1768). Type:

South Africa, Thunberg (U).

Sideroxylon toxiferum Thunb., Trav. ed. 3, 1 : 156 (1795), nom, nud.

Acokanthera lamarckii G. Don, Gen. Syst. 4:485 (1838), nom. illegit. Type: as for

Cestrwn oppositifolium Lam. A. venatorium E. Mey. in Drege, Zwei Pfl. Doc 171

(1843); Sond. in Linnaea 23: 79 (1850); nom. nud. A. venenata sensu Stapf in

F.T.A. 4, 1 : 94 (1902); F.C. 4, 1 : 500 (1907); Sim, For. FI. Cape Col. 270, t.l54,

fig. 1 (1907); Marloth, FI. S. Afr. 3, 1: t.l7 (1932); Brenan & Greenw., Check-list

Tang. Terr. 2: 47 (1949). A. venenata (Burm. f.) G. Don ex C. A. Smith in J. S.

Afr. For. Ass. 20: 42 (1951), nom. illegit. var. scabra (Sond.) Markgf. in NotizbI.

Bot. Gart. Berl. 8: 470 (1949).

Toxicoph/aea thunbergii Harv. in Hook. Lond. Journ. Bot. 1: 24 (1842); Thes. Cap.

10, t. 16 (1859); Sond. in Linnaea, 23 : 78 (1850). Type: as for Cestrum venenatwn

Thunb. T. thunbergii var. scabra Sond., l.c. Type: based on several syntypes.

T. c£’^t/'c>/V/c5 A.DC. in DC., Prodr. 8 : 336 (1844). Type: as for Cestrum venenatum

Thunb.

449

Carissa acokanthera Pichon in Mem. Mus. Hist. Nat. Paris, n.s. 24: 132 (1948). C.

oppositifoUum (Lam.) Pichon in Bull. Jard. Bot. Brux. 22: 109 (1952).

In the protologue to Acokanthera venenata, G. Don cites Cestrum venenatum

Thunb. and C. citrifoUum Retz. It may be assumed that his intention was to take up

Thunberg’s epithet in his genus Acokanthera. As pointed out by Mr. Dandy on p.

428 of this journal, however, C. venenatum Thunb. (1794) must be regarded as an illegiti-

mate homonym of C. venenatum Burm. f. (1768) and therefore has no standing. The

epithet venenatum Burm. f. cannot be taken up in Acokanthera, thus the next legitimate

name which is available, namely, C. oppositifoUum Lam. must be adopted.

Nomenclaturally, because C. venenatum Thunb. is illegitimate, the name Acokan-

thera venenata G. Don becomes a synonym of C. citrifoUum Retz. (for which it is a

superfluous epithet), not of C. venenatum Thunb. C. citrifoUum Retz. (1803) is based

on a plant cultivated at Lund. The opportunity to study the type specimen was made

possible by the kindness of the Keeper of the Lund Herbarium, Dr. Norlindh, who

sent the holotype on loan to Pretoria. An examination of this specimen shows that it

is not conspecific with Cestrum venenatum Thunb., as indicated by G. Don. It probably

belongs to Cestrum or a closely allied genus and it is unlikely that it came originally

from South Africa.

Acokanthera oblongifolia (Hochst.) L. E. Codd, comb. nov. — Carissa oblongifoUa Hochst.

in Flora, 827 (1844). Type: Natal, Krauss (K). C. spectabilis (Sond.) Pichon in

Mem. Mus. Hist. Nat. Paris, n.s, 24: 132 (1948). Toxicophlaea spectabilis Sond.

in Linnaea, 23 : 79(1850). Syntypes: Port Natal, 37 ; 511. Acokanthera

spectabilis (Sond.) Hook. f. in Bot. Mag. t. 6359 (1878); Stapf in FI. Cap. 4, 1:

501 (1907). A. venenata var. spectabilis (Sond.) Sim, For. FI. Cape Col. 270,

t. 154, fig. 2 (1907).

Mr. W. Marais examined type material of Carissa oblongifoUa Hochst. (1844)

at Kew and reports it to be conspecific with Toxicophlaea spectabilis Sond. (1850).

The former epithet must, therefore, be taken up. A. oblongifoUa is closely allied to

A. oppositifolia, but can usually be distinguished by the longer corolla tube, the large,

plum-like fruits and the indistinct secondary nerves on the underside of the leaves.

Acokanthera schimperi {A. DC.) Schweinf. var. rotundata L. E. Codd, var. nov., sed a

typo foliis majoribus rotundioribus plerumque scabridis differt.

Frutex vel arbor parvus ad 3 m altus, ramulis tomentosis vel puberulis. Eolia

coriacea scabrida late elliptica usque subrotunda 4-7 cm longa 3 • 5-5 • 5 cm lata, apice

obtusa usque rotunda mucronata, basi obtusa usque rotunda, nerviis secundariis

obscuris, petiolo robusto 2-6 mm longo. Inflorescentia multiflora subsessilis corymbosa,

bracteis ovatis 1 ■ 5 mm longis caducis, floribus sessilibus. Calyx 2-2 • 5 mm longus

puberulus usque tomentulosus, sepalis ovato-lanceolatis acuminatis. Corolla hypo-

crateriformis, tubo subcylindrato 8-9 mm longo pubescent! vel glabro, lobis 5 ovatis

obtusis 4. 4- 5 mm longis 2 mm latis. Stamina 5 in faucibus corollae inserta leviter

exserta, antheris ovatis pubescentibus breviter mucronatis. Ovarium integre 2-loculare,

loculis 1-ovulatis, stylo filiforme 7-8 mm longo, stigmate parvo conico apice bifido.

Fructus baccatus globosus vel subglobosus 1-5-2 cm diam. Semina 1-2 late elliptica,

endospermo carnoso.

Transvaal. — Sibasa: Kruger National Park, 3^ miles N. of Klopperfontein,

alt. 1,400 ft., flowering and fruiting March 1949, Codd 5432 (type, PRE). Pietersburg:

Chuniespoort, Pole Evans H 19451. Nelspruit: Kruger National Park, Klokwene,

van der Schijjf 758.

S. Rhodesia. — Mangwe, Plowes 1308. Matoppo Hills, near Bulawayo, Galpin

7076; Eyles 6303; Miller 5162.

450

Shrub or small tree up to 3 m high; branchlets tomentose or puberulous. Leaves

coriaceous, scabrid, broadly elliptic to subrotund, 4-7 cm long and 3 -5-5 -5 cm broad,

apex obtuse to rounded, mucronate, base obtuse to rounded; secondary nerves obscure;

petiole stout, 2-6 mm long. Inflorescence many-flowered subsessile, corymbose;

bracts ovate, 1 • 5 mm long, caducous; flowers sessile. Calyx 2-2-5 mm long, puberulous

to tomentulose; sepals ovate-lanceolate, acuminate. Corolla hypocrateriform, tube

subcylindric 8-9 mm long; pubescent or glabrous; lobes 5, ovate, obtuse, 4-4-5 mm

long and 2 mm broad. Stamens 5, situated in the corolla throat, slightly exserted;

anthers ovate, pubescent, shortly mucronate. Ovary entire with 2 locules, 1 ovule in

eachlocule; style filiform 7-8 mm long; stigma small conical, apex bifid. baccate,

globose or subglobose, 1 - 5-2 cm in diameter. Seeds 1 or 2, broadly ellpitic; endosperm

fleshy.

The species of Acokanthera fall into three main groups, namely; {a) A. schimperi,

A. deflersii, A. ouabaio and A. friesiorum; (b) A. oppositifolia, A. longiflora and A.

rhodesica; and (c) A. oblongifolia.

The plant now described, with its subrotund leaves, indistinct secondary venation

and more or less globose fruits clearly belongs to group (a). The material that has

been seen of this group leads to the impression that there may be only one variable

species involved. The oldest name is A. schimperi (A. DC.) Schweinf. and the other

three may welt be reduced to synonymy, or to varieties, as Stapf has already done with

A. defter sii.

In its usually scabrid leaves, var. rotundata resembles A. deflersii from Arabia,

while in leaf shape it is closer to A. friesiorum, though it differs in having larger and less

glossy leaves. In distribution it is completely separated from the A. schimperi complex,

which is recorded from Kenya to southern Arabia, but its characteristics are not

sufficiently clear-cut to warrant separating it as a distinct species.

Carissa

Carissa bispinosa (L.) Desf. ex Brenan in Mem. N.Y. Bot. Card. 8: 502 (1954), aggregate

species.

The C. bispinosa complex is distrubuted from the south-western Cape Province

to Natal, Swaziland, Transvaal and Mocambique, extending northwards to Nyasaland

and westwards to Bechuanaland and South West Africa. There is considerable variation

in leaf size, shape and texture, size of spines, and situation of the anthers in the corolla

tube.

Two main evolutionary trends are recognised and two varieties are upheld. The

two varieties overlap geographically, especially in the southern and eastern Cape

Province; here several intermediate specimens are found and it is for this reason that

the two groups were regarded as not having the status of separate species.

(a) var. bispinosa.

Arduina bispinosa C., Mani. \ \ 52 (1767). Type: the plate in Mill. Ic. 2: t. 300 (1760).

Lycium cordatwn Mill., Card. Diet. ed. 8: No. 10 (1768). Carissa bispinosa (L.)

Desf. ex Brenan, l.c., excl. specimen cited. C. bispinosa Desf., Tabl. Ecol. Bot. 78

(1804), nom. nud. ; Desf. ex Steud., Nom. ed. 2; 298 (1841), in synonymy; Merxm.

in Mitt. Bot. Munchen, 17-18: 399 (1957). C. arduina Lam., Diet. 1: 555 (1785);

Stapf in FI. Cap. 4, 1 : 498 (1907), partly, excl. syn. C. erythrocarpa, C. acuminata,

C. haematocarpa, C. ferox, and C. oblongifolia. C. myrtoides Desf., Cat. Hort.

Paris, ed. 3; 398 (1829). C. cordata (Mill.) Fourc. in Trans. Roy. Soc. S. Afr. 21 :

82 (1934). C. cordata Dinter in Fedde Rep. Beih. 53; 112 (1928), nom. nud.

C. dinteri Markgf. in Notizbl. Bot. Gart. Berlin, 15: 750 (1942). Jasminonerium

bispinosum (L.) O. Ktze., Rev. Gen. PI. 2: 415 (1891).

451

A much-branched, twiggy shrub up to 3 m high. Leaves thickly coriaceous,

1-3 (rarely to 5) cm long, broadly ovate to ovate-elliptic. Spines robust, up to 5 cm

long, bifurcate, twice bifurcate or simple. Anthers situated near the apex of the corolla

tube. Style 3-4 mm long.

This variety is found in karroid scrub, bushveld and coastal scrub vegetation,

usually in relatively hot, dry situations, in the southern and south-eastern Cape Province,

central and western Transvaal and northern South West Africa.

In some specimens from the eastern Cape Province, the robust spines are twice

bifurcate, while from the central and western Transvaal and South West Africa specimens

are found with simple spines. C. dinteri Markgraf is based mainly on the presence of

simple spines; in all other respects it is scarcely distinguishable from var. bispinosa

and is therefore not worth upholding, even as a variety.

The significance of the position of the anthers and length of the style is discussed

later under var. acuminata.

(b) var. acuminata {E. Mey.) L. E. Codd, stat. nov.

Arduina acuminata E. Mey., Comm. PI. 191 (1837); K. Schum. in Pflanzenfam. 4, 2:

126 (1895). Type: between Umzimvubu and Umsikaba Rivers, near large Water-

fall, Drege. A. erythrocarpa Eckl. in S.A. Quart. J. 4: 372 (1830). Carissa

acuminata (E. Mey.) A. DC. in DC. Prodr. 8: 335 (1844). C. erythrocarpa (Eckl.)

A. DC., l.c. C. arduina (non Lam.), Stapf in FI. Cap. 4, 1: 498 (1907), partly,

as to syn. C. acuminata and C. erythrocarpa. C. megaphylla Gdgr. in Bull. Soc.

Bot. France, 65: 59 (1918). C. bispinosa [non (L.) Desf. ex Brenan], Brenan in

Mem. N.Y. Bot. Gard. 8: 502 (1954), as to specimen cited.

Shrub up to 5 m high, often sparingly branched and sometimes semi-scandent.

Leaves thinly coriaceous to coriaceous, ovate to ovate-lanceolate, 2 ■ 5-7 cm long.

Spines slender, bifurcate, 0-5-2 -5 (rarely to 3-5) cm long. Anthers situated near the

middle or towards the apex of the corolla tube. Style 1 • 5-3 (rarely to 3-5) mm long.

This variety is found in fairly dense woodland such as forest margins and scrub

forest in the southern and south-eastern Cape Province, Natal, Swaziland, eastern and

northern Transvaal, eastern Southern Rhodesia and Nyasaland.

The anthers are situated near the middle of the corolla tube or towards the apex

and, as the stigma accompanies the anthers, the style is shorter in this variety than in

var. bispinosa. The distinction is more marked from the Transvaal northwards, rather

than in the Cape Province and Natal, as is shown in Table I and II which summarise

the distribution of style length on a geographical basis in 69 specimens examined.

Table 1

Style length in 28 specimens of C. bispinosa var. bispinosa.

452

Table II

Style length in 41 specimens of C. bispinosa var. acuminata.

An unusual form occurs at high altitudes in the central Transvaal and is represented

by the following specimens; Lydenburg District, summit of Lulu Mountains, 4,500 ft.,

Mogg 16963; Barnard and Mogg 995; Pietersburg District, Wolkberg, 6,000 ft.,

Gerstner 5657. In these specimens the branches are extremely twiggy with very

numerous, relatively short, stout, bifurcate spines; the leaves are small, broadly ovate,

almost cordate-based, 1-1 • 5 cm long; and the flowers are smaller than usual with styles

1 • 5-2 mm long. These specimens have been omitted from the above Tables. They

may represent a distinct variety but, until more information is forthcoming, they are

best regarded as a form of C. bispinosa var. bispinosa.

Adenium

Adenium obesum (Forsk.) R. and S. var. multiflorum (Klotzsch) L. E. Codd, stat. nov. —

A. multiflorum Klotzsch in Peters, Reise Mossamb. Bot. 279, t. 44 (1861). Type:

Tette, Peters s.n.

I am indebted to Mr. W. Marais, our liaison officer at Kew, and Mr. Meikle, of

the Kew staff, for examining the tropical African material of Adenium for me. They

came to the conclusion that an excessive number of species names have been published

and that the following must be regarded as one species complex; A. obesum (Forsk.)

R. and S., Syst. Veg. 4: 411 (1819), based on Nerium obesum Forsk., FI. Aegypt.-Arab.

205 (1775); A. honghel A.DC. in DC., Prodr. 8: 412 (1844); A. multiflorum Klotzsch,

I.C.; A. speciosum Fenzl in Sitzungsber. Kais. Akad. Wiss. Wien, 51; 140 (1865);

A. arabicum Balf. f. in Trans. Roy. Soc. Edinb. 31; 161 (1888); A. micranthum Stapf

in Kew Bull. 334 (1894); and A. coetanum Stapf in FI. Trop. Afr. 4, 1: 227 (1902).

Of these, it is considered that only A. multiflorum is sufficiently distinct to be worth

separating as a variety. It is fairly constant in having acute to acuminate corolla lobes

with crisped margins, while the margins are a darker pink, almost scarlet in colour.

The southern material has thicker leaves with more or less immersed veins, as against

the thinner, markedly veined leaves of material from east and west tropical Africa.

TYPIFICATION OF PACHYPODIUM SUCCULENTUM AND P. BJSPINOSUM

When Linn. fil. described Echites succulenta and E. bispinosa in his Suppl. 167

(1781), he mixed the characters of the two species. For instance, he describes E. succu-

lenta as: “ foliis linearibus subtus tomentosis, corollis infundibuliformibus ” and

E. bispinosa as; “foliis lanceolatis glabris, corollis hypocrateriformibus ”. Actually,

the species with leaves tomentose below has the hypocrateriform corolla and vice versa.

453

Both species were based on Tlwnberg specimens “ ad Cap. bonae Spei ”, and were

transferred to the genus Pachypodium by A. de Candolle in DC. Prodr. 8: 424 (1844).

Thunberg in his Prodromus, 37 (1794) and FI. Cap. ed. Schult. 232 (1823) repeats

the protologue verbatim but, in the latter work, he adds a more detailed description

of the corolla of E. bispinosa which agrees with an infundibuliform, not a hypocrateri-

form, corolla.

E. Meyer, Comm. PI. 188 (1837) created the genus Belonites to accommodate the

above two species. He evidently realised the original discrepancy and placed the

characters correctly, typifying the species as follows:

1. Belonites succulenta: “ folds subtus tomentosis, corollis hypocrateriformibus,

limbi laciniis spathulatis. Echites succulenta Thunb.”

2. B. bispinosa: “ foliis subtus glabris, corollis infundibuliformibus, limbi laciniis

basi latissimus. Echites bispinosa Thunb.”

This emendation by E. Meyer was followed by Stapf in Flora Capensis 4, 1:517

( 1 907) and subsequent workers. It is, however, necessary to assess the available Thunberg

specimens to discover if any specimens bear clear evidence that they were studied by

Linn. fil. There are no relevant specimens in the Linnaean Herbarium, so the Directors

of the Herbaria at Uppsala and Stockholm were approached. Uppsala Herbarium

does not send Thunberg specimens on loan, but the Director informs me that there

are three relevant sheets in the Thunberg Herbarium. One of these is named Echites

succulenta. It has linear leaves, tomentose on the underside and the corolla is badly

pressed but is probably hypocrateriform. This specimen bears a determination by

N. E. Brown: ‘‘^Pachypodium succulentum A. DC.” The other two sheets are named

Echites bispinosa and are designated a and The determination of the first one was

confirmed by N. E. Brown. It consists of a specimen with lanceolate leaves which are

sparingly hispid on the underside, while the flowers appear to be infundibuliform. On

the second sheet, N. E. Brown has placed a note reading: “The specimen marked

X is the only one of E. bispinosa on this sheet, all the others belong to E. succulenta

and have probably been mounted here by mistake ”. None of the specimens bear

any indication of having been studied by the younger Linnaeus.

The Director of the Stockholm Herbarium kindly sent us their specimens on loan.

Five sheets were received, four of which consisted of solitary specimens collected by

Thunberg, while the fifth was a mixed sheet with one Thunberg specimen and two

Sparrman specimens. None of them show any annotation by Linn. fil. and therefore

for purposes of selecting lectotypes they would have the same standing as the Uppsala

specimens. As I have seen the Stockholm specimens, the following lectotypes are

proposed :

Pachypodium succulentum (Linn, f.) A. DC. Lectotype: The sheet in Stockholm

leg. Thunberg, Cap. b. Spei, on which is written at the back of the sheet: “ Echites

(succulenta) aculeis binis extrafoliaceis, foliis linearibus subtus tomentosis, corollis

infundibuliformibus ”. The corolla of the specimen is, in fact, hypocrateriform, and

matches what we have been accusomed to calling Pacypodium succulentum.

Pachypodium bispinosum (Linn, f.) A. DC. Lectotype: The sheet in Stockholm

leg. Thunberg, Cap. b. Spei, on which is written at the back of the sheet: “ Echites

(bispinosa) aculeis binis extrafoliaceis, foliis lanceolatis glabris, corollis hypocrateri-

formibus ”. The corolla is, in this case, clearly infundibuliform and the specimen

matches what we have been accustomed to calling Pachypodium bispinosum.

The other two Stockholm sheets consist each of a somewhat scrappy Thunberg

specimen annotated Echites bispinosa, both of which are what we are calling Pachy-

podium succulentum. The sheets in Stockholm thus agree in essentials with those at

Uppsala, i.e., two specimens (one of each species) are named in accordance with modern

usage, while two are wrongly named. If the first two are selected as lectotypes, the

454

application of epithets will continue as at present, which agrees with the emendation

made by E. Meyer. It is probable that the confusion in the published description by

Linn. fil. resulted from an error in transcription and is not due to flowers of one species

having been mounted in juxtaposition with a vegetative portion of the other species.

Strophanthus

Strophanthus luteolus L. E. Codd, sp. nov., S. petersiano Klotzsch affinis, sed floribus

parvioribus, calyce angustiori, ramulis calyce corolla pubescentibus differt.

Frutex sarmentosus 3-8 m altus, ramulis junioribus pubescentibus sparse lenti-

cellatis. Folia brevissime petiolata ovato-elliptica vel oblongo-elliptica 2 -5-4 -5 cm

longa 1-1 -5 cm lata, apice acuta vel rotundata, sparse pubescentia deinde glabrescentia.

Cymae in ramis abbreviatis terminales, tomentosae, 1-3-florae; bracteae lineares

9-10 mm longae. Calyx tomentosus; sepala lanceolato-linearia 1-1-3 cm longa.

Corolla luteola extus purpureo-striata dense puberula ; tubi pars infra stamina cylindrata

6 mm longa, supra stamina infundibuliformis 7 mm longa 7-9 mm lata; lobi basi

ovato-lanceolati in caudas lineares attenuati 5-8 cm longi; faucium squamae lanceo-

latae 2-5 mm longae. Antherae lanceolatae sagittatae acuminatae 3-5-4 mm longae.

Ovarium dense pubescente. Folliculi anguste fusiformes fusco-brunnei lenticellati

22 cm longi 1 - 5 cm diam.

Transvaal. — Soutpansberg: Wylliespoort, flowering September, 1960, Hardy

and Wells 359 (PRE, holotype); near the Salt Pan, fruiting September, 1960, Hardy 350.

Natal. — Ingwavuma: Maputaland, flowering June, 1914, Maputo Expedition in

TM 14460; Mkuzi-Maputa areas, flowering December 1945, Bell Marley in NH 40920;

Ndumu Game Reserve, flowering October 1959, Finley 499.

Scandent shrub or climber 3-8 m high; branchlets shortly crisped-tomentose,

glabrescent with age; bark reddish-brown, lenticellate. Leaves shortly petioled, drying

dark green or brownish above, paler below, subglabrous or sparingly pubescent with

scattered hairs along the midrib and margin, eventually glabrescent, ovate-elliptic to

elliptic-oblong, 2 -5-4 -5 cm long and 1-1-5 cm broad, base obtuse to acute, apex

acute to rounded, occasionally emarginate; margin incurved; secondary nerves 4-6

on each side, distinct; petiole 2-4 mm long; axillary glands small, tooth-like.

Inflorescence terminal, usually on short lateral shoots, 1-3-flowered, pubescent; bracts

linear, acute, 9-10 mm long, pubescent; pedicels 7-10 mm long, crisped-tomentose.

Flowers yellowish with purple markings in the throat and on the outside of the tube.

Calyx 1-1-3 cm long; sepals linear-lanceolate, tapering gradually to the apex, shortly

pubescent. Corolla shortly and densely pubescent without and within; tube sub-

cylindric for 6 mm then widening gradually to a funnel-shaped portion 7 mm long

and 7-9 mm broad at the mouth; lobes lanceolate at the base, attenuate into linear-

filiform appendages, pendulous, 5-8 cm long, shortly pubescent on both surfaces;

throat scales lanceolate, 2-5 mm long. Stamens included; filaments thick, 1 mm long,

pubescent; anthers connivent, lanceolate, 3-5-4 mm long, acuminate, shortly apiculate,

sagittate, polliniferous in the upper half, hairy on the back. Ovary of two free carpels,

densely pubescent; style 7-8 mm long; stigma enclosed by the anthers, capitate, with

a reflexed frill at the base. Follicles narrowly spindle-shaped, reddish-brown, about

22 cm long and 1 -5 cm in diam., lenticellate, tapering gradually then abruptly thickened

at the apex.

Although superficially resembling S. gerrardii Stapf and S. petersianus Klotzsch,

S. luteolus differs from both these species in the pubescent twigs and flower parts.

From S. gerrardii it differs, in addition, in the longer calyx and erect, not recurved,

sepals, while from 5. petersianus it differs in the smaller, differently coloured corolla

and the narrower sepals, as well as in leaf size and shape. It does not run to any species

in Gilg’s monograph on Strophanthus (Engler, Monogr. Afr. Pfl. Fam. vii, Stroph.,

455

1903), nor in Stapf’s treatment in Flora of Tropical Africa. Among more recently

described species, it appears to come nearest to S. hirsutus H. Hess in Ber. Schweiz.

Bot. Ges. 62: 88 (1952), based on specimens collected in Angola, but dilfers in several

important characters, for example, the longer calyx, the more funnel-shaped corolla

tube and the pubescence of the leaves, which is never densely tomentose on both surfaces

as in 5. hirsutus. A duplicate of Tinley 499 was sent to Mr. Marais at Kew who reports

that it is not matched in the Herbarium of the Royal Botanic Gardens.

The species is known from three gatherings in Maputaland, the north-eastern

extremity of Zululand, and from recent gatherings in the Soutpansberg. In Maputaland

it is found in coastal forest, climbing into the canopy, while in the Soutpansberg it

grows on relatively dry, wooded hillsides, producing numerous slender stems which

climb when they find the support of shrubs or trees. The older stems do not produce

corky, winged out-growths, as is the case with 5. gerrardii. The flowers are creamy-

yellow, with a purplish tinge in the throat and on the outside of the corolla tube.

L. E. CODD

ARACEAE

Zantedeschia jucunda C. Letty, sp. nov., Z. pentlandii (Whyte ex Watson) Wittm. affinis,

sed ita differt: folia trianguli-hastata dense maculata sub-conduplicata, spatha

saturatior lutea interdum intus rugosa subregulariter infundibuliformis ad fauces

valde aperta apice in acumen subulatum attenuata.

Planta decidua habitu rigido. Folia glabra, petiolis 24 cm longis glabris, laminis

saturate viridibus maculatis 17-30 cm longis 5-15 cm latis trianguli-hastatis acutis

apice subulatis subconduplicatis, lobis basalibus obtusis interdum acutis sursum patenti-

bus ad 14 cm longis basi 7 cm latis. Pedunculus foliis aequalis vel longior glabrus viridis.

Spatha 10-16 cm longa, intus saturate lutea leviter rugosa macula purpurea, extus

saturate lutea, limbo subhorizontaliter patenti in acumen subulatum ad 1 • 5 cm longum

attenuate. Spadix ad dimidium longitudinis spathae attingens cylindratus obtusus.

Ovaria c.24 subglobosa angulata sessilia, stigmate 0-5 mm longo sessili discoideo,

staminodiis nullis. Antherae luteae.

Transvaal. — Lydenburg: Lulu mountains, Sekukuniland, Barnard and Mogg

991 (PRE, holotype); Schoonoord, west slope of Lulu mountains, du Plessis in PRE

28835; Barnard 180; summit of Lulu mountains Barnard 181; Magnet Heights,

Thompson in PRE 28836 (cult).

Description. — Plant deciduous, up to 60 cm high, glabrous, fairly rigid in habit.

Leaves with petiole 24 cm long; blade dark green, densely maculate with elongate-

white translucent spots, 17-30 cm long, 5-15 cm broad, triangular-hastate, acute with

a subulate tip, semi-folded along the midrib; basal lobes obtuse or sometimes acute,

upwardly spreading, up to 14 cm long, 7 cm broad at the base often smaller and narrower.

Peduncle as long as, or longer than, the leaves, glabrous, green. Spathe 10-16 cm long,

deep yellow with a purple blotch, slightly rugose within, deep yellow without; limb

sub-horizontally spreading, tapering to a subulate green tip up to 1-5 cm long. Spadix

up to half as long as the spathe, sometimes longer, cylindric obtuse. Ovaries sub-

globose angled from pressure, sessile about 24 in number; stigma 0-5 mm long, sessile,

discoid; staminodes none. Anthers yellow.

Specimens of this most striking Zantedeschia from Sekukuniland have been sent,

from time to time, to the National Herbarium, Pretoria, since 1934. The first arrived

in November 1934 from Mr. W. G. Barnard (No. 180) and further material was received

in January 1939, collected by Barnard and Mogg (No. 991). In 1954 living material

was sent in by Miss S. Thompson of Haenertsburg, grown from plants collected at

Magnet Heights. The Native Commissioner, Schoonoord, kindly supplied plants in

02909—2

456

February 1955 which flowered in Pretoria in January 1956 and in December of the

following year further material from Schoonoord was received from Mr. S. S. du Plessis.

The species has been named Z. jucmda because of the pleasing appearance of

the plants with deep green copiously spotted leaves and the almost campanulate spathe

being brilliant deep yellow both without and within. Its distribution seems restricted

to the Schoonoord-Magnet Heights region of the Lulu mountains, Sekukuniland,

where it flowers from November to January.

It is most nearly related to Z. pentlandii (Whyte ex Watson) Wittm. and Z. tropicalis

(N.E. Br.) C. Letty. It differs from the former in having triangular-hastate copiously

spotted leaves with upwardly spreading basal lobes and an acute tip, the spathe a deeper

yellow without and within and sometimes rugose within, and in the limb more spreading

and tapering to a subulate tip, whereas Z. pentlandii has immaculate (very rarely spotted)

oblong lanceolate leaves, slightly constricted above the short sagittate rounded basal

lobes, apex obtuse, sometimes acute, with subulate tip, the spathe regularly funnel-

shaped, limb lighter in colour, greenish outside, more suddenly rounded into the subulate

tip. Z. jucunda can be distinguished from Z. tropicalis in its less luxuriant growth,

stiffer, more inwardly folded triangular leaves, but chiefly in the less tightly rolled,

consistently more brilliantly yellow more tapering spathes, whereas Z. tropicalis shows

white, cream or salmon pink colour variations, but is never deep yellow.

Zantedeschia tropicalis {N.E. Br.) C. Letty, stat. nov., Richardia melanoleuca Hook. f.

var. tropicalis N.E. Br. in FI. Trop. Afr. 8: 168 (1901). Zantedeschia melanoleuca

Hook. f. var. tropicalis (N.E. Br.) Traub in Plant Life 4; 24 (1948). Lectotype:

Nyasaland: Namasi, Cameron s.n. (K) (2 sheets).

On studying a series of plants of Zantedeschia melanoleuca var. tropicalis both in

the veld and preserved, the differences between these plants and typical Z. melanoleuca

seemed important enough to warrant raising the variety to specific rank. Whereas

Z. melanoleuca seems to be restricted to the coastal regions of Natal, Z. tropicalis occurs

from Nyasaland through Southern Rhodesia into the Transvaal to Potgietersrus,

Letaba and Barberton and to the Mbabane district of Swaziland.

It differs from Z. melanoleuca in being more robust, petioles glabrous, leaf blades

longer and more triangular-hastate, up to 30 cm long and 30 cm across the basal lobes

from tip to tip, lobes spreading, obtuse; spathe larger, cream, pale yellow or suffused

with red, up to 14 cm long, tube sub-cylindric, closed; spadix with many (up to 12

rows) of ovules. Its habitat is at forest margins or in bush in partial shade and its

flowering reaches its peak during December and January.

Nyasaland. — Namasi, 1897, Cameron s.n. (K!); Nivera Hill, Benson 81.

Southern Rhodesia. — Mashonaland, Six mile spruit, Salisbury, Lady Evelyn

Cecil in K. 149; Melsetter. Crook in PRE 28828; Mount Silinda, Obermeyer in PRE

28827.

Transvaal. — Soutpansberg: miles N.E. of Louis Trichardt, Codd 8326;

Entabeni Forestry Station, Codd 8400; four miles N. of Louis Trichardt on Messina

road, Admiraal in PRE 28830; Pietersburg: one mile N. of Haenertsburg, Codd 8415;

on farm Wellstead, Thompson in PRE 28826; Magoebaskloof, fld. Pretoria, O'Connor

in PRE 28825; one mile E. of Haenertsburg, Reynolds 5806a; 5806b; Woodbush,

Mogg in PRE 28832; Mohlakeng, Blaauwberg, Codd Letaba; Tzaneen, Rogers,

12501 ; Duivelskloof, Haar in PRE 28832; The Downs, Murray, PRE 28833; Barber-

ton: Kaapmuiden, Mogg in PRE 28834; Louw’s Creek, Bragshaw 110; eight miles

S.E. of Barberton on Havelock Road, Codd 8159.

Swaziland. — Mbabane: Little Usutu valley, Compton 25405.

Cythna Letty

457

EBENACEAE

Nfw Combinations in the Genus Diospyros

In Bothalia 7, 1: 17-19 (1958) a number of new combinations were made in the

genus Royena. A short while after these notes were submitted for publication and

too late for withdrawal, it was discovered that White (Oxford Univ. Forest Soc. Journ.

4, 6: 31-34, 1958) had come to the conclusion that the genus Royena could not be

regarded as distinct from Diospyros. After some correspondence the author decided

to follow White in uniting the two genera. Characters previously used for distinguishing

Royena from Diospyros have proved to break down when the latter is studied on a

worldwide scale. As pointed out in Bothalia l.c., Royena cannot be regarded as truly

hermaphrodite, and the main difference given in the keys by various authors thus cannot

be used. Attempts were made to associate a number of characters which in combination

could be used to distinguish the two genera. However, as pointed out by White, this

results in a classification in which some closely related species will fall in separate

genera even though not closely related to the other species in the genus in which they

are placed. For these reasons the genus Royena is now united with Diospyros. The

following combinations are necessary.

Diospyros acocksii (de Winter) de Winter comb. nov. Royena acocksii de Winter in

Bothalia 7, 1 : 18 (1958).

Diospyros austro-africana de Winter nom. nov. Royena hirsuta L. Sp. PI. ed. 1 ; 397

(1753).

{a) var. austro-africana. Royena hirsuta L. Sp. PI. ed. 1 : 397 (1753) non Diospyros

hirsuta L. f. Suppl. 440 (1781); Royena angustifolia Willd. Sp. PI. 2; 633 (1800) non

Diospyros angustifolia Audib. ex Spach, Hist. Veg. Phan. 9: 405 (1843); Diospyros

hirsuta Desf. in Ann. Mus. Par. 4: 449, t.62 f.2 (1805) non Linn. f. Suppl. 440 (1781).

{b) var. rugosa (E. Mey. ex A. DC.) de Winter comb, et stat. nov. Royena rugosa

E. Mey. ex. A. DC. Prod. 8; 212 (1844) non Diospyros rugosa Sap. in Ann. Sc. Nat. 4,

17: 264, t.ll f.3 (1862) (fossil species). Royena hirsuta L. var. rugosa (E. Mey. ex

A. DC.) Zahlbr. in Ann. Naturhist. Hofmus. Wien. 28: 397 (1903).

(c) var. microphylla (Burch.) de Winter comb, et stat. nov. Royena microphylla

Burch. Trav. 1: 348 (1822) non Diospyros microphylla Bedd. Forest Man. 145 (1870);

Royena hirsuta L. var. microphylla (Burch.) de Winter in Bothalia 7, 1 : 18 (1958).

{d) var. rubriflora (de Winter) de Winter comb. nov. Royena hirsuta L. var. rubriflora

de Winter in Bothalia 7, 1 : 18 (1958).

As can be seen from the references under the varieties given above all the epithets

available for re-combination in Diospyros were already occupied in this genus, and it

was necessary to give a specific new name.

Diospyros galpinii (Hiern) de Winter comb. nov. Royena galpinii Hiern in F.C. 4, 1 :

450 (1906).

Diospyros glabra (L.) de Winter comb. nov. Royena glabra L. Sp. PI. 1 : 397 (1753).

Diospyros glandulifera de Winter nom. nov. Royena glandulosa Harv. ex Hiern in Trans.

Camb. Phil. Soc. 12: 89, t.2 (1873) non Diospyros glandulosa Lace in Kew Bull.

349 (1915).

Diospyros lycioides Desf. in Ann. Mus. Par. 6: 448, t.62 f.l (1805).

{a) subsp. lycioides. Royena lycioides (Desf.) A. DC. Prod. 8: 214 (1844).

(b) subsp. sericea (Bernh.) de Winter comb, et stat. nov. Royena sericea Bernh. ex

Krauss in Flora 27: 824 (1844); Royena lycioides (Desf.) A. DC. subsp. sericea

(Bernh.) de Winter in FI. PI. Afr. 32, 3: t.l262 (1958).

458

(c) subsp. guerkei (O. Ktze.) de Winter comb, et stat. nov. Royena guerkei O. Ktze.

Rev. Gen. PI. 3, 2; 196 (1898); Royena lycioides (Desf.) A. DC. subsp. guerkei de

Winter in Bothalia 7, 1 : 17 (1958).

{d) subsp. nitens (Harv. ex Hiern) de Winter comb, et stat. nov. Royena nitens Harv.

ex Hiern in Trans. Cambr. Phil. Soc. 12: 87 (1872); Royena lycioides (Desf.) A. DC.

subsp. nitens (Harv. ex Hiern) de Winter in Bothalia 7, 1 : 18 (1958).

Diospyros ramulosa (E. Mey. ex A. DC.) de Winter comb. nov. Royena rainulosa E. Mey.

ex. A. DC. Prod. 8: 212 (1844).

Diospyros scabrida (Harv. ex Hiern) de Winter comb. nov. Royena scabrida Harv. ex

Hiern. in Trans. Cambr. Phil. Soc. 12: 82 (1872).

(a) var. scabrida.

(b) var. cordata (E. Mey. ex A. DC.) de Winter comb, et stat. nov. Royena cor data

E. Mey. ex A. DC. Prod. 8: 211 (1844) non Diospyros cordata (Hiern) Bakhuizen

in Gard. Bull. Straits Settlm. 7: 167 (1933); R. opaca E. Mey. ex A. DC. Prod. 8:

211 (1844) non Diospyros opaca C.B. Cl. in HooL f. FI. Brit. Ind. 3: 567 (1882).

Both the epithets cordata and opaca are older than the epithet scabrida, but are

already occupied in Diospyros hence the next available epithet scabrida \vas used. The

epithet cordata has been used for the variety {b) to insure continuity and avoid more

confusion although both varieties have cordate leaves.

Diospyros simii (O. Ktze.) de Winter comb. nov. Royena simii O. Ktze. Rev. Gen.

PI. 3, 2: 196 (1898).

Diospyros villosa (L.) de Winter comb. nov. Royena vil/osa L. Syst. Nat. 12, 2: 302

(1767).

var. parvifolia (de Winter) de Winter comb. nov. Royena villosa L. var. parvifolia

de Winter in Bothalia 7, 1 : 18 (1958).

D. whvteana (Hiern) F. White comb. nov. Royena whyteana Hiern in Trans. Linn.

Soc. 2, 4: 25 (1894).

D. pallens (Thunb.) F. White comb. nov. Royena pallens Thunb. Prod. 80 (1794).

B. DE Winter and F. White

GENTIANACEAE

Chironia

Chironia stokoei Verdoorn sp. nov., C. jasminoidei affinis sed ab ea et ab aliis speciebus

stigmate peltato, praesertim floribus basi bi-bracteatis differt.

Suffrutex rigidus, adscendens, 30-50 cm altus; caules superne ramosi et foliolati

inferne saepe nudi nodis prominentibus. Folia coriacea, paulo corrugata, elliptica

nunc late nunc angustustate elliptica, interdum ovate vel lanceolati- elliptica, 1 • 7-3 • 5

cm longa, 0-5-1 -8 cm lata, obtusa vel acuta, marginibus anguste revolutis. Cymae

terminales, monochasiales, 3-florae vel ad florem solitarium reductae; pedicelli breves

vel 0; flores basi bi-bractiati ; bracteae 6-18 mm longae basi 1-3-4 mm latae, acumi-

natae, obtusae vel acutae. Calyx 9-16 mm longus infra medium lobatus; lobi a basi

acuminati, obtusi vel acuti, obscure carinati. Corolla “ magenta pink ” (perraro alba),

2-2-2 8 cm longa; tubus submembranaceus, c. 1-2 cm longus in faucibus leviter

constrictus apice ampliatus; lobi elliptici vel subrotundi, basi breviter cuneato-

unguiculares, 1-2-1 -8 cm longi, 0-8-11 cm lati. Stamina paulo infra fauces inserta

459

filamentis c. 1 cm longis, antheris c. 4 mm longis non tortilibus. Ovarium c. 1 cm longum,

in stylum attenuatum; stylus quam ovarium brevior c. 8 mm longus; stigmate peltato

apice crasse stigmatoso.

Cape. — Somerset West: Sneeuwkop, Stokoe 7986 (BOL); between Sneeuwkop

and Landdrost Kop, Stokoe 9211 (BOL). Caledon: Kogelberg, Stokoe 1003 (PRE,

holo.); 1003b; Stokoe in Bol. Herb. 17705; Stokoe in S.A. Museum Herb. 66894;

Palmiet River mountains, Stokoe 982; mountains S. of Sir Lowry’s Pass, Stokoe in

S.A. Museum Herb. 66893; Klein River mountains Stokoe in S.A. Museum 64199.

Rigid ascending suffrutices, 30-50 cm tall; stems leafy and branched above,

leafless below with prominent nodes, branches and leaves notate with shiny, microscopic,

subcutaneous cells. Leaves coriaceous, somewhat corrugated on both surfaces, from

broadly to narrowly elliptic, sometimes oblong-, ovate- or lanceolate-elliptic, I -7-3 -5

cm long, 0-5-1 -8 cm broad, rounded or acute at the apex, margins narrowly revolute.

Cymes terminal, monochasial, 3-flowered or reduced to a solitary flower; pedicels

short or 0; flowers bi-bracteate at the base; bracts 6-18 mm long, 1-3-4 mm broad

at the base, gradually narrowing to an obtuse or acute apex. Calyx 9-16 mm long,

divided beyond the middle; lobes gradually narrowing from a broad base to an obtuse

or acute apex, obscurely carinate. Corolla “magenta pink” (rarely white), 2 -2-2 -8

cm long; tube rather thin textured, slightly narrowed in the throat and widened at

the mouth, about 1 -2 cm long; lobes elliptic to subrotund and cuneate into a claw-like

base, usually obviously longer than broad, 1 - 2-1 - 8 cm long, 0 - 8-1 - 1 cm broad. Stamens

inserted just below the throat; filaments about 4 mm long; anthers about 4 mm long,

not spirally twisted. Ovary about 1 cm long, attenuating into the style; style shorter

than the ovary, about 8 mm long, stigma peltate, thickly stigmatose on top.

When revising the genus Chironia for the Flora of Southern Africa this species

was found to be represented in three herbaria, the Marloth Herbarium (now in the

National Herbarium, Pretoria), the Bolus Herbarium and the Herbarium of the South

African Museum (now at the Herbarium of the National Botanic Gardens, Kirsten-

bosch). All the specimens, nine in number, were collected by Mr. T. P. Stokoe over

a range of about 30 years. The earliest date was August 1924 and the most recent

September 1953. The specimens all came from mountainous country in the Somerset

West and Caledon districts. Unfortunately Mr. Stokoe although a great collector,

never claimed to have an eye for recognizing species. It was after his 91st birthday

(3.3.1959) and just after he had returned from his last trip to the Hottentots Holland

that he was informed of this new species and that it was to be called after him. Had he

lived he would most probably have made a point of investigating it for us. But his

great usefulness to botanists came to a regretted end when he died in April 1959.

Chironia stokoei differs from all the other species with simple styles and a terminal

stigmatic surface, in that the flowers are bi-bracteate at the base. In general it is nearest

C . jasminoides but can be distinguished by the calyx-lobes which are never long attenuate

at the apex and the shorter and broader leaves besides the bracts at the base of the

flowers. In the prominent nodes on the almost leafless lower portion of the stem it

resembles C. arenaria, but besides havi-g basal bracts it differs in the corolla-lobes

being longer than broad and the leaves usually broadly elliptic.

Chironia linoides Complex

C. linoides Linn. Aggregate species.

C. linoides subsp. linoides. Lectotype: specimen cited in Hort. Cliff, p. 54, No. 1.

C. linoides Sp. PI. ed. 1: 189 (1753); Prain in Kew Bull. 1908: 349 and 353 (1908);

Hill & Prain in F.C. 4, 1 : 1103 (1909). Syntypes: specimen cited in Hort. Cliff,

p. 54, No. 1; specimen in Roy. Lugab. 433; and figure in Breyn. Cent. p. 175,

t. 90/1678.

460

[The lectotype is nominated on the evidence of Prain (Kew Bull. 1908: 353) that

the two specimens mentioned still exist and are both typical of the species. The first

mentioned by Linnaeus is here chosen.]

Chironia linoides Linn, subsp. nana Verdoorn, nom. nov. Type: Cape Flats, Ecklon

642 (SAM, holo.!).

C. gracilis Salisbury ex Prain in Kew Bull. 1908: 293 (1908), non Michx. (1803).

Syntypes: many specimens including Ecklon 642 from the Cape Flats.

C. linoides subsp. macrocalyx {Prain) Verdoorn, stat. nov., comb. nov. Lectotype:

Cape Flats, Ecklon 644 (SAM, iso.!).

C. gracilis var. macrocalyx Prain in Kew Bull., 1908: 294 (1908). Syntypes: several

specimens including Ecklon 644 from the Cape Flats.

C. linoides Linn, subsp. emarginata (Jarosz) Verdoorn, stat. nov.

C. emarginata Jarosz, PI. Nov. Cap.: 11 (1821). Type: Peninsula, Berg s.n.

From the synonomy in the Flora Capensis (Vol. 4, 2: 1101-2, 1909) it can be seen

that C. emarginata Jarosz and C. gracilis Salisb. ex Prain have both at some time or

another been classified as C. linoides or a variety or form of that species or one of its

synonyms. Hill and Prain separated these closely related specimens into four categories,

C. linoides, C. emarginata, C. gracilis and C. gracilis var. macrocalyx. After examining

the material now in the South African herbaria it was found that a fair percentage of

this large number of specimens can, on sight, be sorted into one or other of these taxa.

But there remains a significant assortment of intermediates (some may be hybrids)

that do not fit exactly into any one of these. This seems to indicate that, to accommodate

these intermediates, a far more workable treatment would be to look upon the whole

group as belonging to one complex species, C. linoides, comprising four subspecies

and their intermediates. The necessary combinations are made above.

The lectotype nominated for subsp. macrocalyx was selected because a duplicate

of one of the syntypes of the species was seen in a South African herbarium and was

found to answer very well to the original description.

Chironia palustris Complex

C. palustris Burch., aggregate species.

C. palustris Burch, subsp. palustris. Type: Griquatown, Burchell 1925.

C. palustris Burch. Trav. 2: 226 (1824); Hill and Prain in F.C. 4, 1: 1106 (1909).

Type: Griquatown, Burchell 1925. — \ar. foliata (Griseb.) Prain in 1.1. (1909). —

var. radicata (E. Mey.) Schoch in Bot. Centralbl. Beih. 14: 234 (1903). Plocandra

albens E. Mey. Comm. 182 (1837). Type: Between Kachu and Zandplaat, Drege. —

var. /Wico/a E. Mey. l.c. (1837). Type: Mooyplaats, Drpge. P. palustris {^urch.)

Griseb. in D.C. Prod. 9: 43 (1845). — var. foliata Griseb. in l.c. (1845). Type:

Caffraria, Drege.

C. palustris subsp. rosacea (Gilg) Verdoorn. Type; Cape, Pondoland, Bachmann 1038.

C. rosacea Gilg in Bot. Jahrb. 26: 104 (1899); Schoch in Bot. Centralb. Beih. 14:

229 (1903) partly as to Natal specimen; Hill and Prain in F.C. 4, 1 : 1105 (1909).

C. maxima Schoch in Bull. Herb. Boiss. 2nd ser. 2: 1014 (1902). Type: Tugela Riv.

Wood (Z).

C. palustris subsp. transvaalensis (Gilg) Verdoorn. Type: Lydenburg, Wilm 974.

C. transvaalensis Gilg in Bot. Jahrb. 26: 106 (1899); Schoch in Bot. Centralbl. Beih. 14:

227 (1903).

The specimens classified under C. palustris Burch., C. rosacea Gilg and C. trans-

vaalensis Gilg, in the Flora Capensis are obviously closely related. The inflorescences

461

and flowers, especially, are very similar, the flowers varying only in size. After examining

a large number of specimens in South African herbaria and investigating the living

plants in the Transvaal the decision was made to reduce these taxa to subspecific rank.

Reasoning on the following lines led to this decision.

Taking the species C. palustris and C. transvaalensis first, the main distinguishing

features of the most southerly unit, C. palustris, are that it is usually smaller in stature,

the basal leaves are persistent and the cauline usually much reduced while the taller

Transvaal species has well developed cauline leaves and basal leaves which early

disappear. In addition the anthers are strongly twisted in C. palustris and only slightly

so in C. transvaalensis. This latter difference is usually a good auxiliary distinction

but being a matter of degree only, there are cases where it is not of any help.

It was noted from herbarium material that the areas of distribution of these two

“ species ” overlap in the Transvaal. A spot near Kempton Park was visited to examine

the living plants. In the turfy ground along a water course the tall C. transvaalensis

was found with erect pairs of glaucous, linear-lanceolate cauline leaves and, at this

stage, bearing no basal leaves. Nearby in a hardened patch of turf grew the smaller

plant with a rosette of basal leaves pressed fiat on the hardened ground and on the stem

were remote pairs of much reduced leaves. The stamens were a degree more twisted

than in the taller plants. The possibility of the hard condition of the ground being

responsible for the differences was considered. But in the main area of distribution

of the plants with basal leaves, that is the eastern Cape, no examples of plants without

basal leaves have been found although the conditions of soil they seem to require

probably exist in those regions too. Also in the central areas, north of Kempton Park,

the plants are mostly without basal leaves and with well developed cauline leaves

although hard patches of turf, in all probability, abound. So there are two sets of

specimens separable on some vegetative features with separate centres of distribution.

This seems to indicate that taxonomically subspecific status would be more appropriate

for these two “ species ”. From the nomenclatural angle too this status would be an

advantage for in many cases where the lower portion of the plant was not collected it

is difficult to decide whether the cauline leaves are small “ well developed leaves ” or

rather large “ reduced leaves ”. This treatment would also provide a name for some

specimens from the Victoria Falls which have the leaf characteristics of subsp. palustris

but the anther-characters of subsp. transvaalensis.

The third “ species ” in this group for similar reasons is reduced to the same rank.

C. rosacea Gilg agrees in habit with C. transvaalensis but differs in having larger flowers

and broader leaves. These features are merely comparative but again the distribution

is on the whole distinct, C. rosacea occuring in Natal and C. transvaalensis in the central

regions. In Swaziland both are found and here intermediate specimens occur. In

one case one portion of the same gathering has been identified as C. transvaalensis

and another as C. rosacea. But on the whole the specimens are separable into the two

categories.

From C. palustris C. rosacea differs in the well developed broad cauline leaves,

the disappearance of the basal leaves and the larger flowers. It also has a separate

centre of distribution and so bears a similar relationship to C. palustris as does C.

transvaalensis. The necessary combinations for the subspecies are made above.

C H 1 RO N I A PURPURASCENS COMELEX

C. purpurascens {E. Mey.) Benth. and Hook. /., aggregate species.

C. purpurascens (£. Mey.) Benth. and Hook. /., subsp. purpurascens. Type; Natal,

near Umkomaas, Drege 4923.

C . purpurascens (E. Mey.) Benth. and Hook, f., Gen. PI. 2: 805 (1876); Hill and Prain

in F.C. 4, 1 : 1108 (1909). — var. (Gilg) Prain in Kew Bull. 1908 : 350 (1908).

462

— var. impedita Prain Ic. Type: Krook 2028. C. bachmannii Gilg in Bot. Jahrb.

26: 103(1898). Type: Pondoland, 1037. C. tysonii G'\\g\c\ 104(1898).

Type: near Clydesdale, Tyson 1290. F/ocandra pwpurascens E. Mey! Comm. 182

(1895).

C. purpurascens (£. Mey.) Benth. and Hook f. subsp. humilis {Gilg) Verdoorn stat. nov.

Type: Pretoria, Aapies Riv. Zeyher 1193.

C. humilis Gilg in Bot. Jahrb. 26: 105 (1899); Baker and Brown in F.T.A. 4, 1 : 555

(1904); Hill and Prain in F.C. 4, 1: 1107 (1909). — var. wilmsii (Gilg) Prain in

Kew Bull. 1908: 350 (1908); Hill and Prain in F.C. 4, 1: 1107 (1909).— var.

zuluensis Prain in Kew Bull. 1910: 55 (1910). Syntypes: Ginginhlovo Wylie in

Herb. Wood 11, 355; without precise locality, Mrs. McKenzie s.n. C. wilmsii

Gilg in Bot. Jahrb. 26; 105 (1899). Type: Bronkhorstspruit, Wilms 973.

A study of the species of Chironia shows that C. purpurascens (E. Mey.) Benth. and

Hook. f. and C. humilis Gilg are more closely related to each other than to any of the

rest of the species. In the Flora Capensis this is borne out by Hill and Prain who put

these two “ species ” in a section by themselves. To the distinguishing characters

which they use for the section, namely the deeply cut calyx, short corolla-tube, distinctly

spirally twisted anthers and pointed fruits, may be added the following features shared

by the two “species”, the deep colour of the flowers (usually magenta pink); the

acicular apical half of the calyx-lobes and bracts; and the central flowers of the cymes

usually having a much shorter pedicel than those of the lateral. The features which

separate these units are: the smaller stature of C. humilis, about 30 cm as against over

50 cm tall; the pedicel of its central flower usually being under 6 mm long as against

being up to 25 mm long; and flowers frequently with a pair of bracts near the base

of the calyx while in the taller plants they are only rarely present. These differences

are principally in the size of the plant itself or certain parts of it; but the main area

of distribution of the two groups is distinct; the taller plants come from the extreme

eastern Cape, Natal, and Swaziland while the smaller plants are recorded from Zululand

(northern Natal), the Transvaal and northwards to Southern Rhodesia.

These findings indicate that the relationship which the two groups bear to each

other is rather subspecific than specific and C. humilis is therefore here reduced to a

subspecies under C. purpurascens which is the older name.

Enicostemma

Enicostema hyssopifolium {Willd.) Verdoorn comb. nov. Type: specimen 328.30

(Fabricius) in Linnean Herbarium (L, lecto.; PRE, photo.).

E.xacum hyssopifolium Willd. Spec. 1; 640 (1798). Name for Gentiana verticillata

(non Linn.) Linn. f.

Gentiana verticillata (non Linn. 1753) Linn. f. Supple.; 174 (1781) applied when naming

a Fabricius specimen from India; Vahl, Symb. 3: 46 (1794) partly as to Indian

and Arabian references. G. verticillaris (sic) Linn. var. Retz. Obs. Bot. 2; 15

(1781) based on a Konig specimen.

Hippion hyssopifolium (Willd.) Spreng. Syst. 1: 589 (1824).

Enicostema littorale Blume, Bijdr. 848 (1826). Type: Blume, Java.

E. verticillare (Retz.) Baill. Hist. PI. 10: 131 (1891) partly.

E. verticillatum Engl, in PfI. Cat. Afr. C: 313 (1895) as to specimens cited.

Lepinema verticilata Raf., FI. Tellur, 3: 26 (1837) partly as to Vahl reference.

Adenema hyssopifolium (Willd.) G. Don. Gen. Syst. 4: 201 (1837).

Slevogtia verticillata D. Don. in Trans. Linn. Soc. 17: 532 (1837) nom illeg. based on

Gentiana verticillata Linn, fil.; S. orientalis Griseb. in DC. Prod. 9: 65 (1845),

nom illeg.

463

Hippionum verticiUatum O. Ktze. Rev. Gen. PI. 428 (1891) for the greater part, excluding

the American reference.

Lectotype: I nominate the Fabricius specimen No. 328.30 in the Linnean Herbarium

as the lectotype of Enicostema hyssopifolium (Willd.) Verdoorn. Willdenow’s

first reference under Exacum hyssopifolium is “ Linn. f. Suppl. 174 ”, where the

Fabricius specimen from India is described.

The first record of this species was that of Linnaeus filius in the Supplementum

1781 when describing a Fabricius specimen from India. It was classified by him as

Gentiana verticillata Linn., an American species. Willdenow in 1798 recognized this

as an error and not only distinguished between the American and Indian specimens

but transferred them both to the genus Exacum. The American plants he placed under

Exacum verticiUatum (L.) Willd. and gave the name of Exacum hyssopifolium to the

specimen recorded by the younger Linnaeus in Supplementum. No mention is made

by Linnaeus or Willdenow of the unusual double hooded gland at the base of the

filaments. This has proved to be a diagnostic character and in 1826 Blume described

the genus Enicostema meaning “ singular stamen ” for a specimen from Java. Indepen-

dently three other genera were described to take the same taxon, but Enicostema is

the earliest legitimate generic name. This name is therefore combined above with the

earliest legitimate specific epithet “ hyssopifolium ” of Willdenow.

I. C. Verdoorn

Sebaea

Sebaea fourcadei W. Marais sp. nov. S. ramosissimae affinis sed inflorescentia compac-

tiore, calycis segmentis stramineis fragilis, corollae lobis quam tubum longioribus

differt.

Herba annua, gracilis, erecta ad 25 cm alta. Caulis simplex vel basi ramosus.

Eolia ovato-orbiculata vel superne ovata, ad 1 cm longa, 1 cm lata, obtusa vel subacuta,

paribus paucibus, internodiis longis. Inflorescentia corymbosa, pauci- vel pluriflora.

Calyx 5-fidus; segmenta 4-75-5 mm longa, 1-1-5 mm lata, lanceolata vel elliptico-

lanceolata, acuminata mucronata, carinata, marginibus membranaceo-opacis, fragilibus.

Cora//flfiava; tubus 3-4 mm longus; lobi 2-75-3-5 mm longi, 1-1 -25 mm lati, anguste-

elliptici vel anguste ovato-elliptici, oblanceolati, obtusi, nonnumquam leviter cucullati.

Eilamenta 0-25 mm longa in faucibus inserta; anthera 0-875-1 -25 mm longa, glandulis

3 parvis flavis globosis breviter stipitatis ornata. Stylus 2-75-3-75 mm longus, infra

vel plus minus medio callo instructus stigmate capitato 2-lobato.

Cape. — Knysna: Belvedere, Duthie 1175 (K). Humansdorp: Geelhoutboom ,

Eourcade 4880 (K, holo!).

Slender, erect, annual herbs, simple or branched near the base, up to 25 cm high.

Leaves up to 1 cm long and 1 cm broad, ovate-orbicular or the upper ones ovate, obtuse

or subacute, in relatively few pairs, the lower ones, soon dying off; internodes long.

Inflorescence corymbose, few to several flowered. Calyx of 5 segments each 4-75-5 mm

long, 1-1-5 mm broad, lanceolate or elliptic-lanceolate, acuminate-mucronate, their

sides membranous-opaque, brittle and strawlike when dry, keeled. Corolla yellow;

tube 3-5 mm long; lobes 2-75-3-5 mm long, 1-1-25 mm broad, narrowly-elliptic

or narrowly ovate-elliptic to oblanceolate, rounded, sometimes slightly cucullate.

Eilaments 0-25 mm long, inserted in the corolla-sinuses. Anthers 0 -875-1 -25 mm long,

each with three small round, shortly stipitate, pale yellow glands. Style and stigma

2 - 75-3 - 75 mm long, with a stigmatic swelling below or near the middle ; stigma capitate,

2-lobed.

S. capitata Cham, et Schlechtdl. var. sclerosepala {Schinz) Marais, stat. nov.

Basionym; S. sclerosepala Gilg ex Schinz in Mitt. Geogr. Ges. Liibeck 17: 23

(1903).

464

S. micrantha Cham, and Schlectdl. var. intermedia {Cham, and Schlechtdl.) Marais,

comb. nov.

Basionym: S. cordata var. intermedia Cham, and Schlechtdl. in Linnaea 1: 191

(1826).

S. Zeyheri Schinz subsp. acutiloba (Schinz) Marais, stat nov.

Bassionym; S’, acutiloba Schinz in Bull. Herb. Bois. 3; 412 (1895).

S. Zeyheri Schinz subsp. cleistantha {R. A. Dyer) Marais, stat. nov.

Basionym: S. cleistantha R. A. Dyer in K.B. 1933: 461.

S. pentandra E. Mey var. burchellii {Gilg) Marais, stat. nov.

Basionym: S. burchellii Gilg in Bot. Jahrb. 26: 89 (1898).

S. sedoides Gilg var. confertiflora {Schinz) Marais, stat. nov.

Basionym: S. confertiflora Schinz in Mitt. Geogr. Ges. Liibeck 17: 51 (1903).

S. sedoides Gilg var. schoenlandii {Schinz) Marais, stat. nov.

Basionym: S. schoenlandii Schinz in Bull. Herb. Boiss ser 2, 6: 741 (1906).

Lectotypes in Sebaea

Sebaea laxa N.E. Br. in Kew Bull. 1901: 128 (1901).

Syntypes: Burchell 7089 and Galpin 4337. Lectotype: Galpin 4337 (K).

On the sheet of Galpin 4337 which N. E. Brown examined there are some dissected

flowers indicating that this specimen was studied in more detail than Burchell 7089.

Sebaea compacta Hill.

Lectotype: Barber 21 (K). Selected because it is the best of the syntypes as far

as state of preservation and amount of material is concerned.

S. Zeyheri Schinz subsp. acutiloba {Schinz) Marais.

Lectotype: Schlechter 3045 (K).

Schinz cites two specimens with his description of S. acutiloba in Bull. Herb.

Boiss 3: 415 (1895), namely, Schlechter 3045 and Wood 4950, both from Clairmont,

Natal. As Schlechter’s specimens are usually represented in several herbaria and

there is only a fragment of Wood 4950 in the Kew Herbarium, I select the Kew sheet

of Schlechter 3045 as lectotype.

Belmontia and Exochaenium synonymous with Sebaea

In the Llora Capensis (4, 2: 1057, 1909) Hill and Prain follow Schinz and sink

Belmontia E. Mey. under Sebaea R. Br. while maintaining Exochaenium Griseb. as

a distinct genus. Phillips in his Genera of South African Flowering Plants (1951)

follows Bentham and Hooker who treat Exochaenium as being synonymous with

Belmontia, which genus they maintain as distinct from Sebaea.

In my work on the “ Flora of Southern Africa ”, I have come to the conclusion

that both Belmontia and Exochaenium are inseparable from Sebaea and they will be

treated as synonyms of the latter genus in the “ Flora ”. Some of the reasons that

led to this decision are as follows : The main distinguishing features between Belmontia

and Exochaenium collectively on the one hand and Sebaea on the other, are, the presence

in the former group of disc-glands between the calyx and the corolla, the style being

without a two glandular swelling and the stamens included in the corolla-tube instead

of being inserted in, or shortly below, the sinuses of the lobes and exserted from the

tube. In several instances these characteristics have proved to be diagnostically

unreliable. For instance, in the species Sebaea thomasii, disc-glands are sometimes

present and sometimes not. Then again in S. micrantha and S. exacoides, while the

stamens are included in the corolla-tube, a characteristic of Exochaenium and Belmontia,

glandular swellings are present on the styles, features of Sebaea.

W. Marais

465

GESNERIACEAE

Streptocarpus rimicola Story, sp. nov., S. pumilo Burtt affinis, sed stylo breviore et

floribus albis diflferens.

Herba acaulis unifoliata, raro folio altero minimo. Folium plerumque c. 4 x 2 cm

(1-5 X 1-6-5 X 2-5), late ovatum vel angustum et acuminatum, leviter serratum,

chartaceum vel coriaceum, utrinque villis eglanduliferis, venis supra impressis subtus

prominentibus, hypocotylo plerumque minutissimo, raro ad 1 cm longo. Inflorescenlia

villis glanduliferis et eglanduliferis, ad 7 cm alta, plerumque 3-4. PeduncuH numero

ad 4 plerumque solitarii, quisque ad 9 plerumque 2-3 flores ferens. Bracteae minutae.

Pedicelli 1 ad 2-5 cm longi interdum ramosi. Calyx ad basin partitus, lobis 2x0-5

mm. Corolla alba 1 cm longa; tubus cylindratus basi saccatus, leviter supinus, 8 mm

longus 2-3 mm diam., intus glaber (crista villorum eglanduliferorum pellucidorum vel

purpureo-tinctorum super antheris excepta); limbus vix obliquus, lobis orbicularibus

c. 3 mm longis et latis. Orbis annularis leviter lobatus viridis. Androecium staminibus

corollae basi ab 1 mm insertis; filamenta glabra pellucida 4 mm longa; staminodia 3,

pariter basi ab 1 mm inserta, 2 c. 1 mm longa, 1 brevius. Gynoecium ovario 2-2 - 5 mm

longo; stylus 3-3-5 mm longus, apicem versus glaber et subito deflexus; stigma

capitatum, paulum ultra antheras eminens. Capsula robustior, 8-10 mm longa, ad

2-5 mm lata; stylus persistens 3-3-5 mm longus.

Transvaal. — Warmbaths district: Farm Groothoek 1246, western extremity of

Waterberg mountains, steep south-facing mountain side, under overhanging ledges,

6,000 ft., Codd 3974; Codd 6486 (PRE, holo.); Story 6517.

A stemless unifoliate herb, rarely with a second much smaller leaf. Leaf dark

green above, paler and sometimes reddish below, usually about 4x2 cm (in flowering

specimens ranging from 1-5 x 1 to 6-5 x 2-5), sometimes broadly ovate, sometimes

narrowly oblong and acuminate, often withered at the tip, slightly serrate, chartaceous

to coriaceous, with non-glandular hairs on both surfaces, veins sunken above, prominent

below, hypocotyl usually not noticeable, rarely up to 1 cm long. Inflorescence with

glandular and non-glandular hairs, up to 7 cm high but usually 3 to 4. Peduncles up

to four but usually solitary, each bearing up to nine but usually two or three flowers.

Bracts minute. Pedicels 1-2-5 cm long, sometimes branched. Calyx lobed to the base,

lobes 2x0-5 mm. Corolla white, 1 cm long; tube saccate at base, slightly curved

upwards, 8 mm long, 2-3 mm diam., glabrous within except for a patch of light purple

or colourless 2-celled smooth-walled; eglandular hairs above the anthers; limb scarcely

oblique, lobes rounded, about 3 mm long and as wide. Disc annular, fleshy, slightly

lobed, light green. Androecium with stamens inserted 1 mm from the base of the corolla-

tube; filaments glabrous, colourless, 4 mm long; anthers colourless; staminodes 3,

inserted 1 mm from the base of the tube, two of them about 1 mm long, the third smaller.

Gynoecium in freshly-opened flower with ovary 2-2-5 mm long; style 3-3-5 mm long,

glabrous near the tip which is sharply bent down; stigma capitate, projecting a little

beyond the anthers. Capsule erect, fairly stout, 8-10 mm long, up to 2-5 mm wide;

style persistent, 3-3 - 5 mm long.

The highest peak in the Waterberg (6,841 feet) is at the western edge of the range,

which at this point falls abruptly away into the sandy flats of the north-western Transvaal

and the Bechuanaland Protectorate. According to available records, Streptocarpus

rimicola occurs only on this peak, and even here the plants are apparently restricted

to the southern aspect and to the sheer cliffs which make up the last three hundred

feet or so. There is an annual rainfall of about 30 inches which falls mainly in summer,

but the plants depend less on direct rainfall than on seepage and mist, for they grow

mostly in deep shelter under rock ledges. They have so far not been found near the

bottom of the cliffs, although there are as many crevices and ledges there and these

provide apparently ideal conditions for their growth. Presumably there is less mist

466

near the bottom and therefore not enough moisture for them. It was interesting to

find that the only other member of the genus recorded from this locality also has a

very small altitudinal range. Below the cliffs are steep grassy slopes covered with

boulders of all sizes, and a little way above the trees which clothe the lower slopes of

the mountain is a narrow zone of Streptocarpus vande/euri growing on the shady side

of the boulders wherever there is shelter and a little soil. The plants occupy about

200 yards of mountain slope, roughly 300 feet of altitude.

Fig. 1. — Streptocarpus rimicola; 1, front view of flower; 2, side view, both x 8.

Streptocarpus rimicola is often massed together in small irregular patches, but as

it blooms sporadically it does not make a show. By about the middle of April the

flowering season is over. The flower is tilted upwards and has the anthers only lightly

connivent and often separate by the time it is fully open.

The plant is about the same size as Streptocarpus pumilus, and closely related.

Flower colour is a rather unreliable character in the genus but the specific rank of

Streptocarpus rimicola appears justified by the other differences which are set out in

tabular form at the end of this paper. The hypocotyl is as a rule imperceptible when

the plant is growing and the bracketed figure in the table gives the length of the only

one of that size seen among the hundreds of plants which were examined. Our only

457

specimen of Streptocarpus pumilus indicates also that it has a more slender capsule

than Streptocarpus rimicola and a narrower corolla-tube. This last point was confirmed

by Mr. B. L. Burtt, the author of Streptocarpus pumi/us, who kindly reported on a

specimen sent to him.

Streptocarpus pumilus

1. Flower mauve.

2. Hypocotyl up to 3 cm long.

3. Stamens inserted 6 mm from base

of corolla tube.

4. Ovary 2 mm long, style 8 mm long.

Streptocarpus rimicola

1. Flower white.

2. Hypocotyl less than 0-5 cm long

(rarely up to 1 cm).

3. Stamens inserted I mm from base

of corolla tube.

4. Ovary 2-2-5 mm long, style 3-3-5

mm long.

R. Story

GRAMINEAE

Digitariella de Winter, genus novum, afhnis Digitariae, sed ita differt: spicula basi in

pseudo-callum producta, gluma inferior ab superiore internodio distincto separata,

gluma superior et lemma inferius in apicem attenuatum subaristatum sensim

producta.

Annua, culmis ramosis ascendentibus, nodis villosis, vaginis nodis superantibus,

ligula brevi truncata membranacea, laminis linearibus vel lineari-lanceolatis glabris

margine crenulatis. Inflorescentia 2-3 racemis spiciformibus digitate confertibus.

Spiculae binae vel solitariae. Gluma inferior parva, squamiformis, enervis, longe infra

glumam superioram inserto internodio distincto separata. Gluma superior anguste

lanceolata 3 mm longa et 0-75 mm lata, sensim in apicem subaristatum 2-5 mm longum

producta, 7-nervis cum linea pilorum inter nervos uterque latere costae median! et

ad margines. Flos inferior ad lemma reductus. Lemma inferius 3 mm longum

0-75 mm latum anguste lanceolatum 5-nerve, hand aristatum. Flos superior fertilis,

lemmate coriaceo lanccolato-naviculari 2-5 mm longo, breve mucronato, palea coriacea

navicular! 2-nervi, lodiculis truncatis, cuneatis enervibus. Stamina 3, antheris basi

obtuse sagittatis oblongo-ellipticis 0-5-0 -6 mm longis, ovario glabro stylis distinctis

stigmatibus plumosis. Caryopsis libera ambitu oblanceolato-elliptica; embryo

1 /3-3/5 longitudinis caryoptidis; hilum subbasale, ellipticum.

Species unica, Africae tropicae austro-orientalis incola.

Digitariella remotigluma de Winter, sp. nov.

Annual with ascending branched culms, often rooting at the nodes. Nodes sparsely

villous. Leafsheaths exceeding the nodes, striate, slightly keeled, glabrous. Ligule a

shoT membrane. Collar glabrous. Leaf blade linear to linear-lanceolate, glabrous;

margin distinctly undulate-crenulate. Inflorescence consisting of 2-3 digitately arranged

spikelike racemes; racemes 2-5-6 cm long. Spikelets usually in pairs, one shortly

pedicelled, the other pedicel longer, very occasionally single, abaxial in relation to the

axis. Lower glume a small nerveless, membranous scale, distant from the upper, clasping

the calluslike prolongation of the base of the spikelet. Upper glume produced down-

wards into the callus, connate with it, narrowly lanceolate, gradually produced into

an awn-like apex 2-2-5 mm long, 7-nerved, body of the glume about 3 mm long and

0-75 mm wide, nerves on each side of the midnerve approximate, with a line of hairs

between them and along the margin; margins somewhat inflexed. Lower floret reduced

to the lemma. Lower lemma narrowly lanceolate, 3 mm long and 0-75 mm wide,

acuminate, not awned, 5-nerved, hairy between the lateral nerves, and on the margins.

468

Upper floret fertile. Upper lemma coriaceous, white, lanceolate in outline, boat-shaped,

2-5 mm long and 1 mm wide when spread out, acuminate and shortly mucronate.

Palea coriaceous, lanceolate in outline, boat-shaped with inflexed margins, 2-nerved.

Lodicules truncate, cuneate, nerveless. Stamens 3; anthers bluntly saggitate at the

base, oblong-elliptic in outline, about 0-5-0 -6 mm long. Ovary glabrous; styles

distinct; stigmas plumose, purple. Caryopsis tightly enclosed by the lemma and

palea but free, oblanceolate-elliptic, whitish, opaque; embryo about 1/3-3/5 the length

of the grain; hilum sub-basal, narrowly elliptic, brownish.

South West Africa. — Okavango Native Territory: Junction of Mpungu

Omuramba and Okavango River between Tondoro and Lupala, soft annual with yellow-

green inflorescences on swampy edges of lagoon, 17-12-1955, B. de Winter No. 3978

(PRE, holo; K,M,B, BM, SRGH, Windhoek, iso.); River flats below visitors camp

at Runtu, peaty black soil, annual grass growing in moist places on floodplain of river,

22-12-1955, B. de Winter No. 4037. Eastern Caprivi: Mpilila Island, frequent on

banks of Zambesi River, Killick and Leistner 3370.

This interesting new genus belongs in the somewhat loosely defined group of the

Paniceae sometimes referred to as the Digitariastrae. The Digitariastrae are charac-

terised as follows: fertile valves thinly chartaceous to cartilaginous with margins

hyaline to subhyaline, inflexed but not inrolled, and the position of the spikelet abaxial

in relation to the rhachis (upper glume and upper lemma with their surfaces adjacent

to the axis).

The closest affinity of Digitariella is Digitaria as implied by the name. It agrees

with Digitaria in the loosely paired spikelets, the digitate inflorescence and the approxi-

mate lateral nerves of the lower lemma with a line of hair between them. The main

differences are the elongated calluslike prolongation of the base of the spikelet and the

much attenuated awnlike apices of the lower glume and lemma.

It also agrees in general in the structure of the spikelets with genera such as

Oryzidium and Louisiella. These genera however differ in the adaxial orientation of

the spikelet, that is the lower glume and lower lemma are turned towards the axis.

Though allied to these genera the affinity is obviously less close than to the members

of the Digitariastrae.

The basal prolongation of the spikelet of Digitariella is not a true callus but is

formed by the fusion of the base of the upper glume and rhachilla. This calluslike

prolongation strongly reminds one of that found in the South American genus Chaetium

from which Digitariella differs by the orientation of the spikelet and by the minute

remote lower glume. In Chaetium the callus is formed by the fusion of the rhachilla

and the bases of both the glumes. The glumes moreover are large and equal or exceed

the spikelets in length. Chaetium cannot therefore be regarded as very closely allied

to Digitariella.

Eragrostis aristata de Winter, sp. nov., aff. E. crassinervi Hack, sed lemmatibus 3-lobatis,

nervo mediano in aristam brevem terminanti, nervis lateralibus in mucrones

breves excurrentibus, omnibus nervibus et carinis palearum spinis brevibus hyalinis

scaberrimis, gluma superiore 3-nervata.

Annua 30-75 cm alta. Vaginae internodiis breviores, carinatae, carinis cum glandulis

minutis demersis vel crateriformibus. Ligula pilis longis hyalinis ciliata. Lamina infra

galbra, nervis glandulosis scabridis exceptis. Panicula ad 30 cm longa 15 cm lata,

ramis primariis patentibus, spiculis in ramis secundariis, subdense aggregatis, ramis

rhachidibusque scabridis glandulosis. Spiculae breve pedicellatae vel subsessiles ca.

6 mm longae 4 mm latae, 5-12-florae, rhachilla persistent!, glumis inequalibus, inferiori-

bus 1-nervatis, superioribus 3-nervatis, frequenter plusminusve 3-lobatis, ad nervos

glandulosis, lemmatibus 3-lobatis, nervo mediano in aristam ad 1-5 mm longam

excurrenti, nervis lateralibus in mucrones terminantibus, nervis scabridis glandulosis.

469

paleis carinis scabridis bicarinatis. Stamina 3, antheris oblongis 0-5-0 -6 mm longis.

Ovarium glabrum. Lodiculae truncatae glabrae. Caryopsis oblonga lateraliter leviter

compressa.

South West Africa. — Omaruru: Namib, Brandberg, annual, along river bed,

abundant, Schweickerdt 2252 (PRE, holo. K, iso); Brandberg, Hungarob ecke, IViss

1504; Brandberg valley, Liebenberg 5019.

Annual, 30-75 cm high, with soft erect or geniculate 2-4-nodes culms. Sheaths

pallid, fairly tight, usually shorter than the internodes, soft, glabrous, distinctly ribbed

with small sunken or crateriform glands on the ribs especially below the collar and

above the nodes. Ligule a fringe of long hyaline bristly hairs; auricles glabrous.

Leafblade flaccid, glabrous below, with sunken glands on the nerves, scabrid on the

nerves on upper surface, usually flat; margin scabrid and with small crateriform glands.

Panicle exserted when mature, up to 30 cm long and 15 cm wide, primary branches

spreading, not whorled, single or opposite, spikelets clustered on the short secondary

branchlets, axis and branches ribbed with small crateriform glands on the ribs and

scabrid with short hyaline spines. Spikelets shortly pedicelled or subsessile on the

secondary branchlets, about 6 mm long and 4 mm wide, green to deep purple in colour,

5-12-flowered, breaking up from below, paleas usually persistent on the rhachis. Florets

hermaphrodite. Glumes unequal to subequal, lower one-nerved, about 1 mm long,

upper 3-nerved, often somewhat tri-lobed at apex, 1-1 -75 mm long, usually with raised

glands on or near the nerves, nerves usually scabrid. Lemmas tri-lobed, 2-3 • 5 mm long,

ovate, 3-nerved, midnerve excurrent into an awn up to 1-5 mm long, lateral nerves

very shortly awned to mucronate, body of lemma 1 • 5-2 mm long, strongly keeled

along midrib and lateral nerves, scabrid on the nerves and with small raised glands.

Palea strongly curved, bi-keeled, keels scabrid with hyaline spines. Stamens 3 ; anthers

oblong, 0- 5-0-6 mm long, yellow. Ovary glabrous; styles free; stigma plumose.

Lodicules somewhat fleshy, truncate, glabrous. Caryopsis oblong, somewhat laterally

compressed, pale to deep brown; embryo about 1/2-3/5 the length of the caryopsis;

hilum punctiform, basal.

This species has so far only been collected in the vicinity of the Brandberg in South

West Africa. It can be expected to occur in the marginal Namib desert right up to

the Kunene and possibly a little way beyond into Angola, this area offering similar

ecological conditions as those encountered at the Brandberg.

At the first glance E. aristata can hardly be recognised as a species of Eragrostis,

especially when the specimens are not fully mature. It is so obviously related with

E. crassinervis Hack, however that including it in the same genus as the latter is the

only logical way to treat it. E. crassinervis has the nerves of the lemmas excurrent into

very short mucros, and the awns of E. aristata can only be regarded as a further develop-

ment in this direction. Except for the awned lemmas and 3-nerved upper glume, all

the other characteristics agree with those of typical Eragrostis species. Three-nerved

glumes occur also in E. bergiana, E. truncata and several other species.

Eragrostis glandulosipedata de Winter, sp. nov., aff. E. porosae Nees, sedspiculis majori-

bus, lemmatibus latioribus et longioribus, pedicello glanduloso differt, necnon

afiinis E. omahekensi de W. sed ab ea lemmatibus brevioribus apice obtusis vel

subobtusis recedit.

Annua vel subperennis 4-100 cm alta. Culrni erecti geniculati vel adscendentes,

simplices vel ramosi, 2-4 nodosi. Vaginae internodiis breviorae, laxae, glabrae vel

sparse pilosae, costis glandulosis. Ligulae dense ciliatae. Laminae flaccidae plerumque

planae, glabrae vel sparse pilosae. Panicula angusta, plus minusve contracta vel laxa,

ramis basi verticillatis apicem versus solitariis vel binis vel ternis plerumque prope

basim glandulis immersis. Spiculae 6-10-florae, 4-6 mm longae 1-5-2 mm latae,

glumis inaequalibus, 1-nervatis acutis, carinis scabridis. Stamina 3, antheris 7-10 mm

longis. Caryopsis oblonga vel late oblonga, subtranslucens.

470

Transvaal. — Potgietersrus : Roedtan, Barenbrug s.n. ; Singapore Cash Store,

Grass Valley, de Winter 2332 (holotype, PRE, isotype, K, BM, M, B, SRGH, EA,

BOL); Roedtan, de Winter 2224; Roedtan, de Winter 2290.

South West Africa. — Grootfontein: Awagobib, Vo/k A. 11; Asis, Vo/k A. 60

and 656; Hairabib, Vo/k 504; Auros, Vo/k 616 and 633 (635); Ossa, Vo/k 669; 23-5

m, E. of Otavi on rd. to Grootfontein, de Winter 2865; Rietfontein, Schweickerdt 2091 ;

Kumkauas, Kinges 2797; Auros, Schoenfelder S 563. Okahandja; Teufelsbach, de

Winter 2694; Okakuja, Grossart/t s.n. . Tsumeb: EXosha Van, de Winter 2965; Between

Tsumeb and Namutoni, de Winter 2950. Outjo: Gainatzeb, Vo/k 2670.

Kenya. — Nairobi, Bogdan A.B. 3119 (K).

Annua/ 4-100 cm high, culms erect, geniculate-ascending or the lower internodes

prostrate but not rooting at the nodes, simple or branched, 2-4 noded, internodes

hollow, compressible, with a glandular ring just below the nodes, ultimate internode

long exserted when mature. S/teat/is shorter than the internodes, lax, glabrous or

hairy with bulbous-based hairs, shiny, glandular on the ribs, especially on the midrib

above the node and below the junction of the blade and sheath. Co//ar with a fringe

of stiff hyaline hair. Ligu/e a short dense fringe of hair. Leafb/ade linear, flat or

somewhat rolled, distinctly veined below, striate above, scaberulous on the striae,

glabrous or sparsely hairy, glandular on the midrib or without glands, tapering to a

fine point. Panic/e rather dense or lax, narrowly ovate-oblong to narrowly-oblong,

branches whorled at the base, three or two together or single towards the apex, not

spreading and somewhat appressed to the rhachis; rhachis glabrous, often glandular

below the lower whorls of branches. Spike/ets pedicelled, 6-10-flowered, 4-5 mm long

and 1 -5-2 mm broad, blackish or greenish grey to ashgrey in colour; rhachilla tardily

disarticulating between the florets. F/orets awnless, hermaphrodite. G/umes 1 -nerved,

unequal, lanceolate in side view, membranous, acute, scabrid on the keels; lower

about 1 mm long, upper about 1-5 mm long. Lemmas 1-5-1 -7 mm long, broadly

elliptic to broadly oblong-ovate, scabridulous especially near the apex, 3-nerved, nerves

rather prominent, apex obtuse to subacute. Pa/ea as long as lemmas, membranous,

hyaline, bi-keeled, scaberulous. Lodicu/es truncate, cuneate, small. Stamens 3;

anthers 0-7-1 mm long, cultrate, yellow. Ovary glabrous; styles free; stigmas 2,

plumose. Caryopsis oblong to broadly oblong, 0-6-0 -9 mm long, pallid to brownish,

semitranslucent; embryo half the length of the grain, brownish green; hilum puncti-

form, basal.

E. g/andu/osipedata has an interesting distribution. It is common in the Groot-

fontein, Tsumeb and Outjo districts of South West Africa and has also been recorded

from the sandy parts of the Springbuck Flats near Roedtan and Grass Valley in the

Transvaal. One specimen which undoubtedly also belongs here, was collected by

Bogdan at Nairobi, Kenya. Although very interrupted this distribution follows a

pattern observed in the case of a number of other species, such as Tetrapogon tene//us,

Eragrostis pseudo-sc/erait/ia and others. Like in many other annual grasses there is a

great variation in the size of the individuals of this species. In good years they may

reach a height of over 1 m while the author has collected specimens barely reaching

10 cm growing on dry brackish soils at Namutoni. The spikelets however remain

characteristic and show little variation even in size.

Eragrostis kingesii de Winter, sp. nov., aff. E. procumbenti Nees, sed ab ea spiculis minori-

bus, lemmatibus brevioribus, carinis et nervis lateralibus scabridioribus recedit.

Annua, decumbens vel suberecta, parva. Cu/mi geniculati, 2-4 nodosi, ramosi,

glabri. Kog/hoc glabrae, carinis costisque glanduloso-punctatis. L/gw/o ciliata. Lamina

plana, costa media glanduloso-punctata. Panicu/a subdense contracta, ramis solitariis

el binis vel ternis e basi ramosis, omnibus glandula subbasali, pedicellis apice glandula

annulari. Spicu/ae plumbeae ca 5 mm longae 1-1-5 mm latae, 7-14-florae paleis

persistentibus, glumis et lemmatibus post maturitatem deciduis, glumis subaequalibus

471

quam lemma superius brevioribus, lemmatibus depresse cymbiformibus late ellipticis

ambitu oblongis scabridis ad nervos cum glandulis paucibus demersis, paleis quam

lemmata brevioribus. Stamina 3, antheris globosis 0-3 mm longis. Caryopsis late

oblongo-elliptica, subtranslucens.

South West Africa. — Luederitz: Farm Klein-Aus, decumbent annual around

farm houses, 12/5/49, Kinges 2236 (holotype, PRE; isotypes, K, M, BM, Windhoek);

8 m. W. of Aus on rd. to Luederitz, de Winter and Giess 6083.

Decumbent or semi-erect annual, only a few cm high. Culms geniculate, 2-4-noded,

freely branched, internodes often exposed, semi-terete to flattened, distinctly grooved,

glabrous, with an annular glandular ring just below the node. Sheaths chartaceous,

lax and slipping from the culms, glabrous, ribbed, dotted with sunken glands on the

ribs and keel, margins subhyaline, membranous. Ligule a fringe of short stiff hairs.

Collar glabrous, auricles bearded with long stiff hyaline hairs. Leafblade up to 2 cm

long and 3-0 wide, soft, more or less flat, subglabrous, dotted with sunken glands on

the midrib, secondary veins and margin. Panicle fairly densely contracted, rigid, erect,

4-5 cm long and 2-2 -5 cm wide, branches single or 2-3 together, divided from the base,

branchlets and peduncles short, all divisions with a gland below each branching point,

and an annular gland on each pedicel. Spikelets leadgrey, about 5 -0 mm long and 1-1-5

mm broad, narrowly lanceolate-cultrate in shape, 7-14-flowered, breaking up from

below upwards, paleas persistent, lemmas and glumes falling after maturity. Glumes

subequal, shorter than the subtended lemma, one-nerved, boat-shaped; lower about

three-quarters the length of the upper, lanceolate and acute in profile, keel scabrid

occasionally with one or two glands; upper obliquely ovate in profile, keel scabrid.

Lemmas shallowly boatshaped, broadly elliptic-oblong in dorsal view, 3-nerved, scabrid

and with a few sunken glands on the nerves. Pales about three-quarters the length

of the lemmas, strongly bikeeled, scabrid on the keels, persistent. Lodicules truncate,

cuneate, 0-2 mm long, fleshy. Stamens 3, anthers subglobose, small, 0-3 mm long,

yellow. Ovary glabrous, styles distinct, stigmas plumose. Caryopsis semi-translucent,

finely striate, broadly oblong-elliptic in ventral view. Embryo half the length of the

grain. Hilum punctiform, basal.

This species prefers disturbed soil and is usually found on roadsides or around

farm houses where it grows as a weed. Like other annual semi-desert species it shoots

up after rain, flowers and fruits within a few weeks and dies. It is quite an amazing

sight to see these little plants hanging by thin taproots from the side of a bone dry

sandbank and still being fresh and green.

This small annual is closely related to E. procumbens, which likewise occurs in the

dry areas of the Free State, Cape and South West Africa. It differs from the latter

as set out in the description and is a smaller plant with less dense inflorescences.

E. procumbens has as yet not been recorded from the coastal Namib and the areas of

distribution therefore do not overlap.

Eragrostls lamprospicula de Winter, sp. nov., E. membranaceae Hack, afiinis, sed perenni,

cetera spiculis minoribus, inflorescentia minus ramosa ab ea distinguenda.

Perennis, erecta. Culmi simplices 2-5-nodosi, ad 65 cm alti, infra nodos annulo

glanduloso, nodis glabris. Vaginae glabrae carinatae, glanduloso-punctatae. Ligula

breve ciliata. Laminae lineares 5-10 cm longae. Panicula laxa, sparse ramosa, ad

12 cm longa 6 cm lata, ramis gracilibus flexuosis, pedicellis 5-12 mm longis apice glandula

annulari, spiculis ovato-ellipticis ad anguste oblongis 6-15 mm longis, 6-25-floris,

glumis subcoriaceis inaequalibus, uninervatis, lemmatibus coriaceis levibus nitidis

latissime ovatis acutis 2-2-5 mm longis 1-75-2 mm latis nervis 3 inconspicuis, palea

elliptica 2 mm longa bicarinata truncata membranacea,lodiculis 2 carnosis cuneiformibus

2-3-nervatis, staminibus 3, antheris anguste oblongis 1 mm longis, ovario oblongo

stylis 2. Caryopsis lateraliter compressa, 1 -5 mm longa opaca; embryo dimidio quam

granum longiore.

472

Fig. 2. — Eragrostis omahekensis de Winter; a, habit X 1; b, spikelet X 10.

473

Transvaal. — Waterberg: Mosdene Farm near Naboomspruit; bare brack flats,

occasional, 65 cm high, de Winter 734 (PRE, holo, K, BM, B, M, BOL, SARH, NY, iso.).

Southern Rhodesia. — Salisbury: Mrs. Kimpton s.n.

Erect and usually somewhat geniculate perennial. Culms simple, flattened towards

the base, up to 65 cm high, glabrous, 2-5 usually 3-noded; internodes gradually

decreasing in length downwards, exserted from the sheaths, with a ring of glands below

the nodes; nodes glabrous usually purplish. Leaf sheaths striate, glabrous, lower

ones often tinged with purple, compressed and keeled, usually dotted with glands on

the midrib below the collar and above the nodes, glabrous, bearded at the mouth, outer

margin hairy with bulbous-based hairs. Ligule shortly ciliate. Collar inconspicuous,

sometimes purplish, glabrous. Leaf blade linear, up to 10 cm long, smooth and glabrous

on lower surface, grooved and scaberulous on the nerves on upper surface. Inflorescence

rather lax, sparsely branched, elliptic to ovate in outline, up to 12 cm long and 6 cm

wide; basal branches subtended by a glandular patch, 2-4 together but not in a true

whorl, paired and opposite or single upwards, fine and flexuous, purplish, bearing the

spikelets on long fine alternating pedicels; pedicels 5-12 mm long with an annular

often conspicuous gland below each spikelet. Spikelets ovate-elliptic to narrowly

oblong, 6-15 mm long 6-25-flowered, pallid to gunmetal grey in colour. Glumes

subcoriaceous, unequal, one-nerved; lower lanceolate, more or less 2 mm long, upper

ovate-lanceolate, about 2-5 mm long. Lemmas coriaceous, shiny, broadly boatshaped,

very broadly ovate in outline, acute, 2-2-5 mm long and 1-75-2 mm wide, 3-nerved,

nerves inconspicuous. Palea elliptic in outline, 2 mm long, 2-keeled, membranous,

margins indexed, apex truncate. Lodicules 2, fleshy, truncate-cuneate, 2-3-nerved.

Stamens 3; anthers narrowly oblong, about 1 mm long, yellow or purplish. Ovary

shortly stalked, oblong; styles 2, distinct; stigmas plumose. Caryopsis somewhat

laterally compressed, 1 - 5 mm long, obliquely ovate in profile ; pericarp swelling and

becoming mucilaginous when placed in water; endosperm whitish, quite opaque;

embryo half the length of the grain, greenish.

E. lamprospicula is only known from two collections, one from Salisbury in Southern

Rhodesia and the other from the Waterberg district of the Northern Transvaal. In

spite of our scant records it is doubtful whether the species is rare. It is more likely

that it has been overlooked due to its superficial similarity to E. racemosa.

The specific name is derived from lampro = shiny and spicula = spikelet an illusion

to the glossy spikelets of the species.

Eragrostis omahekensis de Winter, sp. nov., affinis E. horizontali Peter, sed paniculis

multo contractis lemmatibus longioribus et acutioribus glumis chartaceis olivaceis

differt.

Herba annua caespitosa, culmis erectis vel ascendentibus. Vagina g\ahva vel sparse

villosa, glandulosa vel eglandulosa. Ligula fimbriata. Eolia plana, apice setacea.

Panicula contracta, ramis glandula subbasali, brevibus, spiculis ramis appressis. Spiculae

angustae oblongae, 5-7 mm longae 1 - 5 mm latae, 4-9-florae, lemmatibus et glumis

deciduis. Glumae subequales, 1 - 5-2 mm longae. Lemmata chartacea, oblique lanceo-

lata carinata, acuta, nervis lateralibus prominentibus. Stamina 3, antheris oblongis

5 mm longis. Caryopsis obovato-oblonga subtranslucens levis.

South West Africa. — Gobabis: 12-7 m. W. of Gobabis; de Win ter 2498 (holotype,

PRE, isotype, K, M, BM, Windhoek); 25 m. W. of Gobabis, de Winter 2498; near

Steinhausen Police Station, de Winter 2435; Okasondana, Schwerdtfeger 4136; do

Liebenberg 4560, 4663; Babibabi, Liebenberg 4663. Okahandja: Bradfleld 387;

Teufelsbach, de Winter 2695; Oukongo, Dinter 3364. Otjiwarongo: Quickborn,

Bradfleld 425 and 365.

Erect caespitose annual. Culms straight or somewhat geniculate at the base,

single or branched, hollow, 2-4 noded, ultimate internode long exserted. Leafsheaths

lax, mostly shorter than the nodes, keeled, finely to strongly ribbed, glabrous or sparsely

474

hairy with bulbous-based long hairs, glanddotted on the keel and ribs with a short

transverse fringe of stiff hyaline hairs below the auricles. Ligule a dense fringe of short

hairs. Leaf blade usually flat, somewhat flaccid, linear, up to 15 cm long and 0-5 mm

broad, tapering to a fine point, glabrous or hairy with long bulbous-based hairs; veins

slightly raised below, upper surface striate, scaberulous. Panicle usually contracted,

very narrowly oblong to narrowly oblong-elliptic, 6-20 cm long 2-5 cm wide; branches

short with spikelets densely clustered on them, ascending or more or less appressed

to the rhachis; rhachis angular or nearly smooth, branches each with a gland at the

base. Spikelets very narrowly oblong 5-7 mm long and 1 • 5 mm broad, dark-grey,

4-9-flowered, rhachilla not disarticulating, lemmas caducous after fruiting, paleas

and glumes persistent on the rhachis. Florets hermaphrodite. Glumes subequal,

one-nerved, lanceolate, scaberulous on the keels, 1 • 5-2 mm long. Lemmas chartaceous,

obliquely lanceolate in profile, keeled, acute, 3-nerved, nerves prominent. Lodicules

fleshy, cuneate, truncate, about 0-3 mm long. Stamens 3. Anthers 0-5 mm long,

yellow. Ovary glabrous, styles distinct, stigma plumose. Caryopsis obovate-oblong,

semi-transparent when mature, smooth; embryo 2/5 the length of the grain; hilum

basal, punctiform.

This annual grass is almost exclusively found on disturbed places especially along

roadsides. On old lands it often forms extensive, practically pure stands. Where

present in large quantities it should make an excellent hay.

E. omahekensis can fairly easily be confused with E. liorizontalis but the much

more contracted inflorescence, the longer and more pointed lemmas, as well as the

firmer grey-green glumes makes it fairly easy to distinguish.

The specific name is derived from the Heroro word “ Omaheke ” a term used to

describe the sandy tree savanah of north-east South West Africa. Up to the present

the species has not been recorded outside of this region.

Eragrostis X pseud-obtusa de Winter, sp. nov. (£. obtusa MunroxE. echinochloidea

Stapf.)

Planta perennis, caespitosa. Culmi 2-3-nodosi, 20-60 cm alti. Vaginae striatae,

carinatae, carinis saepae glanduloso-punctatis. Ligula dense et breviter fimbriata.

Lamina 5-14 cm longa, 2-3 mm lata, anguste spicata. Paniculum plerumque ovato-

oblongum, ramis solitariis. angulis glanduloso-punctatis. Pedicelli spicularum orbi

glanduloso. Spiculae late ovato-oblongae, 3-5 mm longae, 2 -5-3 -5 mm latae, 8-20-

florae, olivaceae, lateraliter compressae, rhachilla disarticulata. Glumae subaequales,

inferiores 1-nervatae superiores 3-nervatae, carinis glanduloso-punctatis. Lemmata

late et profunde cymbiformia, prominenter 3-nervata. Palea 2-carinata, carinis sub

media parte anguste alatis. Stamina 3, antheris anguste oblongis 0-7-0 -8 longis.

Caryopsis elliptica, 1 mm longa, bis vel ter quam embryo longior. hilo basali punctiformi.

Cape. — Hopetown: Liebenberg 4150. Kimberley; Moran s.n. (Bolus H. No.

13905); Kameelhoek, Bruckner 21; Kenilworth, Levy s.n. (Galpin Herb. No. 6324);

Swan s.n.; Wilman s.n. (N.H. No. 28349). Barkly West: Newlands, Wilman s.n.

(Bolus H. No. 25457); Wilman s.n. (N.H. No. 28336); Acocks 140; Brueckner 823.

Vryburg: Tiger Kloof, Brueckner 320 (PRE, holo); Armoedsvlakte, Mogg 3960;

Mogg 3668; Stent s.n. (H. 21516); Benauwdheidsfontein, Marloth 863.

Orange Free State. — Fauresmith: C. A. Smith 3879. Kroonstad; Bothaville,

Schweickerdt 1113. Jacobsdal; Schweickerdt 1149.

Transvaal. — Christiana: Burtt Davy 11411. Bloemhof: Lombard Nature

Reserve, Leistner 91. Wolmaransstad: Sutton 68.

Cultivated specimens. — Prinshof Experimental Station: Story 1940; de Winter

710. Johannesburg: ex Prinshof Experimental Station, R. de V. Pienaar s.n. (sp. 16

plant 4 and plant 2).

Dense erect caespitose perennial, culms straight, or geniculate at the base, mostly

simple, occasionally branched, 2-3-noded, 20-60 cm high, nodes mostly exserted from

the sheaths. Sheaths pallid often hairy on the margins, striate, keeled, finely gland-

475

dotted on the keel. Auricles softly bearded. Ligiile a dense line of short hairs. Leaf-

blade rolled or flat, linear, 5-14 cm long and 2-0-3 -0 mm broad when flattened, primary

nerves raised below, slightly raised above and scaberulous, tapering to a fine point,

midnerve often finely gland-dotted below. Panicle lax to dense, more or less ovate-

oblong in outline, branches divided once or twice, spikelets densely crowded on the

branches, divisions angular, densely gland-dotted on the angles, pedicels of spikelets

often with a glandular ring. Spikelets broadly oblong-ovate, 8-20-flowered ; pallid

or grey to gunmetal grey, rather plump, laterally compressed, 3-5 mm long and 2 -5-3 -5

mm broad; rhachilla readily disarticulating above the glumes and between the florets.

Florets hermaphrodite falling entire. Glumes sub-equal boatshaped, keeled dark

metallic green, darker in colour than the glumes, obliquely lanceolate in side view, keels

distinctly gland-dotted, apex acute to subacute; lower 1-3-nerved; upper 3-nerved.

Lemmas broadly and deeply boatshaped, chartaceous, broadly elliptic in back view

(not flattened), 3-nerved, nerves green, raised. Palea elliptic-oblong, slightly shorter

than the lemmas, 2-keeled, keels narrowly winged below, shortly ciliate on the keels

and less so on the wings, apex emarginate or more or less truncate to rounded, wings

and keels firmer in texture. Lodicules small, oblong-cuneate, truncate, 0-5 mm long.

Stamens 3, anthers 0-7-0 -8 mm long, cultrate, yellow. Caryopsis elliptic, 1 mm long,

brown, slightly dorsally flattened; embryo one-third to half the length of the grain;

hilum basal, punctiform.

SUID-AFRIKA SOUTH AFRICA

Fig. 3. — Distribution of: A, Eragrostis echinochloidea; B, E. obtusa\ C, E.xpseud-

obtusa (E. echinochloidea X E. obtusa).

476

Fig. 4. — Erogrostis pseiicJ-obtusa; a, habit, X 1 ; b, spikelet, x 10; c, anther, x 10;

d, caryopsis, X 10; e, palea of E. pseud-obtusa, x 10; palea of E. obtusa,

X 10; f, palea of E. echinochloidea, x 10.

477

E. pseud-obtusa occupies an intermediate position between E. obtusa and E.

echinochloidea. The intermediates form a distinct group, situated, when judged on

spikelet characters, halfway between the two species. No evidence could be found

that these three species naturally grade into one another, in spite of the fact that a large

number of specimens of each was available for study. From the rather limited distribu-

tion which almost completely covers the area where E. obtusa and E. echinochloidea

overlap (fig. 3), as well as from the completely intermediate character of the spikelet

(fig. 4) one can only conclude that the origin of E. pseud-obtusa must have been that

of a hybrid between the former two species. There is no evidence however that intro-

gressive hibridization has taken place. What the mechanism genetically separating

E. pseud-obtusa from its parents is, is not clear. It probably is not stabilisation due to

doubling of chromosomes, since E. echinochloidea and E. pseud-obtusa (as sp. aff.

obtusa) are both reported to have 2n = 40 by R. d. Y. Pienaar in Grasses and Pastures

of South Africa 561 (1955). The genome of E. obtusa is still unknown.

The great majority of specimens of E. pseud-obtusa have in the past, been referred

to E. obtusa. This is no doubt due to the fact that E. pseud-obtusa more strongly

resembles E. obtusa in the external appearance of the inflorescence and the spikelets

than it does E. echinochloidea. When the spikelet is examined the winged palea is

immediately evident so that the specimens can easily be distinguished from E. obtusa.

Due to the very strongly congested branches of the inflorescence, as well as the acuminate

glumes of E. echinochloidea, E. pseud-obtusa cannot easily be confused with it (fig. 4).

The following key will facilitate the identification of the species.

Pales shortly ciliate on the symmetrically rounded keels E. obtusa.

Pales with a small flap or wing in the lower half of the keels, upper part of keels scaberulous

or shortly ciliate.

Branches of the inflorescence secund, very densely contracted with the spikelets densely

clustered; glumes distinctly acuminate with the apices diverging and not appressed

to the spikelet; flap or wing in the lower half of the keels of the palea usually with

a short tooth on the upper part; upper part of keels scaberulous.... E. echinochloidea.

Branches of the inflorescence more or less secund, fairly strongly contracted or somewhat

lax with the spikelets laxly to fairly densely clustered; glumes acute to subacute

mostly appressed to the spikelet, flap or wing on the lower half of the keels of the

palea without a tooth on the upper part ; upper part of keels very shortly ciliate . E. pseud-obtusa.

Eragrostis remotiflora de Winter, sp. nov., aff. E. micranthae, sed ab ea lemmatibus

brevioribus baud imbricatis, antheris late ovato-oblongis, caryopsis oblonga dorse

canaliculata praecipue differt.

Annua vel subperennis 10-55 cm alta. Panicula plerumque basi vaginata, demum

exserta, 5-20 cm longa 3-14 cm lata ambitu late ovata, laxa, ramis patentibus inferioribus

subverticillatis, superioribus solitariis vel binis, rhachide et ramibus primariis saepe

glanduloso-punctatis. Spiculae solitariae ca 5 mm longae 1 mm latae cinereo-virides

vel plumbeae, 4-8-florae, lemmatibus baud vel pauce imbricatis, glumis valde inaequali-

bus, 1-nervatis, lemmatibus late ovato-oblongis. Stamina 3; antberae 0-3-0 -4 mm

longae. Caryopsis oblonga dorsaliter canaliculata.

Transvaal. — Pretoria: in depressis bumidis prope Apiesrivier, Jan., 1894,

Schlechter 4164 (K); Kaalfontein, Mogg 3741. Bloemhof: Christiana, Burtt Daw

12803 and 12795.

Orange Free State. — Bosboff: 16 m. W.S.W. of Dealesville, Acocks 14016.

Fauresmitb: Pole Evans 1583. Betbulie: 1\ m. N.W. of, Acocks 13521 (PRE, bolo.

K, M, B, BM, BOL, iso). Bloemfontein: Potts 2423, 2522.

Cape. — Cradock: 12-2- m. N.N.W. of, Acocks 15764. Tarka: N.N.E. of Tarkastad,

Acocks 17953. Middelburg: Acocks 15807. Barkly West: Danielskuil, Esterhuizen

2019 and 1061; Danielskuil E. Fm-ar 62. Vryburg: Dry Hartz, Mogg 8500; Progress,

Mogg 9007; Armoedsvlakte, Mogg 3960.

478

f iG. 5. — Eragrostis remotiflora; a, habit, X 1; b, spikelet, X 10; c, anther, x 30;

fAcocks 13521).

479

Annual or subperennial, glabrous, 10-55 cm high, culms erect or sprawling and

geniculate, 1-4-noded, slender, rather soft. Sheaths pallid, pinkish or purplish, weakly

striate, usually exceeding the internodes but lax and stripping from the culms,

occasionally shorter, chartaceous especially when withered, keeled, glandular on the

keels, margins membranous. Ligule a very short dense hairy rim; auricles bearded.

Leaf blade linear, glabrous, flaccid, flat or somewhat rolled, up to 17 cm long and 3 mm

wide, midrib prominent below, often dotted with glands, primary nerves prominent

below, smooth, upper surface finely striate, scaberulous on the striate, blade tapering

to a fine point. Panicle usually sheathed at the base by the uppermost sheath, finally

exserted, as long as or longer than the rest of the plant, 5 X 3 cm to 20 X 14 cm usually

broadly ovate in outline, lax, branches patent or spreading at slightly less than a right

angle, subwhorled at the base, single or in two’s upwards, axils glabrous or with a

few hairs, rhachis and primary branches often dotted with sunken glands especially

towards the base of the panicle. Spikelets solitary, pedicelled, linear, about 5 mm long

and 1 mm wide, greenish-grey to lead-grey, occasionally suffused with purple, 4-8-

flowered, lemmas usually not overlapping the florets above on the same side or only

shortly overlapping at the base. Florets hermaphrodite, lemmas and paleas both falling,

but not together leaving the undulating rhachis intact. Glumes very unequal, lower one-

nerved, lanceolate, 0-5 mm long; upper one-nerved, about 1-5 mm long, lanceolate.

Lemmas broadly ovate-oblong, sub-membranous, 3-nerved. Pales equal to or slightly

shorter than the lemma, bifid at the apex. Lodicules truncate-cuneate. Stamens 3;

anthers purple, broadly-oblong, 0-3-0 -4 mm long. Ovary glabrous, styles free; stigmas

plumose. Caryopsis oblong, brown, finely longitudinally striate, slightly grooved on

the back; embryo two-thirds the length of the grain; hilum punctiform, basal.

E. remotiflora resembles E. pilosa but can be distinguished from this species by the

shorter broader lemmas which do not overlap the base of the lemma directly above it.

The lemmas of E. pilosa are lanceolate in profile, acute and overlap the base of the

lemmas directly above. The grains of E. remotiflora are grooved on the back and the

embryo is about two-thirds the length of the grain while those of E. pilosa taper towards

the apex, are rounded on the back and the embryo is about half the length of the grain.

This species is often found growing in association with E. homomalla Nees = (£

hygrophila. Hubb. and Schweick.) and like the latter is a hygophilous grass growing

in wet or damp soil on the edge of pans and vleis, in tne semi-arid areas of the Cape,

Transvaal and Orange Free State. It shows a superficial resemblance with E. homomalla

but can easily be distinguished by the laxer panicle and the smooth nerves of the lemmas

which are prominent and gland dotted in the latter.

Its closest relative probably is E. micrantha which differs by longer lemmas which

overlap each other, the larger cultrate anthers and the long narrow grain. The inflo-

rescence moreover, although lax is much more branched and not nearly as scanty as

in E. remotiflora.

Kaokochloa de Winter, genus nov., Schmidtiae aff. sed lemmatibus 2-3-aristatis, duobus

nervium lemmatis lateralium (interdum necnon nervo mediano) in aristam planam

rectam excurrentibus, apice lemmatis inflexo inter aristas in lobos 2 breves latos

nigros membranaceos producto praecipue differt

Annua. Culmi erecti geniculati vel prostrati, ad nodos radicantes. Vaginae

internodiis breviores. Ligula longe ciliata. Laminae planae. Panicula satis densa,

ramis solitariis villosis breve glanduloso-pubescentibus, spiculis 3-6-floris. Glumae

9-1 1-nervatae subaequales late ellipticae velova to-ellipticae. Elores hermaphroditi,

lemmatibus 9-nervatis late cymbiformis in parte inferiore inter nervos pubescentibus

apice inflexo saepe nigro, nervis lateralibus duobus, interdum necnon nervo mediano,

in aristam planam rectam excurrentibus, paleis ambitu ellipticis marginibus inflexis

carinis fimbriatis. Lodiculae 2 cuneatae apice truncatae. Stamina 3, antheris lineari-

oblongis flavis. Ovarium glabrum, stylis 2 liberis, stigmatibus plumosis. Caryopsis

480

obovata, basibus stylorum asymmetrice insertis coronata; embryo 4/5 longitudinis

caryoptidis attingens ; hilum basale, punctiforme. Species unica : Kaokochloa nigriros-

tris de Winter, spec. nov.

Kaokochloa has up to the present only been found in the Northern Kaokoveld,

where it inhabits the more western dryer parts of the territory. It often forms large

dense stands. The grazing value is unknown but probably is fairly high since the

grass is a soft annual.

Kaokochloa nigrirostis de Winter, sp. nov.

Annual, 20-60 cm high. Culms erect, geniculate or prostrate at the base, and rooting

at the nodes, nodes 3-7. Sheaths shorter than the internodes, striate, villous with

gland-tipped hairs. Ligule a fringe of long stiff hyaline bristles. Auricles villous.

Leaf blade linear to linear-lanceolate, 5-12 cm long and 5-10 mm wide, tapering to a

fine tip, flat, pilose with long slender hairs mixed with shorter gland-tipped hairs.

Panicle rather dense and contracted, branches single, rhachis and branches grooved,

densely villous with a mixture of long villous hairs and shorter gland-tipped hairs.

Spikelets shortly pedicelled or subsessile and clustered on the rather short branches,

sub-globose, about 6 mm wide and 7 mm long, with 3-6 florets. Florets hermaphrodite.

Glumes 9-11 -nerved, subequal, equalling the spikelet, broadly boatshaped, 6-7 mm

long, broadly elliptic or ovate-elliptic, subacute, villous with long thin hairs mixed

with shorter pinshaped glandular hairs. Lemmas elliptic in outline, broadly boatshaped,

5 mm long and 3 mm wide, coriaceous, densely hairy in the lower half between the

nerves, 9-nerved, one lateral nerve on each side and occasionally the central nerve

excurrent in a broad flat straight scabrid awn; lateral awns 4-6 mm long; central

awn usually much shorter, 1-3 mm long or not developed; awns usually much shorter

or absent in the immature and apical florets; apex of the lemma glabrous, incurved,

ending in an awn flanked by two broad short membranous lobes or awn absent and

ending in two lobes, lobes usually black in colour. Palea elliptic in shape, about 5 mm

long and 2 mm wide, thinly coriaceous shallowly concave on the back, keels somewhat

thickened, densely fimbriate with short stiff bristles, margins membranous, indexed,

long villous at the base. Lodicules 2, somewhat fleshy, wedge-shaped with a more or

less truncate apex. Stamens 3, anthers linear-oblong, 3-4 mm long, yellow. Ovary

glabrous, obovate-cuneate in outline, styles 2, stigmas plumose. Grain obovate in

outline, crowned by the asymmetrically attached base of the style, somewhat convex

dorsally and usually flat on the other side; embryo four-fifths the length of the grain,

obovate in outline; hilum basal, punctiform, brownish to black.

South West Africa. — Kaokoveld: 18-5 miles West of Otju (Otjihu), mica schist

hillocks and mountain slopes with coarse quartz and limestone gravel, de Winter and

Leistner 5679 (PRE, holo; K, BM, M, B, BOL, Windhoek); do. 5679a, cult, at Division

of Botany from seed of 5679 (K, BM, M, B, BOL, Windhoek, NY); Warmbad

(Warmquelle) near Sesfontein, dolomite slope with quartz intrusions, erect annual

forming large almost pure stands, de Winter and Leistner 5848 (K, BM, M, Bolus,

NY).

The specific epithet is an allusion to the characteristic black apices of the lemmas.

B. DE Winter

HYPOXIDACEAE

Rhodohypoxis palustris Killick, sp. nov. R. baurii (Bak.) Nel affinis, sed ita differt:

folia conduplicata semi-carnosa rigidula marginibus exceptis glabra flores semper

purpureo-rubicundi.

Cormus globosus 5-7 mm diam. tunicis apice copiose setosis. Folia erecto-arcuata

anguste linearia acuminata 2-5-10 cm longa 2-3 mm lata conduplicata semi-carnosa

nonnihil rigida flavo-viridia marginibus pilis longis albis alioqui glabra. Pedicelli 1-2

481

simplices erecti graciles 1-6 cm longi sericei. Perianthium purpureo-rubicundum tubo

5 mm longo sericeo segmentis oblongis 7-1 1 mm longis. Stamina 6 antheris 2 mm longis.

Ovarium turbinatum 3 mm longum stigmate trilobato.

Cape Province. — Maclear District : seepage on upper eastern slopes of Drakens-

berg, locally frequent, 8,000 feet, Naude’s Nek Pass, Acocks 12183; boggy slopes,

Naude’s Nek, 20-5 miles N.E. of Rhodes, Marais 1372.

Natal. — Bergville District : locally abundant among small stones in shallow stream,

6,050 feet. Cathedral Peak Forest Influences Research Station, Killick 956; locally

common among small stones in stream in Catchment 1, 6,015 feet. Cathedral Peak

Forest Influences Research Station, Killick 1602 (PRE, type). Estcourt District:

Giants Castle, 8,000 feet, Symons 156. Impendhle District: locally abundant in vlei

with roots in water, 6,000 feet, “ Tillietudlam ”, Huntley 460.

A perennial herb with a globose corm 5-7 mm diam. with fleshy roots and crowned

with persistent bristles. Leaves radical, erecto-arcuate, narrowly linear, 2-5-10 cm

long, 2-3 mm wide, sheathing at base, conduplicate, semi-succulent, somewhat rigid,

margins with long white hairs, otherwise glabrous, yellowish green. Pedicels 1-2,

simple, erect, 1-6 cm long, sericeous. Perianth purplish-pink, tube 5 cm long, sericeous;

segments oblong, 7-11 mm long, 3-4 mm wide. Stamens 6, arranged in two series;

anthers 2 mm long. Ovary turbinate, 3 mm long; stigma minute, trilobed.

This small plant from the Drakensberg Mountains is an attractive addition to

Rhodohypoxis, a genus of only two species. For nearly fifty years this plant has passed

as R. baurii (Bak.) Nel, but it can be distinguished on the following grounds; —

A. — The leaves are conduplicate, semi-succulent, yellowish-green, rather rigid

and only hairy on the margins, whereas in R. baurii they are more or less flat with a

median groove, coriaceous, greyish-green and hairy all over. The leaves of R. palustris

are narrower than those of typical R. baurii, but about equal to those of R. baurii var.

milloides. B. — The flowers are purple-pink, red or crimson. In R. baurii they vary

from white in forma platypetala to red in the typical form. C. — R. palustris, as its

name indicates, is a vlei or marsh plant, whereas R. baurii occupies comparatively

dry habitats, for example mountain grassveld and rock outcrops.

D. J. B. Killick

LILIACEAE

Allium rotundum L. Sp. PI. ed. 2; 423 (1762).

A. ampeloprasum var. B. Gawler in Curt. Bot. Mag. t. 1560 (1813); Thunb. Prod. 65

(1794). A. dregeanum Kunth, Enum. 4: 382 (1843); FI. Cap. 6: 407 (1897).

Type: Cape, Queenstown, Stormberg Range Drege 8660a (K?).

In the Flora Capensis Baker accepted Kunth’s species, A. dregeanum, as the only

indigenous species found in South Africa. He rejected Regel’s conclusion that it should

be sunk under an old well-known European species, A. scorodoprasum L. (cf. Monogr.

All. 42 1875). Recently material of this species was sent to Munich where it was identified

as A. rotundum L. a native of central and southern Europe and the near East.

Ornithogalum seineri {Engl, and Krause) Oberm. comb. nov. Bulbine seineri Engl, and

Krause in Engl. Bot. Jahrb. 45: 124 (1910). Type: Bechuanaland, Litauani, on

grey sand rich in humus between limestone. Seiner II, 98 (B, holo. ! PRE, photo).

Anthericum seineri (Engl, and Krause) Poelln. in Fedde, Rep. 53: 136 (1944).

482

Ornithogalum fiUbracteatum Oberm. in Ann. Transv. Mus. 17: 194 (1937). Type:

Transvaal, Letaba: Mbayinbayi, 28 m. south-west of Punda Maria, Lang TM

31099 (PRE, holo!). O. wilmaniae Leight. in Journ. S.A. Bot. 11: 168 (1945).

Type: South West Africa, Gobabis district, Sandfontein, Wilman in BOL 15280

(holo) in KMG 1599 (iso).

Urginea langii Brem. in Ann. Transv. Mus. 15: 237 (1933). Type: Transvaal, Pieters-

burg: Brak Rivier vlei, Bremekamp and Schweickerdt 25 (PRE, holo!).

This Ornithogalum, like many others, may look deceptively different in the her-

barium because of several factors. In a wet season the uppermost flowers develop,

giving the raceme a cylindrical appearance but in dry conditions these remain sterile

thus giving it the pyramidal shape so typical of Urginea langii. The type of O. fili-

bracteatum shows a very young inflorescence with short pedicels and long filiform bracts

protruding far beyond the buds. As the pedicels lengthen only during anthesis none

had attained their maximum length when the specimen was preserved. The long filiform

bracts giving the young raceme a brushlike appearance, seem characteristic but it was

seen in other species, for instance O. pulchrum Schinz, that they may be long or short

in specimens found in the same area ; moreover the thin upper part of the bract often

dries up at an early stage.

Bremekamp had classified this species as an Urginea because of the flat round

seeds. The seed of Urginea however is very different; it is long, narrow-elliptic, with

a narrow, often winged membrane around its perimeter. There are moreover some other

very good characters which typify the genus Urginea and which are absent in our species;

{a) the inflorescence and the often hysteranthous leaves evolve from different buds

whilst in Ornithogalum the central raceme terminates the leafy rosette; {b) the lowest or

central bracts are spurred; (c) the bulb consists of loose scales. Dr. J. M. J. de Wet

who studied the chromosome pattern of Ornithogalum wilmaniae Leight. (cf. Cytologia

22: 145-159, 1957) found that it possessed large chromosomes with the number n = 10

and they showed affinity to other South African species of Ornithogalum. Dr. de Wet

suggested that it might form a stepping stone between Urginea and Ornithogalum.

The type of Bulbine seineri Engl, and Krause has no underground parts preserved

but Seiner described it as a bulbous plant. The filaments are described as “ tenua ”,

and are not bearded.

The species is found in the warmer parts of the Kruger National Park, in the

Transvaal Bushveld, the Kalahari region and the northern districts of South West

Africa.

Ornithogalum pulchrum Schinz. In her revision of the South African species of Orni-

thogalum [J.S.A. Bot. X: 169 (1944)], Leighton mentions that O. pulchrum Schinz

[Verb. Bot. Ver. Prov. Brandenb. 31 : 221 (1890)] might be the same as O. wilmaniae

but as she had not seen the type and as it had no leaves, she hesitated to associate

her species with O. pulchrum. The Director of the Zurich herbarium kindly sent

us the Schinz type on loan. O. pulchrum is a tall plant, not identical with O. seineri

{0. wilmaniae) and the following specimens at PRE match the type.

South West Africa. — Grootfontein: Tsumeb, Naegelsbach 9. Omaruru: Spitz-

kopje, Boss TM 36006. Kaokoveld: Kaoko-Otavi, de Winter and Leistner 5573.

Okavango Native Territory: near Okavango River at Mupini, de Winter and Marais

AA91 (with shorter bracts).

Southern Rhodesia. — Sabi-Lundi River Junction, Wild 3490.

Baker in the Flora of Tropical Africa 7: 545 (1898) placed O. pulchrum under

O. longibracteatum Jacq. This species comes from the eastern Cape and O. caudatum

Jacq. is a synonym. The length of the bracts varies and is unimportant.

483

Some differences between O. piilchrwn, O. longibracteatum and O. seweri are as

follows

O. pulchrwn

Tall plants up to m.

Leaves up to 70 cm long.

Raceme cylindrical bearing ca

300 flowers; usually with

sterile apical flowers.

Scape 200 cm high, straight.

Bracts up to 7 cm long

(occasionally shorter).

Pedicels slender, up to 2-5

cm in fruit.

Perianth segments lanceolate,

20 mm long.

Filaments ovate, attenuated to

the apex.

Ovary oblong, 3-lobed.

Capsule 3-angled.

Seeds discoid, elliptic, 8 mm

long.

Solitary plants.

Sweet smelling.

Northern South West Africa,

Southern Rhodesia.

Usually in rocky crevices.

O. seineri

Plants up to 50 cm.

Leaves up to 30 cm long.

Raceme usually pyramidal bear-

ing ca 100 flowers; usually

with sterile apical flowers.

Scape ca 20 cm high arcuate.

Bracts 2 cm long (occasionally

longer).

Pedicels sturdy, up to 6 cm in

fruit.

Perianth segments lanceolate,

17 mm long.

Filaments linear-lanceolate,

acute near the apex.

Ovary oblong, 3-lobed.

Capsule 3-angled.

Seeds discoid, round, 1 1 mm in

diam.

Gregarious plants.

Faintly scented.

Northern Transvaal, Kalahari

Region, central and northern

South West Africa.

Sandy flats.

O. caudatum

Plants up to 1 m.

Leaves up to 70 cm long.

Raceme cylindrical bearing ca

80 flowers; no sterile apical

flowers seen.

Scape ca 80 cm, straight.

Bracts variable in length, 2-4

cm long.

Pedicels slender up to 1-2 cm in

fruit.

Perianth segments ovate, 10 mm

long.

Filaments ovate, attenuated to

the apex.

Ovary rounded.

Capsule rounded.

Seeds angled, narrow-oblong in

outline.

Solitary or in clumps, producing

bulbils

Said to be scentless.

Eastern Cape, Natal.

In shady kloofs and grassy

valleys.

The type of Urginea dimorphantha Bak. [Bull. Herb. Boiss. 2nd Ser. No. 8, p. 663

(1903)] from South West Africa, Ovamboland, Ondonga, Rautanen (Z, holo PRE,

photo), kindly sent on loan to us by the Director of the Zurich Herbarium, proved

to be a synonym of Ornitliogalwn pulchnim Schinz. Baker had placed this specimen

under Urginea as the seeds are discoid.

A. Amelia Obermeyer

MALVACEAE

Abutilon flanaganii A. Meeuse, sp. nov., A. pseudocleistogamo Hochr. affinis, sed inter

alia floribus majoribus, petiolis longioribus differt.

Frutex ramosus 25-40 cm altus verisimiliter perennis. Cattles teretes cinereo-

velutini. Folia late ovato-cordata, interdum indistincte 3-lobata, plus minusve acumi-

nata, obtusa minutissime mucronata, irregulariter crenata vel sinuata vel subserrata

interdum subintegra, supra intense viridia minutissime stellato-velutina, subtus cinerea,

stellato-velutina, nervis subtus prominentibus; lamina 2-5 cm longa 1-5-4 cm lata,

petiolis gracilibus cinereo-velutinis interdum parce pilosis 2-5 cm longis, stipulis

subulatis 4-6 mm longis. PediceUi graciles axillares, solitarii, cinereo-velutini, ad

60 m longi apicem versus articulati. Calyx cinereo-velutinus, 5-fidus, tubo cupuliformi

c. 3 mm longo, lobis lanceolatis vel lanceolato-deltoideis carinatis aristato-apiculatis

c. 7 mm longis. Petala flava subglabra basi dense ciliata c. 20 mm longa. Columna

staminalis apice stellato-hirsuta. Infructescentia discoidea, 4-5 mm alta, c. 9 mm

diam., mericarpiis 9-13 monospermis subquadratis muticis vel minute mucronatis

stellato-hirsutulis c. 4 mm longis 5 mm latis. Semen minutissime punctato-verruculosum

c. 2 X 1-5 mm.

Cape Province. — Komgha: near Komgha, Flanagan 340 (PRE, holo.!, NBG,

iso.!).

484

A low much-branched shrub 25-40 cm tall apparently perennial, covered on the

vegetative parts, pedicels and calyces with a short smooth velvety tomentum which is

canescent except on the upper leaf-surface, in addition longer more or less patent pilose

hairs are sometimes found at the very base of the lower leaf-surface, stems, on the

petioles and on the pedicels, more often in tufts at the apex of the petiole. Stems

terete, slender but firm to wiry, glabrescent, the bark ultimately somewhat wrinkled.

Leaves broadly ovate-cordate, sometimes very faintly 3-lobed, shortly acuminate obtuse

and generally minutely mucronate, irregularly crenate, repand or serrate to subentire,

2-5 cm long and 1 • 5-4 cm broad, dark green (drying dark olive brown) and very shortly

velvety above, the paler lower surface with fine prominent main veins; petioles slender,

2-5 cm long; stipules subulate, 4-6 mm long. Flowers solitary, axillary; pedicels

slender, up to 5 cm long articulated in upper 10 mm. Calyx deeply dissected; tube

cupuliform, about 3 mm long, lobes lanceolate to triangular-lanceolate, carinate by

the prominent midrib, apiculate-aristate, about 7 mm long. Petals yellow, glabrous

or nearly so except the densely ciliate narrow base. Staminal column stellate-hirsute

at the apex of the conical dilated basal portion. Fruit discoid, 4-5 mm high and about

9 mm in diam., of 9-13 subquadrate 1 -seeded, 3-4 mm high and 5 mm broad mericarps

which are horizontally truncate at the upper edge forming an acute or somewhat mucro-

nate angle of about 90° with the dorsal side, are rounded at the basal dorsal angle and

have a subterminal ventral tooth which is almost level with the apical edge. Seed

finely punctate-verruculose, about 2x1-5 mm.

This species differs from all other African species in the small mericarps which

are broader than they are high. The only other species with similar mericarps is A.

pseudocleistogamum, a Madagascan species from which it differs in several respects,

especially in the much larger flowers and considerably longer petioles. In habit it is

not unlike A. fruticosum and A. sonneratianum, but its small 1-seeded mericarps dis-

tinguish it at once. A. flanaganii is named after Henry G. Flanagan, who contributed

so much to our knowledge of the flora of the Eastern Cape Province, mainly of the

Komgha and surrounding districts, through his extensive collections of neatly prepared

herbarium specimens of which the main set is in the National Herbarium, Pretoria.

A. flanaganii is apparently a very local and rare species, because only the single

cited gathering was found among all the material from the South African and some

European herbaria.

A. galpinii A. Meeuse, sp. nov., A. grantii A. Meeuse (A. indico sensu Harv., non Don)*

et A. piloso-cinereo A. Meeuse (vide infra) arete affinis, sed mericarpiis minoribus,

foliis majoribus caudato-acuminatis praecipue differt.

Suffrutex ramosus c. 75 cm altus, breviter stellato-tomentosus vel subvelutinus

in partibus junioribus sparse glandulosus. Caules interdum breviter stellato-pilosi.

Folia late ovato-cordata vel cordato-suborbicularia, 3-16 cm longa 2-12 cm lata,

abrupte acuminato-caudata, subtriloba, margine irregulariter dentata, crenata vel

subserrata, supra intense viridia in siccitate atrobrunnea dense strigoso-subvelutina

demum glabrescentia et scabrida, subtus pallidiora subarachnoideo-tomentosa, petiolis

laminis subaequilongis, stipulis lineari-lanceolatis ad 6 mm longis. Flores solitarii,

axillares, pedicellis gracilibus ad 5 cm sub fructu ad 7 cm longis. Calyx cupuliformis,

infra medium lobatus, molliter stellato-tomentosus 9-12 mm longis, lobis ovatis vel

ovato-deltoideis apiculato-caudatis, apiculo subulati-filiformi ad 2 -5 mm longo. Petala

lutea vel dilute ochracea, glabra, basi ciliata, 11-13 mm longa. Columna staminalis

basin versus subsparse stellato-hirsutula. Infructescentia breviter cylindrato-semiglobosa

interdum subcampanulata, apice truncato-concava, c. 10 mm longa 12-14 mm diam.

Mericarpia c. 16, 3-sperma, oblique truncata, dorso et apicem versus subsparse molli-

terque stellato-pubescentia 8-10 mm longa 5-6 mm lata, acuta vel mucronata. Semina

2-5x2 mm, atrobrunnea, minute verrucoso-punctata.

* Vide Meeuse in FI. Zamb. part 2 (1960), in press.

485

Transvaal. — Barberton: Barberton, Umvoti Creek, Galpin 767 (PRE, holo. !,

GRA, iso.!); Valley near Edwin Bray Battery, Galpin 1197 (PRE); Barberton without

precise locality, Williams s.n. = TRY 7671 (PRE).

Swaziland. — Bremersdorp, P. Hutchinson 8 (PRE).

Portuguese East Africa. — Lourenco Marques Distr. : Goba, Lebombo escarp-

ment, Fidalgo de Carvalho 257 (LM, PRE).

Natal. — Lower Tugela: Tugela Valley below Sans Souci, Edwards 1688 (NU,

PRE).

SufFrutex attaining a height of at least 75 cm, much branched in upper portion,

covered with a short stellate-tomentose to velvety pubescence on vegetative parts,

pedicels and calyces; the youngest parts, apical portions of pedicels and calyx-tube

sometimes also somewhat glandular. Stems slender, wiry, terete, rather soon woody,

glabrescent, the pubescence sometimes somewhat pilose. Leaves broadly ovate-

suborbicular with cordate base and a triangular long-attenuate to acuminate or caudate

apex, often abruptly narrowed into the acumen from a broad base which gives the

blade a low, but distinct, 3-lobed appearance; the tip of the acumen subobtuse

to acute, minutely mucronate; the basal sinus more or less triangular, usually rather

deep; basal lobes rounded, the small lateral lobes, if present, more or less triangular,

obtuse or acute; the margin irregularly dentate, serrate or crenate, generally shallowly

so to subentire but the apical portion of the acumen always entire; upper surface

deep green often drying a dark ohve-brown, when young densely velvety strigose,

later sparsely stellate-strigose, glabrescent and turning slightly scabrid, lower surface

distinctly paler, dirty greyish-yellow, sometimes faintly glaucous, at first very densely

stellate-tomentose, later with a cob-webby stellate tomentum; petioles terete, in young

leaves tomentose, later usually with short spreading to deflexed somewhat stiff stellate

hairs especially near the apex, a little shorter to a little longer than the 3-16 cm long

and 2-12 cm broad blades; stipules linear-lanceolate, up to 6 mm long. Flowers

solitary, axillary on main stems and/or on short lateral shoots; pedicels slender, terete,

in flower up to about 5 cm, in fruit up to about 7 cm long, articulated near the apex.

Calyx cupuliform, lobed to a little beyond the middle, 9-12 mm long, with ovate to

ovate-triangular acute lobes each terminating in a subulate to filiform, up to 2-5 mm

long, apiculus. Corolla described as orange-yellow and buff; petals 11-13 mm long,

glabrous or nearly so, the narrowed base ciliate. Staminal column rather coarsely and

sparsely beset with many-rayed stellate-hairs in basal, conical portion. Fruit shortly

subcylindric to semi-globose or slightly campanulate about 10 mm high, 12-15 mm in

diam., truncate-concave at the apex; mericarps about 16, 3-seeded, 8-10 mm long

measured along the back and 5-6 mm broad; much compressed, papery, softly and

rather sparsely stellate-pubescent along the back and apical portion; the back straight

or somewhat bulging in lower half, then inwardly rounded into the truncate base,

the apical edge convex or nearly straight, slanting upwards towards the shortly toothed

mucronate dorsal apical angle, ventral tooth small. Seeds 2-5x2 mm, dark brown,

finely verruculose-punctate with minute, usually orange-brown protuberances.

The much compressed relatively broad, and papery mericarps indicate that this

species is related to A. grantii (= A. indicum sensu Harv. in FI. Cap., non Don) and

A. piloso-cinereum, and consequently to A. sonneratianum, but the mericarps are very

much smaller, the leaves larger and caudate-acuminate and the calys-lobes have a

subulate-filiform up to 2-5 cm long apiculus.

This species remained unrecognised for a long time although it was collected by

Galpin as early as 1896. Burtt Davy, in his Manual FI. Transv. 2: 275 (1932), refers

Galpin 767 to A. mauritianum (Jacq.) Medic, to which it is not remotely related, and

Galpin 1197 to “ indicum ” (= A. grandiflorum Don) which it does not resemble.

The species under discussion seems to be mainly restricted to the Barberton area

and the Lebombo range. It is found in light shade on lower mountain slopes between

600-1,200 m altitude.

486

A. piloso-cinereum A. Meeuse, spec, nov., A. grantii A. Meeuse (= A. indico sensu

Harv. non Don) arete affinis, sed plantis canescentibus ceteris caulibus petiolis

pedicellisque pilis longis patentibus subsparsis obtectis praecipue diflfert.

Suflfrutex probabiliter annuus pauce ramosus ad 1-50 m altus. Caules stellato-

tomentosi vel subvelutini, pilis longis patentibus subsparsis pilosi. Folia late ovato-

cordata velsuborbiculari-cordataad triangulato-cordata,interdum plus minusve 3-lobata,

apice acuminata vel attenuata, margine subregulariter serrata vel crenato-de tita,

supra saturate viridia minute adpresse stellato-pubescentia demum glabrescentia,

subtus molliter velutinosa albido-canescentia conspicue prominenter venosa, 2-7 cm

interdum ad 15 cm longa, 1 -3-5 cm interdum ad 10 cm lata, petiolis laminis subaequi-

longis tomentosis vel subvelutinis pilis patentibus sparse pilosis. Flores axillares

solitarii, pedicellis tomentosis vel subvelutinis pilis patentibus sparse pilosis. Calyx

campanulato-cupuliformis, infra medium lobatus, 9-12 mm longus, dense velutinus,

lobis ovato-lanceolatis vel oblongo-lanceolatis acutis vel acuminatis vel breviter apicu-

latis. Fetala flava 14-18 mm longa, glabra. Infructiscentia subcylindrato-semiglobosa,

10-12 mm longa, 15-20 mm diam. Mericarpia 10-18, 3-sperma, 10-12 mm longa,

6-7 mm lata, apice oblique subtruncata angulo dorsali dentato vel rostrato, dorso et

apicem versus stellato-pubescentia vel tomentosa. Semina c. 2 x 2 mm, minute

verruculosa.

Transvaal. — Pietersburg: Chunies Poort Police Station, Meeuse 10352 (PRE,

rolo.!, BM, BOL, EA, K, L, LD, M, SRGH, isos.!). Potgietersrust : near Potgieters-

rust, Maguire 1499 (NBG). Waterberg: farm Mosdene near Naboomspruit, Galpin

M 23. Brits: Hartebeespoort, Lotsy and Goddijn 348 (L); Hekpoort, J. Phillips 519.

Pretoria: 18 m N.E. of Pretoria near Roodeplaat, Repton 4309; about 12 miles from

Pretoria on road to Zeekoegat (Roodeplaat), Repton 919 \ Pretoria North, Crawley

PRE no. 5181; Pretoria, Arcadia, Leendertz 463 (L, PRE); Pretoria, Curtis’ Hill,

Pole Evans 75; Pretoria, Esterhuysen h. no. 26321 (BOL); Fountains Valley, Repton

223; Hennopsrivier, Brernekamp TRV no. 29048. Krugersdorp: Waterval Kloof,

Mogg 20353 (J, PRE). Vereeniging: Kaalplaats, Mogg 10222. Klerksdorp: Klerks-

dorp, "'Convent'" 67 (GRA). Rustenburg. Zwartruggens Ridge, Sutton 827, 849;

Witkranskloof, Rose Innes 41. Marico: Zeerust, Leendertz h. no. 11312.

Orange Free State. — Kroonstad: Kroonstad, Pont 681; Fauresmith: Faure-

smith, Verdoorn 946, 1180, Verdoorn in herb. Henrici 2396, Smith 3980, Henrici 2008,

Leistner 1104 (KMG, PRE). Bloemfontein: Bloemfontein, Bolus 11047 (BOL),

Wasserfall 842 (NBG); Naval Hill, Potts 8022; Glen, Potgieter 60. District unknown:

” Sepani ”, Brierley 21 (BM).

Cape Province. — Aliwal North: Ruigtefontein, Theron A1795 (NH, PRE).

(All specimens, if not otherwise indicated, in PRE).

A low suflfrutex or soft shrub usually scantily branched (from the base or only

higher up), probably annual, covered on stems, petioles and pedicels with a short dense

tomentose or velutinous pubescence which is usually greyish or whitish and interspersed

with long thin patent hairs. Stems terete, ultimately somewhat woody with a rather

large pith and a thin somewhat wrinkled grooved bark. Leaves in outline suborbicular

cordate or broadly ovate-cordate to triangular-cordate, sometimes faintly 3-lobed

by projecting lateral lobules near the middle of the blade, acuminate or gradually

attenuate at the apex, rather regularly crenate-dentate to serrate but usually only

shallowly so, 2-7 cm, occasionally up to 15 cm long and 1 -3-5(-10) cm broad; upper

surface dark green, minutely adpressed-stellate pubescent, glabrescent, the lower surface

canescent, softly tomentose to velutinous with prominent nervation; petioles about as

long as the corresponding blades; stipules subulate, tomentose or velutinous. Flowers

axillary, solitary; pedicels in flower up to 7 cm long, in fruit up to 9 cm. Calyx cupuli-

487

form-campanulate, densely velutinous and the tube sometimes with additional patent

hairs, greyish, grey-green or canescent, deeply lobed, 9-12 mm long; the lobes oblong-

lanceolate to lanceolate, acute, shortly acuminate and/or shortly apiculate. Petals

yellow, 14-18 mm long, glabrous. Stamina! column stellate-pubescent. Fruit semi-

globose-subcylindric, 10-12 mm long and 15-20 cm in diam. Mericarps 10-18, 3-seeded,

10-12 mm X 6-7 mm, the apical edge somewhat convex, slanting upwards and outwards

and meeting the dorsal side at an acute angle, produced into a tooth or subulate up to

2 mm long awn; the ventral tooth small, the keel on the back and the apical edge with

a row of many-rayed stellate hairs flanked on either side by a zone which is rather

sparsely tomentose or stellate-pubescent with adpressed, smaller and fewer-rayed

stellate hairs. Seeds about x 2 mm, minutely verruculose-punctate.

This plant is obviously closely related to A. grantii, a coastal species, which it

resembles very much in the morphology of the mericarps. It differs in a number of

points such as in the leaf-shape which is usually abruptly acuminate in A. grantii from

a broad basal portion and hence more distinctly 3-lobed, in the leaf-margin which is

usually more entire in A. grantii and in the pubescence of the lower leaf-surface which

is very short and smoothly velutinous in A. grantii, more loosely stellate-velutinous in

A. piloso-cinereum, but mainly in the presence of long patent hairs which are never

numerous in A. grantii (and restricted to the very young parts and the apices of the

petioles), conspicuous in A. piloso-cinereum, and in the colour of the stems which are

not canescent and often dark purple in A. grantii, canescent in the other species anb

later brownish or greyish but never dark.

A. grantii is a perennial and A. piloso-cinereum apparently an annual, but this

difference is not always evident from herbarium specimens. A. piloso-cinereum is by

no means rare; the numerous cited specimens also indicate that it is wide-spread.

The specimens were lying in the herbaria usually under A. sonneratianum or “ A.

indicum ”.

It is a plant of rocky slopes in areas with a fairly low annual rainfall, found at

altitudes between 900 and 1,200 m and its ecology is, therefore, quite different from that

of A. grantii which is a plant of coastal bush and lowland forests below 300 m altitude,

not usually found on rocky soil but generally on alluvial deposits.

Pavonia transvaalensis (Ulbr.) A. Meeuse, stat. nov. — P. schumanniana Gurke var.

transvaalensis Ulbr. in Engl. Bot. Jahrb. 57: 178 (1921). Type of variety: Trans-

vaal, Magalakwin River, Schlechter 4270 (B, holo.f, PRE, iso.!).

P. schumanniana GUrke var. parviflora Schinz in Bull. Herb. Boiss. 2me ser. 3: 829

(1903). Syntypes of variety: Transvaal, Pretoria Distr., Rehmann 4185, 4365,

4938; Potgietersrust Distr., Rehmann 5492 (all in Z).

P. commutata Conr. in Kew Bull. 1908: 220 (1908); Schinz in Vtljschr. Naturf. Ges.

Zurich 68: 428 (1923); Burtt Davy, Man. Flow. PI. Transv. 2: 278 (1932), non

Garcke. Type: Transvaal, Pretoria, Conrath 42 (K, holo. !).

The status and synonymy of Pavonia commutata Conr. in relation to some other

species of Pavonia, especially P. clathrata Mast. (= P. schumanniana Giirke), was

cleared up and discussed by Schinz (1923), but this author overlooked the fact that the

epithet commutata was pre-occupied in Pavonia. The epithet selected from the two

varietal names cited above is not the oldest, but the epithet “ parviflora ” is inappro-

priate, whereas “ transvaalensis ” is very suitable for this species which has not been

recorded from outside the Transvaal.

A. Meeuse

202909—3

488

PLUMBAGINACEAE

Limonium

In Flora Capensis 4, 2: 419 (1906) Wright retained the generic name Statice for

the species from southern Africa. Sprague in Journ. Bot. 62; 267 (1924) reaffirmed

that Limonium should be restored to generic status and gave a fairly comprehensive

index to relevant literature. The result of the restoration of Limonium, excludes Statice

from the indigenous flora of southern Africa. However, for convenience of discussion,

existing names will be used in the following notes.

The distribution of Limonium species is generally accepted as being predominantly

maritime. It is interesting to note, therefore, that one variable species, L. dregeanum

(Presl.) O.K., has a wide distribution inland and extends to relatively high atlitudes

in the central Karoo. On the other hand no species occurs on our coast further east

than about Kentani in the Cape Province.

In working on the genus for the Flora of Southern Africa the usual crop of problems

had to be faced and I wish to thank very sincerely the Curators and Directors of the

several institutions which supplied material for study. These include K, UPSV, S,

SBT, G, M, BOL, GRA, STE, SAM and NBG.

As regards the generic description of Limonium, in none of the southern African

herbarium material dissected could I distinguish a measurable corolla tube and all

petals appeared to be free to the base and to have the filaments of the stamens attached

slightly above the base. The anthers of all material dissected were divided up to the

point of attachment of the filament and in no case was this appreciably less than half

the length of the anther. The ovary was invariably five-angled with five free styles,

one from the margin of each angle at the top.

An early problem was the identity of Statice perigrina Bergius, Descript. PI. Cap.

80 (1767), which has priority of publication over all other names, including S. purpurata

L. in his Mantissa, also published in 1767 (see Sprague in Kew Bull. 1929: 88). Boissier

in DC. Prod. 12; 667 (1848) regarded 5. perigrina Berg, as a synonym of S. rosea

Smith (1819) and Wright F.C. l.c. 420, followed suit but they chose to retain the epithet

rosea. Smith, however, when describing his S. rosea place, S. perigrina doubtfully

under S. purpurata L. thus indicating that he excluded it from his S. rosea. Photographs

and notes from Stockholm prove that S. perigrina does not agree specifically with

S. purpurata. One of the features of S. purpurata is the smoothness of the leaves,

whereas those of S. perigrina are in fact roughish on both surfaces, although Bergius

described them as glabrous on the upper surface and scabrid on the lower. My research

supports the taxonomy of Boissier and Wright but the epithet perigrina must be restored

to priority.

The position is complicated further by the presence in Malmesbury district, with

the above-mentioned species, of 5. longifolia Thunb. (1794), which was regarded by

Boissier (1848) as a variety of S. purpurata L. Unlike the others, S. longifolia is constant

in having dense adpressed hairs almost to the tips of the calyx ribs. A specimen,

Compton 19361, from Bellville Division, nearest S. purpurata L. has a few hairs near

the tips of the calyx ribs and another, Acocks 19785, from Clanwilliam, regarded by

me as a form of S. perigrina also has a few hairs towards the tips of the calyx ribs.

In this feature they indicate some relationship with S. longifolia, which has a wide

range of distribution along the Cape western districts.

With S. longifolia Thunb. in the picture, it seems that S. purpurata falls somewhere

between the more common and more widely spread species S. perigrina, a leafy shrub,

and S. longifolia a more tufted subacaulescent perennial.

It could be suggested that S. purpurata arose by hybridisation between S. longifolia

and S. perigrina but at this stage proof is lacking.

489

A broad view is taken of the variation within the species 5'. longifoUa Thunb.

which results in the loss of specific status of L. fergusonae Bolus. The distribution

range is from the coastal districts of the west via the mountainous region of Worcester

and Robertson to Riversdale on the south coast and omitting the intervening coastal

area of the Peninsula to Swellendam.

The type specimen of L. amoenum (C. H. Wright) R. A. Dyer, collected at Touws

River by Bolus (BOL 1080) has not been matched exactly by other collections from

neighbouring districts but several are regarded as specifically equal. A feature of the

main specimen on the type sheet is the straight scape with up to 15 sessile spikelets.

The scapes in most specimens are more branched and somewhat flexuous and with

only about five spikelets. The type was obviously grazed short before it produced

the new inflorescences and it is assumed that this caused the variation in habit. This

view is supported by the habit of an isotype in the Kirstenbosch herbarium.

L. decumbens (Boiss.) O.K. was based on a specimen collected by Drege, no. 9374,

without exact locality. Boissier stated that it was probably a monstrous form of an

incompletely known species. It has fascicles of leaves on the inflorescence. No sub-

sequent collection has been found to agree with it exactly. It is said to be distinctive

in the pubescence of short tufted hairs on the scape and spikelets, while the calyx is

described as quite glabrous. It seems that it is mainly the proliferous character which

distinguishes it from L. equisetum (Boiss.) R. A. Dyer but in view of the circumstances

no good purpose would be served by speculating further on the relationship between

them.

As regards L. scabrum (Thunb.) O.K., I tried to establish a clear division between

it, L. corymbuhsum (Boiss.) O.K., L. avenaceum (C. H. Wright) R. A. Dyer and L.

penicUlatum Adamson, but failed to do so after the examination of a large number of

specimens. However, because of the wide difference between extreme forms, three

varieties are recognized.

On the other hand several specimens collected in S.W.A. in the vicinity of Luderitz

and Angra Pequena and previously regarded as forms of L. scabrum appear sufficiently

distinct to warrant specific rank. They differ from L. scabrum in the almost completely

membranous bracts on mature scapes and their branches, in having 3-4-flowered

spikelets and pedicellate flowers. In these latter characters and the pedicellate flowers

the species shows an affinity with L. dregeanum but again the bracts are distinctive.

The specimens are described under the new name L. membranaceum.

Other unusual specimens associated with L. scabrum and producing tufts of leaves

on the flowering scapes, have been described under the name L. foliosum R. A. Dyer.

Compton (18168) refers to the tufts of leaves within the inflorescence as a form of

“ vivipery ”. This feature is found occasionally in other species also, such as L. scabrum.

Had L. kraussianum (Buchnig ex Krauss) O.K. been recorded with L. scabrum, I would

have been tempted to suggest that hybridisation had been at play to produce L. foliosum.

In the case of 5. linifolium (L.f.) O.K., two varieties are recognized. The identity

of the specimen described by L.f. has not been confirmed, but the speciment in Thun berg’s

herbarium (?type) matches several specimens collected later in the Port Elizabeth-

Uitenhage districts. But this, the probable nomenclatural type form has a limited

distribution, and the second variety given the name maritimum is far more widespread

and occurs further east in the Cape (Transkei) than any other species. The two varieties

cover much the same field of distribution as L. scabrum. The inter relationship

between L. linifolium and L. scabrum is obscure but there are specimens, for instance

Britten 5022, and others collected by South and Compton, at Port Alfred (Kowie) which

seem to have some characters of both, with a greater leaning to L. scabrum.

The circumscription of L. dregeanum (Presl.) O.K. and L. pedicellatum (Wallr.

ex Boiss.) O.K. entailed similar difficult decisions. Specimens described by Boissier

under the name Statice pedicellata, were at one time considered to be specifically distinct

490

from L. dregeanum because of the differences in the branching of the scapes and general

habit, one tufted and the other mainly cushion-shaped, but after several changes in

opinion L. dregeanum alone has been maintained.

The taxonomy of the genus Limonium in southern Africa seems to bristle with

problems, most of which require intensive field work to crystalize them out, let alone

discover their explanation. Hybrid populations are suspected as occurring frequently

on the south western coast. L. anthericoides (Schlechter) R. A. Dyer seems unique

among the species. Although it exhibits a considerable degree of morphological

variability, it does not seem to have been involved in any of the reproductive problems

of any of its several neighbours in the rich Caledon, Bredasdorp, Swellendam region.

The new names and name changes consequent on the above notes are summarized

below in alphabetical order.

Limonium depauperatum R. A. Dyer, comb, nov., stat. nov., Statice equisetina var.

depauperata Boiss. in DC. Prod. 12: 658 (1847); Wright in F.C. 4, 1: 422 (1906)’

Type: Cape; Burchell 512 (G, lects.).

L. linifolium (L./.) OK. var. linifolium.

L. linifolium var. maritimum {E. and Z. ex Boiss.) R. A. Dyer, comb. nov.

Statice linifolia var. maritima E. and Z. ex Boiss. in DC. Prod. 12: 657 (1848); Wright

in F.C. 4, 1 : 421. 5”. linifolia var. brachyphylla Boiss. in DC. Prod. l.c. 657;

Wright in F.C. l.c. 421.

L. longifolium (Thunb.) R. A. Dyer, comb. nov.

Statice longifolia Thunb. Prod. 54 (1794). Type: Cape: Swartland, Thunberg (UPSV,

holo.).

S. purpurata L. var. longifolia Boiss. in DC. Prod. 12: 667 (1848); Wright in F.C. 4,

1: 420 (1906) as to citation of S. longifolia Thunb.

Limonium fergusonae L. Bolus in J. S.A. Bot. 24: 124 (1934). Type: Riversdale,

Ferguson BOL 20081 (holo.).

Limonium membranaceum R. A. Dyer, sp. nov., L. scabro (Thunb.) O.K. affine, sed

bracteis scapi et ramulorum scapi fere omnino membranaceis, spiculis 3-4-floris differt.

Perenne caespitosum demum basi lignosum. Folia obovata vel oblanceolata vel

lineari-oblanceolata, 2-4 cm longa 4-10 mm lata, obtusa, supra tuberculis centro

punctato-impressis scabra, subtus levis vel tuberculis similibus paucis. Scapi nonnulli,

tuberculis centro punctato-impressis scabri, patentes 10-20 cm longi, erecti etiam

infra medium ramos steriles articulatim multifidos primum breves sursum sensim

ampliatos edentes, superioribus floriferis dense fastigiato-corymbosis, articulis omnium

densis brevibus strictis. Spiculi 3-4-flori, bractea exterior 3 • 5-4 mm longa, obtusa,

margine membranacea glabra, pedicellis persistentibus 0-5-0 -75 mm longis, truncatis.

Calyx 4 mm longus, subcylindratus glaber 5-costatus, limbo membranaceo, lobis 5,

c. 5 mm longis. Petala caerulea oblineari-lanceolata, plus 1 cm longa.

Endemic in southern South West Africa.

South West Africa. — Luderitz: Pole Evans W \92>55', Kinges 2031; Merxmuller

2251 (PRE, holo. M, iso.). Angra Pequena: Marloth 1160; Galpin and Pearson 7490.

This species is allied to L. scabrum (Thunb.) O.K. and is distinguished by the almost

completely membranous bracts on the scape, by the 3-4-flowered spikelets and pedicellate

flowers. It is distinguished from L. dregeanum (Presl.) O.K. also by the membranous

bracts, by the dense asperities on the scape, and by the broader leaves with dense

asperities on the upper surface.

L. perigrinum (Bergius) R. A. Dyer comb. nov.

Statice perigrina Bergius, Descr. PI. Cap. 80 (1767) excl. syn. Type C.B.S. Grubb.

(SBT, holo) [collected by Auge fide Thunb. FI. Cap. Vll (1823)].

491

L. scabrum Thunb. var. avenaceum (C. H. Wright) R. A. Dyer.

Statice avenacea C. H. Wright in FI. Cap. 4, 1 : 423 (1906). Type: Bredasdorp, Ratels

River Mouth, Bolus 8576 (K, holo. BOL iso.!).

Limonium avenaceum (C. H. Wright) R. A. Dyer in Kew Bull. 1932: 155.

L. scabrum (Thunb.) O.K. var. corymbulosum (Boiss.) R. A. Dyer.

Statice corymbulosa Boiss, in DC. Prod. 12: 658 (1848). Type: Camps Bay, Krauss

(G, holo I).

Limonium corymbulosum (Boiss.) O.K. in Rev. Gen.: 2: 395 (1891).

L. penicillatum Adamson in S.A., Journ. Bot. 7: 202 (1941); FI. Cap. Penin. 666 (1950)

Type: Cape; Chapmans Peak, Adamson 859 (BOL holo ).

L. scabrum {Thunb.) O.K. var scabrum.

L. j'cuZj/'WOT (Thunb.) O.K. Rev. Gen. PI. 2: 396(1891). Type: Cape; Thunberg {\}¥Sy

holo.).

Statice scabra Thunb., Prod. 54 (1794).

R. A. Dyer

SCROPHULARIACEAE

Sutera dentatisepala Overkott, sp. nov., a S. cooperi Fliern planta annua minore, folds

non rigidis neque pallidis neque cordati-rotundatis, tubo corollae longiore lobis

emarginatis; a 5". pristisepala Hiern folds non pinnatifidis, floribus maioribus non

purpureis, lobis emarginatis differt.

Radix annua, fibrosa, griseo-brunnea. Herba foetida (teste coll.) ad 22 mm alta

e basi multiramosa. Caules decumbentes vel ascendentes, foliosi, basin versus saepe

radicantes, leviter quadranguli, superne fere teretes et florigeri; ubique pilis satis longis,

perspicuis, glandulosis, fusci-capitulatis et paucis pilis simplicibus obtecti. Folia

membranacea, opposita, omnia breviter petiolata, petiolis planis ad 6 (-12) mm longis,

glandulosi-pilosis, in axillis saepe ramulos parvos foliosos gerentia, rotundati-triangulata,

basi late cuneata, margine irregulariter dentata vel leviter pinnatisecta, utrinque glandu-

losa, ad 10 (-20) mm longa, ad 8 (-17) mm lata, nervis alternantibus pinnatis, superne

immersis, subtus prominentibus, pilos conspicuos gerentibus. Internodia 12-27 mm

longa. Flores plerumque alternantes, albi vel rosei, lutei-ocellati in axillis bractearum

foliis similium quamquam minorum. Pedicellus ad 5 mm longus, calyce brevior.

Calyx ad 7 mm longus, ut caules intus et extus glandulosi-pilosus, ad basin fere divisus.

Sepala cuneata, apice 1-7 dentata vel mucronata. Tubus corollae calyce 2- 5-plo longior,

ad 18 mm longus, extus glandulosi-pilosus, apice leviter ampliatus. Limbus pro rata

latus, subtus glandulis sessilibus insitus, lobis obcordatis, leviter vel distincte emarginatis,

ad 5 • 5 mm longis, ad 3 mm latis. Faux glandulosa et pilosa. Stamina inclusa. Antherae

reniformes inter se aequales, superiores paulo minores. Thecae confluentes. Filamenta

glandulosa, superiora breviora, ad 1-2 mm, inferiora ad 2 mm longa et in tubus

decurrentia, omnia in parte tertia superiore tubi inserta. Ovarium angustum, ad 2 - 5 mm

longum, apice dense glandulosum. Stylus crassus, plerumque glaber, apice brevissime

bilobatus, longe persistens. Fructus non vidi.

Natal. — Bergville District: on boulder bed of Tseketseke River, Cathedral Peak

Forestry Station, 6,700 feet, Killick 1827 (PRE, holo.). Estcourt District: Giants

Castle, Symons (Transvaal Museum No. 25157, PRE).

Basutoland. — Likoloberg, grassy gravelly patch on hill, 9,300 feet, Guillarmod

716 (PRE); Mamalapi, streambank, 8,000 feet, Guillarmod 677 (PRE); Mamalapi,

streambanks, 9,000 feet, Compton s.n. (NBG); Machochi, Seligman s.n. (BM); without

precise locality. Staples 259 (PRE).

flower, X 3; C, calyx and

493

Decumbent or ascending, probably annual herb, about 25 cm high, much branched

from the base. Leaves opposite, rotundate-triangulate, broadly cuneate at the base;

petioles shorter than the leaves, sometimes bearing short foliate branchlets in the axils.

Bracts leaflike, but smaller. Flowers white or mauve with orange centres, alternate,

axillary, subterminal, not very numerous; peduncles shorter than the bracts; calyx

deeply 5-cleft, segments cuneate at the base, with one to seven teeth at the ends, as

branches and leaves covered on both sides with short, gland-tipped hairs and globose

sessile glands; tube about 18 mm long, glandular puberulous outside, lobes ± emargi-

nate; Stamens and style enclosed in the tube; filaments glandular, the upper one shorter

than the lower ones. Ovary small, with numerous sessile glands; style stout, persistent.

O. OVERKOTT

TURNERACEAE

Turnera thomasii (Urb.) Story, comb, et stat. nov. T. ulmifoUa L. var. thomasii Urb. in

Engl. Bot. Jahrb. 25 Beibl. 60: 11. Loewia thomasii (Urb.) Lewis in Kew Bull.

1953: 431.

Turnera oculata Story, sp. nov., T. ulmifoliae L. et T. thomasii (Urb.) Story a ffinissed

ita dilferens: a T. ulmifolia ramis rigidibus, a T. thomasii petalis flavis, indumento

breviori, antheris multo crassioribus.

Frutex erectus tenuis ad 2 m altus, foliis ramulisque junioribus villis simplicibus

stellatisque dense pubescentibus. Folia alternantia simplicia; lamina ad 3 cm longa

2 cm lata, obovata vel late elliptica, apice rotundato vel acuto, praeter basin cuneatam

serrata, pinninervata, nervis utrinque plus minusve 6, basi saepe glandulis prominentibus

1 vel 2; petiolus ad 1 cm longus; stupuli minuti. Inflorescentia terminalis. Bracteae

2 lineares 1 cm longae, superne profunde canaliculatae. Flos axillaris solitarius, pedicello

petiolo adnato. Calyx intus paene glaber; tubus 1-2 cm longus 3 mm diametro,

nervis 10; lobi tubo calycino interdum aequilongi saepius longiores, acuminati, basi

3 mm lati, interdum margine tenue pellucido ad 2 mm lato. Corolla in faucibus tubi

calycini inserta; petala alterna ac lobi calycini, obovata, apice acuto, imbricata, 2-5 cm

longa, in parte latissima 1 ■ 8 cm lata, superne laete flava, inferne saturate rubri-brunnea,

glabra. Androecium staminibus 5, fertilibus; filamenta lobis calycinis opposita, circiter

3 mm infra calycis fauces inserta, villis basi paucis alibi glabra, 2-2 cm libera, inferne

1 cm tubo calycino adhaerentia; antherae c. 7 mm longae, basi emarginatae, paullo

super basin dorsifixae summum filamentum saepientes. Gynoecium stylis 3 glabris

liberis quam stamina paullo longioribus; stigma fimbriatum; ovarium superius

cylindratum apice obtuso, in siccitate 5-costatum, ovulis ad 120 in placenta stipatis.

Capsula 3-valvata, valvis ovatis 9 mm longis in parte latissima 5 mm latis, placenta

parietali; funiculus prope basin seminis affixus; semina matura curvata, 4 mm longa;

arillus incisus membranaceus, semen superans.

South West Africa. — Kaokoveld: sandy bed of watercourse 12 miles south of

the Kunene River, latitude 17° 22' longitude 12° 30' Story 5778; sandy banks of

Kunene River at Otjinungua, de Winter and Leistner 5770 (PRE, holo.).

An erect slender shrub up to 2 m high, younger parts densely pubescent with

simple and stellate hairs, branches sometimes abbreviated, with crowded small leaves,

afterwards deciduous or elongating and permanent. Leaves velvety, dull light green,

simple, very variable in size; lamina up to 3 cm long and 2 cm broad, obovate to broadly

elliptic, rounded or acute at the apex, apiculate-serrate except at the cuneate base,

alternate, divergent, pinnately nerved with about six pairs of nerves more prominent

below, often with one or two conspicuous hollow glands (2 X 1 mm) on the margin

at the base; petiole up to 1 cm long; stipules minute, free, exuding a drop of resin

494

from the glandular tip. Inflorescence terminal; bracts 2, linear, 1 cm long and deeply

channelled adaxially, with sometimes one or several glands like those found on the

leaves but much smaller. Flower axillary, solitary, with pedicel fused to petiole and

thus at first sight apparently sessile and borne at the distal end of the petiole. Calyx

almost glabrous within; tube about 12 mm long and 3 mm in diameter, widening at

the throat, and with 10 veins of which five continue down the centres of the lobes and

five fork between the bases of the lobes to form two marginal veins; lobes sometimes

as long as the calyx tube, but more often 3 or 4 mm longer, acuminate, 3 mm wide at

the base, sometimes with a thin translucent margin up to 2 mm wide. Corolla inserted

in the throat of the calyx tube; petals alternating with the calyx lobes, obovate, acute

at the apex, abruptly narrowed below, imbricate, 2-5 cm long, 1-8 cm wide at the

widest part, bright yellow above, dark red-brown towards the base, glabrous. Androe-

ciiim of 5 stamens, all fertile; filaments inserted about 3 mm below the throat of the

calyx tube and alternating with the petals, glabrous except for a few hairs near the

base, free for about 2-2 cm, below this with thin transparent margins adherent to the

calyx-tube and a thicker central part free to the base, this fixed portion being about

1 cm long; anthers 7 mm long, narrowly oval, emarginate at the base, dorsifixed a

little above the base and enclosing the tip of the filament; thecae dihiscing longitudinally.

4, leaf-glands, X 10; 5, flower, x 1-5.

495

Gynoecium with 3 glabrous free styles a little longer than the stamens; stigma fimbriate;

ovary superior, cylindric, obtuse, 5-ridged when dry, ovules up to 120, crowded on

the placenta. Capsule 3-valved; valves ovate, 9 mm long, 5 mm broad at the broadest

part, thickened on the margin, reticulate within, with parietal placenta; funicle attached

just above the base of the seed and remaining on the placenta after the seed is shed,

thickened near the hilum; mature seeds curved, 4 mm long, with hilum near base on

concave side, abortive seeds conspicuous by reason of the rudimentary aril; aril lobed,

incised, membranous, roughly triangular, with one lobe overtopping the seed on the

concave side and two almost encircling the base.

This plant was first recorded in 1956 in a sandy gully about 12 miles south of the

Kunene River in South West Africa at longitude 12° 30’, latitude 17° 22’, and has so

far been found only in that vicinity. It is not common and usually occurs in small

patches of a dozen or so. It has been found in bloom in May and August, conspicuous

because of the large bright flowers and silvery leaves and erect habit. As one of the

several young plants collected in the field has survived a severe winter in Pretoria and

seems well established, it is likely that Turnera oculata could be cultivated fairly widely

in the Union.

The plant runs to Turnera in Thonner’s key and its other characters are also strongly

in support of its inclusion under Turnera, and I therefore do not feel justified in placing

it in the closely related genus Loewia, even though Loewia is upheld in the Flora of

Tropical East Africa (Lewis 19/2/54, l.c.: 1 et seqq.). The two genera are separable

as follows: —

1. Turnera has a 10-nerved calyx-tube, Loewia has the calyx-tube traversed by

35-40 bundles of very slender vessels (Urban 1897, Ann. R. Inst. Bot. Rom. 6: 189).

2. Turnera has no secretory hairs, Loewia has tubercles secreting resin (Urban l.c.).

(These “ tubercles ” are resin-like nodules, probably swollen hair-bases.)

3. Turnera has the stigma fimbriate — “ capillaceo-multifidum ” — (Linn. Gen. ed. 2:

105 para. 297), Loewia according to Urban (l.c.), Thonner, and Hooker (Icon. 3015)

has it capitate or with the margin sub-entire. (Lewis says of L. tanaensis that the stigma

is fimbriate but shows it lobed in the sketch.)

In all these characters the plant here described agrees with Turnera. Additional

reasons for keeping it separate from Loewia are as follows: —

1. In Loewia the calyx-tube is about two-thirds the length of the whole calyx, in

this plant it is less than half the total length.

2. Loewia has the seeds bi-seriate on the placenta (Urban l.c.), this plant has

them crowded.

3. Turnera may have the pedicels fused with the petioles (R. Hort. Soc. Die. of

Card. 1951 Vol. 4: 2170; Hooker’s Jour. Bot. 4, 1842: 115), in Loewia they are free.

This plant has them fused.

4. Turnera may have conspicuous hollow glands on the basal margins of the leaves,

in Loewia they are absent. This plant has them, though not consistently so.

Two inconsistencies in the literature should be pointed out. They are as follows: —

1. Urban (l.c.) says that in Loewia the margin of the aril is sub-entire, but Hooker

(Icon. 3015) says of Urban’s species L. tanaensis that the margin is lacerate. Hooker’s

statement has been checked at Kew and found to be correct.

2. Lewis (Flor. Trop. E. Afr. 1954) says the pits on the seed of Loewia are two-

pored, Urban (l.c.) says they have no pores. The specimens at Kew indicate that

Urban is correct.

496

For the following reasons I am including Loewia thomasii (Urb.) Lewis under

Turnera: —

1. It has no secretory hairs.

2. The calyx is tubular for less than half the total length.

3. The pedicels are fused with the petioles.

4. There are conspicuous hollow glands on the basal margins of the leaves.

5. The calyx-tube is 10-nerved.

6. The stigma is fimbriate (described incorrectly by Urban as very shortly lobed).

The available records indicate that Turnera thomasii has been collected only twice.

The first collection is by F. Thomas (No. 47) from Witu in the Lamu district of Kenya,

and is the type. The holotype was destroyed in Berlin, and there remains only the

isotype, which is at Kew. The second collection is by P. Bally (No. 2092) at Mahoney

Road, near Muddo Gasha in the Northern Province of Kenya, and is represented in

the Kew and Nairobi herbaria.

I am grateful to Mr. W. Marais, who went through the material and literature at

Kew on my behalf and gave me much useful information.

R. Story

497

The Virus Diseases of Crotalaria, Glycine

and Medicago Species

by

Patricia J. Klesser

Division of Plant Pathology

This report concerns the naturally occurring virus diseases of the following plants: —

Crotalaria juncea (sunnhemp), C. spectabilis. Glycine max (soybean), G. javanica

and Medicago sativa (lucerne or alfalfa).

Throughout South Africa, these genera are used either for fodder, as cover crops,

or, to a lesser extent, for oil extraction.

Further details of some of the diseases found on these plants, will be found in the

reports on bean and broad bean virus diseases, as the viruses were apparently related

to those groups.

Methods and materials

The viruses were isolated from plants collected mainly in the Transvaal, but also

from the Stellenbosch district and from near Salisbury, Southern Rhodesia.

The standard test plants were: Phaseolus vulgaris var. Canadian Wonder, Vida

faba var. Aquedulce, Pisum sativum var. Greenfeast and Vigna unguiculata, var. Dr.

Saunder’s Upright.

Carborundum powder was used for the mechanical sap inoculations, and Aphis

craccivora was used in the insect transmission tests.

There are three separate sections, dealing with the viruses of (A) Crotalaria spp.

(B) Glycine max and (C) Medicago sativa and Glycine javanica.

Each section has its own bibliography and plates.

CROTALARIA VIRUS DISEASES

Crotalaria juncea, sunnhemp, is cultivated extensively in South Africa as a green

manure crop. Crotalaria spectabilis is also used for this purpose but on a smaller scale.

Both species are subject to natural infection by several viroses in the field, and are

also susceptible to many others when inoculated artificially.

With two exceptions, these virus diseases have been described in previous papers,

as they were obviously related to other viruses viz. the bean local chlorosis, the bean

chlorotic ringspot, the bean necrosis and the bean necrotic speckle viruses. Also, a

strain of the broad bean mosaic virus was found on naturally infected C. spectabilis.

The two exceptions, which will be described in detail in this paper, are the alsike

clover mosaic virus and the white clover mosaic virus complex. Both are well-known

elsewhere, but are recorded for the first time in South Africa.

The symptoms caused on most host plants, by what is considered to be the white

clover mosaic virus, are very similar to those described by Zaumeyer and Wade (19).

No attempt has been made to split the components [see Johnson (13)], but the virosis

as found in the field is described as a whole.

498

The symptoms caused by the several viruses on C. juncea are very similar — the

differences are found in the degree of malformation, mottling or necrosis rather than

in the type of symptom. They are only differentiated by the effect of the virus on

other plants.

On C. spectabilis the symptom picture does vary — some viruses cause vivid yellow

streaks, while others result in mottling with necrosis.

Nevertheless, they are not characteristic enough for certain identification, and

alternate hosts must be used.

Review of literature

The first report of probable virus infections of Crotalaria species was that of

Schwartz, 1927, who described a witch’s broom and crinkle disease of C. juncea and

C. anagyroides in Java (17).

In 1931, Illingworth (11) noted a mosaic of Crotalarias in pineapple fields in Hawaii,

and Cook (5, 6) worked on a mosaic disease of C. striata in Puerto Rico. Other mosaic

diseases of Crotalaria spp. were found in Japan by Fukushi (8), and in Trinidad by

Dale (7).

Raychaudhuri (15) noted a mosaic disease which is confined to sunnhemp only

(C. juncea) and Capoor (4) described the Southern sunnhemp mosaic virus which with-

stands very high temperatures and ageing.

Gates (9) reported mosaic and “ little leaf ” diseases of C. spectabilis, Jenson

(12) worked on the aphid transmission of a mosaic of C. incana and Raychaudhuri

and Pathanian (16) found a mosaic of C. striata.

Viruses causing a witch’s broom symptom on Crotalaria spp. were recorded by

Johnson and Lefebvre (14) who were able to transmit the virus to broad bean; and

Hadiwidjaja (10) who transmitted the virus by grafting and by Cuscuta.

Bose and Misra (3) reported a graft-transmissible virus which caused structural

abnormalities of sunnhemp fiowers. This was later proved to be due to two viruses —

one sap-transmissible, and the other, probably tobacco leaf-curl transmitted by Bemisia

t abaci (1).

According to Zaumeyer (18), mosaic symptoms can be induced on Crotalaria

spp. by artificial inoculation with several legume viruses viz. alsike clover mosaic viruses

1 and 2 and pea mosaic viruses 4 and 5.

ALSIKE CLOVER MOSAIC VIRUS 1 Zaumeyer and Wade.

Natural source of virus: Crotalaria juncea.

The young leaves showed a chlorotic speck mottle and they had wavy margins.

The middle leaves had constrictions across the laminae, and irregular-shaped dark

green islands on a light green background. The lower leaves were normal in shape

but had dark green veinbands or midrib. Fig. lA.

Physical properties: Thermal inactivation point 62-64°C. Longevity in vitro

1-2 days. Dilution end point 1 :3000-l :5000.

Transmission: Mechanical sap inoculation. Aphis craccivora.

Host range: Crotalaria juncea L., C. spectabilis Roth., Dolichos lablab L., Glycine

javanica L., G. max (L.) Merr., Lathyrus odoratus L., Lupinus albus L., L. luteus L.,

L. mutabilis Sweet., Medicago lupulina L., Melilotus officinalis Willd., Phaseolus acutifolius

Gray var. latifolius Freem., P. lunatus L., P. mungo L., P. vulgaris L., Pisu/n sativum

L., Trifolium hybridum L., T. incarnatum L., T. pratense L., Vida faba L., Vigna sesqui-

pedalis (L.) Wight., V. unguiculata (L.) Walp., Voandzeia subterranea Thouars.

Reaction of susceptible species

Crotalaria juncea.

Local. Necrotic lesions may develop.

Systemic. In nine days chlorotic spots develop on the young leaves. Later leaves

have a mottle or mosaic with dark green blisters or veinbands. These leaves are

puckered, malformed and elongated. There may also be necrotic specks. Fig. IB.

The plant is stunted and rosetted.

C. spectabilis.

Local. Chlorotic spots may develop.

Systemic. There are chlorotic spots on the young leaves in 19 days. Later leaves

are chlorotic with dark green veinbands and there is a necrotic sheen. These leaves

are malformed and puckered.

Dolichos lab lab.

A symptomless carrier.

Glycine javanica.

Local. There are chlorotic spots in seven days.

Systemic. There may be a chlorotic spotting or mottle.

G. max.

Local. Chlorotic spots develop in seven days.

Systemic. There may be chlorotic areas, but usually the virus is latent.

Lathyrus odoratus.

Local. No reaction.

Systemic. The leaves are rolled and mottled.

Lupinus albus var. Sweet.

Local. The leaves drop after a severe necrosis in three days.

Systemic. In 16 days the young leaves develop a vein clearing and spotting and

they remain folded, and have wavy margins. Later leaves are mottled and malformed

and may have necrotic specks. The plant is rosetted and many leaves drop.

L. albus var. Bitter.

Local. No reaction.

Systemic. The young leaves remain folded and have curled tips. Later leaves

are mottled or generally chlorotic and the plant is rosetted.

L. luteus.

Local. No reaction.

Systemic. The young leaves remain folded and have wavy margins. Chlorotic

spots develop in 20 days. Later leaves are small, malformed (stringlike) and generally

chlorotic. There may be necrosis and many leaves drop. The plant is rosetted.

L. mutabilis.

Local. No reaction.

Systemic. The leaves are malformed and stringlike, and very chlorotic. The

plant is rosetted.

Medicago lupulina. •

Local. No reaction.

500

Systemic. After a vein clearing there are vivid chlorotic streaks and areas, and

the leaves are crinkled. Fig. 1C.

Melilotus officinalis.

Local. No reaction.

Systemic. The young leaves develop chlorotic spots and streaks. Later leaves

are mottled, small and malformed.

Phaseolus acutifolius.

Local. In six days there are chlorotic spots and necrotic veins.

Systemic. The young leaves develop a vein clearing and then necrosis in the veins.

Later leaves are mottled.

P. lunatus.

Local. Chlorotic ringspots develop in six days.

Systemic. There is a chlorosis of the veins which soon becomes necrotic.

P. mungo.

Local. There are chlorotic spots in eight days.

Systemic. There is a chlorotic spotting on most leaves.

P. vulgaris var. Canadian Wonder.

Local. No reaction.

Systemic. A chlorotic network develops on the young leaves, and later ones

have a chlorotic spotting and then a mottle. There are dark green areas and veinbands

and the leaves are puckered,

var. Haricot.

Local. No reaction.

Systemic. After a chlorotic network and spotting, the leaves develop a mottle.

The leaf surface is puckered and many leaves drop. There may be a slight malformation.

Pisum sativum.

Local. No reaction.

Systemic. The young leaves develop a chlorotic network in 1 5 days. Later leaves

have chlorotic spots or a mottle, and are malformed. The old leaves have a white

etching. Fig. ID. The tendrils are abnormally curled, and the plant is stunted.

Trifolium hybridum.

Local. No reaction.

Systemic. There is a vein clearing and chlorotic spotting followed by a chlorotic

streaking.

T. incarnatum.

Local. There may be dark green rings on the chlorotic leaves.

Systemic. The young leaves show a vein clearing after six days, then chlorotic

spotting or vein flecking. Later leaves have a mosaic and are crinkled and rosetted.

T. pratense.

Local. No reaction.

Systemic. There may be chlorotic streaks or the plant may be a symptomless

carrier.

Vida faba.

Local. No reaction.

Systemic. Only a diffuse chlorotic mottle develops.

501

Vigna sesquipedalis.

Local. There are chlorotic spots in seven days.

Systemic. In 15 days the young leaves develop a vein clearing and chlorotic

spotting. Later leaves are mottled.

V. unguiculata.

Local. Chlorotic spots may develop.

Systemic. There may be a vein clearing follow'ed by a chlorotic mottle, or the

leaves may be symptomless.

Voandzeia subterranea.

Local. There are chlorotic areas in seven days.

Systemic. The symptoms are masked.

IDENTIFICATION.

This virus is very similar to the alsike clover mosaic virus 1 Zaumeyer and Wade

(19). The only difference is that on the broad bean, Zaumeyer’s virus induces a severe

mottle and malformation, whereas this virus causes only a diffuse mottling. This diffe-

rence may be due to temperature or varietal factors.

Nevertheless, it is considered the same as, or closely related to the alsike clover

mosaic virus 1.

WHITE CLOVER MOS.\lC VIRUS Zaumeyer and Wade.

Natural source of virus; Crotalaria spectabilis.

The younger leaves showed a chlorotic mottle, and the older ones, a dark green

vein banding. There were necrotic stem streaks. Fig. 2C.

Physical properties: Thermal inactivation point 56-58 °C. Longevity in vitro

1-2 days. Dilution end point 1:1000-1:2000.

Transmission: Mechanical sap inoculation.

Host range: Arachis hypogaea L., Crotalaria juncea L., C. spectabilis Roth.,

Dolichos lablab L., Glycine javanica L., G. max (L.) Merr., Lathyrus odoratus L., Lupinus

albus L., L. luteus L., L. mutabilis Sweet., Medicago lupulina L., M. sativa L., Melilotus

officinalis Willd., Phaseolus acutifolius Gray var. latifolius Freem., P. lunatus L., P. mango

L., P. vulgaris L., Pisum sativum L., Trifolium hybridum L., T. incarnatum L., T. pratense

L., T. repens L., Vida faba L., Vigna sesquipedalis (L.) Wight., V. unguiculata (L.)

Walp., Voandzeia subterranea Thouars.

Reaction of susceptible species

Arachis hypogaea.

Local. No reaction.

Systemic. Chlorotic spots develop on the young leaves; on later formed ones,

there is a chlorotic streak mottle.

Crotalaria juncea.

Local. Within a week there are chlorotic spots and necrotic veins. The leaves

soon drop.

Systemic. In nine days the young leaves develop a vein clearing and chlorotic

spotting. Later leaves are small and crinkled, and have a mosaic and a necrotic sheen.

There are often necrotic stem streaks which cause the plant to curve to one side.

502

C. spectabilis.

Local. No reaction.

Systemic. The young leaves develop a chlorotic flecking and the next formed

have a mottle. Later leaves develop a vein banding with a slight necrosis which causes

a crinkling of the leaves.

Dolichos lablab.

Local. Chlorotic spots develop in 5-6 days.

Systemic. No reaction.

Glycine javanica.

A symptomless carrier.

Glycine max.

Local. Small necrotic specks with chlorotic haloes develop in 10 days.

Systemic. At the same time the young leaves develop a vein-clearing and chlorotic

spotting. Later leaves have a mottle with necrosis developing in the chlorosis. This

results in a puckering of the leaf surface.

Lathyrus odoratus.

Local. No reaction.

Systemic. After two weeks most leaves develop a chlorotic streak mottle and

they are rolled.

Lupinus albus.

Local. In nine days necrotic specks and veins develop, leaving a chlorotic back-

ground. The leaflets soon drop.

Systemic. The young leaves show a veinclearing and spotting with necrotic specks.

Later leaves are mottled and rolled, and the necrosis spreads until most leaves have

dropped leaving the rosetted petioles. Any secondary shoots have small malformed

leaves with dark green blisters.

L. luteus.

Local. In 10 days there are chlorotic spots.

Systemic. At the same time the young leaves develop a chlorotic network or

vivid rings. Later leaves are mottled and malformed, and they remain folded. The

older leaves have chlorotic spots with necrotic rings. The plant is severely stunted

and rosetted.

L. mutabilis.

Local. No reaction.

Systemic. The leaves curl up, and are small and chlorotic. The plant is rosetted

and stunted.

Medicago lupulina.

Local. No reaction.

Systemic. The leaves show chlorotic streaks and a dark green midrib. The plant

is stunted.

M. saliva.

Local. No reaction.

Systemic. The young leaves develop chlorotic spots in three weeks. Subsequent

ones show almost yellow spots or irregular areas.

Melilotus officinalis.

Local. There are dark green rings on chlorotic leaves.

503

Systemic. The young leaves show diffuse chlorotic spots. Later ones develop

an irregular chlorotic mottle.

Phaseolus acutifolius.

Local. There is no reaction but the leaves drop.

Systemic. In 8-10 days there is a veinclearing and curling of the young leaves.

Later ones are small and malformed, and have dark green blisters.

P. lunatus.

Local. In 10 days there are necrotic specks with chlorotic haloes, and a veinal

necrosis.

Systemic. The first and second trifoliates develop necrotic rings, and the young

leaves have chlorotic specks with necrosis setting in.

P. mungo.

Local. In 10 days there are necrotic lesions.

Systemic. Some leaves have chlorotic specks which later become necrotic.

P. vulgaris var. Canadian Wonder.

Local. There are necrotic lesions or ringspots and a slight veinal necrosis.

Systemic. No reaction,

var. Haricot.

Local. Within a week there may be small chlorotic spots with necrotic rings, or

only a veinal necrosis.

Systemic. The young leaves develop a veinal necrosis which spreads to the petioles

and stem, and the plant collapses.

vars. Black Wonder, Long Tom, S.A. Black and White and Tendergreen

develop large local necrotic spots and veins in three days. There are also areas of

almost yellow chlorosis on which the necrotic lesions are vividly marked. Fig. 2B.

var. Idaho Refugee has only a local necrotic vein etching.

Pisum sativum.

Local. There are no lesions, but in 10 days the leaves have become necrotic and

wilted.

Systemic. After a veinclearing and spotting of the young leaves, the next formed

develop a mosaic. Necrosis sets in, in the chlorotic areas and in the veins, and the plant

collapses. Secondary shoots have small, folded and mottled leaves, and the tendrils

are abnormally curled.

Trifolium hybridum.

Local. There are chlorotic spots.

Systemic. After three weeks chlorotic spots develop on the young leaves while

later ones have a chlorotic streak mottle.

T. incarnatum.

Local. A veinal necrosis results in a collapse of the leaves.

Systemic. In 18 days the veins of the young leaves become chlorotic then necrotic.

Later ones are severely crinkled and puckered and have a mottle. The plant is extremely

stunted and normally collapses.

T. pratense.

Local. No reaction.

Systemic. There are chlorotic streaks which become necrotic and cause a mal-

formation of the leaf surface.

504

T. repens.

Local. No reaction.

Systemic. There are chlorotic streaks but less necrosis and malformation than

on T. pratense.

Vida faba.

Local. In two weeks chlorotic spots with almost purple necrotic rings develop.

The leaves wilt.

Systemic. A week later the young leaves show a veinclearing and chlorotic spotting.

On later formed leaves these spots have necrotic haloes (Fig. 2A), and there is a necrosis

of the growing point, which may collapse. Any secondary shoots which develop have

leaves which are rolled and have a mosaic.

Vigna sesquipedalis.

Local. Chlorotic specks develop in two weeks. These fuse to a general chlorosis

and the leaves wilt.

Systemic. The second trifoliates have chlorotic rings and a necrotic sheen. The

young leaves are spotted and later mottled.

V. unguiculata.

Local. In two days there are necrotic specks and in six there are chlorotic areas.

Systemic. A week later the young leaves develop necrotic veins and they are

curled down, and soon absciss. If more leaves form they are slightly malformed with

irregular areas of chlorotic spots along the veins.

Voandzeia subterranea.

Local. No reaction.

Systemic. A dark green blistered mottle develops on most leaves.

IDENTIFICATION

There is a great resemblance between this virus and the white clover mosaic virus

as described by Zaumeyei and Wade (19).

Although the symptoms on all host plants are not identical, they are very similar.

Also, the physical properties and host ranges of the two viruses are alike.

Therefore, this virus is considered the same as the white clover mosaic virus.

(No attempt has been made to isolate the two components, but the virus is described

as it was found in the field.)

SUMMARY

In South Africa, three viruses occur naturally on Crotalaria juncea, viz. the alsike

clover mosaic virus and strains of two bean viruses (described in the report on bean

virus diseases).

On C. spectabilis there are four viruses which occur in the field — the white clover

mosaic virus (as a complex), two bean viruses and one related to the broad bean mosaic

virus (described under their respective headings).

Of these seven viruses, not one can be correlated with the Crotalaria viruses already

reported in the literature.

The viruses of Bose and Misra (3) and Hadiwidjaja (10) both cause a witch’s broom

effect, which is not a feature of any of these viruses.

Gate’s two diseases of C. spectabilis are not sap-transmissible (9), and the sunnhemp

505

mosaic of Raychaudhuri (15) has only one host viz. sunnhemp. Although the viruses

of Capoor (4) and Raychaudhuri and Pathanian (16) have much wider host ranges,

their physical property values are very high e.g. Capoor’s virus withstands heating up

to 90°C, and ageing to 557 days.

Too few details are available of the other Crotalaria diseases, for an accurate

comparison to be made.

Schwartz (17), Cook (5), Illingworth (11), Fukushi (8) and Jenson (12) give only

the symptoms on Crotalaria species. Dale (7) does include cowpea as a susceptible

host, and Johnson and Lefebvre (14) describe only the reaction on broad bean.

It is therefore concluded that none of these Crotalaria viroses have been found

during this survey in South Africa.

Literature Cited

1. (1949-50) Report of the Division of Mycology and Plant Pathology.

Sci. Rept. Agric. Res. Inst. New Delhi. 1949-50, p. 81-88.

2. Anderson, C. W. (1955) Vigna and Crotalaria viruses in Florida. PI. Dis. Reptr.,

39: 346-357.

3. Bose, R. D. and Misra, S. D. Studies in Indian fibre plants No. 6. Phyllody and some

(1938) other abnormalities in the flower of sunnhemp. Ind.

Jour. Agric. Sci., 8: 417^23.

4. Capoor, S. P. (1950) A mosaic disease of sunnhemp in Bombay. Curr. Sci., 19:

22.

5. Cook, M. T. (1931)

6. Cook, M. T. (1931)

7. Dale, W. T. (1943)

8. Fukushi, T. (1932)

9. Gates, C. M. (1950)

10. Hadiwidjaja, T. (1952)...

11. Illingworth, J. F. (1931)

12. Jenson, D. D. (1950)

13. Johnson, F. (1942)

New virus diseases in Puerto Rico. Phytopathology

(Abstract), 21: 124.

The effect of mosaic on cell structure and chloroplasts.

Jour. Dept. Agric. Puerto Rico, 15: 177-181.

Preliminary studies of the plant viruses in Trinidad. Trop.

Agric., 20: 228-235.

A contribution to our knowledge of virus diseases of plants

in Japan. Trans. Sapporo Nat. Hist. Soc., 7: 13()-141.

(Abstr. Rev. Appl. Myc., 11: 797).

A mosaic and little leaf disease of Crotalaria spectabilis in

South Florida. PI. Dis. Reptr., 34: 203-204.

Die krul- of heksenbezemziekte van Crotalaria. Tijdschr.

over Plantenz., 58: 1-4.

Yellow spot of pineapples in Hawaii. Phytopathology,

21: 865-880.

A Crotalaria mosaic and its transmission by aphids. Phyto-

pathology, 40: 512-515.

The complex nature of white clover mosaic. Phytopathology

32: 103-116.

14. Johnson, H. W. and Lefebvre, Crotalaria mosaic. Phytopathology (Abstr.) 28: 10.

C. L. (1938)

15. Raychaudhuri, S. P. (1947) A note on mosaic virus of sunnhemp (Crotalaria juncea

Linn.) and its crystallisation. Curr. Sci., 16: 26-28.

16. Raychaudhuri, S. P. and Patha- A mosaic disease of Crotalaria mucronata Desv. (C. striata

NiAN, P. S. (1950) D.C.). Curr. Sci., 19: 213.

17. Schwartz, M. B. (1927) Eenige ziekten van onbekenden aard bij groenbemesters.

Korte Meded. Inst, voor Plantenziekten, 5: 1-19.

18. Zaumeyer, W. J. (1940) Three previously undescribed mosaic diseases of pea. Jour.

Agric. Res., 60: 433^52.

19. Zaumeyer, W. J. and Wade, B. L. The relationship of certain legume mosaics to bean. Jour.

(1935) Agric. Res., 51: 715-747.

506

SOYBEAN VIRUS DISEASES

In the field, soybean {Glycine max) is apparently susceptible to three viruses — one

of which is the seed-borne soybean mosaic virus.

This soybean mosaic virus has been found in the Northern, Eastern and Western

Transvaal — but probably occurs wherever soybeans are cultivated, as the percentage

of infected seed can be high.

The other two viruses found on naturally infected soybeans are described in detail

in the report on bean virus diseases; as they were proved to be strains of the bean

chlorotic ringspot and bean local chlorosis viruses.

The field symptoms of the latter two viruses include mottling, puckering and

malformation of the leaves, and stunting of the plants.

Review of literature

Clinton 1916, was the first to report a chlorosis and crinkling of soybean leaves.

The virus origin of this chlorosis was established by Gardner and Kendrick in 1921.

Pierce (11) named the causal organism soybean virus 1, and in Kenneth Smith’s classi-

fication it is Soja virus 1.

The virus can be transmitted in the seed of infected plants where it can survive

for two years, Kendrick and Gardner (8). It is also aphis transmitted.

Heinze and Kohler (6) and Conover (3) were able to infect Phaseolus vulgaris

and Vida sativa, but these hosts have not been confirmed by other workers.

Savulescu et al (12) described three types of symptom on soybeans — leaf curl,

brown mosaic and yellow mosaic. These variants have a thermal inactivation point

of 80°C., which is much higher than is generally accepted for the soybean mosaic virus.

According to Ling (10) the only legume viruses found on naturally infected soybeans

are the bean yellow mosaic virus and the cowpea mosaic virus of Dale.

However, many other legume viruses can induce symptoms on soybeans when

inoculated artificially.

The top necrosis, streak and budblight diseases of soybean are now attributed

to strains of the tobacco ringspot virus. Allington (1) and Johnson (7). This virus

is also seed-transmitted in soybeans. Desjardins et al (4).

The tomato spotted wilt virus results in local necrotic specks surrounded by

chlorotic haloes. The rest of the leaf may be almost orange in colour. There is no

systemic reaction. (See report on pea virus diseases.)

SOYBEAN MOSAIC VIRUS Gardner and Kendrick.

Transmission: Mechanical sap inoculation. By seed. (One land had approxi-

mately 35%.) Aphis. (Species not determined.)

Hosi range: Glycine max (L.) Merr.

Symptoms on Glycine max:

{a) Artificial infection in the glasshouse:

Local. No reaction.

Systemic. The young leaves develop chlorotic spots in three weeks. This

is followed by a mottle and slight malformation on the next formed leaves.

Often, necrotic rings develop round the chlorotic spots as the leaves mature.

The plant is stunted, and has a proliferation of secondary shoots which

remain rosetted.

507

{b) Natural infection in the field:

In a land of maturing soybeans, infected plants are noticeable for their

lush green foliage, when other plants have dropped most of their leaves.

These plants are stunted, and have a great number of lateral branches.

The leaves are mottled, crinkled and slightly malformed. They later

become leathery.

(c) Seed infection in the glasshouse:

The primary leaves are mottled and crinkled with wavy margins. The

trifoliates have the usual symptoms.

The plants are spindly and etiolated.

IDENTIFICATION

As this virus is seed-borne, and has only one host plant viz. the soybean, there

can be little doubt that it is the soybean mosaic virus.

Literature Cited

1. Allington, W. B. (1946) Budblight of soybean caused by the tobacco ringspot virus.

Phytopathology (Note), 36: 319.

2. Clinton, G. P. (1916) Report of the botanist for 1915. Conn. Agric. Exp. Sta.

Ann. Rept., 1915. p. 446.

3. Conover, R. A. (1948) Studies of two viruses causing mosaic diseases of soybean.

Phytopathology, 38: 724-735.

4. Desjardins, P. R., Latterell,

R. L. AND Mitchell, J. E. (1954)

5. Gardner, M. W. and Kendrick,

J. B. (1921)

6. Heinze, K. AND Kohler, E. (1940)

Seed transmission of tobacco-ringspot virus in Eincoln

variety of soybean. Phytopathology (Note) 44: 86.

Soybean mosaic. Jour. Agric. Res., 22: 111-114.

Die Mosaikkrankheit der Sojabohne und ihre Ubertragung

durch Insekten. Phytopath. Zeitschr., 13: 207-242.

7. Johnson, E. (1943) Soybean streak in Ohio. PI. Dis. Reptr., 27: 86-87.

8. Kendrick, J. B. and Gardner,

M. W. (1924)

9. Kreitlow, K. W., Boyd, H. C.,

Chamberlain, D. W. and Dun-

LEAVY, J. M. (1957)

10. Eing, L. (1951)

11. Pierce, W. H. (1935).

12. Savulescu, T., Sandu-Ville, C.,

Aronescu, a. and Alexandri,

A. V. (1936)

Soybean mosaic: seed transmission and effect on yield.

Jour. Agric. Res., 27: 91-98.

A bibliography of viruses infecting the soybean [Glycine

max (E.) Merr.]. PI. Dis. Reptr., 41: 579-588.

Bibliography of soybean diseases. PI. Dis. Reptr. Suppl.,

204: 111-173.

The identification of certain viruses affecting leguminous

plants. Jour. Agric. Res., 51: 1017-1039.

E’etat phytosanitaire en Roumanie en 1934-1935. Publ.

Inst. Cere. Agron. Romaniei, 25: 1-97. (Abstr. Rev.

App. Myc., 16: 19.)

508

LUCERNE (ALFALFA) MOSAIC VIRUSES

The typical lucerne mosaic virus occurs naturally on lucerne in many districts

of the Union.

One variant of this virus was found in the Western Cape, also on lucerne; and

another was isolated from Glycine javanica growing in experimental plots in Pretoria.

These three viruses will be described below.

The necrotic ring strain of lucerne mosaic virus, which was commonly found on

naturally infected peas and sweet peas, has been described in the report on the virus

diseases of peas and sweet peas.

The word “ lucerne ” is used in preference to “ alfalfa ”, as it is the common

name for Medicago sativa in this country.

Natural source of virus:

(a) Lucerne mosaic virus — type, (Type*) Medicago sativa. Some leaves had a

chlorotic network, spotting or streaking, while others had dark green blisters

and were severely puckered.

{b) Lucerne mosaic virus — necrotic strain. (Necr. Str.) M. sativa. The leaves

were diffusely mottled and slightly malformed.

(c) Lucerne mosaic virus — Glycine strain, (Glyc. Str.) Glycine javanica. Most

leaves had irregular, vivid yellow markings. There was no malformation.

See Fig. 5A.

Physical properties:

Thermal inactivation point (all 3) 62-64 °C.

Longevity in vitro. Type and Necr. Str. 2-4 days.

Glyc. Str. 6-8 days.

Dilution end point (all 3) 1 :3000-l :5000.

Transmission:

All three are mechanically sap transmissible.

Type and Necr. Str. Aphis craccivora.

Aphis fabae.

Glyc. str. Results inconclusive.

Host ranges.

*Abbreviations to be used in the text.

509

Reaction of susceptible species

Arachis hypogaea.

Local. No reaction.

Systemic. Type. A dark green mottle develops.

Necr. Str. There is an irregular chlorotic streak mottle and occasional

patterns.

Glyc. Str. Not tested.

Crotalaria juncea.

Local. Necrotic specks develop with alt three. Fig. 4B.

Systemic. Type. The growing point curves to one side and the young leaves

have a chlorotic network. Later formed leaves develop dark green

blisters, and are malformed or even stringlike.

Necr. Str. Most leaves are mottled and elongated, with wavy margins.

Later, necrotic specks develop.

Glyc. Str. The leaves show a chlorotic network and then a mottle.

The chlorotic areas extend across the leaf, thereby constricting the

lamina.

Crotalaria spectabilis. Only the Glyc. Str. was tested.

Local. No reaction.

Systemic. A chlorotic speckling develops.

Dolichos lablab.

Local. Type. Small necrotic specks develop.

Glyc. Str. There is a veinal necrosis and collapse.

Systemic. Type. Some leaves develop a chlorotic spotting.

Glyc. Str. A vein clearing of the young leaves, is followed by yellow

vein flecks on later formed leaves. The plant is stunted.

Glycine javanica.

Local. Glyc. Str. Small necrotic specks and veins develop in two days.

510

Systemic. Glyc. Sir. In three weeks the young leaves show a vein-clearing.

Later formed leaves have irregular yellow areas or concentric patterns.

The symptoms in the glasshouse are not as vivid as those in the field’

Glycine max.

Local. Type. No reaction.

Glyc. Sir. There are necrotic specks, rings and veins.

Systemic. Type. There are chlorotic spots and later a mottle.

Glyc. Str. In three weeks the young leaves develop a chlorotic net-

work or star flecks. Later formed leaves have vivid yellow areas,

Fig. 4F and the plant is stunted.

Lathyrus odoratus. Only the Glyc. Str. was tested.

Local. No reaction.

Systemic. A chlorotic mottle develops.

Lupinus albus.

Local. Type. Necrotic specks develop in 5-6 days.

Necr. and Glyc. Str. There are necrotic specks with chlorotic haloes.

The necrosis spreads and the leaves absciss. Fig. 4G.

Systemic. Type and Necr. Str. The leaves remain folded, and they have wavy

margins. There is a mottle, and fine necrotic speckling on most

leaves. The plant is rosetted.

Glyc. Str. As above, but no necrosis. In addition, the leaves are

malformed and almost stringlike and there are dark green blisters.

L. luteus.

Local. Type and Necr. Str. The leaves collapse after necrosis.

Glyc. Str. No reaction.

Systemic. Type. The leaves have broad dark green veinbands, and the tips are

twisted.

Necr. Str. The leaves are mottled and have necrotic specks. They

remain small and the plant is rosetted.

Glyc. Str. The young leaves develop chlorotic spots. Later formed

ones have irregular dark green areas or blisters. They are also

malformed or stringlike, and the plant is stunted.

L. mutabilis. Only the Glyc. Str. was tested.

Local. No reaction.

Systemic. The leaves are chlorotic with dark green marginal blisters. They are

severely rolled.

Medicago lupulina.

Local. No reaction with the Type and Glyc. Str.

Systemic. Type. Most leaves are mottled and puckered.

Glyc. Str. There is a diffuse chlorotic streaking.

M. sativa.

Local. Type. Necrotic specks develop in 5-6 days.

Necr. and Glyc. Str. No reaction.

Systemic. Type. The young leaves develop chlorotic spots and later ones are

mottled and slightly crinkled.

Necr. Str. The young leaves have chlorotic, almost white spots,

which may become necrotic. On later formed leaves there are

irregular yellow areas which cause a malformation. The old leaves

have chlorotic blotches.

511

Glyc. Sir. Only a diffuse mottle develops.

Melilotus officinalis.

Local. Glyc. Sir. Chlorotic blotches develop.

Systemic. Glyc. Sir. The young leaves have chlorotic spots or streaks along

the veins, which cause a distortion. Later leaves are small, crinkled

and rolled, with irregular chlorotic areas. The plant is stunted.

This plant is a symptomless carrier for the Type virus.

Phaseolus acutifolius.

Local. Type. There are chlorotic spots and rings.

Necr. Str. There are many necrotic lesions which fuse to give a “ scorch ”

effect.

Glyc. Str. No reaction.

Systemic. With all three, there is a veinclearing followed by a mottle.

With the Type virus there are also dark green veinbands.

P. lunatus.

Local. Type. Chlorotic spots develop in 4-5 days.

Glyc. Str. There are large chlorotic blotches.

Systemic. Only with the Glyc. Str. are there chlorotic vein flecks on the older

leaves.

P. nmngo.

Local. Type. No reaction.

Glyc. Str. The veins become necrotic.

Systemic. Type. The young leaves develop chlorotic spots with necrotic centres.

Glyc. Str. There are vivid yellow vein flecks.

P. vulgaris var. Canadian Wonder.

Local. Type. There are necrotic rings with almost white centres. Fig. 3A.

Necr. Str. The lesions are similar, but they have irregular outlines.

Fig. 3B.

Glyc. Str. There are necrotic lesions and considerable veinal necrosis.

Fig. 3C.

Systemic. Type and Glyc. Str. No reaction.

Necr. Str. Isolated necrotic lesions develop along the veins of the

trifoliate leaves. (These lesions are of the same type as occur on

the inoculated leaves.) They result in the uneven growth of the

laminae. Fig. 3D.

var. Haricot.

Local. The necrotic lesions are similar to those on Canadian Wonder; but large

areas of these leaves become chlorotic, and the leaves drop.

Systemic. Necr. Str. There is often a severe veinal necrosis which results in

the collapse of the plant.

Pisum sativum.

Local. Type and Glyc. Str. No reaction.

Necr. Str. There are necrotic lesions in 5-6 days.

Systemic. Type. After a vein clearing of the young leaves the next ones develop

chlorotic spots. Older leaves have a chlorotic vein etching.

Necr. Str. No reaction.

Glyc. Str. The leaves are mottled and malformed, and they remain

folded. The tendrils are abnormally curled.

512

Trifolium hybridum.

Local. Glyc. Sir. No reaction.

Systemic. Glyc. Str. There are chlorotic streaks and rings, and on later formed

leaves, a general chlorosis. As the leaves mature, necrosis sets in,

in the chlorosis.

This plant is a symptomless carrier of the Type virus.

T. incarnatum.

Local. Type. The leaves become chlorotic leaving dark green rings with central

necrotic specks.

Necr. and Glyc. Str. No reaction.

Systemic. With all three there is a chlorotic network followed by a mosaic, and

the leaves are crinkled.

Type. In addition, the plants are severely stunted and rosetted. Fig. 4E.

Glyc. Str. There are well-defined dark green veinbands on most

leaves. Fig. 4D.

T. pratense.

Local. Type and Glyc. Str. Necrotic specks develop in 5-6 days.

Systemic. Type. Some plants have a chlorotic mottle, but with others, the virus

is latent.

Glyc. Str. The virus is latent but can be recovered.

T. repens.

Local. Type. No reaction.

Systemic. Type. Chlorotic streaks develop.

This plant is a symptornless carrier for both the Necr. and Glyc.

strains.

Vicia faba.

Local. With all three there are red necrotic lesions which may fuse with closely

adjacent ones. Fig. 4A.

The leaves become flaccid and absciss.

Systemic. Type. No reaction.

Necr. Str. After an initial vein-clearing there may be a lethal necrosis.

Glyc. Str. A diffuse chlorotic mottle develops.

Vigna sesquipedalis.

Local. Type. Chlorotic spots develop.

Necr. and Glyc. Str. There are necrotic specks.

Systemic. Type and Necr. Str. After a vein clearing there is a chlorotic speck

mottle.

Glyc. Str. No reaction.

V. unguiculala.

Local. Type. There are small necrotic lesions. Fig. 4C.

Necr. and Glyc. Str. In addition, the rest of the leaves become chlorotic,

and they soon drop.

Systemic. Type and Glyc. Str. There is a diffuse mottle.

Necr. Str. Chlorotic spots and necrotic specks develop on the young

leaves. On the older ones there are necrotic rings, and the leaves

absciss.

Voandzeia subterranea.

Local. Necr. and Glyc. Str. No reaction.

513

Systemic. Necr. Str. There are chlorotic spots and later a mottle.

Glyc. Str. The leaves are small and puckered, and they are mottled

with dark green veinbands and blisters.

Nicotiana glutinosa.

Local. Type and Necr. Str. There are small necrotic rings with almost white

centres. Fig. 5B.

Systemic. Type and Necr. Str. A chlorotic spot mottle develops.

N. tabacum.

Local. Type. Chlorotic spots develop in six days.

Necr. Str. There are irregular almost white lesions.

Systemic. Type. No reaction.

Necr. Str. There is a diffuse chlorotic spotting.

Solarium capsicum.

Local. Type. There are chlorotic spots in six days.

Necr. Str. Necrotic lesions develop.

Systemic. Type. The leaves are mottled and malformed or cupped.

Necr. Str. There are necrotic rings or irregular patterns which result

in a malformation of the leaf.

IDENTIFICATION.

{a) Type. — On the basis of host range (including families other than Leguminosae),

symptoms and physical properties, this virus is classified as lucerne (alfalfa)

mosaic virus. Weimer (22, 23).

{b) Necrotic Strain. — This virus results in both local and systemic necrosis on

bean, and it is therefore firstly compared with others causing a similar symptom.

(i) On some varieties of bean, both the bean Southern mosaic virus, Zaumeyer

and Harter (26) and the bean pod mottle virus, Zaumeyer and Thomas

(27) cause a local and systemic necrosis. However, both have very restricted

host ranges, and the physical properties values differ considerably from

those for this virus. Further the pod mottle virus causes a far more severe

necrosis on bean.

(ii) The pea New Zealand streak virus, Chamberlain (3) differs in the type of

necrosis on bean, pea and lupin, and causes only local lesions on the

TrifoUum species. The physical properties also differ.

(iii) There are several other viruses and strains of lucerne mosaic virus which

result in both local and systemic necrosis on bean, but the differences in

the type of necrosis and symptoms on other hosts are too great for closer

comparisons to be made.

(iv) There is apparently a relationship with the lucerne (alfalfa) mosaic virus

group, as the symptoms on many hosts and the physical properties are

very similar.

However, this virus cannot be wholly identified with any previously

described strain, and it is therefore named lucerne mosaic virus. Necrotic

strain.

(c) Glycine Strain. — There are many similarities between this virus and those

in the lucerne mosaic virus group, in respect of symptom expression on several

hosts and physical property values.

However, it differs from them in being non-infectious to the solanaceous

plants, and the reaction of broad bean, soybean and pea etc. is different. More-

over, many species show a vivid yellow mottle instead of the usual chlorosis.

Relationship with the lucerne mosaic virus is probable and it is named

the Glycine strain.

514

1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

Literature Cited

Berkeley, G. H. (1947)

Black, L. M. and Price, W. C.

(1940)

Chamberlain, E. E. (1939)

A strain of alfalfa mosaic virus on pepper in Ontario. Phyto-

pathology, 37: 781-789.

The relationship between viruses of potato calico and alfalfa

mosaic. Phytopathology, 30: 44AA47.

Pea streak (Pw«w virus 3). N.Z. Jour. Sci. Tech., 20: 365 A-

381A.

Houston, B. R. and Oswald,

J. W. (1953)

Johnson, E. M. (1946)

Kovacevski, I. C. (1942)

Kreitlow, K. W. and Price,

W. C. (1949)

McWhorter, F. P. (1949)

Milbrath, J. a. (1952)

Milbrath, j. a. and McWhorter,

F. P. (1954)

Oswald, J. W. (1950)

Pierce, W. H. (1934)

(1935)

Porter, D. R. (1931)

(1935)

Price, W. C. (1940)

Ross, A. F. (1941)

Smith, K. M. (1957)

Snyder, W. C. and Rich, S. (1942)

Thomas, H. R. (1951)

(1953)

Weimer, j. L. (1931)

(1934)

Zaumeyer, W. j. (1938)

(1953)

AND Harter, L. L. (1943)

AND Thomas, H. R.

(1948)

AND Wade, B. L. (1935)

The mosaic virus disease complex of ladino clover. Phyto-

pathology, 43: 271-276.

Two legume viruses transmissible to tobacco. Phyto-

pathology, 36: 142-147.

Die Buntblattrigkeit der Paprikapflanze (Capsicum annuum)

(Medicago virus 2 K. Smith var. typicum Black u. Price).

Zeitschr. Pflanzenkr. 52: 533-540.

A new virus disease of ladino clover. Phytopathology, 39:

517-528.

Alfalfa virus N. Phytopathology (Abstr.) 39: 861.

Variations in potato calico caused by strains of alfalfa mosaic.

Phytopathology (Abstr.) 42: 516-517.

Response of cowpea varieties to strains of alfalfa mosaic,

virus. Phytopathology (Abstr.) 44: 498.

A strain of alfalfa-mosaic virus causing vine and tuber necrosis

in potato. Phytopathology, 40: 973-991.

Viroses of the bean. Phytopathology, 24: 87-115.

The identification of certain viruses affecting leguminous

plants. Jour. Agric. Res., 51: 1017-1039.

The infectious nature of potato calico. Hilgardia, 6: 277-294.

Insect transmission, host range and field spread of potato

calico. Hilgardia, 9: 383-394.

Comparative host ranges of six plant viruses. Amer. Jour.

Bot., 27: 530-541.

Purification and properties of alfalfa-mosaic virus protein.

Phytopathology, 31: 394-410 and 410-420.

A textbook of plant virus diseases. Published by Churchill,

Ltd., London.

Mosaic of celery caused by the virus of alfalfa mosaic.

Phytopathology, 32: 537-539.

Yellow dot, a virus disease of bean. Phytopathology, 41 :

967-974.

Isolation of alfalfa mosaic virus strains from field-grown

beans. PI. Dis. Reptr., 37: 391-392.

Alfalfa mosaic. Phytopathology, 21 : 122-123.

Studies on alfalfa mosaic. Phytopathology, 24: 239-247.

A streak disease of peas and its relation to several strains of

alfalfa mosaic virus. Jour. Agric. Res., 56: 747-772.

Alfalfa yellow mosaic virus systemically infectious to beans.

Phytopathology, 43: 38^2.

Two new virus diseases of beans. Jour. Agric. Res., 67 :

305-327.

Pod mottle, a virus disease of beans. Jour. Agric. Res., 77:

81-96.

The relationship of certain legume mosaics to bean. Jour

Agric. Res., 51: 715-749.

515

Fig. 1. — Alsike Clover Mosaic Virus 1.

A. Crotalaria juncea — natural infection. B. C. juncea — artificial infection.

C. Medicago lupulina. D. Pisum sativum.

202909

516

Fig. 2. — White Clover Mosaic Virus.

A Vida faba. B. Phaseolus vulgaris. C. Crotalaria spedabilis.

517

Fig. 3. — Lucerne Mosaic Virus Strains.

A-D. Phaseolus vulgaris — local lesions with A. Type virus. B. Necrotic strain.

C. Glycine strain. D. Systemic lesions with the necrotic strain.

Ik

518

Fig. 4. — Lucerne Mosaic Virus Strains.

A. Vicia faba. B. Crotalaria juncea. C. Vigna unguiculata with the type virus.

D-E. Trifoliiim incarnatwn. D. Glycine strain. E. Type virus. F. Glycine max

with the glycine strain. G. Lupinus albus with the necrotic or glycine strains.

519

Fig. 5. — Lucerne Mosaic Virus Strains.

A. Glycine javanica — natural infection. B. Nicotiana glutinosa — local lesions.

202909

521

The Virus Diseases of Beans

by

Patricia J. Klesser

Division of Plant Pathology

In South Africa several types of bean are cultivated extensively — the runner,

bush and haricot types being the most popular. Lima beans are grown on a small

scale, but other Phaseolus species are rarely found. Apparently, no natural infections

occur in the latter group.

The common bean mosaic has already been reported in this country (9); but,

although various other disorders were known to be of virus origin, the causal organisms

had not been identified.

Of the viroses described here, only bean mosaic and bean yellow mosaic are well

known. The others are apparently new viruses or new variants of existing viruses.

An earnest attempt has been made to correlate these diseases with those already

described, but with the information available this has not been possible, in all cases.

The differences in symptom expression may be due to the manifestation of known

viruses on other varieties of test plants, or to climatic or environmental factors.

Many plant viruses occur in specific localities, and it is also possible that the diseases

described here, are limited to this part of the continent.

A detailed account of each virosis is given here, in the hope that this will enable

other workers to place them in the known plant virus classification.

Methods and materials

Naturally infected plants were collected from many parts of the Union — notably

Pretoria and district, Potchefstroom, the Northern Transvaal, Bethal and district and

Stellenbosch.

Some sunnhemp plants were sent from Salisbury, Southern Rhodesia.

The standard test plants used were: Phaseolus vulgaris vars. Canadian Wonder

and Haricot, Pisum sativum var. Greenfeast, Vida faba vars. Aquedulce and Long

Pod and Vigna unguiculata var. Dr. Saunder’s Upright.

Many other legume species were also tested and some solanaceous plants.

Carborundum powder was used to facilitate sap inoculation, and Aphis craccivora

was the insect vector in limited experiments.

Although some of the isolates used here came from species other than bean (e.g.

sunnhemp and soybean) they are placed with the bean viruses. Some caused symptoms

on bean, which closely resembled those of the bean isolates; while others resulted in

symptoms on bean which were unusual, whereas on the original host, they were insigni-

ficant or similar to those caused by other viruses.

1. BEAN MOSAIC VIRUS Stewart and Reddick.

Physical properties : Thermal inactivation point 56-58 C.

Longevity in vitro 1-2 days.

Dilution end point 1:1000.

522

Transmission; Mechanical sap inoculation.

Aphis craccivora.

Seed — average of 25-3 per cent for eight varieties.

Host range: Phaseolus acutifolius Gray var. latifolius Freem., P. lunatus L.,

P. vulgaris L.

Occasionally, plants of the following species would show symptoms. These were

checked by sub-inoculation to Phaseolus vulgaris.

Pisum sativum L., Trifolium incarnatum L., Vida faba L., Vigna unguiculata (L.)

Walp.

Reaction of susceptible species

Phaseolus acutifolius.

Local. There are chlorotic spots in seven days.

Systemic. The young leaves develop a veinclearing or chlorotic speckling in

12 days, and they are curled. Later leaves are malformed and mottled with dark green

blisters.

P. lunatus.

Local. Large chlorotic blotches develop.

Systemic. In three weeks the older leaves show chlorotic vein flecks and the young

leaves a chlorotic network.

P. vulgaris var. Canadian Wonder.

Local. Usually there is no local reaction, but occasionally inoculated leaves

show large chlorotic blotches some time after the young leaves show systemic symptoms,

i.e. about three weeks after inoculation.

Systemic. Rarely do the first trifoliates show any symptoms. If so, there are

small chlorotic spots. The second trifoliates curl down from the margins (inverse

cupping) and are twisted. They may be slightly puckered and have a diffuse dark

green mottle. Later formed leaves are lighter green with dark green puckered veinbands.

The tips of these leaflets are long and narrow, and may be malformed. The plant is

stunted.

Plants grown from infected seed.

Primary leaves have large dark green areas and the leaf surface may be rugose.

There is a slight downward curl of the margins. Fig. lA.

Trifoliate leaves have varying degrees of inverse cupping and twisting, with broad

dark green bands along the midrib of older leaves, and dark green mottling of the

younger. Fig. IB. Later formed leaves may show a clearing of the veins or small

chlorotic spots. They may also be malformed and elongated. The plants are generally

stunted.

The pods of some varieties show no symptoms, but most are mottled with slight

depressions on the surface. The variety Rooikrans, if infected early, has pods with

sunken red-brown necrotic areas. The virus is readily transmitted from these lesions,

var. Haricot.

Local. No reaction.

Systemic. The older leaves are inversely cupped, rolled down or twisted. They

have raised dark green areas. Later formed leaves have a dark green mottle or vein-

banding and may also be curled down, with narrow tips.

Plants grown from infected seed.

The primary leaves have large irregular chlorotic areas, and the surface is uneven.

The margins may be curled down. Fig. 2.

The trifoliate leaves have a dark green veinbanding or mottle with raised areas.

The young leaves are slightly smaller than normal and may be tly puckered, and/or

malformed. These have small chlorotic areas. The plant is stunted. Fig. 2.

523

The following varieties were also tested and found to be susceptible: Black Wonder,

Idaho Refugee (Fig. ID), Kentucky Wonder (Fig. IE), Long Tom (Fig 1C), Refugee,

S.A. Black and White, Tendergreen and Victor.

A survey was made of some experimental plots and the incidence of natural infection

can be seen in Table 1.

Pisum sativum.

Local. No reaction.

Systemic. Some plants developed chlorotic veinbands and a mottle similar to

those on the bean. On sub-inoculating to bean, typical symptoms were induced.

Trifolium incarnatum.

Local. No reaction.

Systemic. Some leaves show a diffuse veinbanding and they are rolled.

Vicia faba.

Local. No reaction.

Systemic. Some plants develop chlorotic blotches.

Vigna unguiculata.

Local. No reaction.

Systemic. There may be a chlorotic network followed by a diffuse chlorotic mottle

on some varieties.

Natural source of virus: Phaseolus vulgaris vars. Canadian Wonder, Kentucky

Wonder, Long Tom, Rooikrans and Haricot from Pretoria and district, Nylstroom,

Bethal, Rustenburg and Potchefstroom.

IDENTIFICATION.

From the characteristic symptoms on beans and the amount of seed transmission

there can be little doubt that this is the common bean mosaic virus. (34, 42).

Table 1. — Incidence of Bean Mosaic Virus in Field Plots

Roodeplaat — March 1956

524

(b) Runner Beans.

As the runners were tangled on the trellises it was impossible to count the individual

plants, so the percentage is only approximate.

2.— BEAN MOSAIC VIRUS STRAIN

A second seed-borne virus was found on Phaseolus vulgaris var. Canadian Wonder

which is used extensively in the insect proof glasshouses. In several batches of seed,

about 12 per cent of the seedlings showed symptoms on the primary leaves immediately

on unfolding. Conspicuous symptoms later developed on the trifoliate leaves. Fig. 3A.

The effect simulates a heritable abnormality, but positive sub-inoculations to

many plant species disprove this possibility.

Physical properties: Thermal inactivation point 60-62°C.

Longevity in vitro 2-3 days.

Dilution end point 1:1000-1:5000.

Transmission: Mechanical sap inoculation.

Seed-borne approx. 12 per cent.

Not by Aphis craccivora.

Host range: Arachis hypogaea L., Crotalaria juncea L., C. spectabilis Roth.,

Dolichos lablab L., Glycine max (L.) Merr., Lathyrus odoratus L., Lupinus albus L.,

vars. sweet and bitter, L. luteus L., L. mutabilis Sweet., Medicago lupulina L., M. sativa

L., Melilotus officinalis Willd., Phaseolus acutifolius Gray var. latifolius Freem., P.

lunatus L., P. mungo L., P. vulgaris L., Pisum sativum L., Trifolium hybridum L., T.

incarnatum L., T. pratense L., T. repens L., Vicia faba L., Vigna sesquipedalis (L.) Wight.,

V. unguiculata (L.) Walp., Voandzeia subterranea Thouars.

Reaction of susceptible species

Arachis hypogaea.

Local. No reaction.

Systemic. After 14 days there are chlorotic spots on the young leaves. Later

ones are chlorotic with dark green islands.

Crotalaria juncea.

Local. No reaction.

Systemic. There are chlorotic spots in 10 days and later leaves are mottled and

slightly puckered.

C. spectabilis.

Local. No reaction.

Systemic. The young leaves develop a veinclearing, and chlorotic spotting, and

flecking. Later leaves are mottled with dark green veinbands, and they are crinkled.

Dolichos lablab.

A symptomless carrier.

525

Glycine max.

Local. Chlorotic stars develop in five days.

Systemic. The trifoliates have isolated yellow areas which may cause a malforma-

tion.

Lathyrus odoratus.

Local. No reaction.

Systemic. A chlorotic spot mottling develops on most leaves.

Lupiniis albus var. sweet.

Local. Necrotic specks develop.

Systemic. The young leaves remain folded and have a chlorotic network. Later

leaves develop a mosaic and have wavy margins.

L. albus var. bitter.

Local. No reaction.

Systemic. Most leaves are chlorotic with dark green blisters. They are small

and curled. The plant is rosetted.

L. liiteus.

Local. No reaction.

Systemic. The leaves are mottled, elongated and malformed.

L. mutabilis.

Local. No reaction.

Systemic. The leaves are chlorotic with dark green blisters and necrotic specks.

They are small and stringlike and the plant is rosetted.

Medicago lupidina.

Local. No reaction.

Systemic. After an initial spotting, later leaves develop a chlorotic streak mottle.

M. sativa.

Local. No reaction.

Systemic. There may be diffuse chlorotic spots.

Melilotus officinalis.

Local. No reaction.

Systemic. There are chlorotic streaks leaving dark green veinbands. The leaves

are slightly puckered.

Phaseolus acutifolius.

Local. Chlorotic spots develop which later become necrotic.

Systemic. There is a chlorotic mottle with dark green blisters which result in an

uneven leaf surface.

P. Iiinatus.

Local. There are chlorotic spots in five days.

Systemic. Most leaves develop chlorotic vein flecks with occasional chlorotic spots.

P. nningo.

Local. Chlorotic spots develop in five days.

Systemic. The young leaves have chlorotic specks.

P. vulgaris var. Canadian Wonder.

Local. There may be isolated chlorotic specks in five days.

Systemic. The young leaves are malformed and curl to one side. They develop

an irregular chlorotic speck mottle or vein flecking. Some plants are stunted. Fig. 3B.

var. Haricot.

Local. Small chlorotic spots may develop in five days.

Systemic. The leaves are crinkled, curled and malformed. They develop a chlorotic

speck mottle.

Pisuni sativum.

Local. No reaction.

Systemic. The young leaves develop a veinclearing and chlorotic spotting in

10 days. Later leaves are mottled, and the plant is rosetted.

202909—1

526

Trifolium hybridwn.

Local. No reaction.

Systemic. A chlorotic streaking develops.

T. incarnatum.

Local. No reaction.

Systemic. The young leaves show a chlorotic network. Later leaves are chlorotic

with dark green areas and also crinkled. The plant is stunted and rosetted.

T. pratense.

Local. No reaction.

Systemic. There is a chlorotic mottle leaving a dark green band along the midrib.

T. repens.

A symptomless carrier.

Vicia faba.

Local. Necrotic lesions develop, and the rest of the leaf becomes chlorotic.

Systemic. No reaction.

Vigna sesquipedalis.

Local. Chlorotic spots develop in five days.

Systemic. After a veinclearing there is a chlorotic spot mottle.

V. unguiculata.

Local. There are chlorotic spots in five days.

Systemic. The young leaves develop a chlorotic network and later ones a chlorotic

spotting.

Voandzeia subterranea.

Local. No reaction.

Systemic. After a chlorotic flecking on the young leaves, later ones develop

mottle with necrotic specks.

Natural source of virus: Phaseolus vulgaris.

Canadian Wonder plants sown in an insect-free glasshouse. The primary leaves

had chlorotic, almost white, specks, uniformly scattered over the leaf surface. The

trifoliates were crinkled, slightly malformed and often curved to one side. There were

areas of chlorotic speck mottling with varying amounts of necrosis in the chlorosis.

The plants were severely stunted. Fig. 3A.

IDENTIFICATION.

The symptoms caused by this virus on bean seem to resemble those described by

Horsfall (17, 18) in a brief report on the one-sided mosaic — a strain of common bean

mosaic virus. However, Zaumeyer (51) working with a disorder similar to that found

by Horsfall, attributed the symptom to a heritable abnormality.

When sap from the bean plants used in the experiments described here, was mechani-

cally inoculated to several other legumes, obvious virus symptoms developed. Thus,

it was concluded that in this case there was an infective organism involved and not a

genetical mutation. Further tests proved that a virus was present.

Bean mosaic virus and its strains are known to be easily seed-transmissible, but

there are also reports of the possible seed transmission of the red node virus, Thomas

and Graham (46) and bean Southern mosaic virus, Zaumeyer (55). Neither of the

latter two viruses show further similarities with the one described here.

The type bean mosaic virus and most of its strains have very restricted host ranges,

whereas with this virus, 26 species of Leguminosae are susceptible.

527

Snow (41) reported that a strain from gladiolus, which was 100 per cent seed-borne

in beans, was able to infect several other legumes; but not broad bean, red, white or

sweet clovers — all of which are hosts to this virus.

As this virus is also seed-transmissible it is provisionally placed with the bean

mosaic virus group.

3.— BEAN CHLOROTIC RINGSPOT VIRUSES

Although none of these viruses originated from naturally infected Phaseolus vulgaris,

the symptom expression on the bean is common to all forms described here. The local

reaction consists of unusual, very marked chlorotic ring and line, or ringspot patterns,

and the systemic symptom includes a mosaic and dark green blistering accompanied

by a necrotic speckling.

The four isolates described here differ from one another on other host plants,

but they are provisionally grouped together as strains of one virus, as their reactions

on bean are identical.

Natural source of virus and locality:

Strain A. — Crotalaria juncea. Pretoria district and Groblersdal.

Strain B. — Glycine max. Pretoria and Bethal districts and Potchefstroom.

Strain C. — Voandzeia subterranea. Pretoria and district.

Strain D. — Eryilirina cajfra. Pretoria.

Physical properties

Transmission: All are transmissible by mechanical sap inoculation. Aphis

craccivora is not a vector.

Symptoms on naturally infected plants:

Strain A. — Crotalaria juncea. Most leaves had a mosaic mottle with dark green

veinbands. They were also malformed and rolled. The older leaves

had necrotic specks.

Strain B. — Glycine max. There were chlorotic specks, spots, veinbands or a

mottle on the leaves, and they were slightly malformed and crinkled.

The plants were rosetted and stunted.

Strain C. — Voandzeia subterranea. The leaves had a chlorotic mottle and the

plants were rosetted.

Strain D. — Erythrina caffra. The young leaves were reduced in size and had

chlorotic spots. Older leaves showed a chlorotic, almost yellow,

mottle.

528

Host ranges

Reaction of susceptible species

Not all host plants were inoculated with all four strains, but available results

are tabulated below. The letters A-D indicate the strains used.

Arachis hypogaea. B.C.

Local. B.C. No reaction.

Systemic. B.C. The leaves are chlorotic with dark green islands or veinbands

Crotalaria juncea. A-D.

Local. B. There are chlorotic spots in 10 days.

C.D. Necrotic spots or rings develop.

A. A combination of the above, viz., chlorotic spots which later become necrotic.

Systemic. A. After a chlorotic spotting of the young leaves later ones are mottled

with dark green blisters. They are also malformed and almost stringlike. The old

leaves have necrotic specks. The plant is stunted.

C. There is less malformation than with A but more necrosis — some plants may

collapse.

B. D. There is a veinclearing and chlorotic spotting followed by a mottle with

dark green veinbands or concentric chlorotic patterns. Fig. 5A. With D the leaves

have wavy margins and may be slightly malformed.

C. spectahilis. A-D.

Local. A. D. No reaction.

B. C. There are isolated necrotic lesions.

529

Systemic. All show a veinclearing followed by a mottle with dark green islands,

veinbands or blisters. Fig. 5B. With A-C the leaves are also crinkled and rolled;

and the plant is rosetted with A.

Dolichos lablab. A. C. D.

A symptomless carrier for the three strains.

Glycine javanica. A.

Local. There are chlorotic spots in six days.

Systemic. A diffuse chlorotic spotting develops.

G. max. A-D.

Local. C. No reaction.

A. B. There are chlorotic specks or ring and line patterns.

D. A slight veinal necrosis develops.

Systemic. C. D. The leaves have chlorotic patterns and irregular dark green

areas. They are also crinkled. Fig. 5H.

A. B. In addition there are necrotic specks, Fig. 51, and with A *he plant is

stunted and rosetted.

Lathyrus odoratus. A. C. D.

Local. None show a reaction.

Systemic. A. D. The leaves are mottled with daik j,.een veinbanoj), and rolled

or cupped.

C. There are chlorotic streaks along the veins and the leaves are malformed.

Lupinus albus. A-D.

Local. C. No reaction.

D. There are first chlorotic spots, but later the leaves are uniformly chlorotic.

A. There are chlorotic spots which later become necrotic and the leaves drop.

B. The chlorotic spots have a necrotic speck in the centre.

Systemic. D. The leaves develop a mosaic and the plant is slightly stunted.

B. C. As well as a mosaic, the leaves have dark green blisters and they are small

and malformed. With B many leaves drop, leaving an almost bare stem and petioles.

A. In addition there are necrotic specks on the leaves and the plant is severely

stunted.

L. luteus. A-D.

Local. C. D. No reaction.

A. B. Chlorotic spots develop.

Systemic. C. D. The young leaves remain folded and have wavy margins. Later

formed leaves are mottled, and malformed or elongated. The plant is stunted.

A. B. In addition there are necrotic specks.

L. mutabilis. A. B. D.

Local. No reaction.

Systemic. With all three strains the leaves are mottled, malformed and reduced

in size. The plant is rosetted.

Medicago lupulina. A. C. D.

Local. No reaction.

Systemic. A. C. After a veinclearing there is a mottle with chlorotic streaks.

The leaves are crinkled.

D. A symptomless carrier.

Melilotus officinalis. A-D.

Local. A-C. No reaction.

D. There are chlorotic spots.

Systemic. B-D. The leaves are mottled and crinkled.

A. There is a chlorotic ring spotting on the young leaves, and streaking on later

ones, which are also small and malformed. Fig. 5F.

Phaseolus acutifolius. A-D.

Local. A. Chlorotic spots and veins develop.

530

B. C. D. There are necrotic lesions and veins plus chlorotic spots. These leaves

usually collapse.

With C. there is also a marked epinasty.

Systemic. A. The young leaves curl back and have chlorotic veins and spots.

Later leaves have an almost white speck mottling.

C. In addition to the above mottle there are dark green blisters. The pods are

also mottled.

B. C. D. The young leaves curl back and develop a chlorotic network or patterns

which become necrotic. Fig. 5D. This results in a collapse of plants inoculated with D.

Should secondary shoots develop they have malformed, mottled leaves.

P. lunatus. A-D.

Local. D. No reaction.

B. Large chlorotic blotches develop.

A. C. There are chlorotic spots or ringspots which with C. become necrotic later.

Fig. 5E.

Systemic. With all strains there is a chlorotic network or vein flecking. With

A and D this is accompanied by a necrosis in the chlorosis.

P. mungo. A-D.

Local. C. No reaction.

D. Chlorotic spots develop.

A. B. There are necrotic spots and veins.

Systemic. A-C. There is a chlorotic spotting.

D. Only a diffuse mottle develops.

P. vulgaris. A-D.

(a) var. Canadian Wonder.

Local. All induce chlorotic ringspots, ring and line patterns or a vein etching.

Fig. 4A.

Systemic. The young leaves develop a veinclearing and chlorotic spots. Fig. 4D.,

and the leaves curl down. On some newly formed leaves there are yellow specks or

distinct rings Fig. 4E., but later there is a mottle. The leaves are crinkled and have

many small necrotic specks. Fig. 4C.

With D. the pods are mottled and with C. they are malformed and have necrotic

rings. Fig. 5G.

(h) var. Haricot.

Local. All strains result in chlorotic ringspots, or ring and line patterns. Fig. 4B.

Systemic. After an initial veinclearing and chlorotic spotting the next formed

leaves are mottled with dark green blisters, and they are malformed. The plants are

rosetted.

With A. there is also a necrotic speckling which may cause a collapse of the plant.

(c) vars. Black Wonder, S.A. Black and White, Long Tom and Victor develop

symptoms similar to Canadian Wonder.

(d) vars. Tendergreen and Idaho Refugee do not produce a distinct local

reaction, but the systemic symptoms are similar to those on Canadian Wonder.

Pisum sativum. A-D.

Local. B. C. No reaction.

A. Chlorotic specks develop and the leaves drop.

Systemic. B. C. Some leaves are mottled, while others have chlorotic areas which

lead to a constriction of the laminae. The plants are rosetted and stunted.

A. The young leaves remain folded and have a chlorotic network and spotting.

Later formed leaves develop a mosaic, while the old leaves have a white vein etching.

The plant is rosetted and the tendrils are abnormally curled.

D. A symptomless carrier.

Trifolium hybridum. A. B. D.

Local. No reaction.

531

Systemic. All three strains cause chlorotic spots or streaks and occasionally

a mottle.

A. can be latent on this host.

T. incarnatum. A-D.

Local. A. B. C. No reaction.

D. There is a chlorotic vein etching and spotting.

Systemic. A. B. C. After a veinclearing there is a mottle with dark green veinbands,

and the leaves are crinkled.

D. In addition the older leaves develop chlorotic spots with necrotic rings.

T. pratense. A-D.

Local. No reaction.

Systemic. B. C. D. There are chlorotic spots, streaks or blotches. With C there

are also yellow concentric patterns on the older leaves.

A. A symptomless carrier.

T. repens. A. D.

A symptomless carrier with both strains.

Vida faba. A-D.

Local. D. No reaction.

A. Necrotic lesions develop.

B. There are purple necrotic rings and the remaining tissue becomes yellow.

Systemic. D. Diffuse chlorotic spots develop.

A. There is a chlorotic spotting with a necrotic network.

B. C. No symptoms but the virus can be recovered.

Vigna sesquipedalis. A-D.

Local. D. No reaction.

A. There are chlorotic spots but later the whole leaf becomes chlorotic.

B. There are necrotic lesions and the rest of the leaf becomes chlorotic. They

soon absciss.

C. In addition to the symptoms with B there is a marked epinasty.

Systemic. All induce a veinclearing, chlorotic spotting and mottle. B. may cause

a stunting of the plants.

V. unguiculata. A-D.

Local. D. No reaction.

A. After a chlorotic spotting the leaves become flaccid.

B. There are necrotic lesions and the remaining leaf tissue becomes yellow. The

leaves soon drop.

C. There is a marked epinasty as well.

Systemic. B. C. D. Chlorotic spots develop and later a mottle.

A. In addition there are dark green blisters with the mottle, and later a necrotic

sheen. The leaves are malformed and there may be necrotic stem streaks.

Voandzeia suhteiranea. A-D.

Local. Only with C are there chlorotic blotches.

Systemic. With all strains there may be a veinclearing, mottle and crinkle. Fig. 5C.

Some plants with A and D are symptomless carriers.

IDENTIFICATION.

The characteristic feature of this group of viruses is the well-defined chlorotic ring

and line or ringspot reaction on the inoculated leaves of bean. Similar symptoms on

some varieties of bean are described by Cheo and Zaumeyer (8) with their strain of

tobacco ringspot virus from bean, and with the green strain of tobacco ringspot.

However, none of the four isolates recorded here can be fully identified with either

of these strains. There may be similarities in symptom expression on several hosts.

532

e.g. on Glycine max the D isolate resembles the bean strain of Cheo and Zaumeyer,

and on white lupin the A isolate is similar to the green strain, but on other plants there

are considerable differences.

Further, Dolichos lablab which is used as a differential host by Cheo and Zaumeyer

when comparing four strains of tobacco ringspot virus, is a symptomless carrier for

three of the isolates described here, viz. A. C. D. and is not susceptible to the fourth.

On cowpea, the bean strain induces a systemic necrosis, which does not occur

with any of these strains; and on broad bean it causes systemic necrotic rings which

are not present with any of these isolates.

The type tobacco ringspot virus on bean usually induces local necrotic spots,

rings or veins followed by systemic necrosis and collapse. Pierce (29) and Price (32).

The effect of the viruses described here is mainly chlorotic with only necrotic specks

on the trifoliates.

Stubbs (43) described the symptoms of the type virus on peas, which is also necrotic,

whereas with these strains it is only chlorotic.

In addition, the host ranges differ considerably.

Although there can be little doubt that the viruses recorded here are related to

the tobacco ringspot group, there is no complete correlation with any of the strains

previously described. They are therefore reported as additional strains and named

bean chlorotic ringspot viruses A, B, C and D.

4.— BEAN LOCAL CHLOROSIS VIRUSES

The characteristic feature of the viruses in this group is the local reaction on the

bean, viz. a distinct chlorotic spotting. This reaction occurred with isolates from

several plant species. In other details these seven isolates differ from one another.

They are therefore not considered to be one virus, but strains of the virus linked by

their common local reaction on the bean. (Only strains F and G differ slightly in the

systemic reaction on bean.)

Natural source of virus and locality:

Strain A. — Medicago denticulata. Stellenbosch.

B. — Lathyrus odoratus. Pretoria and district.

C. — Phaseolus vulgaris. Rustenburg.

D. — Crotalaria juncea. Salisbury.

E. — Glycine max. Pretoria, Nylstroom, Potchefstroom.

F. — Phaseolus vulgaris. Pretoria district.

G. — Trifolium incarnatum. Stellenbosch.

Physical properties

533

Transmission: All are transmissible both by mechanical sap inoculation and by

Aphis craccivora.

Symptoms on naturally infected plants (i.e. source of virus).

Strain A. — Medicago denticulala. The leaves showed varying degrees of chlorosis — ■

spots, streaks, mottle and mosaic. They were small and crinkled and

the plants were stunted. The incidence of infection was high.

Strain B. — Lathyrus odoratus. The leaves had chlorotic streaks or veinbands

and they were small and puckered.

Strain C. — Phaseolus vulgaris var. Black Wonder. Most leaves were puckered

and had small chlorotic specks. The plants were slightly stunted.

Strain D. — Crotalaria juncea. Some leaves had chlorotic spots, but most showed

a mosaic, with dark green veinbands. They were also slightly malformed

and rosetted; and the plants were stunted.

Strain E. — Glycine max. The leaves were mottled and puckered, and the plants

were slightly stunted with a proliferation of the lateral shoots.

Strain F. — Phaseolus vulgaris {a) var. Long Tom. The leaves were mottled and

severely curled and malformed. There were occasional necrotic stem

streaks, and the pods had marked necrotic lesions, {b) var. Haricot.

The young leaves had a chlorotic network and were curled. Older

ones were crinkled and mottled with dark green blisters.

Strain G. — Trifolium incarnatum. The leaves showed irregular chlorosis and they

were small, crinkled and malformed. The plants were severely stunted.

Host ranges.

534

Reaction of susceptible species

Not all strains were tested on every host plant, but the available results are compared

below. The letters, A-G indicate the strains.

Arachis hypogaea. A. E. G.

Local. No reaction.

Systemic. A. The young leaves develop a mottle with a dark green midrib.

E. Chlorotic spots develop on the young leaves in 14 days. Later leaves are

chlorotic with dark green islands.

G. There is a chlorotic mottle on most leaves.

Crotalaria juncea. A. C. D. F. G.

Local. A. C. G. No reaction.

D. Chlorotic spots develop in eight days.

F. In seven days large necrotic lesions develop which fuse later to give a “ scorch ”

effect.

Systemic. A. G. After a veinclearing of the young leaves, the next formed are

mottled with dark green blisters. They are also crinkled and curled and the plant is

stunted.

C. In addition to the above there is a severe malformation and some leaves are

almost stringlike. Fig. 6A.

D. F. Similar to C. with necrotic specks. With F, also a vascular necrosis which

results in the death of the plant.

Crotalaria spectabilis. A. B. C. D. F. G.

Local. Only with D are there small chlorotic spots.

Systemic. A. D. The young leaves develop a chlorotic network and speckling.

The next leaves are crinkled with wavy margins, and are chlorotic with irregular dark

green islands.

C. F. After a veinclearing the leaves develop dark green veinbands.

B. The new leaves have conspicuous white streaks which result in a malformation.

(See report on sweet pea virus diseases for photograph.)

G. There is a mild speck mottling.

Dolichos lablab.

A symptomless carrier for A. C. D.

Glycine javanica. A. C. D. F. G.

F. G. Symptomless carriers.

Local. A. D. Chlorotic spots develop in 4-5 days.

C. No reaction.

Systemic. A. D. A diffuse mottle develops.

C. A chlorotic spotting is followed by chlorotic streaks.

Glycine max. B-G.

Local. B. C. D. No reaction.

G. There is a general chlorosis.

E. F. Chlorotic stars, rings and spots develop. With F. there are also necrotic

specks.

Systemic. B. C. Only a diffuse mottle develops.

D. After a chlorotic spotting, there is a mottle of the leaves which are crinkled.

G. There may be chlorotic spots with necrotic rings on the older leaves, while

the younger ones are mottled.

E. F. There is first a veinclearing, then chlorotic spotting and a mottle. Later,

necrotic specks develop. The leaves are crinkled and the plants are stunted.

Lathyrus odoratus. A. B. C. D. F. G.

Local. No reaction.

Systemic. C. There is only a diffuse mottle.

F. The leaves are elongated and rolled and show chlorotic streaks.

535

A. D. G. The early symptoms are similar to F, but later formed leaves have a

mottle with well-defined dark green veinbands.

B. There is a mosaic mottle on most leaves and the stems show necrotic streaks.

(See report on the virus diseases of sweet pea.)

Lupinus albus. A-G.

Local. A. B. C. No reaction.

D. F. There are small necrotic specks.

E. After the necrotic specks have developed, the leaves become chlorotic and

soon drop.

Systemic. A. After a chlorotic spotting of the young leaves, there are dark green

blisters on later formed leaves which are reduced in size. The plant is stunted.

C. E. There is an initial mosaic but later leaves are quite chlorotic. They are

small, have wavy margins and remain folded. Many absciss.

B. D. G. In addition to the symptoms for C. E. there are necrotic specks on the

leaves; and necrotic stem streaks with D.

Lupinus luteus. A-G.

Local. A. B. C. F. G. No reaction.

D. Chlorotic spots develop.

E. The leaves are chlorotic with necrotic specks and they soon drop.

Systemic. A. B. The leaves are mottled and crinkled and they remain folded.

C. E. G. There is also a malformation of the leaves and the plant is stunted.

D. F. The malformation is severe and the leaves are stringlike with dark green

blisters.

With F. there are also necrotic stem streaks.

Lupinus mutabilis. A. C. D. G.

Local. No reaction.

Systemic. A. D. The leaves are chlorotic with dark green blisters, and they are

malformed. The plant is rosetted.

C. G. The plants collapse after necrosis.

Medicago lupulina. A. C. D. F. G.

Local. Only with F is there a reaction. Necrotic lesions develop, and fuse, and

the leaves collapse.

Systemic. The symptoms are similar for all strains, viz. a chlorotic streak mottle

and/or dark green veinbands. With D the leaves are crinkled.

Melilotus officinalis. A. C. D. F. G.

Local. A. C. G. No reaction.

D. F. There are chlorotic spots with dark green rings.

Systemic. C. D. A chlorotic spot mottling develops.

A. In addition the leaves are malformed.

F. G. There are distinct ring and line patterns and the leaves are malformed.

With F there may also be necrotic specks.

Phaseolus acutifolius. A. C. D. E. F. G.

Local. All these strains induce a chlorotic and necrotic reaction — usually chlorotic

spots with necrotic veins. C and E also induce necrotic spots and G also causes necrotic

rings round the chlorotic spots.

Systemic. A. C. F. After a veinclearing there is a mottle with slight malformation.

D. E. G. In addition there is a necrosis of the veins; or necrotic specks with G.

P. lunatus. A. C. D. E. F. G.

Local. Except for D, these strains induce chlorotic spots or large blotches. D shows

no reaction.

Systemic. D. After a veinal necrosis the leaves drop.

The other strains cause a chlorotic vein flecking.

P. mungo. A. C. D. E. F. G.

Local. A. D. There are chlorotic spots.

536

F. There are necrotic spots.

E. G. There are both chlorotic spots and necrotic specks or veins.

Systemic. A. E. After a veinclearing there are chlorotic spots or irregular areas.

D. F. G. In addition, necrosis sets in, in the chlorosis and the leaves have a necrotic

sheen.

C is latent in this plant.

P. vulgaris var. Canadian Wonder. A-G.

Local. All strains cause distinct, vivid chlorotic spots. Fig. 6C. These may

fuse with closely adjacent ones. Fig. 6D, but usually they retain their individuality.

Systemic. A-E. After a clearing of the veins there is a chlorotic spotting and

then a mottle.

With D and E there are also dark green veinbands.

F. G. In addition the young leaves develop necrotic specks and they are malformed.

The pods have sunken dark green areas and they are misshapen.

vars. Black Wonder, Idaho Refugee, Long Tom, S.A. Black and White.

Tendergreen and Victor show symptoms similar to those on Canadian Wonder.

var. Flaricot may develop local chlorotic specks with strains A. D. E. F., but,

on the trifoliate leaves the reaction is the same as that on Canadian Wonder.

Pi sum sativum. A-G.

Local. B. C. E. G. No reaction.

A. F. After a veinal necrosis the leaves collapse.

D. There may be chlorotic specks.

Systemic. A. B. F. After a veinclearing and chlorotic spotting there is a mottle

on later formed leaves. Necrosis sets in, in the chlorosis, or there are necrotic stem

streaks and the plant collapses.

C. G. There is a veinclearing, chlorotic spotting and mottle, and the plants are

rosetted.

D. E. In addition to the mottle, the leaves are malformed and small, and the

tendrils are abnormally curled. The older leaves have an almost white vein etching.

Trifolium fragiferum. F.

Local. Necrotic spots develop after six days.

Systemic. A diffuse chlorotic (almost white) spotting occurs.

T. hybridum. A-G.

Local. A. C. D. E. G. No reaction.

B. F. There is a chlorotic spotting.

Systemic. A. B. C. G. Chlorotic streaks following the veins develop.

D. F. There is a chlorotic spotting on the young leaves, and on later formed ones

there is a vivid yellow streak mosaic. With D the plants are also stunted.

E. A symptomless carrier.

T. incarnaturn. A-G.

Local. B. C. E. F. No reaction.

A. D. Chlorotic spots develop in 5-6 days.

G. There are chlorotic spots with necrotic rings and necrotic centres.

Systemic. B. C. E. After a veinclearing there is a mosaic mottle, and slight

puckering of the leaves.

D. G. There is a chlorotic network, followed by a chlorotic spotting and later

a mosaic. The leaves are crinkled and malformed, and the plant is rosetted and stunted.

F. In addition to the symptoms for D and G, necrosis sets in, in the chlorotic

spots.

A. After 10 days the young leaves show a veinclearing. Later formed leaves

have a mosaic with dark green veinbands and blisters. They are severely crinkled,

with wavy margins. The plant is extremely stunted and rosetted, and may collapse.

T. pratense. A. D. F. G.

Local. A. D. F. G. No reaction.

537

Systemic. A. In 10 days the veins are chlorotic, and later a mosaic develops.

The leaves are crinkled and may be malformed with necrotic areas.

D. F. There is a chlorotic streak mottle, but D may be latent in this host.

G. After a veinclearing there are short dark green veinbands and a dark green

midrib. The leaves are crinkled or slightly malformed.

T. re pens. F. G.

A symptomless carrier for both strains.

Vida faba. A-G.

Local. B. C. No reaction.

A. There are chlorotic spots with small necrotic specks in four days.

D. Small necrotic lesions develop.

F. In 4-5 days there are large necrotic lesions which spread and fuse with adjacent

ones. As the leaves mature, the background becomes yellow. Fig. 6B.

Systemic. A. B. C. D. There is a chlorotic network and spotting, with a mottle

on later formed leaves.

F. Occasionally the local necrosis will spread down the petiole and thence into

the stem. This results in the collapse of the plant.

E. G. Symptomless carrier.

Vigna sesquipedalis. A. D. E. F. G.

Local. A. E. F. G. Chlorotic spots develop in 5-6 days.

D. There are first chlorotic ringspots, but later the whole leaf becomes chlorotic.

Systemic. A. No reaction.

D. E. F. G. There are chlorotic spots and later a mottle.

V. unguiculata. A-G.

Local. D. No reaction.

A. B. C. E. G. There are necrotic lesions in 4-5 days.

F. Although there is no necrosis, the leaves become flaccid and absciss.

Systemic. B. No reaction.

A. C. E. F. G. After a veinclearing there are chlorotic specks, and, on later formed

leaves, a mottle.

D. The mottle is accompanied by dark green blisters.

Voandzeia subterranea. A. C. D. F. G.

Local. A. C. F. G. No reaction.

D. There are chlorotic spots with necrotic rings in seven days.

Systemic. All induce a chlorotic spotting followed by a mottle. In addition,

D causes dark green veinbands.

IDENTIFICATION.

The feature which links these seven strains is the vivid chlorotic spotting on the

inoculated leaves of the bean.

A similar symptom has been described by Ainsworth with the “ sweet pea streak ”

virus (1), Zaumeyer with alfalfa yellow mosaic virus (53) and Smith and Markham with

a virus from Ligusticum (40). Van der Want (50) reports that on the variety Servus,

local chlorotic spots are sometimes produced with two strains of bean yellow mosaic

virus.

However, although there may be other similarities between these four viruses

and some of the strains described here, there is no complete correlation, e.g. strain G

resembles alfalfa yellow mosaic in its reaction on bean, pea and broad bean, but on

other hosts it differs considerably. Further, the alfalfa yellow mosaic can infect

solanaceous plants, whereas the host range of this virus is restricted to the Leguminosae.

During a survey of legume virus diseases in England (22) three strains causing

this characteristic local reaction on bean were isolated from naturally infected species

of Phaseolus, Fisurn, Vida, Trifolium, Medicago, Melilotus and Lathyrus. These strains

538

were designated A. B. and C, and they have also been found in this country on Phaseolus,

Lathyms and Medicago spp. A further four strains have been isolated from other

naturally infected host plants, viz. Crotalaria, Glycine, Phaseolus and Trifolium spp.

and are named bean local chlorosis virus, strains D. E. F. and G.

5. BEAN NECROSIS VIRUS

A virus found on naturally infected Crotalaria spectabUis caused extremely severe

local and systemic necrosis on bean. A similar reaction on bean developed with isolates

from DoHchos lablab and Glycine javanica, but on some other host plants these isolates

differed.

In this account the virus is described as a whole, and the variations are given under

the appropriate species.

Physical properties; Thermal inactivation point 65-70°C.

Longevity in vitro 6-7 days.

Dilution end point 1:5,000-1:10,000.

Transmission: Mechanical sap inoculation.

Not by Aphis craccivora.

Host range: Arachis hypogaea L., Crotalaria juncea L., C. spectabilis Roth.,

DoHchos lablab L., Glycine javanica L., G. max (L.) Merr., Lathyrus odoratus L., Lupinus

albus L., L. luteus L., L. mutabilis Sweet., Medicago lupulina L., Melilotus officinalis

Willd., Phaseolus acutifolius Gray var. latifolius Freem., P. lunatus L., P. mungo L.,

P. vulgaris L., Pisum sativum L., Trifolium hybridum L., T. incarnatum L., T. pratense L.,

Vida faba L., Vigna sesquipedalis (L.) Wight., V. unguiculata (L.) Walp., Voandzeia

subterranea Thouars.

Reaction of susceptible species

Arachis hypogaea.

Local. No reaction.

Systemic. The leaves have chlorotic areas with dark green islands.

Crotalaria juncea.

Local. There are necrotic lesions and the leaves drop.

Systemic. After 12 to 14 days there is a vein clearing of the young leaves, followed

by a chlorotic spotting and a mottle mosaic. The leaves are rolled and the plant is

stunted.

C. spectabilis.

Local. No reaction.

Systemic. After a veinclearing and spotting of the young leaves, later ones are

very chlorotic and rolled and crinkled. The plant is stunted.

DoHchos lablab.

Local. No reaction.

Systemic. The young leaves develop a chlorotic network. The next formed leaves

are mottled and slightly crinkled.

Glycine javanica.

Local. There are small necrotic specks and the leaf becomes chlorotic.

Systemic. Chlorotic spots and vein flecks develop on most leaves. On some there

are large irregular chlorotic blotches, and the leaves are puckered.

Glycine max.

Local. Small necrotic specks appear in 10 days.

Systemic. The young leaves develop chlorotic veins, spots and flecks. Necrosis

develops in the chlorosis of the old leaves, resulting in a malformation. The plant

is stunted.

539

Lathyrus odoratus.

Local. No reaction.

Systemic. Most leaves show a dark green streak mottle and they are rolled down-

wards.

Lupinus albus.

Local. There are chlorotic spots.

Systemic. In 11 to 12 days the young leaves develop a veinclearing with chlorotic

spots. Later leaves are small and malformed with dark green blisters. They remain

folded and the margins are irregular. The growing point is rosetted and the plant is

stunted. Many older leaves drop.

L. luteus.

Local. Small chlorotic spots may develop.

Systemic. A veinclearing and chlorotic spotting of the young leaves develops

in eight to nine days. New leaves remain folded with the tips twisted. They are mottled

and small. The plant is stunted.

L. mutabilis.

Local. No reaction.

Systemic. The young leaves curl up, and they are spotted. Later ones are very

small and chlorotic. Many drop.

Medicago lupulina.

A symptomless carrier.

Me/i/otus officinalis.

Local. No reaction.

Systemic. Occasional irregular vivid chlorotic areas and some dark green ringspots

develop. Fig. 7B. The virus is recoverable from symptomless leaves.

Phaseolus acutifolius.

Local. There may be necrotic specks.

Systemic. The young leaves curl back and have veins which are chlorotic then

necrotic. The plant collapses.

P. lunatus.

Local. Small necrotic lesions appear in four days. These have chlorotic haloes.

Systemic. There are chlorotic vein flecks on most leaves in three weeks. Fig. 1C.

P. mungo.

Local. In five days there are necrotic rings.

Systemic. There is a chlorotic spotting and then a mosaic with a slight distortion.

P. mungo with the B isolate.

A symptomless carrier.

P. vulgaris var. Canadian Wonder.

Local. There is a severe necrosis of the veins and petioles. Fig. 7E.

Systemic. In 12 days the young leaves start to curl round and back, and have a

veinclearing. Necrosis sets in, in the veins causing a malformation. These leaves

drop and very few more are formed. Fig. 7 A. The old leaves have large chlorotic

blotches.

var. Haricot.

Local. In seven to eight days the leaves have necrotic veins.

Systemic. This necrosis spreads along the petioles and up the stem. The young

leaves show a necrotic sheen and the growing point collapses in 14 to 15 days.

Pisum sativum.

Local. The leaves collapse after a general necrosis.

Systemic. In eight to nine days a veinclearing of the young leaves develops. Later

leaves have a mottle or mosaic with necrosis. The growing point is rosetted and the

plant is stunted and may collapse.

P. sativum. With the B isolate.

Local. No reaction.

540

Systemic. Only a chlorotic network of the young leaves develops in 13 days.

Trifolium hybridum.

Local. No reaction.

Systemic. After a chlorotic spotting on the young leaves, the next show alternate

streaks. Necrosis sets in, in the chlorosis. Fig. 7D.

The B isolate is latent.

T. incarnatum.

Local. There are necrotic veins on chlorotic leaves.

Systemic. In a week the young leaves show a chlorotic network, and vein flecks.

Later leaves have a mosaic with dark green veinbands and they are crinkled. The

plant is very stunted and usually collapses.

The B isolate is latent.

T. pratense.

Local. No reaction.

Systemic. There are diffuse dark green veinbands.

Vida faba.

Local. There are necrotic rings on chlorotic leaves.

Systemic. After 10 days the young leaves show a chlorotic network with small

chlorotic spots. Later leaves are mottled.

Vigna sesquipedalis.

Local. No reaction.

Systemic. The young leaves develop chlorotic spots and flecks, and later veinbands.

V. imguiculata.

Local. In five days the leaves show necrotic spots and they drop.

Systemic. Only a diffuse mosaic develops on the young leaves of some varieties.

The virus is recoverable from symptomless plants.

Voandzeia subterranea.

A symptomless carrier.

Natural source of virus:

(fl) Crotalaria spedabilis. Potchefstroom, Pretoria and district. There was a

chlorotic mottle with dark green veinbands and blisters. On older leaves

there was necrosis in the chlorosis, and there were also necrotic stem streaks.

The plants were stunted and rosetted.

{b) Dolichos lablab. Prinshof Experimental Station, Pretoria. The younger

leaves had chlorotic spots; and the older ones, a mottle with dark green vein-

bands and blisters.

(c) Glydne javanica. An experimental plot adjacent to (b). There were irregular

chlorotic areas on most leaves, which were also puckered.

IDENTIFICATION.

The dominant feature of this virus is the severe local and systemic necrosis on bean.

This type of symptom occurs with many other viruses, with which this one will be

compared: —

(a) The reaction appears to be very similar to photographs of black root, Jenkins

(19, 20) which is the symptom expression of common bean mosaic virus on

particular variety lines of bean. However, as Canadian Wonder has been

used throughout these experiments, and has shown the more typical reaction

of bean mosaic (veinbanding, rolling etc.) that virus is not the causal agent.

This is further confirmed by the great differences in host range and physical

properties.

541

(b) There are several strains of alfalfa mosaic virus which may result in similar

symptoms on some varieties of bean, e.g. the tuber necrosis and potato calico

strains of Oswald (26), alfalfa mosaic N of McWhorter (23) and alfalfa yellow

mosaic virus of Zaumeyer (53). However, on other hosts the symptoms differ

considerably from those caused by this virus, and, with the first three-named

strains, the property values differ also.

(c) Of the several bean yellow mosaic virus strains, only the X-strain of McWhorter

(24) and the severe yellow strain of Thomas and Zaumeyer (49) induce both

local and systemic necrosis on some varieties of bean. Nevertheless, the

differences in symptom expression on other host plants, host ranges and physical

properties are too great for this virus to be identified with either strain.

(cl) On bean, pea and soybean this virus induced symptoms similar to those caused

by the pea streak virus, Chamberlain (7), but on other plants they are dissimilar.

Further, the pea streak virus withstands higher temperatures and dilutions and

longer ageing than this one.

(e) This virus cannot be identified with the bean Southern mosaic, Zaumeyer

and Harter (55) as the latter has a very restricted host range and high physical

property values.

(/) There are too few details of the bean yellow necrosis virus Johnson (21) and

the Dolichos mosaic virus Capoor and Varma (5) for an accurate comparison

to be made.

(g) There are other non-legume viruses which result in a similar type of reaction

on certain varieties of bean, e.g. tobacco necrosis virus causing stipple streak

of bean, van der Want (50) and a tobacco streak strain causing red node of

bean, Thomas and Zaumeyer (47, 48). Also the tobacco ringspot virus Pierce

(30), tobacco broken ringspot virus. Smith and Markham (39) and a cucumber

mosaic virus strain isolated from spinach by Fulton (12). With none of these

is there a complete correlation.

From this account of the data available, it seems that the virosis described above

cannot be wholly identified with any previously recorded disease.

It shows some similarities with several viruses and is probably related to one of

them — meanwhile it is named the bean necrosis virus.

6.— BEAN VIRUS COMPLEX

When beans were inoculated with the sap from several naturally infected Pliaseolus

vulgaris and Crotalaria spectabiHs plants, two distinct types of local lesions developed,

viz. chlorotic spots or blotches and necrotic specks. After several transfers to and from

other hosts, it became evident that more than one virus was involved.

For the convenience of others who may have to identify bean viruses, the complex

is firstly described as such, and then the symptoms of the individual components are

compared.

The virus causing the local chlorosis appears to be identical with the bean local

chlorosis virus A, which has already been described and which occurs naturally on

its own.

The second component, which results in the local necrotic lesions on bean, is

named bean necrotic speckle virus. Apparently, it is not found alone in the field,

but only in the complex.

542

Natural source of virus complex and locality:

(a) Phaseolus vulgaris. Pretoria and district, Groblersdal and Bronkhorstspruit.

Several varieties showed large chlorotic blotches on the older leaves, and

a mottle on the younger ones. These were also crinkled and slightly malformed.

Most plants were stunted. On some varieties e.g. Rooikrans there were necrotic

specks on the leaves and sunken necrotic areas on the pods.

{h) Crotalaria spectabilis. Potchefstroom.

The leaves had dark green veinbands with irregular chlorotic areas inter-

veinally. There were short necrotic stem streaks and the plants were stunted.

Throughout this description, the following abbreviations will be used: — ■

Comp. — virus complex.

BLCV. A. — bean local chlorosis virus A.

BNSV. — bean necrotic speckle virus.

Physical properties: Thermal inactivation point 60-64°C.

Longevity in vitro 2-3 days.

Dilution end point 1:1,000-1:5.000.

These values refer to the complex and to the two components.

Transmission: Mechanical sap inoculation.

Aphid transmission tests were inconclusive.

Host range: Arachis hypogaea L., Crotalaria juncea L., C. spectabilis Roth.,

Glycine max (L.) Merr., Lathyrus odoratus L., Lupinus albus L., L. luteus L., L. mutabilis

Sweet., Medicago lupulina L., M. sativa L., Melilotus officinalis Willd., Phaseolus

acutifolius Gray var. latifolius Freem., P. lunatus L., P. mungo L., P. vulgaris L., Pisum

sativum L., Trifolium hybridum L., T. incarnatum L., T. pratense L., T. repens L., Vida

faba L., Vigna sesquipedalis (L.) Wight., V. unguiculata (L.) Walp., Voandzeia subterranea

Thouars.

BLCV. A is not able to infect Glycine max, Medicago sativa or Trifolium repens.

Reaction of susceptible species

Arachis hypogaea.

Local. No reaction.

Systemic. With all three viruses a mottle develops.

Crotalaria juncea.

Local. Comp, and BNSV. In nine days there are chlorotic spots with necrotic

centres.

BLCV. A. No reaction.

Systemic. Comp. The growing point curves strongly to one side. The leaves

are puckered with wavy margins and they have dark green blisters. The plant is rosetted

and stunted.

BLCV. A. There is a crinkling of the leaves with a mottle and dark green blisters.

BNSV. After a veinclearing of the young leaves, later formed ones are mottled

with a necrotic sheen. The plant is stunted.

C. spectabilis.

Local. No reaction.

Systemic. The symptoms with all three viruses include a veinclearing and/or

chlorotic spotting, followed by a mottle with dark green blisters. The leaves are crinkled

and the plant is slightly stunted.

Glycine max.

Local. Comp, and BNSV. Chlorotic spots develop in nine days.

Systemic. Both viruses cause a chlorotic spotting on the young leaves, and a

mottle and malformation on later formed ones. Fig. 8D.

With BNSV. there are also necrotic specks on the old leaves.

543

Lathyrus odoratus.

Local. Comp, and BNSV. Small necrotic lesions develop.

BLCV. A. No reaction.

Systemic. With all three, the leaves are rolled and mottled. There are also dark

green veinbands with BLCV. A.

Lupinus albus.

Local. Comp, and BNSV. There are small necrotic lesions and the leaflets absciss.

BLCV. A. No reaction.

Systemic. The three viruses induce a veinclearing or chlorotic spotting on the

young leaves which remain folded. Later formed leaves are small and chlorotic with

dark green blisters, and the plant is rosetted.

With the complex, the stem may become necrotic and the plant collapses.

L. luteus.

Local. No reaction.

Systemic. Comp. The young leaves are mottled and remain folded. Later

formed leaves are malformed and almost stringlike, with large dark green blisters.

The plant is stunted and rosetted.

BNSV. Chlorotic spots may develop. Later leaves have a dark green band along

the midrib which is puckered, and the margins are rolled upwards. The plant is rosetted.

BLCV. A. The leaves are mottled, folded and crinkled.

L. mutabUis.

Local. Comp. There is a general necrosis and the leaflets absciss.

BNSV. and BLCV. A. No reaction.

Systemic. Comp. Necrotic stem streaks develop, the growing point curls to one

side, and the plant eventually collapses.

BNSV. If the plants are young when infected, the reaction is similar to that of

the complex. On older plants, there is a mottle on the leaves which are small and

stringlike, and rosetted.

BLCV. A. The leaves are severely curled and very chlorotic with dark green

blisters.

Medicago lupuUna.

Local. No reaction.

Systemic. With the complex there are areas with mosaic markings, but with the

other two viruses there is a mottle with dark green veinbands.

M. saliva.

Local. No reaction.

Systemic. Both the complex and BNSV cause a diffuse chlorotic spot mottle,

and a slight malformation of the leaves.

Melilotus officinalis.

Local. No reaction.

Systemic. Comp, and BLCV. A. There is a mottle with chlorotic streaks and

the leaves are malformed.

BNSV. Only a chlorotic mottle develops.

Phaseolus acutifolius.

Local. All three viruses induce a chlorotic spotting with necrotic veins or rings.

The leaves collapse.

Systemic. Comp. The young leaves develop a veinclearing with necrosis, and

they are severely curled. Later leaves have a speck mottle.

BNSV. This necrosis usually results in the collapse of the growing point. Should

any secondary shoots develop, they have small, malformed and mottled leaves.

BLCV. A. There is a yellow network only.

544

P. lunatus.

Local. Comp. Chlorotic spots with necrotic specks develop.

BNSV. There are small necrotic specks in nine days.

BLCY. A. Chlorotic spots develop in four days.

Systemic. Comp, and BLCY. A. After a veinclearing, there are chlorotic flecks

along the veins on most leaves.

BNSY. On the first trifoliates to show symptoms, there is a chlorotic network

which soon becomes necrotic. Later formed leaves are speckled.

P. mungo.

Local. Same as for P. lunatus.

Systemic. Comp. The younger leaves develop chlorotic spots, each with a necrotic

ring.

BNSY. After a chlorotic spotting on the young leaves, the next formed have

dark green veinbands with irregular shaped chlorotic areas, which cause a distortion

of the laminae. There are necrotic specks on the older leaves.

BLCY. A. Irregular chlorotic areas develop on some leaves.

P. vulgaris var. Canadian Wonder.

Local. Comp. In 4-5 days small necrotic specks develop which may fuse with

closely adjacent ones. After six days there are also chlorotic spots, which develop

independently from the specks. Fig. 8A.

BNSY. In 2-3 days necrotic specks develop.

BLCY. A. There are chlorotic spots in 4-5 days.

Systemic. Comp. The first trifoliates show large chlorotic blotches in 12-14 days.

Young leaves have chlorotic spots, and later ones develop a mottle with dark green

blisters. They are slightly puckered and may have twisted tips. The pods are mottled.

Fig. 8B.

BNSY. After a chlorotic spotting of the young leaves, later ones develop a mottle,

with dark green blisters. These leaves are malformed and elongated, and the plant

is stunted.

BLCY. A. There is a veinclearing of the young leaves; later formed ones are

mottled, crinkled and slightly malformed.

vars. Black Wonder and S.A. Black and White.

Local. Comp, and BLCY. A. Same as for Canadian Wonder.

BNSY. There are necrotic rings in addition to the necrotic specks.

Systemic. Same as for Canadian Wonder.

vars. Idaho Refugee, Long Tom, Tendergreen and Yictor.

Local. Comp. Yery few necrotic specks develop.

BNSY. and BLCY. A. Similar to that on Canadian Wonder.

Systemic. Same as for Canadian Wonder.

var. Haricot.

Local. Comp. In addition to the symptoms described for Canadian Wonder,

there is a necrosis of the veins and petioles.

BNSY. There are necrotic specks, veins and petioles. Fig. 8C.

BLCY. A. Same as Canadian Wonder.

Systemic. Comp, and BNSY. There are necrotic specks and stem streaks which

may result in the collapse of the plant.

BLCY. A. Similar to Canadian Wonder.

Pisu/n sativum.

Local. Comp. After a general necrosis the leaves collapse.

BLCY. A. The veins become necrotic and the leaves collapse.

BNSY. Necrotic spots develop.

Systemic. Comp, and BLCY. A. There is a veinclearing of the young leaves.

545

Later formed ones are mottled with necrosis setting in, in the chlorosis. The plant

usually collapses.

BNSV. Most leaves develop chlorotic spots and veins, which soon become necrotic.

The leaves remain folded, the tendrils are abnormally curled and the plant is rosetted.

Later, necrotic stem streaks develop, which may lead to the ultimate collapse of the

plant.

Trifolium hybridum.

Local. BNSV. There are diffuse chlorotic spots, with dark green rings.

Comp, and BLCV. A. No reaction.

Systemic. BNSV. There are chlorotic spots, streaks and ringspots.

Comp, and BLCV. A. A chlorotic streaking develops.

T. incarnatum.

Local. Comp. Chlorotic spots with necrotic rings develop.

BNSV. There are necrotic rings and a general chlorosis.

BLCV. A. Chlorotic spots develop.

Systemic. All three viruses cause a veinclearing, and later a mottle with dark

green areas, veinbands or blisters. With the complex and BLCV. A. the leaves are

severely crinkled. With BNSV. necrosis may set in, in the chlorosis and the plant may

collapse.

T. pratense.

Local. Comp, and BNSV. Small necrotic lesions may develop.

BLCV. A. No reaction.

Systemic. Comp. There are chlorotic blotches on the old leaves and chlorotic

streaks on the younger ones. Necrosis may set in, in the latter, resulting in a distortion

of the laminae.

BLCV. A. Similar symptoms develop.

BNSV. This virus is latent in the young leaves.

T. repens.

Local. No reaction.

Systemic. Comp, and BNSV. A chlorotic streak mottle develops, or there are

no symptoms.

Vicia faba.

Local. Comp. Small necrotic specks develop.

BNSV. In addition there are necrotic rings.

BLCV. A. There are necrotic specks and chlorotic spots in 4-5 days.

Systemic. Comp. The older leaves develop large necrotic lesions which spread

and fuse, and the leaves drop. This necrosis may continue into the growing point which

then collapses; or the young leaves may be mottled.

BNSV. Very diffuse chlorotic spots with necrotic specks develop on the young

leaves. Later formed leaves have almost yellow spots which soon become necrotic.

Most leaves drop.

BLCV. A. After a chlorotic spotting, there is a mottle.

Vigna sesquipedalis.

Local. Comp. There are chlorotic spots with necrotic specks and veins. The

leaves soon drop. Fig. 8E.

BNSV. Only the necrosis develops.

BLCV. A. Only the chlorotic spotting occurs.

Systemic. Comp, and BNSV. There is a veinclearing of the young leaves. Later

formed ones have dark green specks, veinbands and blisters, and they are rolled down-

ward.

BLCV. A. No reaction.

V. unguiculata.

Local. With all three viruses there are necrotic specks and/or veins.

546

Systemic. The three viruses induce a chlorotic network followed by a mottle on

later formed leaves.

With BNSV. there are also dark green blisters.

Voandzeia subterranea.

Local. Comp, and BNSV. Small chlorotic spots develop.

BLCV. A. No reaction.

Systemic. Comp. There are chlorotic spots and flecks on the young leaves.

Later leaves have dark green veinbands and a necrotic sheen.

BNSV. The young leaves have chlorotic spots, with necrosis setting in.

BLCV. A. There is a chlorotic mottle.

IDENTIFICATION.

This virus complex consists of two distinct components, which differ considerably

jn their individual symptom expression on most of the host plants tested.

The component causing the local chlorosis on bean has already been designated

bean local chlorosis virus A — one of a group of viroses which result in this characteristic

reaction.

On some varieties of bean this same type of local chlorosis occurs with the alfalfa

yellow mosaic virus of Zaumeyer (53), but on many other host plants the symptoms

do not agree.

With the alfalfa yellow mosaic, bean varieties develop either a local necrosis or

a local chlorosis — never both symptoms simultaneously.

The second component, which causes small local necrotic lesions on bean, is named

bean necrotic speckle virus. This symptom occurs both locally and systemically on

many host plants, and on some species it is lethal.

Again, there are similarities in symptom expression on some varieties of bean,

with those caused by alfalfa yellow mosaic virus, but differences on other hosts preclude

complete correlation. Further, the host range of alfalfa yellow mosaic includes more

susceptible species and it can withstand higher temperatures.

There are several other viruses, which, on some varieties of bean, cause the same

combination of symptoms as occur with bean necrotic speckle virus, viz. local necrosis

with systemic chlorosis.

The bean mosaic virus 4, Zaumeyer (56) and pod mottle virus, Zaumeyer (58)

both have very restricted host ranges, and strong resistance to heating and ageing.

The pea streak virus, Chamberlain (7) also withstands a very high dilution.

Although the red node virus, Thomas and Zaumeyer (47), yellow dot virus, Thomas

and Zaumeyer (44) and white clover mosaic virus complex, Zaumeyer and Wade (60)

do cause both local necrosis and systemic chlorosis, the descriptions of the type of

symptom do not agree with those induced by bean necrotic speckle virus.

The local lesion strain of bean yellow mosaic virus, Thomas and Zaumeyer (49)

results in similar symptoms on some bean varieties, but there are considerable differences

in the symptoms on other plants, the host ranges and physical properties.

In England, a virus similar to the bean necrotic speckle virus was found on broad

bean, also in combination with one causing local chlorotic spots on bean (22).

At the time, it was considered a new virus, but from recent literature, it now appears

to be related to the lucerne (alfalfa) mosaic virus group.

There is a considerable resemWance between this virus and some of the many

strains of lucerne mosaic, in respect of symptoms on several hosts, legume host range,

and some physical property values; but it cannot be completely identified with any

one of these variants.

It is therefore described here as a possible new strain and named bean necrotic

speckle virus.

547

BEAN YELLOW MOSAIC VIRUS

This virus is common on naturally infected Lupinus albus and L. angustifoUus, in

the Western Province. It has therefore been described in the paper on lupin virus

diseases.

BEAN YELLOW MOSAIC VIRUS, NECROTIC STRAIN

This strain was found on naturally infected Lathyrus odoratus in Pretoria and

district. It is described under sweet pea virus diseases.

LUPIN VIRUS B

This virus was also found on beans growing near various species of Lupinus in

the Western Province. Details are given in the paper on lupin virus diseases.

In the field the symptoms are mild — only a chlorotic mottle and slight crinkling

of the leaves.

PEA WILT VIRUS STRAIN

In an experimental plot of Kentucky Wonder beans in the Rustenburg area, many

plants showed vivid yellow spots on the leaves. There was also mottling, veinbanding

and puckering, and the plants were stunted. This virus is described under pea virus

diseases.

Summary

Details are given of bean virus diseases in South Africa, viz. four viruses and their

several variants, and one virus complex. Mention is also made of a further two viruses

and two strains, which have been described under other headings.

(fl) Bean mosaic virus — the presence of this virus was confirmed.

{b) Bean mosaic virus strain — about 12 per cent seed transmission in the variety

Canadian Wonder of unknown origin. Differs from the type virus in having a wider

host range; and thermal inactivation point of 60-62°C., and longevity in vitro of 2-3

days.

(c) Bean chlorotic ringspot virus with four strains — probably related to the tobacco

ringspot group. As the name suggests, this virus causes distinct ringspot patterns on

the inoculated leaves of bean. It was isolated from Crotalaria juncea, Erythrina cajfra.

Glycine max and Voandzeia subterranea. The strains withstand heating to between

56-62°C., ageing between one and four days and dilutions of between 1 : 1000-1 ; 10,000.

{d) Bean local chlorosis virus and seven variants — isolated from species of Crota-

laria, Glycine, Lathyrus, Medicago, Phaseolus and Trifolium. The virus is named after

the marked local chlorotic spotting on bean. The thermal inactivation points of the

strains vary between 56-65°C., the longevity in vitro is between one and four days

and the dilution end points are between 1:1000 and 1:5000.

(e) Bean necrosis virus — isolated from Crotalaria spectabilis, Dolichos lablab and

Glycine javanica. Although the symptoms on the original host plants are mild, on

bean there is a lethal necrosis. The virus is inactivated at temperatures of 65-70°C.,

after 6-7 days ageing, and at dilutions of between 1:5000-1:10,000. The relationship

of this virus to others is uncertain.

(/) Bean virus complex. With isolates from Phaseolus vulgaris, Crotalaria specta-

bilis and Glycine javanica, two distinct local reactions developed on bean. One com-

ponent was identified as a variant of the bean local chlorosis virus, and the other was

named bean necrotic speckle virus. The latter is probably related to the lucerne mosaic

virus group. For both components the thermal inactivation point is between 60-64°C.,

the longevity in vitro 2-3 days, and the dilution end point between 1 :1000 and 1 :5000.

548

Literature Cited

1. Ainsworth, G. C. (1940)

2. Bawden, F. C. and V. d. Want,

J. P. H. (1949)

3. Bridgmon, G. H. and Walker,

J. C. (1951)

4. Burkholder, W. H. and Muller,

A. S. (1926)

5. Capoor, S. P. and Varma, P. M.

(1948)

6. Capoor, S. P. and Varma, P. M.

(1948)

7. Chamberla[n, E. E. (1939)

8. Cheo, P. C. and Zaumeyer, W. J.

(1952)

9. Doidge, E. M., Bottomley, A. M.,

van der Plank, J. E. and Pauer,

G. D. (1953)

10. Fajardo, T. G. (1930)

11. Fajardo, T. G. (1930)

12. Fulton, J. P. (1950)

1 3. Grogan, R. G. and Walker, J. C.

(1948)

14. Grogan, R. G. AND Walker, J. C.

(1948)

15. Hagedorn, D. j. (1950)

16. Harrison, A. L., Horsfall, J. G.

AND Burkholder, W. H. (1932)

17. Horsfall, J. G., Burkholder,

W. H. AND Reinking, O. A. (1937)

18. Horsfall, J. G. (1930)

19. Jenkins, W. A. (1940)

20. Jenkins, W. A. (1941)

21. Johnson, E. M. (1946)

22. Klesser, P. j. (1951)

23. McWhorter, F. P. (1949)

24. McWhorter, F. P. and Boyle, L.

(1946)

25. Milbrath, j. a. (1952)

26. Oswald, J. W. (1950)

27. Parker, M. C. (1933)

28. Parker, M. C. (1935)

The identification of certain viruses found infecting leguminous

plants in Great Britain. Ann. Appl. Biol., 27: 218-226.

Bean stipple-streak caused by a tobacco necrosis virus.

Tijdschr. over Plantenziekten, 55: 142-150.

The relation of southern bean mosaic to black root. Phyto-

pathology, 41: 865-871.

Hereditary abnormalities resembling certain infectious diseases

in beans. Phytopathology, 16: 731-737.

Enation mosaic of Dolichos lablab Linn., a new virus disease.

Current Sci. (India), 19: 57-58.

Yellow mosaic of Phaseoliis lunatus, L. Current Sci. (India),

17: 152-153.

Pea streak (Pisuin virus 3). New Zeal. Jour. Sci. and Technol.,

20: 365A-381A.

A new strain of tobacco ringspot virus isolated from bean.

PI. Dis. Rptr., 36: 459-465.

A revised list of plant diseases occurring in South Africa.

Science Bulletin No. 346. (Botany and Plant Pathology

Series No. 16).

Studies on the mosaic of the bean (Phaseoliis vulgaris L.).

Phytopathology, 20 : 469—494.

Studies on the properties of the bean-mosaic virus. Phyto-

pathology, 20: 883-888.

Studies on strains of cucumber virus 1 from spinach. Phyto-

pathology, 40: 729-736.

A pod-distorting strain of the yellow mosaic virus of bean.

Jour. Agr. Res., 77: 301-314.

The relation of common mosaic to blackroot of bean. Jour.

Agr. Res., 77: 315-331.

A cucumber virus strain with a wide leguminous host range.

(Abstract) Phytopathology, 40: 11.

Diseases of canning beans in New York. Plant Dis. Rptr.,

16: 164-166.

Diseases of Green Refugee beans in New York in 1937.

Plant Dis. Rptr., 21: 318-319.

Diseases of canning crops. Plant Dis. Rptr. Sup., 76: 83-89.

A new virus disease of snap beans. Jour. Agr. Res., 60:

279-288.

A histological study of snap bean tissues affected with black-

root. Jour. Agr. Res., 62: 683-690.

Two legume viruses transmissible to tobacco. Phytopatho-

logy, 36: 142-147.

A study of the viruses affecting leguminous plants and the

diseases they cause. Ph. D. thesis, Cambridge Univ.

Alfalfa virus N. (Abstract) Phytopathology, 39: 861.

X-disease and other forms of yellow bean mosaic in Western

Oregon. Oreg. Agr. Expt. Sta., Sta. Cir. Inform., 379, 6 pp.

Variations in potato calico caused by strains of alfalfa mosaic.

(Abstract) Phytopathology, 42: 516-517.

A strain of the alfalfa-mosaic virus causing vine and tuber

necrosis in potato. Phytopathology, 40: 973-991.

The inheritance of a yellowspot character in the bean. Jour.

Hered., 24: 481-486.

Inheritance of a leaf variegation in the common bean. Jour.

Hered., 25: 165-170.

549

29. Pierce, W. H. (1934)

30. Pierce, W. H. (1935)

31. Price, W. C. (1938)

32. Price, W. C. (1940)

33. Quantz, L. (1955)

34. Reddick, D. and Stewart, V. B.

(1919)

35. Richards, B. L. (Jr.) and Burk-

holder, W. H. (1945)

36. Smith, K. M. (1937)

37. Smith, K. M. (1937)

38. Smith, K. M. and Bald, J. G.

(1935)

39. Smith, K. M. and Markham, R.

(1944)

40. Smith, K. M. and Markham, R.

(1944)

41. Snow, G. F. (1955)

42. Stewart, V. B. and Reddick, D.

(1917)

43. Stubbs, M. W. (1937)

44. Thomas, H. R. (1951)

45. Thomas, H. R. (1953)

46. Thomas, W. D. and Graham,

R. W. (1951)

47. Thomas, FI. R. and Zaumeyer,

W. J. (1950)

48. Thomas, H. R. and Zaumeyer,

W. J. (1950)

49. Thomas, H. R. and Zaumeyer,

W. J. (1953)

50. VAN DER Want, J. P. H. (1948).

51. Zaumeyer, W. J. (1938)

52. Zaumeyer, W. J. (1942)

53. Zaumeyer, W. J. (1953)

54. Zaumeyer, W. J. and Fisher,

H. H. (1953)

55. Zaumeyer, W. J. and Harter,

L. L. (1943)

56. Zaumeyer, W. J. and Harter,

L. L. (1944)

Viroses of the bean. Phytopathology, 24: 87-115.

The identification of certain viruses affecting leguminous

plants. Jour. Agr. Res., 51: 1017-1039.

Studies on the virus of tobacco necrosis. Amer. Jour. Bot.,

25: 603-612.

Comparative host range of six plant viruses. Amer. Jour.

Bot., 27: 530-541.

Fin Ringfleckenvirus von Buschbohnen. Phytopath. Ztschr.,

23: 209-220.

Transmission of the virus of bean mosaic in seed and observa-

tions on thermal death-point of seed and virus. Phyto-

pathology, 9: 445-450.

A new mosaic disease of beans. Phytopathology, 33:

1215-1216.

Studies on a virus found in the roots of certain normal-looking

plants. Parasitology, 29: 70-85.

Further studies on a virus found in the roots of certain normal-

looking plants. Parasitology, 29: 86-95.

A description of a necrosis virus disease affecting tobacco

and other plants. Parasitology, 27: 231-245.

Two new viruses affecting tobacco and other plants. Phyto-

pathology, 34: 324-329.

A virus disease of lovage {Ligusticiim scoticum). Phyto-

pathology, 34: 335-340.

A virus from Gladiolus similar to common bean mosaic.

(Abstract) Phytopathology, 45: 696.

Bean mosaic. (Abstract) Phytopathology, 7: 61.

Certain viroses of the garden pea, Pisum sativum. Phyto-

pathology, 27 : 242-266.

Yellow dot, a virus disease of bean. Phytopathology, 41:

967-974.

Isolation of alfalfa mosaic virus strains from field-grown

beans. Plant Dis. Reptr., 37: 390-391.

Seed transmission of red-node virus in Pinto beans. Phyto-

pathology, 41 : 959-962.

Inheritance of symptom expression of pod mottle virus.

Phytopathology, 40: 1007-1010.

Red-node, a virus disease of beans. Phytopathology, 40:

832-846.

A strain of yellow bean mosaic virus producing local lesions

on tobacco. Phytopathology, 43: 11-15.

Het stippelstreep van den Boon, (Phaseoh/s vulgaris), een

ziekte veroorzaakt door een virus, dat in de grond overblyft.

Tijdschr. over Plantenziekten, 54: 85-90.

A heritable abnormality of beans, resembling mosaic. Phyto-

pathology, 28: 520-522.

The inheritance of a leaf variegation in beans. Jour. Agr.

Res., 64: 119-127.

Alfalfa yellow mosaic virus systemically infectious to beans.

Phytopathology, 43: 38-42.

A new necrotic lesion — producing strain of yellow bean

mosaic. Phytopathology, 43: 45-49.

Two new virus diseases of beans. Jour. Agric. Res., 67:

305-327.

A severe necrosis caused by bean-mosaic virus 4 on beans.

Phytopathology, 34: 510-512.

550

57. Zaumeyer, W. J. and Thomas,

H. R. (1947)

58. Zaumeyer, W. J. and Thomas,

H. R. (1948)

59. Zaumeyer, W. J. and Thomas,

H. R. (1950)

60. Zaumeyer, W. J. and Wade, B. L.

(1935)

Bean diseases in some of the mountain states in 1947. Plant

Dis. Rptr. 31: 432^2.

Pod mottle, a virus disease of beans. Jour. Agr. Res., 77:

81-96.

Yellow stipple, a virus disease of bean. Phytopathology, 40:

847-859.

The relationship of certain legume mosaics to bean. Jour.

Agr. Res., 51: 715-749.

551

Fig. 1. — Bean Mosaic Virus.

Fhaseolus vulgaris vars.

A and B. Canadian Wonder, from infected seed. A. Primary leaves. B. Trifoliate

leaves. C. Long Tom. D. Idaho Refugee. E. Kentucky Wonder. C-E. Artificial

infection.

202909

552

Fig. 2. — Bean Mosaic Virus.

Phaseolm vulgaris var. Flaricot, from infected seed.

553

k

Pjg_ 3 _bean Mosaic Virus Strain.

Phaseohis vii'garis var. Canadian Wonder.

From infected seed. B. Artificial infection.

554

Fig. 4. — Bean Chlorotic Ringspot Virus.

Phaseolus vulgaris.

A. var. Canadian Wonder, local lesions. B. var. Flaricot, local lesions. C-E van

Canadian Wonder, systemic symptoms.

555

Fig. 5. — Bean Chlorotic Ringspot Virus.

A. Crotalaria juncea. B. C. spectabilis. C. Voandzeia subterranea. D. Phaseolus

acutifolius. E. P. lunatus. F. Melilotus officinalis. G. Pod of Phaseolus vulgaris var.

Canadian Wonder. H_and I. Glycine max.

556

Fig. 6. — Bean Local Chlorosis Virus.

A. Crotalaria jimcea. B. Vida faba. C and D. Phaseolus vulgaris var, Canadian

Wonder.

557

Fig. 7. — Bean Necrosis Virus.

A. Phaseolus vulgaris var. Canadian Wonder, systemic necrosis. B. Melilotiis

officinalis. C. P ...^ssolus lunatus. D. Trifolium hybridum. E. Phaseolus vulgaris var.

Canadian Wonuci, local reaction.

558

Fig. 8. — Bean Virus Complex.

A-C. Phaseolus vulgaris. A. var. Canadian Wonder, local reaction with the

complex. B. Pods of var. Canadian Wonder. C. var. Haricot, local lesions with the

bean necrotic speckle virus. D. Glycine max. E. Vigna sesquipedalis.

559

A Contribution to our Knowledge

of the Freshwater Algae of the Transvaal Province

by

M. I. Claassen,

Department of General Botany, University of Pretoria,

with the Cooperation of the Division of Botany,

Department of Agriculture

{Being an abridged form of the thesis presented in partial fulfilment of the degree of Master

of Science at the University of Pretoria)

CONTENTS

I. Introduction.

11. Enumeration of the Taxa.

HI. Vegetative Reproduction in Desmids.

IV. Summary.

V. Literature.

I. INTRODUCTION

The freshwater algae occurring in the Transvaal districts of Warmbaths, Nylstroom

and Naboomspruit have apparently not been investigated before. This study consisted

of an enumeration of the taxa found, an investigation of vegetative reproduction in

Desmids and a few hydrobiological observations.

The classification adopted here is essentially that of Wettstein (1935), W. and

G. S. West (1904) and Hustedt (1930). Soon after the investigation was begun, it

became apparent that many of the taxa found did not correspond with any already

known and described. The new ones are described and illustrated in this paper.

It was possible to investigate cell division in the Desmids in living material; the

rate of cell division proved to be of particular interest.

As it was impracticable to collect samples regularly from the various localities,

no systematic hydrobiological observations could be made on variations in pH,

periodicity, diurnal and annual fluctuations in temperature, etc., but where possible

the pH of the water was measured at the time the material was gathered.

Collection and Preservation of Samples. — Samples were taken with a tablespoon

from deposits on submerged rocks and mud in shallow pools, springs and small streams.

Glass vials of 30 ml. capacity with plastic screw caps were used for keeping the samples.

Twenty-seven ml. of the sample and 3 ml. of 40 per cent formalin were added to a vial.

Where living material was to be studied, the samples were placed in one-quart

Ball jars together with a large volume of the water in which they had been found. The

mouth of each jar was covered with a piece of waxed paper held in place by a rubber

band.

Mounting of the Material. — It was found best to mount the algae in the liquid

in which they were preserved or, to prevent drying out, in glycerine. Living material

was mounted in the water in which the sample was collected or in distilled water. Where

necessary a drop of distilled water was added to compensate for evaporation during

microscopic examination.

560

Drawings. — Accurate scale drawings were made by means of a micrometer-net-

ocular on specially printed squared paper.

Type material. — Owing to the fact that it is practically impossible to preserve the

type material successfully over many years in a liquid medium, the author desires that

the drawing plus the description of the new taxa described in the present paper constitute

the type.

Enumeration of the Samples and Localities. — The samples were collected in the

districts of Warmbaths, Nylstroom and Naboomspruit in the Northern Transvaal.

The average altitude of the country exceeds 3,500 feet above sea-level, the maximum

being about 5,000 feet. The area consists largely of sandstone with scattered groups

of red-granite, diabase and quartz; but the lower country, south of Warmbaths and

Naboomspruit, consists of basalt.

The following list of samples collected serves as a key; in the systematic portion

of this paper, only the numbers of the samples are cited. Most of the samples were

collected over a period extending from April, 1954 to July, 1955. The principal localities

where collections were made are shown in Figure 1.

A. — Doornfontein and Leeupoort

Situated about 13 miles N.-N.W. of Nylstroom. The samples were collected

from the Doornfontein Spruit and Leeupoort Dam. The locations are shown in

Figure 2.

1. Greyish-brown deposit covering submerged stones in the spruit near drift,

about half a mile S.W. of Leeupoort Dam, in slow-running water, 16-4-54;

pH 6.2.

2. Greenish-brown ooze from side of road near drift, 16- 4- 54.

3. Reddish-brown ooze from side of road halfway between drift and Leeupoort

Dam, 16 •4- 54.

The following samples were collected mainly from pools on rocks below the overflow

on the S.W. side of Leeupoort Dam, 16-4-54.

4. Greenish-brown, stagnant water.

5. Orange masses floating on surface of small pool.

6. Reddish-brown deposit, stagnant water.

7. Attached to inner side of embankment of Leeupoort Dam, below surface of

the water.

8. Decaying plants on moist part of embankment.

9. In pool on rocks, yellowish-brown.

10. Orange ooze on rocks.

11. Dirty orange-brown scum.

12. Greyish-brown deposit on bottom of pool next to spruit; pH 6.15.

Samples 13 to 17 were collected from Doornfontein Spruit about one mile N.E.

of Leeupoort Dam (underneath bridge on asphalt road between Nylstroom and Vaal-

water), 16-4-54.

13. Orange-brown growth among stones, N.E. side of stream under bridge, pH 6.35.

14. Orange- to greenish-brown scum next to 13.

15. Dirty orange-brown deposit, N.W. side under bridge, slow-running water.

16. Orange deposit from S.W. side under bridge, in placid water.

17. Greyish-brown deposit, from flowing water-furrow.

Samples collected near school about three miles N.E. of Leeupoort Dam, 16- 4- 54.

18. From pond, dirty orange- to greenish-brown scum, pH 5.75.

19. In furrow near pond, greyish-brown deposit.

561

Samples from Doornfontein Spruit about three miles N.E. of Leeupoort Dam,

near school, 16-4-54.

20. From pool near stream, pH 6.1.

21. Another pool near the stream.

22. Orange-brown ooze from side of same pool as in 21.

23. Greenish-brown deposit on stones at drift, in running water, pH 6.25.

B. — Moddemek

Situated about nine miles N.-N.W. of Nylstroom. The part of the farm from which

collections were made belongs to Mr. P. W. Botes. The samples were collected from

springs, smalt ponds and water-furrows. The locations are shown in Figure 3.

24. Floodwater S.W. of spring A, 22-4-54.

25. Bottom of small pond (S. side), almost black deposit; about 90 yards S. of

spring A, 22-4.54; pH 5.8.

26. Same as 25, 19-11-54.

27. Furrow between spring A and small pond, about 85 yards from spring A,

22-4-54.

28. Spirogyra species floating on surface of small pond, 22-4-54.

29. On E. side of larger pond, about 145 yards S. of spring A, 22-4-54.

30. On W. side of same pond, 2-4-55.

31. 32 and 33. Sides, surfaces and bottoms of furrows S. of the pond, 22-4-54.

34 and 35. From a vlei about three-quarters of a mile S.E. of spring A, 19-11-54;

pH 6.31.

36. Near 25 and 26, 2-4-55; pH 5.85.

37. W. side of small pond, 2-4-55.

38. N. side of small pond, 2-4-55.

39. Spring B, 2-4-55; pH 5.45.

40. Spring A, orange scum from side, 2-4-55; pH 5. (This sample did not contain

any algal flora.)

41. Furrow halfway between spring B and small pond, 2-4-55.

C. — Rietfontein 288

Situated about 28 miles N.-N.W. of Nylstroom. The farm belongs to Mr. J. E.

Theron.

42. From spring, 6-4-55; pH 5.7.

43. Water from spring flowing into pond N.W. of spring, 6-4-55.

44. From N.E. side of vlei, W. of pond, 6-4-55.

D. — Glentig

A farm belonging to Mr. L. J. Groenewald; situated about seven miles E. of

Rietfontein 288.

45. From spring, 21-11-54; pH 5.5.

46. From furrow, about 30 yards N. of spring, 21-11-54.

E. — Wannbaths

Samples collected between the old and new bridges over the Plat River, about

3-4 miles W. of Warmbaths; leg. Cholnoky.

47. Almost stagnant water in deep part of river, 30-5-54.

48. Slightly lower down, deposits on a flat stone, slow-running water, 30-5-54.

49. In shade of new bridge in slow-running water, brown covering on sand, 30 - 5 - 54.

50. Near 49, deep bend in river in slow-running water, 30-5-54.

562

F. — Between Warmbaths and Nylstroom

Samples from a tributary of the Nyl River and other places near the road about

halfway between Warmbaths and Nylstroom; leg. Cholnoky.

51. Side of waterfall W. of road, swift-running water over sandstone, 6 -6 ‘54.

52. A small tributary, slow-running water, 6-6-54.

53. On rocks, stagnant water, 6- 6 -54.

54. Shallow water from river above the waterfall, 6‘6'54; pH 6.35.

Samples collected in a vlei E. of the road; pH 6.2.

55. About 200 yards from bridge, beginning of a vlei, stagnant water, 6 -6 '54.

56. Small pools further down the vlei, 6-6-54.

57 and 58. Two places in the vlei on Gramineae and Cyperaceae, 6 -6 -54.

G. — K ran shop

Situated about seven miles E. of Nylstroom; leg. Cholnoky.

59. Small pond with partly broken bank, about 100 yards in diameter, stagnant

water with Marsilea and Nymphaea, 6-6-54; pH 6.75.

60. Overflow of pond, swift-running water, 6- 6- 54.

Samples from a tributary of the Nyl River near Kranskop.

61. Shallow water near the bank, 6-6-54.

62. Deeper, slow-running water from the main part of the stream, 6 '6 '54.

H. — Modderpoort

The farm belonging to Mr. G. Moerdyk between Warmbaths and Nylstroom,

not far from the locality where samples 51-58 were collected; leg. Cholnoky.

63 and 64. Further from the road to the west, two samples collected on stones

in the river, 6 •6-54; pH 6.2.

1. — Mosdene

Samples collected on the farm of Mr. E. A. Galpin, near Naboomspruit, 12 ‘7 '55;

pH 6. 1-6.4; leg. Cholnoky.

65. From the portion of the farm called Masoga, in a swimming pool among

Myriophyllum.

66. From the shallow part of a vlei with a sandy bottom on Masoga.

67. Same locality as 66 but in deep water among Green Algae.

68. Wall of Premier Dam, overflow, slow-running water, among Marsilea.

69. From Premier Dam; the bottom of the dam was sampled at a place far from

the bank, and where the water was one metre deep.

70. A vlei situated between the Premier Dam and the Nyl River; in water about

10 cm. deep.

71. Near 70, in a furrow near the Nyl-road.

72. From the Nyl River, in swift-running water, on marshy sand; clearly oligo-

trophic.

73. From the Nyl Dam near the Nyl River, large algal masses.

74. Premier Dam, living material; leg. Dr. S. Saubert.

Geographical Distribution. — In the taxonomic part of this paper, the general

geographical distribution of each taxon of the Desmidiaceae encountered has been

added. As the author was unable to study all literature dealing with Desmids, the

details regarding their geographical distribution must needs be considered incomplete.

563

II.— ENUMERATION OF THE TAXA

TRIBUS: SCHIZOPHYTA

Class: Schizophyceae (Cyanophyceae).

Order: Chroococcales.

Fam. : Chroococcaceae.

Merismopedia Meyen

1. M. convolutaBrib. (Geitler & Pascher : Susswasserflora, 12, 1925, p. 106.) Sample

20.

2. M. glauca (Ehrenb.) Nag. (Geitler & Pascher; Susswasserflora, 12, 1925, p. 106,

fig. 125.) Samples 47, 48.

3. M. punctata Meyen. (Geitler & Pascher: Susswasserflora, 12, 1925, p. 106, fig.

124.) Samples 39, 48.

Order: Hormogonales.

Fam.: Stigonemataceae.

Stigonema Ag.

4. S. pseudominutum spec, nova, S. minuto (Ag.) Hass, affinis, sed filamentis et hormo-

goniis tarn angustus ut vel 5 jx aequantibus, vaginis filamentorum vetustiorum

luteo-fulvis et heterocystis nonnumquam etiam terminalibus differt.

Except for the considerably narrower threads and hormogonia this agrees with

5. minutum (Ag.) Hass. (Geitler & Pascher: Susswasserflora, 12, 1925, p. 186, figs.

224, 225); the heterocysts occupy terminal as well as lateral and intercalary positions.

The cells are arranged in 1-4 series, but mainly in a single series. The threads are

well-branched, the branches being narrower than the main axis. The older sheaths

are yellowish-brown in colour whereas the younger sheaths are colourless. Diameter

of threads 5-20 p. (Tab. I, figs. 1, 2.) Sample 27.

Fam.: Scytonemataceae.

Scytonema Ag.

5. S. myochrous (Dillw.) Ag. (Geitler & Pascher: Susswasserflora, 12, 1925, p. 275,

figs. 324, 325.) Threads narrower than in the typical form and some of the cells

are shorter than broad; parts of the sheath lack the diverging layers. Breadth

of thread 10-24 p\ breadth of cells 5-7 p\ length of cells 3-6 p\ breadth of hetero-

cysts 8 p; length of heterocysts 9-10 p. (Tab. 1, figs. 3-7.) Sample 22.

Fam.: Nostocaceae.

Nostoc Vaucher

6. N. commune Vaucher (Geitler & Pascher: Susswasserflora, 12, 1925, p. 301, fig.

350). Sample 29.

7. N. nylstromicum spec, nova, N. muscorum Kiitz. affinis, sed heterocystis cylindratis

(nunquam apice rotundatis) valde distincta. ,, Dauerzellen ” abesse videntur.

This species comes nearest to N. muscorum Kiitz. (Geitler & Pascher: Suss-

wasserflora, 12, 1925, p. 299, fig. 349), but can be distinguished by the shape of the

heterocysts which are cylindrical in outline; no “Dauerzellen” were observed.

Trichomes loosely twisted, blue-green in colour; cells cylindrical, 2 -8-3 -6 p broad

and 3 -5-8 -75 p long. Heterocysts 3 -85-4 -4 p broad and 5 -6-9 -5 p long. A number

of trichomes within a common mucilaginous sheath. (Tab. 2, fig. 2.) Sample 20.

564

8. N. pseudogelatinosum spec, nova, N. gelatinoso Schousboe affinis, sed differt magni-

tudine aliquanto minore et cellulis interdum non longioribus quam latioribus;

constrictio saeptorum magis conspicua, heterocystorum nulla nisi terminalia sunt,

neque ullae ,, Dauerzellen ” adesse videntur.

This differs from N. gelatinosum Schousboe (Geitler & Pascher: Susswasserflora,

12, 1925, p. 298, fig. 347) in having smaller dimensions, more marked constrictions at

the septa, terminal heterocysts, no “ Dauerzellen ” and cells which can be as long as

they are broad. Trichomes irregularly twisted, blue-green in colour; cells quadrate

or cylindrical, 2 -4-2 -8 /x broad and 2 -4-6 -3 p long; heterocysts always occupy a

terminal position, more or less elliptical in outline, 3- 1-4 /x broad and 5 -5-7 -9 /x long.

Only the common sheath, within which a number of trichomes occur, is distinct and

of a dirty yellowish colour. (Tab. 2, fig. 1.) Sample 20.

Anabaena Bory

9. A. galpinii spec, nova, A. affini Lemm. affinis, sed trichomatis semper separatis,

vaginae mucosae abesse videntur; cellulae breviores quam latiores. ,, Dauerzellen

iuxta heterocysta sitae sunt.

This species differs from A. affinis Lemm. (Geitler & Pascher: Susswasserffora.

12, 1925, p. 320, fig. 374) in having individual trichomes (not containing a number of

trichomes within a common sheath), in the absence of a sheath, in the position of the

“ Dauerzellen ” (adjacent to the heterocysts), and in the cells being shorter than broad.

Trichomes free-fioating, straight, blue-green in colour; cells 8 • 5-9 p broad and 4 •3-6- 8 p

long; heterocysts more or less spherical, 8 • 5-9 p broad and 8-9 p long. “ Dauerzellen ”

single, oval, bright blue-green in colour and more granular than the vegetative cells,

about 11 p broad and 16 p long. (Tab. 2, fig. 3.) Sample 66.

Named after Mr. E. A. Galpin of Mosdene, Naboomspruit.

10. A. mesiana spec, nova, A. catenulae (Kiitz.) Born, et Flah. affinis, sed differt et

trichomatis semper separatis et quod vagina mucosa abest. ,, Dauerzellen ”

breviores adsunt.

This species is near A. catenula (Kiitz.) Born, et Flah. (Geitler & Pascher: Suss-

wasserffora, 12, 1925, p. 318, ffg. 370); however the trichomes are single and the

“ Dauerzellen ” are not as long as in Born, and Flah.’s species. Trichomes free-floating,

straight, blue-green in colour; no sheaths were discernible. Cells spherical or slightly

elliptical, 5-6 p broad and 5-7 p long. “ Dauerzellen ” single, adjacent to the hetero-

cysts, oblong, 7-5-8 p broad and 11-5-19 p long. (Tab. 2, fig. 4.) Sample 66.

This species is named after the late Prof. Dr. M. G. Mes, formerly Head of the

Department of Plant Physiology and Biochemistry, University of Pretoria.

11. A. pseudocatenula spec, nova, A. catenulae (Kiitz.) Born, et Flah. affinis, sed eo

differt quod separata sunt trichomata, vagina mucosa abesse videtur, cellulae

vero minores sunt ac breviores quam latiores neque ullae ,, Dauerzellen ” visae

sunt.

This species closely resembles A. catenula (Kiitz.) Born, and Flah., but the trichomes

are single, the cells are smaller and shorter than broad and no sheaths or “ Dauerzellen ”

were observed. Trichomes free-floating, straight, blue-green in colour. Cells 3- 16-3-65

P broad and 1-58-3-5 p long. Heterocysts elliptical, 3-5 /x broad and 6-63 p long.

(Tab. 2, fig. 5.) Sample 1.

Fain.: Oscillatoriaceae.

Oscillatoria Vaucher

12. O. nylstroinica spec, nova, O. simpUcissimae Gom. affinis, sed trichomatis angus-

tioribus subcaeruleisque, cellulis brevioribus vel tarn longis quam latis bene

distinguenda.

565

This species agrees superficially with O. simplicissima Gom. (Geitler & Pascher,

Siisswasserflora, 12, 1925, p. 364, fig. 429), but the trichomes are narrower, and light

steel-blue in colour, and the cells are shorter than broad, or quadrate. Trichomes

solitary, free-floating, more or less straight, not tapering towards the extremities,

4 -6-4 -75 jj. broad. Cells granular at the septa and not constricted, 1-45-4 -8 /x long.

Apical cell broadly rounded, without calyptra. (Tab. 2, fig. 7.) Samples 1, 11.

13. O. princeps Vaucher. (Geitler & Pascher: Siisswasserflora, 12, 1925, p. 358,

fig. 421.) Sample 47.

14. O. subpristleyi spec, nova, O. pristleyi W. et G. S. West affinis, sed valde differt

quod trichomata extrema non curvantur et quod cellulae paulo angustiores porro

longiores quam latiores sunt neque ad saepta constrictae.

This specimen closely resembles O. pristleyi W. et G. S. West (Geitler & Pascher:

Siisswasserflora, 12, 1925, p. 371, fig. 451), but differs from the latter in the absence

of a curvature near the apex, the narrower trichomes, the cells being longer than broad

and in the absence of constrictions at the septa. Trichomes solitary, free-floating,

bright blue-green in colour, more or less straight, tapering towards the extremities,

cells 3-16-5 -5 p, broad and 4 -7-9 -5 p long, granular at the septa. Apical cell slightly

pointed, without calyptra. (Tab. 2, fig. 6.) Sample 1.

15. O. tenuis Ag. (Geitler & Pascher: Siisswasserflora, 12, 1925, p. 362, figs. 427,

428a.) (Tab. 2, fig. 8.) Samples 1, 5, 7, 9, 11, 12, 13, 14, 15, 18.

16. O. waterbergensis spec, nova, O. annae van Goor affinis, sed differt eo quod cellulae

ad saepta non nisi levissime constrictae, breviores vel paulo longiores quam latiores

et ad saepta granulatae sunt neque trichomata ad apicem sunt curvata.

This species agrees superficially with O. annae van Goor (Huber-Pestalozzi: Die

Binnengewasser, Band XVI, Teil 1, p. 232, fig. 171); but some of the cells are longer

than broad, and the trichomes lack a curvature near the apex, are slightly constricted

at the septa and are granular. Trichomes more or less straight, greyish-green in colour,

not tapering towards the extremities, 7-7-5 p broad. Cells 4-75-8 p long; apical

cell rounded, without calyptra. (Tab. 2, fig. 9.) Sample 1.

Lyngbya Ag.

17. L. uliginosa spec. nova. Nulla affinitate obvia. Cellulae vel tarn longae quam

latae vel etiam longiores, filamentis ad saepta manifeste constrictae sunt; filamenta

separata sunt neque vagina mucosa arte ad trichomata adhaeret.

Threads straight, solitary, free-floating, 4 -5-5-1 p broad; sheath of the filament

thin, colourless, encircling the trichome loosely. Trichomes constricted at septa,

2 -8-3 -3 p broad; cells 2 -8-6 -2 p long, dark green in colour; apical cell rounded.

(Tab. 2, fig. 10.) Sample 3.

TRIBUS: MONADOPHYTA

Class: Flagellatae.

Order: Euglenineae.

Fam.: Euglenaceae.

Euglena Ehrenb.

18. E. spirogyra Ehrenb. var. major var. nova. Haec specimina multo maiora sunt

quam typus et varietates specie! adhuc notae, neque ullae formae intermediae

visae sunt.

This variety is much larger than the type and its varieties, with the exception of

var. suprema Skuja (Huber-Pestalozzi: Die Binnengewasser, Band XVI, Teil 4, p. 102,

fig. 83), from which it differs in shape and breadth of cells. Cells 152-196 p long and

22-26 // broad. (Tab. 32, fig. 1 1.) Samples 25, 39.

202909—,-)

566

As most Flagellates are either completely dissolved or quite unidentifiable in

formalin-treated material, examination of living material must be made for their

recoanition. Several unidentifiable Euglena species were found in samples 2, 23, 24,

25, 26, 27, 34, 35, 36 and 47.

Phacus Dujardin

19. Ph. pleuronectes (O.F.M.) Duj. (Pascher & Lemmermann: Siisswasserflora,

2, 1913, p. 138, fig. 236.) Samples 1, 12, 32, 34, 35.

TRIBUS: CONJUGATOPHYTA

Class: Conjugatae.

Fam.: Mesotaeniaceae.

Cylindrocystis Menegh.

20. C. barbarica spec, nova, affinis C. crassae De Bary, sed cellulis 1-2-1-4-plo

longioribus quam latioribus, etiam forma chromatophorum differt, quippe quibus

processus minus sint radiantes.

This species superficially agrees with C. crassa De Bary (W. & G. S. West, Monogr.

Brit. Desm., Vol. I, 1904, p. 59, PI. IV, figs. 33-38); but the cells are about 1-2-1 -4

times longer than broad and the chloroplasts possess fewer radiating processes. Cells

small, unconstricted, subcylindrical, with the apices broadly rounded; cell wall smooth

and colourless; chloroplasts stellate with a few radiating processes. Length 28-32 p;

breadth 23-24 ijl. (Tab. 3, fig. 1.) Sample 19.

21. C. caffra spec, nova, affinis C. brebissonii Menegh., sed cellulis latioribus, etiam

forma chromatophori differt.

This species somewhat resembles C. brebissonii Menegh. (W. & G. S. West, Monogr-

Brit. Desm., Vol. I, 1904, p. 58, PI. IV, figs. 23-32; PI. V, fig. 10); but the cells are

broader than in Meneghini’s species and the shape of the chloroplast differs. Cells

of medium size, cylindrical, unconstricted, about 1-7-2 -6 times longer than their

diameter, apices rounded, chloroplasts with a few radiating prolongations; cell wall

smooth and colourless. Length 36-58 /x; breadth 20-24 jx. (Tab. 3, fig. 2.) Samples

18, 19.

Netrium Nag.

22. N. digitus (Ehrenb.) Itzigs. & Rothe. (W. & G. S. West, Monogr. Brit. Desm.,

Vol. I, 1904, p. 64, PI. VI, figs. 14-16.) Several of the specimens measured are

considerably smaller than in the typical form namely 92-126 p long, 33-44 /x

broad and 15-20 /x broad near the apices. Samples 2, 19, 22, 25, 35, 39, 41, 66,

68, 69, 70.

Geographical Distribution. — England; Wales; Scotland; Ireland; Generally

distributed in Europe; Faeroes; India; Ceylon; Siam; Java; Central China; Japan;

Australia; New Zealand; Azores; United States; British Guiana; Brazil; Portuguese

East Africa; South Africa.

23. N. oblongum (De Bary) Liitkem. var. cylindricum W. & G. S. West. (W. & G. S.

West, Monogr. Brit. Desm., Vol. I, 1904, p. 67, PI. V, fig. 7.) Sample 25.

Geographical Distribution. — England; Wales; Scotland; Italy; Somaliland;

Japan; South Africa.

567

Order: Desmidiales.

Fam. : Desmidiaceae.

Subfam.: Saccodermae.

Gonatozygon De Bary

24. G. monotaenium De Bary. (W. & G. S. West, Monogr. Brit. Desm., Vol. I, 1904,

p. 30, PI. I, figs. 1-7; PI. V, fig. 5.) Sample 20.

Geographical Distribution. — England; Wales; Scotland; Ireland; France;

Germany; Austria; Hungary; Sweden; Faeroes; N. Russia; Greenland; India;

Ceylon; Siam; Sumatra; West, Central and East Africa; W. Indies; United States;

Brazil (var.); South Africa.

Subfam. : Placodermae.

Penium Breb.

25. P. barbaricum spec, nova, affinis P. margaritaceo (Ehrenb.) Breb., sed granulis

membranae diffuse (neque in seriebus longitudinalibus) dispositis, membranaque

achroa differt.

This species superficially resembles P. margaritaceum (Ehrenb.) Breb. (W. & G. S.

West, Monogr. Brit. Desm., Vol. I, 1904, p. 83, PI. VIII, figs. 32-35); but the cell wall

is colourless and furnished with numerous scattered granules. Cells of medium size,

about 3 '2-6 -3 times longer than their diameter, with a distinct median constriction,

apices rounded. Length 70-144 /z; breadth 18-23 /x. (Tab. 3, fig. 6.) Samples 20,

47, 48.

26. Penium species ad P. cruciferum (De Bary) Wittr. Cellulae cum P. crucifero valde

congruentes praeterquam quod paulo minores sunt neque ullae costae in chromato-

phora visae sunt. Zygospora inventa est sed P. cruciferi zygospora adhuc ignoratur.

This comes very near P. cruciferum (De Bary) Wittr. (W. & G. S. West, Monogr.

Brit. Desm., Vol. I, 1904, p. 100, PI. X, figs. 18-19), but the cells are slightly smaller

and no ridges were observed in the chloroplasts. A zygospore was found, but the

zygospore of P. cruciferum has hitherto been unknown. The cells are very small,

about 1-7-2 -2 times longer than their diameter; ellipsoid-subcylindrical with a slight

median constriction, apices rounded; cell wall smooth and colourless; each chloroplast

with a central pyrenoid. Zygospore subquadrate, sides concave; cell wall straw-

coloured. Length 12-13 u; breadth 6-7 u; diameter of zygospore 12-14 a. (Tab. 3,

figs. 16-17.) Sample 35.

27. P. cucurbitinum Biss. (W. & G. S. West, Monogr. Brit. Desm., Vol. I, 1904, p. 94,

PI. IX, figs. 13, 14.) Sample 39.

Geographical Distribution. — England; Wales; Scotland; Ireland; Japan; South

Africa.

28. P. cucurbitinum f. botesii f. nova, affinis P. cucurbitino formae minori West, sed

cellulis multo minoribus, apice hebetato, punctis in 8 series longitudinales ordinatis

differt.

This species somewhat resembles P. cucurbitinum f. minor West (W. & G. S. West,

Monogr. Brit. Desm., Vol. I, 1904, p. 95, PI. IX, fig. 16); but the cells are much smaller,

with the apices truncate; cell wall punctate, punctae arranged in about eight longitudinal

rows. Cells small, about twice as long as broad, with a slight median constriction;

chloroplasts with about four longitudinal ridges. Length 24-28 p.; breadth 11-14 p;

breadth of isthmus 7-10 p. (Tab. 3, fig. 15.) Sample 27.

Named after Mr. P. W. Botes of Moddernek, Nylstroom.

568

29. P. curtum Breb. (W. & G. S. West, Monogr. Brit. Desm., Vol. I, 1904, p. 97,

PI. X, figs. 21, 22.) Samples 7, 11, 32.

Geographical Distribution. — England; Wales; Scotland; Ireland; France;

Germany; Austria; Italy; Norway; Sweden; Greenland; Spitzbergen; Nova

Zembla; Franz-Joseph Land; India; Burma; Siam; West Africa; Madagascar;

West Indies; United States; South Africa.

30. P. curtum var. waterbergense var. nova, a typo speciei membrana ad apices cellulae

baud incrassata et sculptura membranae differt.

This agrees somewhat with the typical form, but differs in the absence of a cell

wall thickening at the apices and in the sculpture of the cell wall. Cell wall punctate,

punctae arranged in about eight longitudinal rows. Length 27-33 breadth 18-20 /a;

breadth of isthmus 16-18 /x. (Tab. 3, fig. 9.) Samples 30, 51, 52, 53, 54, 58, 64.

31. P. jenneri Ralfs. (W. & G. S. West, Monogr. Brit. Desm., Vol. I, 1904, p. 77,

PI. VII, figs. 20, 21.) Sample 2.

Geographical Distribution. — England; Scotland; Ireland; Germany; Austria;

Norway; Sweden; Poland; South Africa.

32. P. libellula (Focke) Nordst. var. schweickerdtii var. nova, affinis P. libellulae var.

interrupto W. et G. S. West, sed cellulis 3-8-5-plo longioribus quam latioribus

chromatophoris constrictis neque interruptis differt.

This approaches P. libellula var. interruptum W. & G. S. West (W. & G. S. West,

Monogr. Brit. Desm., Vol. I, 1904, p. 74, PI. VII, figs. 9-10); but the cells are shorter,

and 3 ■ 8-5 times longer than their diameter; chloroplasts constricted and not transversely

divided. Cells of medium size, unconstricted, fusiform, apices slightly truncately

rounded; chloroplasts with about 10 longitudinal ridges. Length 68-108 p.\ breadth

16-25 /x; breadth near apices 6-12 /X. (Tab. 3, fig. 7.) Samples 2, 24, 25, 27, 32, 39, 41.

Named after Prof. Dr. H. G. Schweickerdt, Head of the Department of General

Botany, University of Pretoria.

33. P. margaritaceum (Ehrenb.) Breb. (W. & G. S. West, Monogr. Brit. Desm., Vol. I,

1904, p. 83, PI. VIII, figs. 32-35.) Tab. 3, fig. 3 shows a form during cell-division

in which the cell wall is furnished with longitudinal rows of granules, but the granules

are irregularly scattered at the apices. Samples 13, 18, 19.

Geographical Distribution. — England; Wales; Scotland; Ireland; France;

Germany; Austria; Poland; Hungary; Italy; Portugal; Norway; Sweden; Finland;

Lapland in Russia; Faeroes; Greenland; Spitzbergen; Java; New Zealand; East

Africa; Azores; United States; Ecuador; Switzerland; South Africa.

34. P. margaritaceum var. brevior var. nova, a typo speciei cellulis valde latioribus

fere 3-6-4-8-plo longioribus quam latioribus differt.

Length in proportion to breadth differs from the typical form; cells 3 -6-4 -8 times

longer than their diameter. Length 84-96 /x; breadth 20-23 /x. (Tab. 3, fig. 5.) Samples

13, 19.

35. P. margaritaceum var. incognitum var. nova, a typo speciei semicullulis ad basim

tumidis similiter ac apud genus Pleurotaenium Nag. differt.

This variety differs from the typical form in possessing a basal inflation in each

semicell similar to that found in the genus Pleurotaenium Nag. Cells large, 7 -7-8 -4

times longer than their diameter, apices rounded; cell wall yellowish-brown and

furnished with longitudinal rows of granules; chloroplasts with about 10 longitudinal

plates (or ridges) and showing a median interruption. Length of cell 184-185 /x; breadth

at base of semicell 22 /x, near basal inflation 18-18-5 /x, at middle of semicell 22-24 /x,

at apices 16-17 p.. (Tab. 3, fig. 4.) Sample 13.

569

36. P. mesianum spec, nova, affinis P. phymatosporo Nordst., sed membrana cellulae

granulata baud striata differt.

This is near P. phymatosporum Nordst. (W. & G. S. West, Monogr. Brit. Desm.,

Vol. 1, 1904, p. 91, PI. VI, figs. 9-1 1); but the cell wall is furnished with minute granules

and is not striated. Cells small, more than twice as long as broad, with a slight median

constriction ; apices truncately rounded ; cell wall conspicuously covered with granules ;

each chloroplast has a single pyrenoid. Length 32-42 p\ breadth 13-16 p. (Tab. 3,

figs. 10-13.) Samples 3, 24, 27.

Named after the late Prof. Dr. M. G. Mes, formerly Head of the Department of

Plant Physiology and Biochemistry, University of Pretoria.

37. P. navicula Breb. (W. & G. S. West, Monogr. Brit. Desm., Vol. I, 1904, p. 75,

PI. VII, figs. 12-15, 19.) Specimens were found with length 33-80 p; breadth

12-20 p; breadth of apices 6-8 p. Samples 2, 51, 52, 53, 54, 57, 58, 63.

Geographical Distribution. — England; Wales; Scotland; Ireland; France;

Germany; Austria; Hungary; Norway; Sweden; Italy; Russia; Lapland; Faeroes;

Greenland; India; Ceylon; Singapore; United States; Brazil; First record for

South Africa.

38. P. phymatosporum Nordst. (W. & G. S. West, Monogr. Brit. Desm., Vol. 1, 1904,

p. 91, PI. VI, figs. 9-11.) Sample 2.

Geographical Distribution. — England; Scotland; Ireland; France; Austria;

Italy; Argentina; Trinidad (?); Japan; Switzerland; Nantucket, U.S.A.; South

Africa.

39. P. pseudorufescens spec, nova, affinis P. rufescenti Cleve, sed cellula dimidio minore

differt; constrictura media baud visa, membrana achroa.

Near P. rufescens Cleve (W. & G. S. West, Monogr. Brit. Desm., Vol. I, 1904,

p. 99, PI. VI, figs. 12, 13); but the cells are only about half the size of those of Cleve’s

species; no trace of a constriction in the middle of the cell was found; cell wall colourless.

Cells small, about 2-5-3 times longer than their diameter, slightly attenuated towards

the apices, which are rounded; cell wall smooth. Length 28-30 p; breadth 9-12 p.

(Tab. 3, fig. 14.) Sample 24.

40. P. subcucurbitinum spec, nova, affinis P. cucurbitino Biss., sed cellulis fere duplo

longioribus quam latioribus, apice hebetato nec non sculptura membranae differt.

Near P. cucurbitinum Biss. (W. & G. S. West, Monogr. Brit. Desm., Vol. I, 1904,

p. 94, PL IX, figs. 13, 14); however the cells are about two times longer than their

diameter, and the apices are truncate; it also differs in the sculpture of the cell wall.

Cells small, subcylindrical, with a slight median constriction, gradually attenuated

towards the apices; cell wall furnished with granules. Length 39 p\ breadth 20 p.

(Tab. 3, fig. 8.) Sample 39.

Closterium Nitzsch

41. C. abruptum West var. westiorum var. nova, affinis C. abrupto var. breviori W. et

G. S. West, sed cellulis minus arcuatis vel rectis et polis hebetatis differt.

Near C. abruptum var. brevius W. & G. S. West (W. & G. S. West, Monogr. Brit.

Desm., Vol. I, 1904, p. 160, PI. XX, figs. 1 1, 12); but the cells are less curved or practi-

cally straight and the apices are more truncate. Cells about 5 -9-6 -5 times longer than

their diameter. Chloroplasts with about eight ridges and a central series of 2-3 pyre-

noids. Length 75-84 /x; breadth 12-13 /x; breadth of apices 6-7 /x. (Tab. 6, figs. 9, 10).

Sample 2.

Named after Messrs. W. & G. S. West.

570

42. C. acewsum (Schrank) Ehrenb. var. waterbergeme var. nova, affinis C. aceroso

var. elongato Breb., sed cellulis valde minoribus et polis hebetatis differt.

This comes near C. acerosum var. elongatum Breb. (W. & G. S. West, Monogr.

Brit. Desm., Vol. I, 1904, p. 148, PI. XVIII, fig. 1); but differs in having smaller cells

with truncate apices. Cells of medium size, about 9- 5-10- 5 times longer than their

diameter, inner margin slightly concave, gradually attenuated towards the apices,

which are truncate; cell wall yellowish-brown, striated, about 40 striae visible across

the cell; chloroplasts with about eight ridges and a central series of 10-15 pyrenoids;

terminal vacuoles with a number of moving granules. Length 398-432 breadth

41-42 /m; breadth of apices 8 /x. (Tab. 4, figs. 6, 7.) Sample 54.

43. C. acutum (Lyngb.) Breb. var. linea (Perty) W. & G. S. West f. minor f. nova, a

varietate linea (Perty) W. et G. S. West cellulis multo brevioribus (nec angustioribus)

differt.

The cells are considerably shorter than those in var, tinea (Perty) W. & G. S. West

(W. & G. S. West, Monogr. Brit. Desm., Vol. I, 1904, p. 178, PI. XXIIl, fig. 15), but

have the same breadth. Cells 16-18 times longer than their diameter; chloroplasts

with 3-4 pyrenoids; terminal vacuoles with one moving granule. Length 64-88 /x;

breadth 4-5 /x. (Tab. 6, fig. 22.) Sample 22.

44. C. atomicum spec, nova, affinis C. striolata Ehrenb., sed differt cellulis multo

minoribus, margine ventrali medio incrassato, membrana cellulae 14 striis quae

videre possunt ornata.

This is near C. striolatum Ehrenb. (W. & G. S. West, Monogr. Brit. Desm., Vol. I,

1904, p. 122, PI. XIII, figs. 7-16); but the cells are considerably smaller, inner margin

tumid in the median part. Cells of medium size, 7-8 times longer than their diameter,

curved, inner margin concave and tumid in the median portion, gradually attenuated

towards the apices, which are slightly truncate; cell wall yellowish-brown, about 14

striae visible across the cell. Length 160-196 /x; breadth 22- 5-26 |x; breadth of apices

9-12 iJ.. (Tab. 5, fig. 16.) Sample 3.

45. C. barbaricum spec, nova, affinis C. parvulo Nag., sed cellulis minus arcuatis, polis

hebetatis, vacuolis terminalibus granulo singulo se moventi praeditis differt.

This species differs from C. parvulum Nag. (W. & G. S. West, Monogr. Brit. Desm.,

Vol. I, 1904, p. 133, PI. XV, figs. 9-12) in the cells having a smaller curvature, the apices

being acute and in the terminal vacuoles containing only one moving granule. Cells

small, 7-9 times longer than their diameter, curved, inner margin concave, not tumid,

gradually attenuated towards the apices, which are truncately-rounded; cell wall

smooth and colourless; chloroplasts with six ridges and six pyrenoids. Figs. 5 and 6

show cells which have atypical apices, possibly owing to the fixative used. Length

66-92-5 jLx; breadth 8-11-5 /x. (Tab. 6, figs. 4-6.) Sample 2.

46. C. boyanwn spec, nova, affinis C. stellenboschensi Hodgetts, sed cellulis maioribus,

minusque arcuatis et apice utroque poro praedito differt.

Near C. stellenboschense Hodgetts (W. J. Hodgetts, Trans. Roy. Soc. of S.A.,

Vol. XIII, 1926, p. 74, fig. 8); but the cells are larger and less curved, there is a pore

at each apex. Cells of medium size, 7 -3-7 -4 times longer than their diameter, inner

margin concave, slightly tumid in the median portion, gradually attenuated towards

the apices, which are slightly acutely-rounded; cell wall smooth, straw-coloured or

yellowish, with a pore at each apex; chloroplasts with about six ridges and a central

series of five pyrenoids; terminal vacuoles large, with a number of moving granules.

Length 167-174 /x; breadth 23-24 /x. (Tab. 5, fig. 13.) Samples 57, 58, 64.

47. C. caffrorum spec, nova, affinis C. cynthiae De Not, sed cellulis multo minus arcuatis

et polis hebetatis differt. Membrana cellulae plus minusve 9 striis quae videri

possunt ornata.

571

This species is near C. cynthia De Not (W. & G. S. West, Monogr. Brit. Desm.,

Vol. I, 1904, p. 113, PI. XI, figs. 1-3); but the cells are less curved and have truncate

apices. Cells small, curved, inner margin concave, apices truncate; chloroplasts

with six ridges and a central series of about four pyrenoids; cell wall yellowish-brown

and striated, about nine striae being visible across the cell; terminal vacuoles with a

single moving granule. Length 100-104 /u,; breadth 13 - 5-14 (Tab. 8, fig. 4.) Sample

64.

48. C. calosporum Wittr. var. minor var. nova, a typo specie! cellulis multo minoribus

ad apices hebeto-rotundatis et minus curvatis differt.

These specimens differ from the typical form (W. & G. S. West, Monogr. Brit.

Desm., Vol. I, 1904, p. 138, PI. XVI, figs. 1-4) in having smaller cells, truncately-rounded

apices and a smaller curvature. Cells about 7-7 times longer than their diameter;

chloroplasts with a single series of about two pyrenoids. Length 54 /x; breadth 7 p..

(Tab. 6, fig. 20.) Sample 3.

49. C. ceratium Perty var. angustum var. nova, a typo specie! cellulis multo angustioribus

(cellula 49- 1-56- 1-plo longior quam latior) differt. Formae intermediae non

visae.

Cells narrower than in the typical form (W. & G. S. West, Monogr. Brit. Desm.,

Vol. 1, 1904, p. 176, PI. XXlll, figs. 6-8). Cells about 49- 1-56- 1 times longer than

their diameter; straight for almost the whole length of the cell; chloroplasts with

1 1-12 pyrenoids. Length 123 -5-1 24 /x; breadth 2- 2-2- 5 ju. (Tab. 6, fig. 14.) Sample 3.

50. C. cornu Ehrenb. var. minor var. nova, a typo cellulis multo minoribus et poli

minus acutis differt. Formae intermediae non visae.

Cells considerably smaller than in the typical form (W. & G. S. West, Monogr.

Brit. Desm., Vol. 1, 1904, p. 157, PI. XX, figs. 1-5), being about 13-17 times longer

than their diameter; chloroplasts with about three pyrenoids. Length 64-74 ; breadth

4-5 p\ breadth of apices 2 -2-2 -6 p. (Tab. 6, fig. 23.) Sample 22.

51. C. cornu var. angustum var. nova, a var. minori Claassen cellulis paulo angustioribus

et polis multo acutioribus differt.

Near var. minor Claassen; but the cells are narrower and the apices more acute.

Cells 12-5-19 times longer than their diameter. Length 75-77 p-, breadth 3-4-4 p.

(Tab. 6, fig. 24.) Sample 2.

52. C. cymbellaeformis spec. nova. Nulla affinitate obvia. Cellulae naviculares,

margo ventralis tenuiter convexus apices hebetati, et membrana cellulae paulo

incrassata.

Cells of medium size, boat-shaped, inner margin slightly convex, apices truncate

with a slightly thickened cell wall; cell wall smooth and colourless; chloroplasts with

about 10 ridges and a central series of four pyrenoids; subterminal vacuoles large,

with a number of moving granules. Length 168-171 -5 /x; breadth 28-29 p; breadth

of apices 8-10 p. (Tab. 5, fig. 15.) Sample 2.

53. C. cynthia De Not var. waterbergense var. nova, differt a typo specie! cellulis minus

arcuatis pauloque latioribus, polis minus acutis et membrana plus minusve 13

striis quae videri possunt ornata.

The cells are less curved than in the typical form (W. & G. S. West, Monogr.

Brit. Desm., Vol. I, 1904, p. 113, PI. XI, figs. 1-3); they are slightly broader, the apices

are less acute and the cell wall has about 13 striae visible across the cell. Inner margin

slightly tumid; cell wall yellowish-brown; chloroplasts with about six ridges and a

central series of four pyrenoids; terminal vacuoles with one moving granule. Length

115-119 p\ breadth 18-5-20 p. (Tab. 8, fig. 3.) Sample 64.

572

54. C. decorum Breb. var. minor var. nova, differt a typo cellulis multo minoribus et

membrana striis 23-24 quae videri possunt disposita.

Cells considerably smaller than in the typical form (W. & G. S. West, Monogr.

Brit. Desm., Vol. I, 1904, p. 184, PI. XVII, figs. 7, 8; PI. XXVIII, figs. 1-3) and the

cell wall with more striae. Cells of medium size, 10-14-6 times longer than their

diameter, curvature somewhat variable, inner margin concave, slightly but broadly

tumid in the median portion, gradually attenuated towards the apices, which are truncate;

cell wall straw-coloured or yellowish-brown; each chloroplast with a central series

of 6-11 pyrenoids. Length 176-292 p.\ breadth 17-20 breadth of apices 4-5 /la.

(Tab. 5, fig. 12.) Samples 3, 63.

55. C. ehrenbergii Menegh. (W. & G. S. West, Monogr. Brit. Desm., Vol. I, 1904,

p. 143, PI. XVII, figs. 1-4.) Sample 42.

Geographical Distribution. — England; Wales; Scotland; Ireland; France;

Belgium; Germany; Austria and Galicia; Bosnia in Hungary; Italy; Portugal;

Norway; Sweden; Denmark; North, Central and South Russia; Faeroes; Central

China; Japan; India; Siam; Samoa; New Zealand; Central Africa (var.); East

Africa; United States; West Indies; Brazil; Uruguay; Patagonia; South Africa.

56. C. ehrenbergii var. minutissimum var. nova, affinis C. ehrenbergii Menegh., sed

cellulis fere 5-7-plo minoribus, polis plus hebetatis, vacuolis terminalibus granulo

singulo se moventi praeditis differt.

The cells are about 5-7 times smaller than in the typical form, the apices are more

truncate and the terminal vacuoles have a single moving granule. Cells small, 4- 1-4-8

times longer than their diameter; cell wall smooth and yellowish; chloroplasts with

about six ridges and a central series of five pyrenoids. Length 75-5-77 /u.; breadth

16 /la; breadth of apices 5-6 /la. (Tab. 6, fig. 21.) Sample 39.

57. C. gracile Breb. var. brevius var. nova, a typo specie! cellulis multo latioribus

differt. Formae intermediae non visae.

Cells broader than in the typical form (W. & G. S. West, Monogr. Brit. Desm.,

Vol. I, 1904, p. 166, PI. XXI, figs. 8-12). Cells about 11-15-5 times longer than their

diameter; chloroplasts with about six ridges and a single series of 4-6 pyrenoids;

terminal vacuoles with one moving granule. Length 112-136 /a; breadth 8-5-12 p.;

breadth of apices 2-2-3 /a. (Tab. 6, fig. 12.) Samples 51, 52, 54, 64.

58. C. incurvum Breb. var. elaboratum var. nova, a typo specie! cellulis minus arcuatis’

polis minus acutis, vacuolis terminalibus singulo se moventi granulo praeditis

differt.

This differs from the typical form (W. & G. S. West, Monogr. Brit. Desm., Vol. I,

1904, p. 136, PI. XV, figs. 28-30) in the smaller curvature of the cells, in the terminal

vacuoles containing only one moving granule and in the less acute apices. Cells 4-5-6

times longer than their diameter, apices acutely rounded; chloroplasts with six ridges

and a central series of two pyrenoids. Length 51-52 /la; breadth 8-5-11 p; breadth

of apices 1 -5-2-5 p. (Tab. 6, fig. 19.) Samples 58, 59, 63.

59. C. insolitum spec. nova. Nulla affinitate obvia. Cellulae solum plus minusve

2'8-plo longiores quam latiores, valide arcuatae ad apices rotundatae, membrana

achroa glabraque. Chromatophora binis seriebus 2 vel 3 pyrenoidorum ornata.

This differs from other species of Closterium in the cells being rather broad in

proportion to their length. Cells of medium size, about 2-8 times longer than their

diameter, strongly curved, apices rounded; cell wall smooth and colourless; chloro-

plasts with about four ridges and two series of 2-3 pyrenoids each. Length 124 /u,;

breadth 44 p. (Tab. 8, fig. 5.) Sample 39.

573

60. C. intermedium Ralfs var. mesianum var. nova, affinis C. intermedia var. hibernico

West, sed polis non valde arcuatis, membrana cellulae glabra et achroa (quae

colore caret) differt.

This superficially resembles C. intermedium var. hibernicum West (W. & G. S. West,

Monogr. Brit. Desm., Vol. I, 1904, p. 126, PI. XIV, fig. 6); but the poles are less incurved;

cell wall smooth and colourless and the cells are considerably longer than in West’s

variety. Cells of medium size, about 19-22-7 times longer than their diameter,

moderately curved, inner margin straight in the median portion, gradually attenuated

towards the apices, which are truncate and slightly incurved; chloroplasts with about

eight ridges, and 12-13 pyrenoids in one axile series; terminal vacuoles with a number

of moving granules. Length 324-364 breadth 16-18 /x; breadth of apices 4 /x.

(Tab. 4, fig. 3.) Sample 44.

Named after the late Prof. Dr. M. G. Mes, formerly Head of the Department of

Plant Physiology and Biochemistry, University of Pretoria.

61. C. jenneri Ralfs. (W. & G. S. West, Monogr. Brit. Desm., Vol. I, 1904, p. 134,

PI. XV, figs. 23-25.) Sample 27.

Geographical Distribution. — England; Wales; Scotland; Ireland; France;

Germany; Gahcia; Hungary; Italy; Spain; Norway; Sweden; Bornholm; North

and South Russia; Faeroes; Nova Zembla; Greenland; Siam; Java; Central and

East Africa; Azores; United States; Brazil; Switzerland; South Africa.

62. C. jenneri var. dubitabilis var. nova, affinis C. jenneri var. robusto G. S. West, sed

cellulis minus arcuatis differt.

This is near C. jenneri var. robustum G. S. West (W. & G. S. West, Monogr. Brit.

Desm., Vol. I, 1904, p. 136, PI. XV, figs. 26, 27); but the cells are less curved than in

West’s variety. Cells small, about 6 -5-7 -4 times longer than their diameter; cell

wall smooth and yellowish-brown; chloroplasts with about six ridges and five pyrenoids;

terminal vacuoles with 1-2 moving granules. Length 84-86 p\ breadth 11-5-13 /x;

breadth of apices 5-6 /x. (Tab. 6, fig. 7.) Sample 3.

63. C. kranskopense spec. nova. Prima facie forma cellulae plus minusve C. cornu

Ehrenb. congruens, sed cellulis 7-plo maioribus, membrana luteo-brunea apicem

versus incrassata et plus minusve 60 striis disposita differt.

The shape of the cell (low magnification) agrees with that of C. cornu Ehrenb.

(W. & G. S. West, Monogr. Brit. Desm., Vol. I, 1904, p. 157, PI. XX, figs. 1-5); but

the cells are about seven times as large as those in Ehrenberg’s species; the cell wall

is yellowish-brown and striated, with about 60 striae visible across the cell, and the cell

walls are thickened at the apices. Cells large, 15-16 times longer than their diameter,

moderately curved, inner margin concave, gradually attenuated towards the apices,

which are truncately rounded; cell wall with a distinct thickening at each apex; chloro-

plasts with about six ridges, and a central series of 14-23 pyrenoids; terminal vacuoles

with a number of moving granules. Length 765-854 breadth 52-54.5 /x. (Tab. 4,

figs. 4, 5.) Sample 61.

64. C. kiitzingii Breb. var. transvaalense var. nova, a typo specie! differt quod cellulae

media pars multo longius est incrassata.

This differs from the typical form (W. & G. S. West, Monogr. Brit. Desm., Vol. I,

1904, p. 186, PI. XXV, figs. 6-11) in the cells being tumid in the median part for a

greater distance. Cells about 24 times longer than their diameter; about 16 striae

visible across the cell. Length 336 /x; breadth 14 /x; breadth of apices 3 /x. (Tab. 6,

fig. 1.) Sample 3.

65. C. malinvernianum De Not. (W. & G. S. West, Monogr. Brit. Desm., Vol. I,

1904, p. 145, PI. XVII, figs. 5, 6.) Up to 68 striae visible across the cell. Figs. 9-1 1

are cells soon after division. (Tab. 4, figs. 8-11.) Samples 51, 52, 53, 54, 57, 58.

574

Geographical Distribution. — England; Scotland; Ireland; Austria and Galicia;

Italy; Spain; Norway; Sweden; South Africa.

66. C. moniliferum (Bory) Ehrenb. (W. & G. S. West, Monogr. Brit. Desm., Vol. I,

1904, p. 142, PI. XVI, figs. 15, 16.) Samples 4, 6, 14.

Geographical Distribution. — England; Wales; Scotland; Ireland; France;

Belgium; Germany; Switzerland; Galicia in Austria; Hungary and Bosnia; Italy;

Portugal; Norway; Sweden; Denmark; Bornholm; North and South Russia;

Poland; Faeroes; Iceland; China; Japan; Ceylon; New Zealand; Central Africa;

United States; Brazil; Argentina; Uruguay; Patagonia; South Africa.

67. C. moniliferum var. epithemioides var. nova, a typo speciei margine ventrali medio

minus tumido, membrana straminea et cellulis plerumque paulo minoribus differt.

Inner margin slightly but broadly tumid; cell wall straw-coloured; cells usually

somewhat smaller than in the typical form. Cells 5-7 times longer than their diameter;

chloroplasts with about eight ridges and a central series of 5-6 pyrenoids. Length

184-240 p.', breadth 34-38 /x; breadth of apices 6-8 p. (Tab. 5, fig. 14.) Samples

13, 14, 15.

68. C. parvulum Nag. (W. & G. S. West, Monoar. Brit. Desm., Vol. I, 1904, p. 133,

PI. XV, figs. 9-12.) Sample 32.

Geographical Distribution. — England; Wales; Scotland; Ireland; France;

Germany; Switzerland; Austria; Galicia; Hungary; Spain; Norway; Sweden;

Denmark; Finland; North, South and Central Russia; Nova Zembla; Greenland;

Siberia; Japan; Ceylon; Siam; Sumatra; Java; Samoa; Australia; Central and

East Africa; United States; Brazil; Ecuador; Patagonia; South Africa.

69. C. parvulum var. minor var. nova, a typo speciei cellulis multo minoribus et apicibus

aliquanto plus hebetatis differt. Formae intermediae baud visae.

Cells smaller and apices more truncate than in the typical form. Length 76-76 -5 p',

breadth 7-8 p; breadth of apices 3-4 p. (Tab. 6, fig. 8.) Sample 3.

70. C. peracerosum Gay var. elegans G. S. West. (W. & G. S. West, Monogr. Brit.

Desm., Vol. I, 1904, p. 155, PI. XIX, figs. 12, 13.) Ventral margin in the median

portion of the cell slightly more tumid than in the typical form, and the cells are

shorter. Length 134-172 /x; breadth 9-15 /x. (Tab. 6, fig. 1 1.) Samples 13, 14, 15.

Geographical Distribution. — England; First record for South Africa.

71. C. praelongum Breb. forma brevior West. (W. & G. S. West, Monogr. Brit. Desm.,

Vol. I, 1904, p. 165, PI. XXI, figs. 3-5.) Cell wall yellowish and punctate. Sample 6.

Geographical Distribution. — England; Scotland; Ireland; Portugal; Sweden;

New Zealand; South Africa.

72. C. pritchardianum Arch. (W. & G. S. West, Monogr. Brit. Desm., Vol. I, 1904,

p. 172, PI. XXII, figs. 6-14.) These specimens are generally larger than the typical

form. Length 496-825 p\ breadth 48-58 p\ breadth of apices 5-7 p. Samples

6, 42.

Geographical Distribution. — England; Scotland; Ireland; France; Germany;

Austria and Galicia; Italy; Norway; Sweden; North Russia; China; Brazil; Argen-

tina; Switzerland; South Africa.

73. C. pseudoleibleinii spec, nova, affinis C. leibleinii Kiitz., sed differt et cellulis non

tantopere curvatis et polis hebetatis et quod tumor in medio margine ventrali

minor est. Membrana cellulae iuxta ante apices in marginibus ventrali dorsalique

incrassata.

575

This species comes near C. leihlemii Kiitz. (W. & G. S. West, Monogr. Brit. Desm.,

Vol. I, 1904, p. 141, PI. XVI, figs. 9-14); but the cells are less curved, with truncate

apices and the ventral margin less tumid; the cell wall has an annular thickening at

each apex. Cells of medium size, about 10-5 times longer than their diameter; cell

wall smooth and yellowish-brown; the chloroplast has about six ridges and a central

series of 6-8 pyrenoids; terminal vacuoles with one large moving granule. Length

231-5 /x; breadth 21-5 /x; breadth of apices 8-5 /x. (Tab. 5, fig. 8.) Sample 24.

74. C. pseudolunula spec, nova, affinis C. lunulae (Mull.) Nitzsch, sed apicibus cellulae

hebetatis et minus recurvatis, etiam cellulis angustioribus differt.

Near C. lunula (Miill.) Nitzsch (W. & G. S. West, Monogr. Brit. Desm., Vol. 1,

1904, p. 1 50, PI. XVIIl, figs. 8, 9), but the apices of the cells are truncate and less recurved

and the cells are narrower and shorter. Cells of medium size, about 7-8 times longer

than their diameter, straight, inner margin fairly straight or slightly tumid in the median

part of the cell, gradually attenuated towards the apices, which are truncately rounded ;

cell wall smooth and colourless; each chloroplast with about 10 ridges and numerous

scattered pyrenoids; terminal vacuoles large, with a number of moving granules.

Length 372-412 /x; breadth 51-5-58 p.; breadth of apices 14-16 p. (Tab. 4, fig. 1.)

Samples 34, 35.

75. C. ralfsii Breb. var. subralfsii var. nova, affinis C. ralfsii var. hybrido Rabenh.,

sed differt et cellulis minus arcuatis, et quod margo ventralis in media cellula

incrassatus non est; et membrana 25-30 striis, chromatophoraque 11-16 pyre-

noidibus minoribus instructa differt.

Near C. ralfsii var. hybridum Rabenh. (W. & G. S. West, Monogr. Brit. Desm.,

Vol. I, 1904, p. i83, PI. XXIV, figs. 8-13); but the cells are less curved, and are without

a median inflation; 25-30 striae visible across the cell; chloroplasts with a median

series of 11-16 small pyrenoids. Cells of medium size, 9-13 times longer than their

diameter, apices truncately rounded; cell wall yellowish-brown; chloroplasts with

about six ridges; terminal vacuoles with a number of moving granules. Length 314-424

p; breadth 34-38 p; breadth of apices 8-10 p. (Tab. 5, fig. 1.) Samples 34, 35.

76. C. ralfsii var. glentigianum var. nova, affinis var. subralfsii Claassen, sed membrana

cellulae 32-34 striis et chromatophora 5-7 pyrenoidibus magnis instructa differt.

Cell wall with 32-34 striae across the cell; chloroplasts with a series of 5-7 large

pyrenoids. Apices slightly obliquely-rounded. Length 322-344 p; breadth 35-36 p;

breadth of apices 9-10 p. (Tab. 5, fig. 2.) Sample 46.

77. C. ralfsii var. minor var. nova, a typo specie! cellulis multo minoribus differt.

Membrana cellulae 21-23 striis et chromatophora 10-12 pyrenoidibus instructa.

Formae intermediae baud visae.

Cells considerably smaller than in the typical form. Cells of medium size, 7-10

times longer than their diameter, moderately curved, inner margin concave. Cell wall

yellowish-brown, striated, 21-23 striae visible across the cell; chloroplasts with 10-12

pyrenoids. Length 202-264 p\ breadth 26-32 p\ breadth of apices 8-10 p. (Tab. 5,

figs. 4-5.) Sample 3.

78. C. ralfsii var. nodosum var. nova, affinis C. ralfsii var. hybrido Rabenh. sec.

Fukushima et Fujisawa, sed differt et polis leviter capitatis et quod plus minusve

18 striis, quae videri possunt, membrana distincta est.

This is near C. ralfsii var. hybridum Rabenh. sec. Fukushima & Fujisawa (H.

Fukushima and the late K. Fujisawa, Desmids Flora of Oze. Reprinted from Scientific

Researches of the Ozegahara Moor. Tokyo, 1954, p. 596, PI. I, figs. L-M.); but the

apices are slightly clavate and about 18 striae are visible across the cell. Cells of medium

size, 8 -6-8 -8 times longer than their diameter, curved, inner margin concave and

576

straight in the median portion, gradually attenuated towards the apices; cell wall

yellowish-brown; chloroplasts with about 12 ridges and a central series of 12-13

pyrenoids; terminal vacuoles with a number of moving granules. Length 306-310 ft;

breadth 35-35-5 ft; breadth of apices 12-12-5 jj.. (Tab. 5, fig. 3.) Sample 42.

79. C. rostratum Ehrenb. (W. & G. S. West, Monogr. Brit. Desm., Vol. I, 1904, p. 188,

PI. XXVI, figs. 1-5.) Some specimens are slightly narrower (17 /x) than in the

typical form. Sample 21.

Geographical Distribution. — England; Wales; Scotland; Ireland; France;

Germany; Austria and Galicia; Hungary; Italy; Portugal; Norway; Sweden;

Bornholm; Poland; North, Central and South Russia; Finland; Faeroes; Iceland;

Nova Zembla; Greenland; Siberia; Japan; Ceylon (form); East Africa; United

States; Brazil; Ecuador; Paraguay; Switzerland; South Africa.

80. C. schweickerdtii spec, nova, affinis C. stellenboschensi Hodgetts, sed cellulis multo

maioribus et magis arcuatis, polisque acutioribus differt. Apex uterque poro

vacuolaeque terminales paucis granulis se moventibus praeditae.

Near C. stellenboschense Hodgetts (W. J. Hodgetts, Trans. Roy. Soc. of S.A.,

Vol. XIII, 1926, p. 74, fig. 8); but the cells are considerably larger, the apices are more

acute and the cells are strongly curved; each apex furnished with a pore; terminal

vacuoles with a number of moving granules. Cells of medium size, about 6-8-8 5

times longer than their diameter, inner margin tumid in the median portion, gradually

attenuated towards the apices, which are somewhat acute; cell wall smooth and

colourless; chloroplasts with about eight ridges and a series of 5-9 pyrenoids; terminal

vacuoles with 4-5 large moving granules. Length 170-186 breadth 20-26 /x. (Tab. 5,

fig. 18.) Samples 42, 44.

Named after Prof. Dr. H. G. Schweickerdt, Head of the Department of General

Botany, University of Pretoria.

81. C. spetsbergense Borge var. subafricanum var. nova, affinis C. spetsbergensi var.

africano Fritsch et Rich, sed differt et forma polorum et quod membrana cellulae

striata est. Cellulae multo minores sunt; striarum dispositio cum C. pritchardiano

Arch, valde congruit.

This is near C. spetsbergense Borge var. africanwn Fritsch & Rich (F. E. Fritsch &

F. Rich, Trans. Roy. Soc. of S.A., Vol. XVIII, 1930, p. 57, fig. 16C-G); but the shape

of the apices is different, the cells are considerably smaller, and the cell wall is striated;

striae similar to those found in C. pritchardianum Arch. Cells about 7-1 1 times longer

than their diameter; cell wall yellowish or yellowish-brown, 35-40 striae visible across

the cell; striae composed of fine punctae; chloroplasts with 6-10 ridges, and a central

series of 4-9 pyrenoids; terminal vacuoles with a number of moving granules. Length

238-394 /x; breadth 34-52 p.; breadth of apices 5-6 /x. (Tab. 4, fig. 2.) Samples 13,

14, 15.

82. C. striolatum Ehrenb. var. nylstromicum var. nova, a typo cellulis brevioribus et

7-8-plo longioribus quam latioribus, polis plus hebetatis, membrana ad apices

leviter incrassata et 13-18 striis praedita differt.

This differs from the typical form (W. & G. S. West, Monogr. Brit. Desm., Vol. I.

1904, p. 122, PI. XIII, figs. 7-16) in having shorter cells (about 7-8 times longer than

their diameter) with blunter apices; cell wall slightly thickened at the apices, yellowish-

brown, with 13-18 striae visible across the cell. Length 192-220 p; breadth 26-32 p;

breadth of apices 12 p. (Tab. 5, fig. 9.) Samples 34, 35.

83. C. striolatum var. subnylstromicum var. nova, affinis var. nylstromico Claassen,

sed differt striis inter se plus approximatis, plus minusve 19 in membrana dispositis,

quae ad apices est incrassata.

577

Cells of medium size, about 7 • 5 times longer than their diameter, slightly curved,

gradually attenuated towards the apices, which are obtusely-rounded; cell wall yellowish-

brown, striated, with 19 striae visible across the cell; each chloroplast with 6-7 pyre-

noids; terminal vacuoles with one large moving granule. Length 191 /j.; breadth 25 n;

breadth of apices 12 jj,. (Tab. 5, fig. 10.) Sample 54.

84. C. subdecorum spec, nova, affinis C. decoro Breb., sed cellulis multo minoribus,

margine ventrali pro parte maiore tumido, membrana striis carenti differt.

This is near C. decorum Breb. (W. & G. S. West, Monogr. Brit. Desm., Vol. I,

1904, p. 184, PI. XVII, figs. 7, 8; PI. XXVIII, figs. 1-3), but the cells are considerably

smaller, the inner margin of the cell is tumid for a longer distance and the cell wall is

smooth. Cells of medium size, about 10-4-10-7 times longer than their diameter,

curved, gradually attenuated towards the apices, which are obtusely-rounded; cell

wall smooth and colourless; chloroplasts with a central series of three pyrenoids;

terminal vacuoles with a number of moving granules. Length 203-5-208 fj.; breadth

of apices 3-5-4 fi. (Tab. 5, fig. 11.) Sample 2.

85. C. sublagoense spec, nova, aifinis C. lagoensi Nordst., sed cellulis angustioribus,

forma poli et margine ventrali medio incrassato differt.

This species is near C. lagoense Nordst. (W. & G. S. West, Monogr. Brit. Desm.,

Vol. I, 1904, p. 1 14, PI. XI, figs. 5-7); but the cells are narrower, the shape of the apices

differs and the inner margin is tumid in the median portion. Cells small, 7 -5-8 -5

times longer than their diameter, strongly curved, gradually attenuated towards the

apices, which are somewhat acutely rounded; cell wall yellowish, striated, with about

11 striae visible across the cell. Length 160-168 /x; breadth 20-21 /x. (Tab. 5, fig. 17.)

Sample 12.

86. C. subsiliqua spec, nova, affinis C. siliqua W. et G. S. West, sed cellulis haud curvatis,

margine ventrali recto, vel in media cellula leviter incrassata differt.

This somewhat resembles C. siliqua W. & G. S. West (W. & G. S. West, Monogr.

Brit. Desm., Vol. I, 1904, p. 154, PI. XIX, figs. 6-8); but the cells are straight. Cells

of medium size, about 8-9-4 times longer than their diameter, more or less straight,

inner margin straight or slightly tumid in the median portion, gradually attenuated

towards the apices, which are slightly acute, and recurved in some specimens; cell

wall smooth and colourless or straw-coloured; chloroplasts with about 10 ridges and

a central series of 7-11 pyrenoids; terminal vacuoles with one large moving granule.

Length 227-264 /x; breadth 26-32 /la; breadth of apices 3-6 /u,. (Tab. 5, fig. 7.) Samples

13, 14, 15.

87. C. truncatum spec. nova. Nulla affinitate notata. Cellulae plus minusve 6-2-plo

longiores quam latiores. Margo ventralis in media cellula leviter incrassatus;

poli hebetati. Membrana glabra, subflava et apicem versus leviter incrassata.

Cells of medium size, about 6 ■ 2 times longer than their diameter, slightly curved,

inner margin somewhat concave, with the median portion slightly tumid, gradually

attenuated towards the apices, which are truncate; cell wall smooth, yellowish and

with a thickening at the apices; chloroplasts with about 10 ridges and a central series

of four pyrenoids; terminal vacuoles large, with a number of moving granules. Length

224-225 -5 /x; breadth 36 /x; breadth of apices 8-8 -5 /x. (Tab. 5, fig. 6.) Sample 21.

88. C. tumidum Johnson. (W. & G. S. West, Monogr. Brit. Desm., Vol. I, 1904, p. 156,

PI. XIX, figs. 15-18.) Cells up to 9 -9-13 times longer than their diameter. Length

102-114 /x; breadth 9-10-5 /x; breadth of apices 2-5-3 yx. (Tab. 6, fig. 15.)

Sample 9.

Geographical Distribution. — England; Wales; Scotland; Norway; Nova Zembla;

Siam; Celebes; Samoa; United States; Brazil; Paraguay; Patagonia; Japan;

South Africa.

578

89. C. tumidwn var. angustum var. nova, a typo specie! cellulis multo angustioribus

et minus arcuatis differ!. Cellula 15-5-22-plo longior quam latior.

Cells narrower than in the typical form, being about 15-5-22 times longer than

their diameter, and less curved. Chloroplasts with about nine pyrenoids. Length

108-132 /x; breadth 6-8 /x; breadth of apices 1-1-3 /x. (Tab. 6, fig. 13.) Sample 3.

90. C. venus Kiitz. (W. & G. S. West, Monogr. Brit. Desm., Vol. I, 1904, p. 137,

PI. XV, figs. 15-20.) Samples 4, 9, 21.

Geographical Distribution. — England; Wales; Scotland; Ireland; France;

Germany; Austria and Galicia; Hungary; Italy; Norway; Sweden; North Russia;

Faeroes; Nova Zembla; Greenland; Siberia; Central China; Japan; Ceylon;

Burma; Siam; New Zealand; Central Africa; Azores; United States; Brazil;

Patagonia; Switzerland; South Africa.

91. C. venus var. inflatum var. nova, a typo specie! differ! quod margo ventralis cellulae

media in parte leviter inflatus, membranaque achroa est vel subflava.

This differs from the typical form in that the inner margin is slightly tumid. Cells

about 6 -8-7-1 times longer than their diameter; cell wall smooth and colourless or

yellowish; each chloroplast with a central series of two pyrenoids; terminal vacuoles

with 1-2 moving granules. Length 68-70 /x; breadth 9-10 /x; breadth of apices 1 - 5-2- 5/x.

(Tab. 6, figs. 16-18.) Samples 6, 11.

92. C. warmbadianum spec, nova, affinis C. spinosporo Hodgetts, sed cellulis brevioribus

latioribusque, membrana cellulae ad apices non incrassata sed in medio margine

ventral! leviter incrassata differ!. Vacuolae terminates granulis paucis globosis

seque moventibus praeditae.

Near C. spinosporum Hodgetts (W. J. Hodgetts, Trans. Roy. Soc. of S.A., Vol. XIII,

1926, p. 72, fig. 7A-B); but the cells are shorter and broader, the inner margin is slightly

tumid in the median portion. The apices are without a cell wall thickening; and the

terminal vacuoles are large, with a number of spherical moving granules. Cells small,

6-5-8 times longer than their diameter, curved, gradually attenuated towards the

apices, which are slightly acutely rounded; cell wall smooth and colourless; chloroplasts

with about eight ridges and 2-5 pyrenoids. Length 104-118 /x; breadth 14-18 p.

(Tab. 6, fig. 2.) Samples 51, 52, 53, 54.

93. C. warmbadianum var. porulosum var. nova, a typo specie! differ! quod apex uterque

poro est praeditis et vacuolae terminales granula aliquot maiora, elongata, se

moventia habent.

Each apex with a pore in the cell wall; terminal vacuoles with a number of large

oblong moving granules. Cell wall colourless or yellowish-brown; chloroplasts with

about six ridges and 3-4 pyrenoids. Length 104-120 /x; breadth 12-15 /x. (Tab. 6,

fig. 3.) Samples 58, 63, 64.

Pleurotaenium Nag.

94. P. caffrorum spec, nova, affinis P. basiundato W. et G. S. West, sed semicellulis

iuxta tumorem basalem unica undulatione praeditis differ!.

This comes near P. basiundatum W. & G. S. West (W. & G. S. West, Algae of

Madagascar, Trans. Linn. Soc. London, Ser. 2, Bot., Vol. V, 1895-1901, PI. 5, fig. 35);

but the semicells have a single small undulation above the basal inflation. Cells of

medium size, cylindrical, 25 8-30 times longer than their diameter; semicells gradually

attenuated from base to apex, with a distinct basal inflation and a smaller undulation

above it; apices obtusely rounded, bordered by a ring of tubercles, eight in number

(four visible across the apex); cell wall punctate. Length 400-482 breadth at base

of semicells 20 p, at middle of semicells 15-5-16 p, at apices 14-4-15 p. (Tab. 7, fig. 6.)

Sample 20.

579

95. P. ehrenbergii (Breb.) De Bary. (W. & G. S. West, Monogr. Brit. Desm., Vol. I,

1904, p. 205, PI. XXIX, figs. 9-11; PI. XXX, fig. 1.) Sample 20.

Geographical Distribution.— EngXdLwd', Wales; Scotland; Ireland; France;

Germany; Austria and Galicia; Hungary; Italy; Norway; Denmark; Bornholm;

Poland; North Russia; Nova Zembla; Siberia; Central China; India; Ceylon;

Java; Sumatra; Samoa; Australia; New Zealand; Madagascar; East and West

Africa; United States; Brazil; Patagonia; Japan; Switzerland; South Africa.

96. P. ehrenbergii var. waterbergense var. nova, affinis P. ehrenbergii var. undulato

Schaarschm., sed dilfert et membrana luteo-fulva et quod semicellulis iuxta tumorem

basalem undulatio aut deest aut vix perspici potest.

Near P. ehrenbergii var. undulatum Schaarschm. (W. & G. S. West, The Freshwater

Algae of Ceylon, Trans. Linn. Soc. London, Ser. 2, Vol. VI, 1901-1902, PI. 18, fig. 28);

but the semicells are without or with one very slight undulation above the basal inflation,

cell wall yellowish-brown. Cells of medium size, 10- 7-1 1-4 times longer than their

diameter; semicells gradually attenuated from base to apex; apices bordered by a

ring of pear-shaped tubercles, 4-8 in number (three or five visible across the apex);

cell wall punctate. Length 416-428 ,u; breadth at base of semicells 36-4-41 /x, at middle

of semicells 36-36-5 /x, at apices 26-29 /x. (Tab. 7, figs. 7, 8.) Sample 22.

97. P. maximum (Reinsch) Lund. (W. & G. S. West, Monogr. Brit. Desm., Vol. I,

1904, p. 213, PI. XXXI, figs. 1, 2.) Sample 61.

Geographical Distribution. — Wales; Scotland; France; Germany; Italy; Austria

and Galicia; Hungary; Sweden; Japan (var.); Ceylon; Siam; Abyssinia; West

Africa; Brazil; Ecuador (var.); Paraguay; Uruguay; South Africa.

98. Pleurotaenium species ad P. ovatum Nordst. Cells of medium size; 3-8-4 times

longer than their diameter; semicells tumid, a little broader at the middle than at

the base, attenuated from the middle to the apices, without a basal inflation;

apices with one or two rings of tubercles, outer ring with 6-10 tubercles (4-6 visible

across the apex) and the inner ring with 6-8 (3-4 visible across the cell); cell wall

punctate. Length 313-356 /x; breadth 83-90 /x; breadth of isthmus 44-60 /x;

breadth of apices 25-28 /x. (Tab. 8, figs. 1, 2.) Samples 20, 42.

Since only the description of P. ovatum Nordst. was available, it is impossible to

decide with certainty whether these specimens really belong to Nordstedt’s species.

99. P. pseudoehrenbergii spec, nova, affinis P. ehrenbergii (Breb.) De Bary, sed membrana

glabra, apice utrinque tuberculo uno ornato, cellula iuxta tumorem basalem vix

undulata differt.

This species is near P. ehrenbergii (Breb.) De Bary (W. & G. S. West, Monogr.

Brit. Desm., Vol. I, 1904, p. 205, PI. XXIX, figs. 9-11; PI. XXX, fig. 1); but the

undulation above the basal inflation is very slight, there is one tubercle on each side

of the apex, and the cell wall is smooth. Cells of medium size, subcylindrical, 12-4-12-5

times longer than their diameter; semicells with one distinct basal inflation and a small

undulation above it, gradually attenuated towards the apices or slightly tumid and

then gradually attenuated towards the apices; apices truncate, furnished with two

tubercles. Length 408-448 /x; breadth at base of semicells 31-34 /x, at middle of semicells

32-36 /X, at apices 24-26 ’5 /x. (Tab. 7, fig. 2.) Sample 24.

100. P. trabecula (Ehrenb.) Nag. var. angustum var. nova, a typo specie! cellulis multo

angustioribus differt. Cellulae plus minusve 25 - 5-28 - 3-plo longiores quam latiores.

Cells narrower in proportion to their length than in the typical form (W. & G. S.

West, Monogr. Brit. Desm., Vol. I, 1904, p. 209, PI. XXX, figs. 11-13); cells 25-5-28-3

times longer than their diameter. Apices obtusely rounded, without tubercles; cell

wall punctate. Length 514-568 /x; breadth at base of semicell 24-24-5 /x, at middle

of semicell 20-21 /x, at the apices 16-19 ix,. (Tab. 7, figs. 3-5.) Samples, 59, 60.

580

101. P. trabecula var. barbaricum var. nova, affinis P. trabeculae var. recto (Delp.)

W. et G. S. West, sed differt quod apex cellulae circulo 6 tuberculorum instructus

est, cellulae 14-9-16-plo longioribus quam latioribus, at quam variationis rectum

(Delp.) W. et G. S. West paulo latiores sunt.

This closely resembles P. trabecula var. rectum (Delp.) W. & G. S. West (W. &. G. S.

West, Monogr. Brit. Desm., Vol. 1, 1904, p. 212, PI. XXX, figs. 9, 10); but the cells

are a little broader and the apices tuberculate. Cells of medium size, rather narrow

and subcylindrical ; semicells somewhat variable in shape, attenuated from base to

apex or slightly tumid and then gradually attenuated to the apices, with a distinct basal

inflation and one small undulation immediately above it; apices truncate, bordered

by a ring of conical tubercles, six in number (four visible across the apex); cell wall

smooth. Length 402-5-420 /x; breadth at base of semicells 27-28 /x, at middle of

semicells 23-27 /x, at apices 18-20 (Tab. 7, fig. 1.) Sample 61.

102. P. trabecula var. brevis var. nova, a typo specie! cellulis valde brevioribus et apicibus

tuberculis ornatis differt.

Cells shorter than in the typical form, apices with one tubercle. Length 359

breadth at base of semicell 28-28 • 5 /x, at middle of semicells 26-27 ■ 5 /x, and at apices

19- 20 /X. (Tab. 7, fig. 10.) Sample 35.

103. P. trochiscum W. & G. S. West var. galpimi var. nova, a typo cellulis multo

brevioribus et 7-2-8-plo longioribus quam latioribus, margine apical! circulo

tuberculorum fere 8 ornato differt.

The cells are considerably shorter than in the typical form (W. & G. S. West,

North American Desmidieae, Trans. Linn. Soc. London, Ser. 2, Bot., Vol. V, 1895-1901,

PI. 13, figs. 4, 5), and are about 7-2-8 times longer than their diameter; apices bordered

by a ring of tubercles, about eight in number (five visible across the apex). Length

252-264 /u.; breadth at base of semicell 33-35 /x, at middle of semicell 30-32 /x, at apices

20- 23 fi. (Tab. 8, fig. 7.) Sample 71.

Named after Mr. E. A. Galpin of Mosdene, Naboomspruit.

104. P. truncatum (Breb.) Nag. var. mattiei var. nova, affinis P. truncato var. granulate

West, sed cellulis multo brevioribus, semicellulis media in parte minus tumidis,

margine apical! tuberculis paucioribus ornato differt.

Near P. truncatum var. granulatum West (W. & G. S. West, Monogr. Brit. Desm.,

Vol. I, 1904, p. 205, PI. XXIX, figs. 7, 8); but the cells are much shorter, they are less

tumid in the median part of the semicells and the apical margin has fewer tubercles.

Cells small, about 7-7-2 times longer than their diameter; semicells with one basal

inflation, slightly tumid in the middle of the semicells; apices bordered by a ring of

tubercles, 2-4 in number (2-3 visible across the apex); cell wall punctate. Length

272 /x; breadth at base of semicell 36 ii, at middle of semicell 38-39 im, at apices 26-28 /x.

(Tab. 8, fig. 6.) Sample 21.

105. P. westiorum spec, nova, affinis P. monilifero W. et G. S. West, sed apicibus cellulae

multo latioribus (modo non tarn latis quam ad semicellulae basim) differt.

This is near P. moniliferum W. & G. S. West (W. & G. S. West, The Freshwater

Algae of Madagascar, Trans. Linn. Soc. London, Ser. 2, Bot., Vol. V, 1895-1901,

PI. 5, fig. 32); but with broader apices (about as broad as at the base of the semicells).

Cells large, 16-16-5 times longer than their diameter; semicells slightly tumid in the

middle, with a slight basal inflation and two smaller undulations above it; apices

truncate, and bordered by a ring of pear-shaped tubercles, 24 in number (13 visible

across the apex); cell wall smooth. Length 796-800 jtx; breadth at base of semicells

48-50 /X, near base of semicells 44-46 /x, at middle of semicells 48-50 /x, near apices

40-44-5 /X, at apices 44-47 /x. (Tab. 7, fig. 9.) Sample 1.

Named after Messrs. W. & G. S. West.

581

Tetmemorus Ralfs

106. T. laevis (Kiitz.) Ralfs. (W. & G. S. West, Monogr. Brit. Desm., Vol. 1, 1904,

p. 222, PI. XXXII, figs. 11-16.) Sample 2.

Geographical Distribution. — England; Wales; Scotland; Ireland; France;

Belgium; Germany; Austria and Galicia; Italy; Norway; Sweden; Denmark;

Bornholm; North and South Russia; Nova Zembla; Greenland; Singapore;

Australia; New Zealand; Sandwich Islands; Azores; Dominica and Trinidad;

West India; Brazil; Japan; Central Africa; Switzerland; United States; South

Africa.

Euastrum Ehrenberg

107. E. ansatum Ralfs. (W. & G. S. West, Monogr. Brit. Desm., Vol. II, 1905, p. 27,

PI. XXXVI, figs. 10-13.) Fig. 8 shows a cell with dividing pyrenoids. (Tab. 9,

fig. 8.) Samples, 3, 25, 34, 35, 36, 54, 56.

Geographical Distribution. — England; Wales; Scotland; Ireland; France;

Belgium; Germany; Austria and Galicia; Hungary; Italy; Norway; Sweden;

Denmark; Bornholm; Finland; North and South Russia; Faeroes; Iceland; Green-

land; Central China; Japan; India; Ceylon; Burma (var.); Siam; Singapore;

Sumatra; Java; Australia (form); New Zealand; Madagascar; East Africa; Azores;

Sandwich Islands; United States; West Indies (var.); Brazil; Paraguay; Switzerland;

South Africa.

108. E. brasiliense Borge var. theronii var. nova, affinis E. brasiliensi var. africano

Fritsch et Rich, sed cellularum a latere visarum forma diflfert quoniam semicellulae

gradatim a basi apicem versus in cuneum deplanantur, neque tuberculi apparent.

This variety closely resembles E. brasiliense var. africanum Fritsch & Rich (F. E.

Fritsch & F. Rich, Trans. Roy. Soc. of S.A., Vol. XI — Part 4, 1924, p. 330, fig. 9);

but the shape of its semicells differs in being gradually attenuated from base to apex

when seen in lateral view and the cells are slightly smaller. Cells of medium size, about

3-3-4 times longer than broad, deeply constricted; sinus narrowly linear, open towards

the extremity; semicells pyramidate, basal angles rounded, lower part of lateral margins

convex, upper part slightly concave; apex truncate with rounded angles, median incision

not very deep; semicells tumid above the isthmus. Side view of semicell elongate-

pyramidate; cell wall punctate. Length 107-108 breadth 32-35 /x; breadth of

isthmus 16-18 yu.; breadth of apex 22-23 p. (Tab. 9, figs. 6, 7.) Sample 42.

Named after Mr. J. E. Theron of Rietfontein 288, Nylstroom.

109. E. cuneatum Jenner var. minor var. nova, cum typo specie! valde congruens, sed

cellulis multo minoribus et incisura apicali multo angustiore bene distincta. Formae

intermediae baud visae.

Cells much smaller than in the typical form (W. & G. S. West, Monogr. Brit.

Desm., Vol. II, 1905, p. 25, PI. XXXVI, fig. 9) and the apical incision is narrower.

Cell wall finely punctate. Length 64-68 p.; breadth 25-27 p; breadth of isthmus 9-12 p;

thickness 20 p. (Tab. 9, figs. 13, 14.) Sample 3.

110. E. denticulatum (Kirchn.) Gay. (W. & G. S. West, Monogr. Brit. Desm., Vol. II,

1905, p. 56, PI. XXXIX, figs. 1-4.) Sample 24.

Geographical Distribution. — England; Wales; Scotland; Ireland; France;

Germany; Galicia and Austria; Norway; Sweden; Bornholm; Finland; North

Russia (var.); Iceland; Greenland; Central China; Ceylon; Siam; Singapore;

Java; Australia (var.); New Zealand; Madagascar (var.); East and Central Africa;

Azores; United States; West Indies; Brazil (van); South Africa.

582

111. E. divaricatum Lund. var. transvaalense var. nova, cum typo specie! plus minusve

congruens, sed paulo major; lobus apicalis dentibus duobus parvis, lobis lateralis

infimus dentibus 3 parvis munitus. A latere visus processus supra isthmum glaber

singularisque, sed a vertice bipartitus videtur.

The cells are slightly larger than in the typical form (W. & G. S. West, Monogr.

Brit. Desm.,Vol. II, 1905, p. 42, PI. XXXVIII, figs. 3, 4). Lower lateral lobules with

three teeth and upper lobules with two teeth each; protuberance in each semicell

above the isthmus single in lateral view and bilobulate in vertical view. Length 52 /a;

breadth 36 ^ (with teeth 41 /a); breadth of isthmus 8-9 /x; breadth of apex 20 fx. (Tab.

24, figs. 1-3.) Sample 39.

112. E. divergens Joshua var. subbifidum var. nova, affinis E. divergenti var. bifido

Schmidle, sed praecipue forma lobi polaris et cellulis multo minoribus differt.

This variety has a greater resemblance to var. bifidum Schmidle (W. Schmidle,

Fngler Botan. Jahrb., Vol. 26, 1899, p. 44, Taf. 11, fig. 34) than to Joshua’s species

(W. Joshua, Burmese Desmidieae, Journ. Linn. Soc. Bot., Vol. XXI, p. 640, PI. 23,

figs. 8, 9). However, it differs from Schmidle’s variety in having relatively smaller

cells and in the shape of the polar lobes. Cells small, about 1-2 times longer than

broad, deeply constricted; sinus narrowly linear; semicells deeply 3-lobed; granules

above central inflation in concentric arrangement; lateral lobes semifusiform, ends

truncate, upper part of lobes elongated, lobes ornate with granules; polar lobe quadrate

in lower part and widening at the apex. Length 44-50 breadth 36-42 breadth

of isthmus 10-12 /x; maximum breadth of polar lobe 20 /tx. (Tab. 9, fig. 3.) Sample 20.

113. E. divergens var. galpinii var. nova, affinis E. divergenti var. subbifido Claassen,

sed lobis lateralibus multo brevioribus latioribusque (quam prioris varietatis)

et forma lobi polaris differt.

Upper part of lateral lobes shorter and broader than in var. subbifidum Claassen.

and the shape of the polar lobe is different. Length 55 /x; breadth 48 /x (with spines

53 /x); breadth of isthmus 12 /ix; maximum breadth of polar lobe 24 /x. (Tab. 29,

fig. 3.) Sample 68.

Named after Mr. E. A. Galpin of Mosdene, Naboomspruit.

114. E. dubitabilis spec, nova, affinis E. inermi (Ralfs) Lund., sed discrepantia latitudinis

ad basin et ad apicem multo minore bene distincta.

This species is near E. inerme (Ralfs) Lund. (W. & G. S. West, Monogr. Brit.

Desm., Vol. II, 1905, p. 24, PI. XXXVI, figs. 7, 8); but the apices are not very much

narrower than the base of the semicells and the cells are smaller. Cells small, about

2- 1-2-2 times longer than broad, deeply constricted, sinus narrowly linear; semicells

truncate-pyramidate, basal angles rounded; upper part of lateral margins somewhat

concave; apex convex with a narrow median incision. Cell wall smooth. Length

36-5-38 /x; breadth 16-5-18 /x; breadth of isthmus 7-5-8 /x; breadth of apex 12 ^x.

(Tab. 9, fig. 9.) Sample 24.

115. E. elegans (Breb.) Kiitz. var. transvaalense var. nova, affinis E. elegant i var.

rnadagascariensi W. et G. S. West, sed forma apicis, incisura apicali multo angus-

tiore, et tumore supra isthmum granulis 4 ornato bene distincta.

This is near E. elegans var. madagascariense W. & G. S. West (W. & G. S. West.

The Freshwater Algae of Madagascar, Trans. Linn. Soc. London, Ser. 2, Vol. V,

1895-1901, PI. 6, fig. 18); but differs in the shape of the apex, the apical incision is narro-

wer and the protuberance above the isthmus contains four granules; the cells are slightly

smaller than in Messis. West’s variety. Length 23-24 /x; breadth 16 /«; breadth of

isthmus 3-4 /x; breadth of apex 6-5-8 p. (Tab. 9, fig. 18.) Sample 20.

583

116. E. galpinii spec, nova, affinis E. bidentato Nag., sed lobo infimo laterali latiore

quam lobo laterali supremo et margine apicali recto bene distincta.

Near E. bidentatum Nag. (W. & G. S. West, Monogr. Brit. Desm., Vol. II, 1905,

p. 39, PI. XXXVII, figs 16-19); but the semicells are broader at the lower lateral

lobule than at the upper lateral lobule; apical margin straight. The protuberance in

the centre above the isthmus has four granules, and there are a few granules within the

lateral lobules and the apex. Length 38 breadth 27 /x; breadth of isthmus 8 /x;

breadth of apex 16 /x. (Tab. 29, fig. 1.) Sample 72.

117. E. insulare (Wittr.) Roy. (W. & G. S. West, Monogr. Brit. Desm., Vol. II, 1905,

p. 68, PI. XL, figs. 11-13.) Fig. 11 shows a cell during normal cell division and

fig. 12 shows an abnormal one. (Tab. 9, figs. 10-12.) Samples 3, 24, 41.

Geographical Distribution. — England; Wales; Scotland; Ireland; Galicia in

Austria; Norway; Sweden; Siam; United States; Japan; Switzerland; South

Africa.

118. E. obesuni Josh. (W. & G. S. West, Monogr. Brit. Desm., Vol. II, 1905, p. 29,

PI. XXXVI, figs. 16, 17.) Sample 52.

Geographical Distribution. — England; India; Burma; Singapore; Sumatra;

Madagascar; South Africa.

119. E. pseudovalidum spec, nova, affinis E. valido W. et G. S. West, sed forma cellulae

(praecipue ad basin semicellulae) haud multum sed plane differt. Dentes desunt.

Near E. validum W. & G. S. West (W. & G. S. West, Monogr. Brit Desm., Vol. II,

1905, p. 75, PL XL, figs. 21, 22); but differs in the shape of the basal part of the semicells,

and in the absence of teeth. Cells small, about 1 -5-1 -6 times longer than broad, very

deeply constricted, sinus narrowly linear; semicells iruncate-pyramidate, lateral margins

retuse, inferior angles inflated, superior angles somewhat acutely-rounded; apex

broad, truncate-convex, emarginate. Semicells ovate in lateral view, and elliptic in

vertical view. Cell wall smooth. Length 36-36-5 /x; breadth 22-23 /x; breadth of

isthmus 8-9-5 /x; breadth of apex 14-16 p.; thickness 14 p. (Tab. 9, figs. 15-17.)

Sample 24.

120. E. rostratum Ralfs subsp. umbonatum W. & G. S. West var. kranskopense var.

nova, affinis E. rostrato subsp. umbonato W. et G. S. West, sed sculptura membranae

omnino diversa facile distinguenda.

The shape of the cells agrees with that of E. rostratum subsp. umbonatum W. & G. S.

West (W. & G. S. West, The Freshwater Algae of Madagascar, Trans. Linn. Soc.

London, Ser. 2, Vol. V, 1895-1901, PI. 6, fig. 16); but the cell wall sculpture differs.

Apex of polar lobe convex, slightly undulate, with a deep and slightly open median

notch, a short spine at each outer angle, and a thickening of the cell wall at each angle

and below the apical notch. Cell wall punctate, punctae in rows. Length 50-54 p\

breadth 30-5-32 /x; breadth of isthmus 9-10 p\ breadth of apex (without spines)

19-20 p. (Tab. 9, figs. 4, 5.) Sample 61.

121. E. spinulosum Delponte subsp. africanum Nordst. (W. & G. S. West, The Fresh-

water Algae of Ceylon, Trans. Linn. Soc. London, Ser. 2, Vol. VI, 1901-1902,

PI. 19, fig. 21.) Interlobular incisions slightly deeper than in the typical form.

The Doornfontein specimens are considerably larger than the Mosdene specimens.

Doornfontein Mosdene

Length 78-84 p 60 p

breadth 68-72 p 56-60 p

breadth of isthmus 16-18 /ix 16 p

(Tab. 9, fig. 1 ; Tab. 29, fig. 2.) Samples 20, 68, 72.

Geographical Distribution. — Ceylon; South Africa.

584

122. E. spinulosum subsp. africanum var. transvaalense var. nova, affinis var. duplo-

minori W. et G. S. West, sed forma lobi polaris et incisuris interlobularibus minus

profundis distinguenda.

This differs from var. duplo-minor W. & G. S. West (W. & G. S. West, The Fresh-

water Algae of Madagascar, Trans. Linn. Soc. London, Ser. 2, Bot., Vol. V, 1895-1901,

PI. 6, fig. 13) in the shape of the polar lobe, and in having shallower interlobular incisions

and larger cells. Lobes more or less semicircular. Length 81-82 /x; breadth 62 -5-64 /x;

breadth of isthmus 15-15-5 p.. (Tab. 9, fig. 2.) Sample 6.

123. E. schweickerdtii spec, nova, ab omnibus generis speciebus distincta. Cellulae

plus minusve tarn longae quam latae. Semicellulae latae et obtuse pyramidales,

margine apicali baud penitus inciso, supra isthmum dentes magno munitae; a

latere visae planae et ovatae, a vertice ellipticae. Incisura apicali non profunda.

Cells small, about as long as broad, deeply constricted, sinus narrowly linear;

semicells broad and flat, apical margin with a shallow median incision; with a tooth

in the centre of each semicell above the isthmus. Side view of semicell flattened-ovate

and oval in vertical view. Length 27 /x; breadth 27-28 p; breadth of isthmus 7 p;

thickness 12-5-13 p (with teeth 18 p). (Tab. 9, figs. 19-21.) Sample 38.

Named after Prof. Dr. H. G. Schweickerdt. Head of the Department of General

Botany, University of Pretoria.

Micrasterias Ag.

124. M. americana (Ehrenb.) Ralfs var. transvaalensis var. nova, affinis typo speciei,

sed lobo polari bifido et minore differt.

This differs from the typical form (W. & G. S. West, Monogr. Brit. Desm., Vol. II,

1905, p. 117, PI. LlII, figs. 4, 5; PI. LIV, figs. 1-3) in the shape of the polar lobe. Polar

lobe smaller than in the typical form, divided into two equal parts near the base (lateral

view of semicell); apical margin with a slight median depression, angles produced

into small divergent processes, each with a truncate-denticulate apex. Length 104-108 p:

breadth 80-82-5 p\ breadth of isthmus 17-5-18 p\ maximum breadth of polar lobe

34-35 p. (Tab. 16, fig. 4.) Sample 19.

125. M. apiculata (Ehrenb.) Menegh. (W. & G. S. West, Monogr. Brit. Desm., Vol. II.

1905, p. 97, PI. XLVII, figs. 1, 2.) Polar lobe less exserted than in the typical

form, apical angles with one spine and in some specimens without spines (the

typical form has a pair of diverging spines). Some specimens slightly smaller

than in the type. Length 210-236 p (with spines 238-252 p); breadth 173-196 p

(with spines 186-217 p); breadth of isthmus 30-35 /x; maximum breadth of polar

lobe 64-80 p. (Tab. 10, figs. 1-4.) Samples 25, 26, 38, 39.

Geographical Distribution. — England; France; Germany; Galicia in Austria;

Italy; Sweden; Denmark; Bornholm; Finland; Poland; Russia; Japan; India;

Burma; United States; First record for South Africa.

126. M. crux-melitensis (Ehrenb.) Hass. (W. & G. S. West, Monogr. Brit. Desm.,

Vol. II, 1905, p. 116, PI. LlII, figs. 1-3.) The shape of the polar and lateral lobes

differs slightly from those in the type. Cells slightly larger than in the typical

form. Length 116-148 p (with processes 128-168 p); breadth 118-126 p (with

spines 128-140 p); breadth of isthmus 20-23 p; maximum breadth of polar lobe

60-80 p. (Tab. 18, figs. 1-14; Tab. 19, figs. 1-11; Tab. 20, figs. 1-6; Tab. 21,

figs. 1-4; Tab. 22, figs. 1-6; Tab. 23, fig. 1.) Samples 66, 73, 74.

Geographical Distribution. — England; Wales; Scotland; Ireland; France;

Germany; Austria; Galicia; Hungary; Italy; Sweden; Denmark; Poland; North

and South Russia; Japan; North India; Ceylon; Burma; Celebes; Madagascar

(var.); Central Africa; United States; Brazil; Switzerland; South Africa.

585

127. M. crux-melitensis var. evoluta W. B. Turner. (W. & G. S. West, The Freshwater

Algae of Madagascar, Trans. Linn. Soc. London, Ser. 2, Vol. V, 1895-1901, PI. 6,

fig. 1.) Apical margin of the polar lobe with four small teeth; the lateral lobes

with more subdivisions than in the typical form. Length 98-124 /x (with processes

113-144 /x); breadth 101-116 /x (with teeth 1 1 1-128 /x); breadth of isthmus 17-22 /x;

maximum breadth of polar lobe 36-5-45 tx. (Tab. 16, fig. 3; Tab. 17, fig. 1.)

Samples 59, 74.

Geographical Distribution. — Madagascar; East India; South Africa.

128. M. crux-melitensis var. transvaalensis var. nova, typo specie! affinis, sed forma

lobi polaris distincta et margine apical! paene recto quippe cui pars media leviter

sit depressa.

This differs from the typical form in the shape of the polar lobe and in having an

almost straight apical margin. Fig. 2 shows a cell shortly after division. Length

98-104 |tx (with processes 1 10-1 13 /x); breadth 86-98 p (with spines 96-105 /x); breadth

of isthmus 16-19 /x; maximum breadth of polar lobe 40-44 p. (Tab. 16, figs. 1, 2.)

Samples 24, 39.

129. M. denticulata Breb. var. africana var. nova, affinis M. denticulatae Breb. var.

angustosinuatae Gay, sed lobulo polar! latiore et breviore, incisuris minus altis

bene distinguenda.

This variety is near M. denticulata var. angustosinuata Gay (W. & G. S. West,

Monogr. Brit. Desm., Vol. II, 1905, p. 108, PI. L, fig. 5); but differs in having a shorter

and broader polar lobe and shallower interlobular incisions. Some specimens with

slightly exserted polar lobes (fig. 3). Length 234-248 p\ breadth 196-220 p; breadth

of isthmus 32-40 p; maximum breadth of polar lobe 74-95 p. (Tab. 11, figs. 1-4;

Tab. 12, fig. 1.) Samples 24, 25, 36, 39.

130. M. denticulata var. subnotata West forma cornuta forma nova, affinis var. subnotatae

West, sed differt quod in lobis polaribus ad incisuram media utrimque bini sunt

processus, quibus singuli denticuli insunt.

This differs from West’s variety (W. & G. S. West, Monogr. Brit. Desm., Vol. II,

1905, p. 108, PI. L, fig. 7) in the polar lobe possessing a pair of processes on each side

of the median notch; each process furnished with a small tooth. Cells slightly larger

than var. subnotata West, and some lobules with more subdivisions. Length 208-256 /tx;

breadth 180-216 p (fig. 3, 156 p); breadth of isthmus 26-34 p; maximum breadth of

polar lobe 44-52 p. (Tab. 12, fig. 2; Tab. 13, figs. 1-3.) Samples 35, 38.

131. M. decemdentata (Nag.) Arch. (Nageli, Gatt. einzell. Algen, 1849, p. 123, Tab. VI,

H, fig. 2; F. E. Fritsch & F. Rich, Trans. Roy. Soc. S. Afr., Vol. XI — Part 4, 1924,

p. 337, fig. 13.)

Cell wall smooth or punctate. Length 76- 5-80 /tx; breadth 71 • 5-75- 5 /x (with teeth

79-87 /x); breadth of isthmus 15-17 /x; maximum breadth of polar lobe 54-60-5 p.

(Tab. 16, fig. 5-7.) Samples 24, 39.

132. M. decemdentata var. galpinii var. nova. A typo specie! cellulis multo moniiibus

distincta. Formae intermediae baud visae.

Cells smaller than in the typical form; lateral angles of polar lobe acuminate.

Length 48-5-52 p\ breadth 47-51 p (with teeth 56-63 /tx); breadth of isthmus 14-17 p;

maximum breadth of polar lobe 40^4 p. (Tab. 17, figs. 4-7.) Samples 65, 66, 68,

70, 71, 74.

Named after Mr. E. A. Galpin of Mosdene, Naboomspruit.

586

133. M. gwenewaldii spec, nova, affinis M. confertae Lund., sed lobi polaris forma

differt et membranae tres processus supra isthmum sunt conspicui.

This species is near M. confevta Lund. (W. & G. S. West, Monogr. Brit. Desm.,

Vol. II, 1905, p. 88-89, PI. XLIII, figs. 4-9); but differs in the shape of the polar lobe

and each semicell has three processes above the isthmus. Cells fairly large, a little

longer than broad, sub-elliptic, deeply constricted, sinus and interlobular incisions

open outwards; semicells 5-lobed; polar lobe with convex sides in its lower half, upper

half cuneate, apex with a median notch, apical margin with 4-5 spines on each side

of the median hollow; superior lateral lobes slightly larger than inferior lateral lobes,

each divided into two lobules, lobules again divided, the four ultimate divisions of

each lobe being emarginate (or sometimes tridenticulate); the small lobule adjoining

the polar lobe usually tridenticulate. Semicells with three projections across the base,

the middle one immediately above the isthmus larger than the lateral ones. Cell wall

furnished with numerous minute spines, arranged in subradiate rows. Cell in vertical

view fusiform, poles acute, with three projections on each side towards the middle.

Length 184-190 ^ (with spines 197-203 /x); breadth 162-166 fi (with spines 170-174 /x);

breadth of isthmus 29-33 ju; maximum breadth of polar lobe 61-68 /x. (Tab. 14,

figs. 1-4.) Sample 46.

Named after Mr. L. J. Groenewald of Glentig, near Nylstroom.

134. M. mahabuleshwarensis Hobson var. transvaalensis var. nova, affinis M. maha-

buleshwarensi var. tetracero W. et G. S. West, sed et forma lobi lateralis summi

et dispositione dentium distincta. Margo lobi polaris duobus tantum processibus

accessoriis est munitus, qui processibus M. mahabuleshwarensis simules sed multo

breviores sunt.

Near M. mahabuleshwarensis var. tetracerum W. & G. S. West (W. & G. S. West,

The Freshwater Algae of Madagascar, Trans. Linn. Soc. London, Ser. 2, Vol. V,

1895-1901, PI. 6, figs. 2-4); but differs in the shape of the upper lateral lobule, and in

the arrangement of the small spines; polar lobe with only two accessory denticulate

processes like those in var. wallichii (Grun.) W. & G. S. West (W. & G. S. West, Monogr.

Brit. Desm., Vol. II, 1905, p. 122, PI. LIV, figs. 7, 8; PI. LV, figs. 1-3); but the processes

are considerably shorter. Apical margin of polar lobe furnished with four 3-pointed

spines; each lobule of the lateral lobes furnished with singular or branched teeth;

cell wall with a series of small teeth within the lateral lobes, across the lateral sides of

the polar lobe and a few denticulations above the isthmus. Length 128 /x (with apical

processes 148 /x); breadth 1 12 /x; breadth of isthmus 28 /x; maximum breadth of polar

lobe 84 /X. (Tab. 17, fig. 2.) Samples 65, 71.

135. M. nylstromica spec, nova, ex affinitate M. confertae Lund., sed et forma lobi

polaris et loborum superiorum-lateralium differt et quod dentibus est ornata.

This species somewhat resembles M. conferta Lund. (W. & G. S. West, Monogr.

Brit. Desm., Vol. 11, 1905, p. 88-89, PI. XLlil, figs. 4-9); but differs in the shape of

the polar and superior lateral lobes; cell wall furnished with small spines. Cells large,

broadly elliptic, deeply constricted, sinus narrowly linear; semicells 5-lobed, the inter-

lobular incisions being narrowly linear and not very deep; polar lobe sub-cuneate,

lateral angles downwardly curved, apex concave with a median notch, with three spines

on each side of the median notch, and a large curved spine on the apical margin close

to each angle, and a smaller spine near the extremity; lateral lobes unequal, superior

lateral lobes considerably larger than the inferior lateral lobes, each lateral lobe divided

into two more or less similar lobules and each lobule furnished with four marginal

spines except the lobule adjoining the polar lobe; this lobule consists of a lower part

with four marginal spines, and an upper part which consists of a two-spined lower

half and a three-spined upper half, the latter with the top spine upwardly curved to

overlap the apical angle of the polar lobe. Cell wall strongly punctate and furnished

587

with numerous minute spines, arranged in subradiate rows within the polar and lateral

lobes. Length 264-276 /i. (with spines 278-292 /x); breadth 221-226 n (with spines

233-241 fj.); breadth of isthmus 42-43 /x; maximum breadth of polar lobe 102-5-108 /ix.

(Tab. 12, fig. 3; Tab. 13, fig. 4.) Samples 26, 39.

136. M. pimiatifida (KUtz.) Ralfs var. transvaalensis var. nova, affinis M. pinnatifidae

var. divisae W. West formae majori Schmidle, sed diflfert et forma lobi polaris et

quod tuberculi iuxta supra isthmum semicellulae desunt.

This differs from var. divisa W. West forma major Schmidle (W. Schmidle, Engler

Botan. Jahrb., XXXII Bd., 1903, p. 73, Taf. II, fig. 7) in the shape of the polar lobes

and in the absence of tubercles above the isthmus in each semicell. Length 108-1 16 /x;

breadth 104-116 p. (with teeth 114-124 /x); breadth of isthmus 16-18 p; maximum

breadth of polar lobe 74-92 p. (Tab. 23, figs. 2, 3.) Samples 72, 74.

137. M. sol (Ehrenb.) Kutz. (W. & G. S. West, Monogr. Brit. Desm., Vol. II, 1905.

p. 95, PI. XLVI, figs. 1, 2.) These specimens are slightly smaller than in the typical

form. Length 146-152 p; breadth 146-152 p; breadth of isthmus 18-20 /x;

maximum breadth of polar lobe 32-36 p. (Tab. 15, fig. 6.) Samples 24, 25, 39.

Geographical Distribution. — England; Wales; Scotland; Ireland; France;

Italy; Sweden; Denmark; United States; Jamaica; Brazil; First record for South

Africa.

138. M. tropica Nordstedt var. elongatissima var. nova, affinis M. tropicae var. elongatae

Schmidle, sed forma lobi polaris et dispositione dentium valde distincta. Cellula

non duplo longior quam latior.

Near var. elongata Schmidle (W. Schmidle, Engler Botan Jahrb., Vol. 26, 1899.

p. 48, Taf. Ill, fig. 13); but the proportion of length to breadth is not 2:1, and the shape

of the polar lobe and the arrangement of the spines differ. Cells longer than in var.

elongata Schmidle, polar lobes shorter and with a distinct constriction below the apex,

apical margin concave, with the angles produced into short processes. Cell wall furnished

with seven series of small spines within the lateral lobes, and with a number of spines

within the lateral margins of the polar lobe. Length 112-128 p; breadth 76-80 p:

breadth of isthmus \6 p; maximum breadth of polar lobe 22-24 p. (Tab. 17, fig. 3.)

Sample 73.

139. M. truncata (Corda) Breb. (W. & G. S. West, Monogr. Brit. Desm., Vol. II,

1905, p. 82, PI. XLII, figs. 1-8; PI. XLV, figs. 5, 6.) These specimens are usually

slightly shorter than broad. Length 124-132 /x; breadth 124-138 /x; breadth of

isthmus 22-5-27 /x; maximum breadth of polar lobe 82-100 p (without spines

76-94 p). (Tab. 15, figs. 1-3.) Samples 24, 25, 39.

Geographical Distribution. — England; Wales; Scotland; Ireland; France;

Germany; Austria; Galicia; Hungary; Italy; Norway; Sweden; Denmark; Born-

holm; Finland; Poland (form); North and South Russia; Faeroes; North India;

Sandwich Islands; United States; Brazil; Switzerland; Japan; South Africa.

140. M. truncata var. africana Fritsch & Rich. (F. E. Fritsch & F. Rich, Trans. Roy.

Soc. S. Afr., Vol. XI — Part 4, 1924, p. 338, fig. 14B and C.) Polar lobe somewhat

variable in shape, cuneate or with subparallel sides in the lower portion and dilated

in the upper portion. Length 134-152 p; breadth 124-136 p (without spines

120-131 p); breadth of isthmus 25-5-26 p; maximum breadth of polar lobe

62-67 p. (Tab. 15, figs. 4, 5.) Samples 24, 39.

Geographical Distribution. — South Africa.

588

141. M. truncata var. minor var. nova. Cellulae varietatis quam typi specie! aliquanto

minores sunt. Formae intermediae baud visae.

These specimens are considerably smaller than in the type. Lateral angles of polar

lobe acuminate; cell wall punctate. Length 57 • 5-60 ; breadth 56-60 (with spines

62-66 ^l)\ breadth of isthmus 12 -5-13 ju; maximum breadth of polar lobe 44-50

(Tab. 17, fig. 8.) Sample 74.

Cosmarium Corcia

142. C. barbaricum spec, nova, affinis C. entochondro W. et G. S. West, sed membrana

cellulae glabra, sinu undulato et marginibus lateralibus magis undulatis facile

distinguenda.

Near C. entochondrum W. & G. S. West (W. & G. S. West, Monogr. Brit. Desm.,

Vol. Ill, 1908, p. 193, PI. LXXXVll, fig. 17); but the sides have more undulations,

the sinus is undulate and the cell wall is smooth. Cells small, about as long as broad,

deeply constricted, sinus narrow and slightly crenate; semicells semicircular, apex

truncate, basal angles furnished with a spine, sides convex and with 3-5 undulations,

apical margin smooth. Chloroplasts axile, each with two pyrenoids. Length 20-28 /x;

breadth 19-28 breadth of isthmus 9-9-5 /x; breadth of apex 10-14 /x. (Tab. 26,

fig. 16.) Sample 24.

143. C. binum Nordst. (W. & G. S. West, Monogr. Brit. Desm., Vol. Ill, 1908, p. 246,

PI. LXXXVIII, figs. 10-14.) Some specimens with two horizontal series of granules

below the conspicuous tumour immediately adjacent to the isthmus. (Tab. 27,

figs. 1, 2.) Samples 7, 8, 12, 19, 20, 21, 23, 30, 35, 42, 43, 46, 47, 48, 51, 52, 53,

54, 57, 58, 65, 66, 72.

Geographical Distribution. — Scotland; Austria; Poland; Ceylon; Sumatra;

Central Africa; Brazil; Australia; United States; South Africa.

144. C. botesii spec, nova, affinis C. praemorso Breb., sed et dispositione granulorum

omnino dissimili et membrana cellulae leviter undulata differt.

This species differs from C. praemorsum Breb. (W. & G. S. West, Monogr. Brit.

Desm., Vol. Ill, 1908, p. 196, PI. LXXXIV, figs. 1-5) in the cell wall sculpture and in

having slightly undulate margins. Cells of medium size, about 1-1-1 -3 times as long

as broad, deeply constricted, sinus narrowly linear; semicells semicircular, apex truncate,

sides with 3-6 distinct and a number of obscure undulations, apical margin smooth or

with about 3-4 undulations. Cell wall minutely punctate and with a number of large

scattered granules in the upper part of the semicell. Each chloroplast contains two

pyrenoids. Length 46-49 /u ; breadth 39-41 /x; breadth of isthmus 11-12-5 /x. (Tab. 27,

figs. 3, 4.) Samples 24, 25.

Named after Mr. P. W. Botes of Moddernek, Nylstroom.

145. C. cajfrorum spec, nova, affinis C. subundulato Wille, sed semicellulis paulo applana-

tis, membranis lateralibus solum undulatis, ore sinus denticulato, distinguitur.

This comes near C. subundulatum Wille (W. & G. S. West, Monogr. Brit. Desm.,

Vol. 11, 1905, p. 151, PI. LIX, figs. 13-15); but the semicells are slightly flattened,

only the lateral margins are undulate and there is a small spine at each basal angle.

Cells small, 1 -2-1-3 times longer than broad, deeply constricted, sinus narrow, linear

or slightly undulate; semicells truncate-pyramidate, basal angles furnished with a

spine, sides convex with 3-4 undulations. Cell wall slightly thickened at the apex;

chloroplasts axile, with two pyrenoids each. Length 33-42-5 /x; breadth 26-33 /x;

breadth of isthmus 8-5-10-5 /x. (Tab. 25, figs. 7, 8.) Sample 24.

589

146. C. connatum Breb. (W. & G. S. West, Monogr. Brit. Desm., Vol. Ill, 1908,

p. 25, PI. LXVII, figs. 15-17.) Sample 71.

Geographical Distribution. — England; Wales; Scotland; Ireland; France;

Germany; Austria; Hungary; Galicia; Italy; Norway; Sweden; Denmark; Born-

holm; Finland; South Russia; Japan; India; Burma; Singapore; Sumatra (form);

Java; Central Africa; Sandwich Isles; United States; Guiana; Brazil; Switzerland;

South Africa.

147. C. connatum var. subellipticum var. nova, affinis C. connato var. truncato West,

sed semicellulis plus deplanatis et sinu aliquanto profundiore differt.

This is closely allied to C. connatum var. truncatum West (W. & G. S. West, Monogr.

Brit. Desm., Vol. Ill, 1908, p. 26, PI. LXVII, fig. 18); but differs in the character of

the sinus and in having more flattened semicells. Cells of medium size, about 1 -3-1 -4

times as long as broad, deeply constricted, sinus widely open; semicells subelliptic

with a flattened apex. Cell wall smooth and colourless and thickened at the apices.

Chloroplasts with two pyrenoids each. Length 48-49- 5 /x; breadth 37-40 /x; breadth

of isthmus 14-17 p; breadth of apex 14-17 p. (Tab. 27, fig. 8.) Sample 24.

148. C. contractum Kirchn. (W. & G. S. West, Monogr. Brit. Desm., Vol. II, 1905,

p. 170, PI. LXI, figs. 23-25, 34.) Sample 24.

Geographical Distribution. — England; Wales; Scotland; Ireland; Germany;

Galicia in Austria; Poland; Norway; India; Ceylon; Burma; Siam; Australia;

Central Africa; Madagascar; United States; Patagonia; First record for South

Africa.

149. C. contractum forma jacobsenii (Roy) W. & G. S. West. (W. & G. S. West,

Monogr. Brit. Desm., Vol. II, 1905, p. 171, PI. LXI, fig. 26.) Sample 71.

Geographical Distribution. — England; Scotland; Germany; Galicia in Austria;

Norway; Sweden; Denmark; Bornholm; Central Africa; South Africa.

150. C. contractum var. ellipsoideum (Elv.) West. (W. & G. S. West, Monogr. Brit.

Desm., Vol. II, 1905, p. 172, PI. LXI, figs. 28, 35.) Sample 38.

Geographical Distribution. — England; Wales; Scotland; Ireland; Germany;

Sweden; Finland; North Russia; Faeroes; Iceland; Australia; Switzerland; South

Africa.

151. C. contractum var. pseuclogartanense var. nova. Cellulae plus orbiculatae quam

in typo, membrana marginis apicalis incrassata.

The semicells are more rounded than in the typical form; cell wall thickened at

the apices and punctate. This is near var. gartanense W. & G. S. West (W. & G. S. West,

Monogr. Brit. Desm., Vol. II, 1905, p. 173, PI. LXI, fig. 27) in shape; but the apex is

not retuse-emarginate. Length 47-48 p\ breadth 30-31 p\ breadth of isthmus 14 p.

(Tab. 24, fig. 6.) Sample 39.

152. C. debaryi Arch. var. minor var. nova. Differt a typo speciei cellulis dimidio

minoribus et in medio margine apicali incisura minime profunda. Formae inter-

mediae baud visae.

These specimens are about half the size of the typical form (W. & G. S. West,

Monogr. Brit. Desm., Vol. Ill, 1908, p. 61, PI. LXX, figs. 14-16; PI. XCIII, fig. 2);

apical margin with a shallow median incision. Length 46-5-48 p; breadth 22-5-23 p\

breadth of isthmus M p', thickness 18-20 p. (Tab. 25, figs. 15, 16.) Sample 39.

590

153. C. decoratum W. & G. S. West. (W. & G. S. West, The Freshwater Algae of

Madagascar, Trans. Linn. Soc. London, Ser. 2, Vol. V, 1895-1901, PI. 7, fig. 21.)

Cell wall sculpture only distinct in the median parts of the front, vertical and side

views of the semicell as the semicell is not flat but ovate in side view and elliptic

in vertical view. In this respect I do not agree with Messrs. West’s figures. Samples

30, 46.

Geographical Distribution. — Madagascar; First record for South Africa.

1 54. C. decoratum var. galpinii var. nova, a typo speciei cellulis latioribus et membranae

sculptura in media tantum parte conspicua differt.

Cells broader than in the type; cell wall sculpture only distinct in the median part

of the semicell; cells about 1 -2 times longer than broad. Length 84 /x; breadth 70 /x;

breadth of isthmus 24 /x. (Tab. 29, fig. 11.) Sample 72.

Named after Mr. E. A. Galpin of Mosdene, Naboomspruit.

155. C. decoratum var. waterbergense var. nova, a typo speciei hiatibus, quibus sculptura

membranae consistit, plus confertis — qui ornatus non per totam superficiem

membranae apparet — et magnis granulis in ordinem circum marginem cellulae

dispositis bene distincta.

Cell wall sculpture only distinct in the median part of the semicells and with a

series of fairly large granules situated on the margins ; cells about 1 • 4 times longer

than broad. Length 73-82 /x; breadth 52-60-5 /x; breadth of isthmus 20-23-5 fx.

(Tab. 24, fig. 11.) Sample 54.

156. C. galpinii spec, nova, affinis C. isthmochondro Nordst., sed differt quod margo

apicalis duorum ordinum — quorum alter tantum a fronte videri potest — granulis

quaternis magnis est ornatus; porro membranae sculptura toto caelo differt.

This species is near C. isthmochondrum Nordst. (W. & G. S. West, Monogr. Brit.

Desm., Vol. Ill, 1908, p. 173, PI. LXXXI, fig. 7); but differs in that the apical margin

contains two series of four large granules each (only one series visible in front view of

semicell), and in the cell wall sculpture. Cells small, about 1 -2 times as long as broad,

fairly deeply constricted, sinus narrowly linear; semicells truncate-semicircular, apical

margin straight, with a series of four large granules; cell wall colourless or yellowish-

brown, punctate, and with a number of scattered granules. Vertical view more or

less elliptic, with two circular series of granules in the centre, two series of four granules

each on each side, and three smaller granules arranged in a triangle within the poles.

Length 36 /x; breadth 29 /x; breadth of isthmus 14 /x; thickness 24 /x. (Tab. 29, figs.

5, 6.) Samples 65, 70.

Named after Mr. E. A. Galpin of Mosdene, Naboomspruit.

157. C. granatum Breb. (W. & G. S. West, Monogr. Brit. Desm., Vol. II, 1905, p. 186,

PI. LXIII, figs. 1-4.) Sample 43.

Geographical Distribution. — England; Wales; Scotland; Ireland; France;

Germany; Austria and Galicia; Poland; Hungary; Italy; Bosnia; Portugal; Norway;

Sweden; Denmark; Bornholm; Finland; North and South Russia; Siberia; Faeroes:

Iceland; Greenland; Spitzbergen; Nova Zembla; Afghanistan; Kordofan; India,

Ceylon; Burma; Central China; New Zealand; Australia; West, Central and East

Africa; Madagascar; United States; West Indies; Ecuador (var.); Brazil; Paraguay

(var.); Argentina; Patagonia; Switzerland; South Africa.

158. C. granatum var. africanum Fritsch. (F. E. Fritsch & E. Stephens, Trans. Roy.

Soc. of S. Afr., Vol. IX, 1921, p. 32, fig. IIC.) (Tab. 25, figs. 17, 18; Tab. 29,

fig. 4.) Samples 39, 74.

Geographical Distribution. — South Africa; Switzerland.

591

159. C. hammeri Reinsch. var. minor var. nova, a typo specie! cellulis dimidio minoribus

diflfert. Formae intermediae baud visae.

Cells less than half the size of the typical form (W. & G. S. West, Monogr. Brit.

Desm., Vol. 11, 1905, p. 181, PI. LXII, figs. 20, 21); and about 1-3-1 -6 times longer

than broad. Length 16-19 /u.; breadth 12 /x; breadth of isthmus 3-4 fi. (Tab. 24,

fig. 13.) Sample 24.

160. C. impressulum Elfv. (W. & G. S. West, Monogr. Brit. Desm., Vol. Ill, 1908,

p. 86, PI. LXXII, figs. 14-18.) Some specimens are considerably smaller than in

the typical form, viz.: length 16-21 /x; breadth 12-15 jtx; breadth of isthmus

3-4 /X. (Tab. 24, figs. 14, 15.) Samples 7, 19, 42.

Geographical Distribution. — England; Wales; Scotland; Ireland; France;

Germany; Switzerland; Bohemia and Galicia in Austria; Denmark; Finland;

Faeroes; Greenland; Bosnia; Siberia; Japan; India; New Zealand; Australia;

West and East Africa; Azores; United States; Brazil; Argentina; Patagonia; South

Africa.

161. C. laeve Rabenh. (W. & G. S. West, Monogr. Brit. Desm., Vol. Ill, 1908, p. 99,

PI. LXXIII, figs. 8-19.) Samples 37, 38.

Geographical Distribution. — England; Wales; Scotland; Ireland; France;

Germany; Galicia in Austria; Hungary; Italy; Spain; Portugal; Norway; Poland;

North Russia; Faeroes; Nova Zembla; India; Ceylon; Burma; Siam; Australia;

Madagascar; Central and East Africa; Azores; United States; West Indies; Ecuador;

Uruguay; Argentina; Patagonia; Switzerland; South Africa.

162. C. laeve var. majus var. nova. Differt a typo specie! cellulis multo maioribus.

Formae intermediae baud visae.

This differs from the typical form in having much larger dimensions. Cells 1 • 5-1 • 7

times longer than broad. Length 48-50 ju. ; breadth 29-32 /tx; breadth of isthmus 10-1 1

/X. (Tab. 24, fig. 7.) Samples 37, 39.

163. C. maximum (Borg.) W. & G. S. West var. minor West. (W. & G. S. West,

Welwitsch’s African Freshwater Algae, Journ. of Bot., Vol. XXXV, 1897, p. 114,

Tab. 367, fig. 21.) Cells slightly larger than in the typical form. Length 69-5-70 /x;

breadth 55-5-58 p. (with spines 60-5-66 /x); breadth of isthmus 12-14 -5 p. (Tab.

27, fig. 5.) Sample 24.

Geographical Distribution. — Central Africa; First record for South Africa.

164. C. meneghinii Breb. (W. & G. S. West, Monogr. Brit. Desm., Vol. Ill, 1908, p. 90.

PI. LXXII, figs. 29-32.) Sample 38.

Geographical Distribution. — England; Wales; Scotland; Ireland; France;

Germany; Austria (and Galicia); Hungary; Italy; Norway; Sweden; Denmark;

Bornholm; Finland; Poland; Lapland; North, Central and South Russia; Faeroes;

Iceland; Nova Zembla; Greenland; Siberia; Mongolia; China; Japan; India;

Ceylon; Siam; Chatham Island; New Zealand; Madagascar; Central Africa;

Azores; United States; Porto Rico; Jamaica; Brazil; Ecuador; Paraguay; Argentina;

Patagonia; Switzerland; South Africa.

165. C. nanum spec, nova, affinis C. pseudarcto Nordst., sed cellulis ad apicem leviter

deplanatis et sinu profunde constricto distincta.

Near C. pseudarctoum Nordst. (W. & G. S. West, Monogr. Brit. Desm., Vol. III.

1908, p. 32, PI. LXVIII, figs. 12-14; PI. LXXII, figs. 40, 41); but the cells are slightly

depressed at the apices and are deeply constricted at the sinus. Cells very small, 1 - 7-1 - 8

times as long as broad; sinus narrowly linear; semicells semicircular with a slightly

592

truncately-rounded apex. Cell wall smooth. Each chloroplast contains one central

pyrenoid. Length 14-20 /la; breadth 8-12 /a; breadth of isthmus 3-4-5 /la. (Tab. 24,

fig. 12.) Sample 19.

166. C. norimbergense Reinsch. (W. & G. S. West, Monogr. Brit. Desm., Vol. Ill,

1908, p. 52, PI. LXIX, figs. 25-27.) Sample 38.

Geographical Distribution. — England; Wales; Scotland; Ireland; Germany;

Hungary; Sweden; Bornholm; Finland; India; Ceylon; Siam; New Zealand;

Central Africa; Japan; First record for South Africa.

167. C. nylstromicum spec, nova., affinis C. ndgeliano Breb., sed cellulis multo maioribus,

numero undulationum membranae et forma angulorum basalium semicellularum

differt.

This somewhat resembles C. ndgelianum Breb. (W. & G. S. West, Monogr. Brit.

Desm., Vol. Ill, 1908, p. 14, PI. LXVI, fig. 12); but differs in the number of undulations,

the delineation of the basal angles of the semicells, and in having larger dimensions.

Cells of medium size, about 1-2-1 -3 times longer than broad, deeply constricted,

sinus narrowly linear with a slightly dilated extremity; semicells truncate-pyramidate,

basal angles rounded, sides with 4-5 undulations, apical margin with four undulations;

cell wall smooth and colourless. In each semicell there is one axile chloroplast containing

a single central pyrenoid. Length 41-49 /la; breadth 33-36 p; breadth of isthmus

18-19 -5 /la; breadth of apex 18-20 p. (Tab. 26, fig. 13.) Sample 19.

168. C. obliquum Nordst. forma minima West. (W. & G. S. West, Monogr. Brit.

Desm., Vol. Ill, 1908, p. 51, PI. LXIX, figs. 22, 23.) Sample 27.

Geographical Distribution. — England; Norway; First record for South Africa.

169. C. obsoletum (Hantzsch) Reinsch var. transvaalense var. nova. Hae cellulae

praeter membranae crassationem ad semicellularem basium angulos, etiam alteram

membranae crassationem ad apicem ostendunt, polis plus quam in typo applanatis.

This differs from the typical form (W. & G. S. West, Monogr. Brit. Desm., Vol. II,

1905, p. 133, PI. LVI, figs. 1-3) in the cell wall being thickened at the apices; semicells

more depressed. Length 40-46 p; breadth 46-50 p; breadth of isthmus 12-13 p.

(Tab. 28, figs. 1-3.) Samples 24, 25, 27, 38, 39, 41.

170. C. orthostichum Lund. var. compactum W. & G. S. West. (W. & G. S. West,

Monogr. Brit. Desm., Vol. Ill, 1908, p. 169, PI. LXXX, fig. 22.) Samples 7, 43.

Geographical Distribution. — Scotland; First record for South Africa.

171. C. orthostichum var. pseudopumilum var. nova, affinis var. pumilo Lund., sed

semicellulis minus applanatis et sculptura membranae toto caelo differt.

This differs from var. pumilum Lund. (W. & G. S. West, Monogr. Brit. Desm.,

Vol. HI, 1908, p. 169, PI. LXXX, figs. 20, 21) in having less flattened semicells and in

the sculpture of the cell wall. Cells small, a little longer than broad, deeply constricted,

sinus open outwards; semicells subelliptic, with a somewhat flattened apex. Cell

wall granulate, granules without any definite arrangement; chloroplasts with one

pyrenoid each. Length 23-29 p\ breadth 19-5-24 /x; breadth of isthmus 8-5-12 p.

(Tab. 25, fig. 3.) Sample 3.

1 72. C. orthostichum var. transvaalense var. nova, a typo specie! numero et dispositione

granulorum in membrana et sinu angustiore differt.

This differs from the typical form (W. & G. S. West, Monogr. Brit. Desm., Vol. Ill,

1908, p. 167, PI. LXXX, figs. 12-19) in the number and disposition of the granules,

and in having a narrower sinus. Cell wall granulate, granules arranged in 9-10 vertical

593

series, with 4-6 granules in each series; granules more or less of the same size. Length

19-22 fi; breadth 16-18 breadth of isthmus 5-5-6 ijl. (Tab. 26, fig. 8.) Samples

20, 22.

173. C. pachydermum Lund. var. waterbergense var. nova, affinis C. pachydermo var.

aethiopico W. et. G. S. West, sed cellulis aliquanto minoribus et ad apicem deplanatis

differt.

This is near C. pachydermum var. aethiopicum W. & G. S. West (W. & G. S. West,

Monogr. Brit. Desm., Vol. II, 1905, p. 140, PI. LVII, figs. 8, 9); but differs in having

smaller dimensions and flattened apices and in the cell wall sculpture. Cell wall more

or less regularly granulate, granules small. Length 43-54 yx; breadth 36-46 /x; breadth

of isthmus 21-25 /X. (Tab. 28, fig. 4.) Samples 18, 51, 52, 53, 54, 57, 58, 72.

174. C. pseudamoenum Wille. (W. & G. S. West, Monogr. Brit. Desm., Vol. IV, 1912,

p. 31, PI. Cl I, figs. 7-9.) (Tab. 29, figs. 9, 10.) Samples 66, 70, 71.

Geographical Distribution. — England; Ireland; Germany; Galicia in Austria;

Sweden; Bornholm; Finland; North Russia; Central China; New Zealand; East

Africa; United States; Brazil; First record for South Africa.

175. C. pseudoprotractum spec, nova, affinis C. protracto (Nag.) De Bary, sed lobo

polari amplius rotundato, forma cellulae a latere visae et dispositione granulorum

differt.

Near C. protractum (Nag.) De Bary (W. & G. S. West, Monogr. Brit. Desm.,

Vol. Ill, 1908, p. 181, PI. LXXXII, fig. 8; PI. XCLV, figs. 4, 5); but the shape of the

apical lobe, the shape of the semicells in lateral view and the arrangement of the granules

differ. Cells of medium size, longer than broad, deeply constricted, sinus narrowly

linear with a slightly dilated extremity; semicells 3-lobed, with a subrectangular incision

between the apical lobe and each lateral lobe, lateral and apical lobes broadly rounded

and fairly short; lobes granulate, granules large and arranged in 3-4 more or less

distinct horizontal rows and in somewhat irregular vertical rows. Side view of semicell

subovate; vertical view rather narrowly elliptic, with a smooth protuberance at the

middle on each side. Chloroplasts axile, with one pyrenoid each. Length 48 /x; breadth

32-34 /lx; breadth of isthmus 12-5-14 /x; thickness 20-21 /x; breadth of apical lobe

16-19 ;Lx. (Tab. 25, figs. 4-6.) Samples 9, 11.

176. C.pseudopyramidatum Lund. (W. & G. S. West, Monogr. Brit. Desm., Vol. II,

1905, p. 201, PI. LXIV, figs. 9-12.) Samples 19, 25.

Geographical Distribution. — England; Wales; Scotland; Ireland; France;

Germany; Austria and Galicia; Poland; Hungary; Italy; Norway; Sweden; Fin-

land; North and South Russia; Siberia; Spitzbergen; Nova Zembla; Faeroes;

Ceylon; Siam; New Zealand (var.); Australia; West, Central and East Africa;

Madagascar; United States; West Indies; Brazil; Paraguay; Uruguay; Switzerland;

Mozambique; South Africa.

177. C. pseudotaxichondrum Nordst. var. atomicum var. nova, affinis C. pseudotaxi-

chondro var. siamensi W. et G. S. West, sed membrana baud undulata, cellula a

vertice visa ad polum utrimque denticulo ornata distincta est.

This variety is near var. siamense W. & G. S. West (W. & G. S. West, Siamese

Desmids, Botanisk Tidsskrift, Bind 24, 1902, p. 173, Tavle 3, fig. 26); but there are

no undulations; each pole has a small tooth in vertical view. Length 31-32 /x; breadth

28-28-5 /x; breadth of isthmus 11-12 /x; thickness 16-18 /x. (Tab. 25, figs. 1, 2.)

Sample 3.

594

178. C. pyramidatum Breb. (W. & G. S. West, Monogr. Brit. Desm., Vol. II, 1905,

p. 199, PI. LXIV, figs. 5-7.) Samples 24, 25, 27, 39.

Geographical Distribution. — England; Wales; Scotland; Ireland; France;

Germany; Austria and Galicia; Hungary; Italy; Norway; Sweden; Denmark;

Bornholm; Finland; South Russia; India; Ceylon; Burma (form); West Africa;

Azores; United States; Brazil; Paraguay; Uruguay (form); Argentina; Patagonia

(form); Switzerland; Japan; Mozambique; South Africa.

179. C. pyramidatum var. elongatum var. nova. Differ! a typo speciei cellulis dimidio

longioribus, nulla latitudinis discrepantia.

The cells are about one and a half times longer than the typical form; but the

difference in breadth is slight. Cells very large, slightly more than twice as long as

broad ; sides of semicells almost straight, apex slightly concave, cell wall with a thickening

at the apices and densely punctate. Length 149-150 p.; breadth 69-5-70 p; breadth

of isthmus 26-28 p. (Tab. 27, fig. 10.) Sample 22.

180. C. pyramidatum var. majus var. nova, a typo speciei et cellulis aliquantulo maioribus

et quod membranae cellularum in apicibus crassiores fiunt, et semicellularis non

ita constanter truncatis et pyramidalibus differ!.

The shape of the cells in front view is somewhat variable, the cells are larger than

in the typical form, and the cell wall is densely punctate and thickened at the apices.

This may also be compared with C. pseudopyramidatum Lund, subsp. maximum Borgesen

forma minor Fritsch & Rich (F. E. Fritsch & F. Rich, Trans. Roy. Soc. of S. Afr.,

Vol. XI — Part 4, 1924, p. 347, fig. 17-E-G) to which it has a great resemblance as far

as the shape of its semicells in front view is concerned, but from which it differs in

having larger dimensions, in having no thickening of the membrane in the region above

the isthmus, as seen in lateral and vertical views in the majority of the specimens investi-

gated, and in the individuals always containing two pyrenoids per semicell. Length

1 18-130 /x; breadth 70-76 /Lt; breadth of isthmus 26-30 /x. (Tab. 24, figs. 4, 5.) Samples

38, 39, 41.

181. C. quadratum Ralfs var. africanurn Fritsch. (F. E. Fritsch & E. Stephens, Trans.

Roy. Soc. of S. Afr., Vol. IX, 1921, p. 33, fig. 12.)

Cells about 1 • 6-1 • 8 times as long as broad; basal angles of semicells less prominent

than in the type; cell wall distinctly punctate. Length 41-64 p\ breadth 24-36 p:

breadth of isthmus 8-10 p. (Tab. 24, fig. 10.) Sample 24.

Geographical Distribution. — South Africa.

182. C. reniforme (Ralfs) Arch. (W. & G. S. West, Monogr. Brit. Desm., Vol. III.

1908, p. 157, PI. LXXIX, figs. 1, 2; PI. LXXXII, fig. 15.) Specimens usually

slightly smaller than in the typical form. Length 40-46 p: breadth 38-48 p:

breadth of isthmus 13-17 p. (Tab. 28, figs. 5, 6.) Samples 4, 7, 19, 30, 32, 42,

47, 48, 51, 57.

Geographical Distribution. — England; Wales; Scotland; Ireland; France;

Germany; Switzerland; Austria and Galicia; Italy; Norway; Sweden; Faeroes;

Greenland; Spitzbergen; United States; Brazil; Argentina; South Africa.

183. C. schweickerdtii spec, nova, affinis C. cucumi (Corda) Ralfs. Margo apicalis

membranae crenis — binis vel ternis — ornatus est; iuxta marginem apicalem bina

vel terna granula adsunt conspicua ac sub granulis 3 vel 4 depressiones ellipticae.

Near C. cucumis (Corda) Ralfs (W. & G. S. West, Monogr. Brit. Desm., Vol. II,

1905, p. 152, PI. LIX, figs. 18-20); but the apical margin has 2-3 crenations, there are

2-3 large granules within the median part of the apex and 3-4 elliptical depressions

just below the granules. Cells small, about 1-4 times longer than broad, deeply con-

595

stricted, sinus narrowly linear; semicells more or less semicircular. Cell wall punctate.

Length 38-40 breadth 28 fi; breadth of isthmus 10-11 /x. (Tab. 29, figs. 7, 8.)

Sample 71.

Named after Prof. Dr. H. G. Schweickerdt, Head of the Department of General

Botany, University of Pretoria.

184. C. sexangulare Lund. (W. & G. S. West, Monogr. Brit. Desm., Vol. Ill, 1908,

p. 81, PI. LXXII, fig. 3.) (Tab. 25, fig. 9.) Samples 24, 25.

Geographical Distribution.— Scol\&nd\ Germany; Galicia in Austria; Hungary;

Sweden; Poland; North and South Russia; Japan; Australia; East Africa; Azores;

United States; South Africa.

185. C. sexnotatum Gutw. var. simplex var. nova, affinis C. sexnotato var. tristriato

(Liitkem.) Schmidle, sed membrana cellulae omnino glabra differt.

Near var. tristriatum (Liitkem.) Schmidle (W. & G. S. West, Monogr. Brit. Desm.,

Vol. Ill, 1908, p. 228, PI. LXXXVl, figs. 8, 9); but the cell wall is smooth and without

granules. Sides of semicells convex and 5-crenate, apex truncate and straight. Length

28 /x; breadth 22 /x; breadth of isthmus 10 /x; breadth of apex 11-12 /x. (Tab. 24,

figs. 8, 9.) Sample 19.

186. C. speciosum Lund. var. simplex Nordst. forma africanum forma nova, a varietate

simplici Nordst. forma genuinum numero granulorum et cellula apicem versus

minus deplanata differt.

These specimens form a group within the limits of Nordstedt’s variety (W. & G. S.

West, Monogr. Brit. Desm., Vol. Ill, 1908, p. 250, PI. LXXXIX, fig. 6); but which

nevertheless can be distinguished by the number of granules and less attenuated apices.

Margins of semicells crenate, crenations 14-16 (four apical and 5-6 lateral); granulate

within the margins, granules in regular radial series, seven granules in each series,

basal vertical series of granules absent. Chloroplasts axile, with one pyrenoid each.

Length 44-48 /x; breadth 26-28 /x; breadth of isthmus 16-18 /x. (Tab. 26, figs. 9, 10.)

Sample 19.

187. C. subauriculatum W. & G. S. West. (W. & G. S. West, The Freshwater Algae

of Madagascar, Trans. Linn. Soc. London, Ser. 2, Vol. V, 1895-1901, PI. 6, fig. 31.)

Sample 71.

Geographical Distribution. — Madagascar; South Africa.

188. C. subconnatum spec, nova, a latere visa affinis C. connato Breb. var. truncato

West, sed a fronte visa cellulis plus deplanatis et pyramidalibus distincta.

In lateral view, this species resembles the front view of C. connatum Breb. var,

truncatum West (W. & G. S. West, Monogr. Brit. Desm., Vol. II, 1908, p. 26, PI. LXVIl,

fig. 18); but it has rounded apices; the semicells in front view are somewhat truncate-

pyramidate. Cells of medium size, about 1 • 1-1 -3 times longer than broad, moderately

constricted, sinus very widely open with an obtuse apex; semicells transversely subelliptic

with a broad base, apex flattened. Cell wall punctate with two series of minute punctae

on each side of the isthmus. Side view of semicell subovate. Chloroplasts containing

two pyrenoids each. Length 42-66 /x; breadth 34-56 /x; breadth of isthmus 24-38 jx.

(Tab. 26, figs. 14, 15.) Sample 20.

189. C. subcostatum Nordst. var. warmbadianum var. nova, differt a typo speciei

dispositione et numero granulorum in membrana.

This differs from the typical form (W. & G. S. West, Monogr. Brit. Desm., Vol. Ill,

1908, p. 236, PI. LXXXVII, figs. 3-5) in the number and arrangement of the granules.

Sides of semicells convex and crenulate, with about two small entire crenulations near

the basal angles and 3-5 emarginate (or bigranulate) crenulations on the rest of the

596

lateral margin; apex truncate and with a few small undulations; within the margin

minutely granulate, granules radially and concentrically disposed; opposite each of

the three crenulations at the base of the semicell is a series of three granules, the two

series next to the margin opposite the next crenulation binate and the series nearest

the centre single, the first series next to the margin opposite the following two crenu-

lations binate and the second and third series single; next to the apex two series of

two granules each; in the centre, above the isthmus with a granulated tumour, granules

in three vertical series of three each. Length 32-36 /x; breadth 27-30 /x; breadth of

isthmus 10-12 fx; breadth of apex 12-13 /x. (Tab. 25, figs. 12-14.) Sample 48.

190. C. subtumidum Nordst. var. klebsii (Gutw.) W. & G. S. West. (W. & G. S. West,

Monogr. Brit. Desm., Vol. II, 1905, p. 193, PI. LXIII, figs. 21-23.) Sample 7.

Geographical Distribution. — England; Wales; Scotland; Ireland; Germany;

Galicia in Austria; South Africa.

191. C. subtumidum var. theronii var. nova, affinis C. subtumido var. klebsii (Gutw.)

W. et G. S. West, sed cellulis triplo maioribus, membrana cellulae densis punctis

distincta, sinu foras aperto differt.

Near var. klebsii (Gutw.) W. & G. S. West; but differs in the cells being about

thrice the size of var. klebsii, and in having an open sinus and a densely punctated cell

wall. Cells very large, about 1 -2-1 -4 times as long as broad, deeply constricted, sinus

open; semicells broadly pyramidate with truncate apices, basal angles rounded. Cell

wall very densely punctate, with a thickening at the apex of each semicell. Each

chloroplast with two pyrenoids. Length 141-5-148 /x; breadth 104-124 /x; breadth

of isthmus 42-44 /x; breadth of apices 40-43-5 /x. (Tab. 27, fig. 9.) Samples 42, 72.

Named after Mr. J. E. Theron of Rietfontein 288, near Nylstroom.

192. C. tetragonum (Nag.) Arch. var. transvaalense var. nova, affinis C. tetragono

var. eleganti (Roy et Biss.) W. et G. S. West, sed numero undulationum membranae,

sculptura eiusdem, et cellula apicem versus minus vel omnino non deplanata

differt.

Near var. elegans (Roy et Biss.) W. & G. S. West (W. & G. S. West, Monogr. Brit.

Desm., Vol. Ill, 1908, p. 20, PI. LXVI, fig 27); but differs in the number of undulations,

in the cell wall sculpture and in the semicells, which are less attenuated towards the apices.

Cells about 1 -6-1 -7 times as long as broad; sides of semicells (including the angles) 7-8-

undulate, apex with 8-10 undulations; with a series of three granules within each lateral

undulation and a series of five granules within each apical undulation. In each semicell

there is one axile chloroplast containing two pyrenoids. Length 44-51 /x; breadth

29-34 fi; breadth of isthmus 17-20-5 /x; breadth of apices 21-32 jx. (Tab. 26, figs. 11,

12.) Sample 19.

193. C. trachypleurum Lund. var. pseudonatalensis var. nova, affinis C. trachypleuro

var. natalensi Fritsch et Rich, sed hie spinarum vel maxima pars in margine sita,

quum in Fritschii et Richii varietate aliquantulum introrsum ab margine absit.

This differs from var. natalensis Fritsch & Rich (F. E. Fritsch & F. Rich, Trans.

Roy. Soc. of S. Afr., Vol. XI — Part 4, 1924, p. 349, fig. 18-I-J) in having pronounced

spines on the margins and about three series of spines within the margins. Length

60-64 IX (with spines 67-69 /x); breadth 48-51 -5 p (with spines 56-57 /x); breadth of

isthmus 18-19 /x. (Tab. 27, fig. 7.1 Sample 24.

597

194. C. trachypleurum var. pseudonatalensis forma irregulare forma nova. Dentes

temere dispositi. Semicellula dentibus ternis sub margine apicali glabro munita

est.

In these specimens the spines are irregularly disposed, and there are three spines

just within the apical margin, which is smooth. Length 58-59 /x; breadth 44-47 /x;

breadth of isthmus 16-17 /x. (Tab. 27, fig. 6.) Sample 25.

195. C. trachypleurum var. subspinosum var. nova, afirnis C. trachypleuro var. spinoso

West, sed differt quod singularum semicellularum media in parte iuxta supra

isthmum quattuor tantum spinulae adsunt.

The specimens closely resemble var. spitiosum West (G. Nygaard, Trans. Roy.

Soc. of S. Afr., Vol. XX— Part 2, 1932, p. 142, fig. 43; W. & G. S. West, Freshwater

Algae of Madagascar, Trans. Linn. Soc. London, Ser. 2 — Bot., Vol. V, 1895-1901,

PI. 7, fig. 17) except that the median tumour contains only four granules. The shape

of the semicells in front view has a greater resemblance to the figure by Nygaard than

to the type illustrated by Messrs. West. Length 42-42-5 /x; breadth 32-5-34 p.; breadth

of isthmus 12-13 p; breadth of apices 16-18 p. (Tab. 25, fig. 10.) Sample 47.

196. C. transvaalense spec, nova, affinis C. trachypleuro Lund., sed semicellulis minus

applanatis, margine apicali spinis munito, ore sinus spinoso, et omni spinarum

dispositione valde distincta.

This species is near C. trachypleurum Lund. (W. & G. S. West, Monogr. Brit.

Desm., Vol. Ill, 1908, p. 172, PI. LXXXI, figs. 2, 3); but the semicells are less flattened,

there are spines on the apical margin, there is a spine at each basal angle, and the cell

wall sculpture is different. Cells of medium size, about 1 - 1-1 - 3 times longer than broad,

deeply constricted, sinus narrowly linear; semicells semicircular, apex somewhat

flattened. Cell wall punctate and furnished with a number of irregularly disposed

spines. Length 48-50-5 yu,; breadth 36-37-5 /x; breadth of isthmus 15-16 )tx. (Tab. 25,

fig. 11.) Sample 25.

197. C. trilobulatum Reinsch. (W. & G. S. West, Monogr. Brit. Desm., Vol. II, 1905,

p. 185, PI. LXII, figs. 28-30.) Samples 19, 72.

Geographical Distribution. — England; Scotland; Ireland; France; Germany;

Hungary; North Russia; Poland; Norway; Sweden; New Zealand; Madagascar;

Central and East Africa; Brazil; Japan; South Africa.

198. C. tumidum Lund. (W. & G. S. West, Monogr. Brit. Desm., Vol. II, 1905, p. 160,

PI. LX, fig. 18.) Sample 71.

Geographical Distribution. — Wales; Scotland; Ireland; France; Germany

(form); Norway; Sweden; Denmark; Poland; North Russia; Finland; Greenland;

United States; First record for South Africa.

199. C. waterbergense spec, nova, affinis C. holmiensi Lund., sed membrana hand

undulata differt.

Near C. holmiense Lund. (W. & G. S. West, Monogr. Brit. Desm., Vol. Ill, 1908,

p. 1, PI. LXV, figs. 1, 2); but differs in having crenulations. Cells rather small, about

1-5-1 -8 times longer than broad, deeply constricted, sinus narrowly linear; semicells

broadly pyramidate with a flattened apex, basal angles rounded. Cell wall punctate.

Length 35-39 /x; breadth 20-26-5 p\ breadth of isthmus 8-9-5 p\ breadth of apex

12-14 p. (Tab. 26, figs. 1-7.) Sample 19.

202909—6

598

Xanthidium Ehrenberg.

200. X. cristatum Breb. var. delpontei Roy & Biss, forma laevis forma nova, differt

a varietate delpontei et membrana cellulae omnino glabra et quod spinulae supra

isthmum desunt.

A form in which the semicells are entirely destitute of a tliickened or scrobiculated

central area; the shape of the semicells differ slightly from the typical form (W. & G. S.

West, Monogr. Brit. Desm., Vol. IV, 1912, p. 74, PI. CXI, fig. 5). Length 72 /j. (with

spines 90-94 fi); breadth 68 p (with spines 80-88 p); breadth of isthmus 15-5 p. (Tab.

29, fig. 12.) Sample 74.

201. X. cristatum var. transvaalense var. nova, affinis var. delpontei Roy et Biss., sed

cellulis multo angustioribus differt. Formae intermediae non visae.

This differs from var. delpontei in the narrower semicells. Cells about 1-25-1 -5

times longer than broad; each tumour above the isthmus with a circle of 7-10 small

spines; in some specimens there may be a spine in the centre of the circle. Length

48-56 p', breadth 38-44 p', breadth of isthmus 16-18 p\ length of spines 8-12 p.

(Tab. 28, figs. 7-10.) Samples 24, 25.

Straurastrum Meyen.

202. St. barbaricum spec, nova, affinis St. cajfrorum Claassen sed differt quod depressio

in margine apicali deest, porro supra isthmum tantum una series spinularum adest

sculpturaque membranae etiam differt.

This is near St. cajfrorum Claassen; but differs in having no cavity in the apical

margin, and in the cell wall sculpture. Cells small, about 1-3-1 -4 times as long as

broad, deeply constricted, sinus open; semicells more or less quadrangular in vertical

view, with a single series of spines near the base, apical margin and angles furnished

with spines. Length 28 p', breadth 20-21-5 p\ breadth of isthmus 7 p. (Tab. 30,

figs. 10, 11.) Sample 3.

203. St. brevispinum Breb. var. masoganum var. nova, affinis St. brevispino var. obverso

W. et G. S. West, sed sinu latiore et cellularum e vertice visarum lateribus undulatis

neque concavis differt.

This variety is near var. obversum W. & G. S. West (W. & G. S. West, Monogr.

Brit. Desm., Vol. IV, 1912, p. 148, PI. CXXIII, fig. 7); but the sinus is broader, and

the sides in vertical view are undulate and not concave. Cells small, slightly shorter

than broad, deeply constricted, sinus open; semicells more or less fusiform, each apical

angle furnished with a short spine. Vertical view of cell triangular, sides 3-undulate.

Cell wall smooth. One pyrenoid per chloroplast. Length 26 /x; breadth 28 breadth

of isthmus 8 p. (Tab. 29, figs. 19, 20.) Sample 66.

204. spec. nova. Inter species descriptas nulla affinitas obvia. Semicellulae

e vertice visae plus minusve quadratae, margo apicalis depressione praeditus,

quae spinis brevibus circumdata est. Supra isthmum duae series spinularum adsunt

et anguli apicales semicellularum etiam spinulis muniti.

Cells small, about 1 -2 times as long as broad, deeply constricted, sinus open out-

wards; semicells subelliptic, more or less quadrangular in vertical view, apical margin

with a cavity in the median part, margin of cavity furnished with short spines; above

the isthmus in each semicell are two series of small spines, apical angles furnished

with spines. Length 24 p-, breadth 20 p (greatest — fig. 7) and 16 p (smallest — fig. 9);

breadth of isthmus 8 p. (Tab. 30, figs. 7-9.) Sample 3.

599

205. St. connatum Roy & Biss. var. warmbadianum var. nova, affinis St. connato var.

americano W. et G. S. West, sed differt et semicellulis latioribus brevioribusque

et quod a vertice visa latera concaviora sunt.

This closely resembles var. americanum W. & G. S. West (W. & G. S. West, North

American Desmidieae, Trans. Linn. Soc. London, Ser. 2, Bot., Vol. V, 1895-1901,

PI. 16, fig. 7); but the semicells are shorter and broader, and the sides in vertical view

are more concave. Length 20-22 /x; breadth 22-23 breadth of isthmus 6-7-5 /x.

(Tab. 31, figs. 23, 24.) Sample 47.

206. St. crenatum spec, nova, aflfinis St. elliptico West, sed membrana cellulae crenata

differt.

Near St. ellipticum West (W. & G. S. West, Monogr. Brit. Desm., Vol. IV, 1912,

p. 138, PI. CXIX, fig. 7); but the cell wall is crenated. Cells small, about 1-2 times

as long as broad, deeply constricted, sinus widely open; semicells triangular in vertical

view. Length 24-25 /x; breadth 20-20-5 p\ breadth of isthmus 9-10 /x. (Tab. 31,

figs. 21, 22.) Samples 54, 58, 63.

207. St. dickiei Ralfs. (E. Messikommer, Beitrage zur geobotanischen Landesaufnahme

der Schweiz, Heft 24, 1942, Taf. XII, figs. 3, 4.) Spine on apical angle straight

or slightly incurved. (Tab. 31, figs. 7, 8.) Samples 59, 72.

Geographical Distribution. — Davos in Switzerland; Madagascar; United States;

First record for South Africa.

208. St. dilatatum Ehrenb. (W. & G. S. West, Monogr. Brit. Desm., Vol. IV, 1912,

p. 172, PI. CXXVI, figs. 10-15.) Sinus less widely open than in the type and all

the specimens observed were triangular in vertical view. The Rietfontein and

Mosdene specimens are slightly larger than the other specimens. Length 24-40 /x;

breadth 20-36 /x; breadth of isthmus 5-12 p. (Tab. 30, figs. 1-6; Tab. 31, figs.

14-16.) Samples 2, 3, 12, 19, 20, 24, 27, 32, 35, 38, 41, 42, 51, 65, 66, 67, 71, 72.

Geographical Distribution. — England; Wales; Scotland; Ireland; France;

Germany; Austria and Galicia; Hungary; Roumania; Servia; Macedonia; Italy;

Portugal; Norway; Sweden; Denmark; Central and South Russia; Greenland;

Japan; India; Ceylon; Australia; New Zealand; Madagascar; Central Africa;

Azores; United States; Brazil; Bolivia; Paraguay; Argentina; South Africa.

209. St. excavatum W. & G. S. West. (W. & G. S. West, The Freshwater Algae of

Madagascar, Trans. Linn. Soc. London, Ser. 2, Bot., Vol. V, 1895-1901, PI. 8,

fig. 42.) Length 12 /x; breadth 10 /x; length of arms 12-20 p] breadth of isthmus

6 p. (Tab. 32, fig. 7.) Samples 65, 72.

Geographical Distribution. — Madagascar; First record for South Africa.

210. St. furcatum (Ehrenb.) Breb. (F. Rich, Trans. Roy. Soc. of S. Afr., Vol. XX — Part

2, 1932, p. 175, fig. 11 A-C.) (Tab. 32, figs. 3-6.) Samples 65, 66, 74.

Geographical Distribution. — Davos in Switzerland; South Africa.

211. St. galpinii spec, nova, affinis St. rotulae Nordst., sed differt quod semicellula

7 tantum brachiis et margo apicalis 7 papillis est praeditus.

Near St. rotula Nordst. (W. & G. S. West, North American Desmidieae, Trans.

Linn. Soc. London, Ser. 2, Bot., Vol. V, 1895-1901, PI. 18, fig. 11); but each semicell

has only seven arms, and the apical margin is furnished with seven papillae. Cells

of medium size, about 1-75-2-2 times longer than broad, sinus widely open; semicells

with seven slightly incurved arms, each arm containing four small spines at the apex,

sides crenate with a circle of spines around each crenation (not clear in some specimens);

apical margin furnished with seven emarginate projections. Chloroplasts with one

600

central pyrenoid each. Length 38-44 /x (with papillae 46-50 ^i); breadth 18-25 /n

(with arms 56-64 /x); breadth of isthmus 10-13 /x. (Tab. 29, figs. 15-17 w.) Samples

66, 71.

Named after Mr. E. A. Galpin of Mosdene, Naboomspruit.

212. St. gemeUiparum Nordst. var. africanum var. nova, a typo specie! diflfert quod

circulus intimus 6 brachiorum propius mediam cellulae partem dispositus est.

This closely approaches the typical form (W. & G. S. West, The Freshwater Algae

of Ceylon, Trans. Linn. Soc. London, Ser. 2, Vol. VI, 1901-1902, PI. 21, fig. 25); but

in vertical view, the inner circle of six arms is nearer to the axis of the cell. Cells small,

about 1-4 times as long as broad, deeply constricted, sinus open; semicells more or

less elliptical, furnished with six short arms near the base and another six near the apical

margin, apices of arms emarginate. Vertical view of semicell triangular, with two arms

on each angle and a circular series of six arms within the margin, angles truncate. Each

chloroplast containing one central pyrenoid. Length 24 p. (with arms 30 /x); breadth

\1 p (with arms 24 /x); breadth of isthmus 9 p. (Tab. 32, figs. 1, 2.) Sample 66.

213. St. leptocladwn Nordst. var. cornutum Wille. These specimens are larger than

in the typical form (W. & G. S. West, Freshwater Algae of Madagascar, Trans.

Linn. Soc. London, Ser. 2, Vol. V, 1895-1901, PI. 9, figs. 12, 13; F. Rich, Trans.

Roy. Soc. of S. Afr., Vol. XX — Part 2, 1932, p. 175, fig. 13 F). Length 38-41 p\

breadth 11-14 p (with processes 72-108 p)\ breadth of isthmus 8-9 p. (Tab. 32,

fig. 8.) Sample 74.

Geographical DAP-/Z>m//o/7.— Madagascar; South Africa.

214. St. mattiei spec. nova. Nulla affinitate notata. Cellulae parva, aliquanto latiores

quam longiores, sinu foras aperto, membrana spinulis praecipue in angulis apicalibus

praedita, semicellula e vertice visa lateribus concavis triangulata.

Cells small, slightly shorter than broad, deeply constricted, sinus open. Vertical

view of semicell triangular, sides concave. Cell wall furnished with short spines, mainly

on the apical angles. Length 17-5-26 p\ breadth 19-5-28 /x; breadth of isthmus

6-14 p; length of spines 3-4 p. (Tab. 31, figs. 25-27.) Sample 24.

215. St. mesianum spec. nova. Cum sp. St. subgenvnulato W. et G. S. West congruit

quod semicellulis sena brachia sunt, quae tamen glabra sunt, spinula utrimque

ad basim tantum praedita.

Near St. subgemmulatum W. & G. S. West (W. & G. S. West, The Freshwater

Algae of Madagascar, Trans. Linn. Soc. London, Ser. 2, Bot., Vol. V, 1895-1901,

PI. 8, fig. 34) in that each semicell contains six arms, but the arms are smooth and

furnished with a small spine on each side at the base. Cells moderately constricted,

sinus open; semicells with six gradually attenuated arms, with four small spines at

the apex of each arm. Cell wall smooth. Each chloroplast with one pyrenoid. Length

28-29 /x; breadth 20-21 p (with arms 43-44 p); breadth of isthmus 10-12 /x. (Tab. 31,

figs. 3, 4.) Sample 59.

Named after the late Prof. Dr. M. G. Mes, formerly Head of the Department of

Plant Physiology and Biochemistry, University of Pretoria.

216. St. orbiculare Ralfs var. ralfsii W. & G. S. West. (W. & G. S. West, Monogr.

Brit. Desm., Vol. IV, 1912, p. 156, PI. CXXIV, figs. 12-16.) Cells slightly smaller

than in the typical form. Length 26 p; breadth 21 p; breadth of isthmus 7-5-8 p.

(Tab. 30, figs. 14, 15.) Sample 60.

Geographical Distribution. — England; Wales; Scotland; Ireland; France;

Germany; Switzerland; Austria and Galicia; Hungary; Servia; Italy; Portugal;

Norway; Sweden; Denmark; Bornholm; Finland; Poland; North, Central and

601

South Russia; Faeroes; Iceland; Nova Zembla; Spitzbergen; Greenland; East

and North India; Australia; United States; Bolivia; Paraguay; Agentina; Pata-

gonia; Brazil; First record for South Africa.

217. St. orbiculare var. waterbergense var. nova, affinis St. orbiculari var. hibernico

W. et G. S. West, sed lateralibus, a vertice visis, concavis differt.

Near var. hibernicum W. & G. S. West (W. & G. S. West, Monogr. Brit. Desm.,

Vol. IV, 1912, p. 156, PI. CXXIV, figs. 5-9); but differs in that the sides are concave

in vertical view. Length 31-33 /n; breadth 28-30 /x; breadth of isthmus 7-5-9 /x.

(Tab. 31, figs. 5, 6.) Samples 52, 54, 58.

218. St. pseudogenmmlatum spec, nova, affinis St. subgemmulato W. et G. S. West,

sed differt et brachiis leviter incurvatis et quod semicellulis a vertice visis binae

sunt granulorum series, quae circuli modo dispositae ternis consistunt granulis.

Circulo interiori sex euismodi sunt series, exteriori tamen duodecim.

This differs from St. subgemmulatum W. & G. S. West (W. & G. S. West, The

Freshwater Algae of Madagascar, Trans. Linn. Soc. London, Ser. 2, Bot., Vol. V,

1895-1901. PI. 8, fig. 34) in having slightly incurved arms and in the cell wall sculpture.

Cells small, about 1-6-1 -7 times longer than broad, sinus open; semicells with six

arms, apex of each arm emarginate. Cell wall, in vertical view of semicell, with two

circular series of granules arranged in groups of three granules each, inner circle consists

of six groups and outer circle contains 12 groups; and with three circles of granules

around each arm. Each chloroplast contains a central pyrenoid. Length 29-32 /x;

breadth 17-19 /X (with arms 33-34 /x); breadth of isthmus 10-1 1 (Tab. 31, figs. 9, 10.)

Sample 19.

219. St. pseudogemmulatum var. warmbadianwn var. nova, affinis St. pseudogemmulatum

Claassen, sed brachiis quam prioris varietatis longioribus tenuioribusque differt,

quae a vertice visa 4 granula singulis in seriebus ad interiorem circulum habent.

Arms longer and narrower than in St. pseudogemmulatum Claassen; cell wall

with the inner series of granules consisting of four granules per group. Length 28 - 5-32

/x; breadth 16-17 /x (with arms 34-37 p.)\ breadth of isthmus 7-5-9 /x. (Tab. 31, figs.

11-13.) Sample 48.

220. St. pygmaeum Breb. var. botesii var. nova, affinis var. apiculato W. et G. S. West,

sed anguhs apicalibus spinahs munitis, ceteroquin omnino glabra differt.

This differs from var. apiculatum W. & G. S. West (W. & G. S. West, Freshwater

Algae of Madagascar, Trans. Linn. Soc. London, Ser. 2, Bot., Vol. V, 1895-1901,

PI. 8, fig. 26) in the cell wall being smooth except for the apical angles, which contains

a few spines. Ceils small, about as long as broad, deeply constricted, sinus open;

semicells with three spines on the apical angles and a circle of spines near the angles.

Vertical view triangular. Length 26-28 /x; breadth 25-5-26 /x; breadth of isthmus

8-5-10 /X. (Tab. 31, figs. 28, 29.) Sample 24.

Named after Mr. P. W. Botes of Moddernek, Nylstroom.

221. St. quadrangulare Breb. var. subarmatum var. nova, affinis St. quadrangulari var.

armato W. et G. S. West, sed sinu latiore, spinulis longioribus, et cellulis multo

maioribus differt; semicellulae e vertice visae triangulatae.

This differs from var. armatum W. & G. S. West (W. & G. S. West, North American

Desmidieae, Trans. Linn. Soc. London, Ser. 2, Bot., Vol. V, 1895-1901, PI. 16, fig. 18)

in having a broader sinus, longer spines and considerably larger cells ; semicells triangular

in vertical view. Fig. 18 shows the cell in front view; it was impossible to keep the

cell in vertical position in order to illustrate it. Length 32 /x; breadth 41-43 /x (with

spines 48-50 |x); breadth of isthmus 14-14-5 /x. (Tab. 29, fig. 18.) Samples 68, 71.

602

222. St. setigerum Cleve var. pectinatum W. et G. S. West forma australe forma nova.

A var. pectinato W. et G. S. West isthmo multo latiore differt.

This form differs from var. pectinatum (W. & G. S. West, North American Des-

midieae, Trans. Linn. Soc. London, Ser. 2, Bot., Vol. V, 1895-1901, PI. 16, fig. 28) in

that the isthmus is about 2-2 times broader. Length 33 p. (with spines 46 /x); breadth

32 p (with spines 48 p); breadth of isthmus 14 p. (Tab. 29, figs. 13, 14.) Sample 65.

223. St. subgemmulatum W. & G. S. West var. mattiei var. nova, affinis var. gracili

W. et G. S. West, sed cellulae brachiis longioribus tenuioribusque praeditae earn

distinguunt.

These specimens are near var. gracilius W. & G. S. West (W. & G. S. West, The

Freshwater Algae of Madagascar, Trans. Linn. Soc. London, Ser. 2, Bot., Vol. V,

1895-1901, PI. 8, fig. 35); but differ in possessing longer and narrower arms. Length

32-32-5 p; breadth 15-17 p (with arms 43-5-46 p); breadth of isthmus 10-11 p.

(Tab. 31, figs. 1, 2.) Sample 59.

224. St. subtrifurcatwn West forma major W. & G. S. West. These specimens are larger

than in the typical form (W. Schmidle, Engler Bot. Jahrb., XXXII Bd., 1903,

p. 73, Taf. II, fig. 8). Length 72-76 p; breadth 58-62 p; breadth of isthmus

24-26 -5 p; length of spines 20-36 p. (Tab. 30, figs. 16-18.) Samples 24, 25, 39.

Geographical Distribution. — Central Africa; First record for South Africa.

225. St. teliferum Ralfs var. transvaalense var. nova. A typo speciei spinulis hebetatis

differt.

This is near St. teliferum Ralfs (E. Messikommer, Beitrage zur geobotanischen

Landesaufnahme der Schweiz, Heft 24, 1942, Taf. XIV, fig. 8); but differs in the spines

being blunt. Length 30-5-31 p\ breadth 28 p\ breadth of isthmus 11-13 p; length

of spines 2 -5-4 p. (Tab. 31, figs. 19, 20.) Sample 19.

This may also be compared with St. breviaculeatum G. M. Smith (Wm. R. Taylor,

Alpine Algae from the Santa Marta Mountains, Colombia., Am. Journ. of Bot., Vol.

22, 1935, p. 772, PI. 2, fig. 5).

226. St. trihedrale Wolle var. australe var. nova. A typo speciei isthmo multo latiore

differt.

This differs from the typical form (W. & G. S. West, North American Desmidieae,

Trans. Linn. Soc. London, Ser. 2, Bot., Vol. V, 1895-1901, PI. 16, fig. 29) in that the

isthmus is 1 - 5 t i mes broader. Length 44 p ; breadth 30 p ; breadth of isthmus 14-15-5 p.

(Tab. 31, figs. 17, 18.) Sample 27.

221 . St. tumidum Breb. (W. & G. S. West, Monogr, Brit. Desm., Vol. IV, 1912, p. 142.

PI. CXXII, figs. 1-5.) Sample 71.

Geographical Distribution. — England; Wales; Scotland; Ireland; France;

Germany; Galicia in Austria; Norway; Denmark; Sweden; Bornholm; United

States; First record for South Africa.

228. St. xiphidiophcrum Wolle var. westiorum var. nova. E vertice visae cellulae cum

var. brachyacantho W. et G. S. West valde congruunt, sed differt spinis in angulis

apicalibus multo brevioribus.

This closely resembles var. brachyacanthum W. & G. S. West (W. & G. S. West,

North American Desmidieae, Trans. Linn. Soc. London, Ser. 2, Bot., Vol. V, 1895-1901,

PI. 18, fig. 7) in vertical view; but the spines on the apical angles are much shorter

and the cells are smaller. Cells small, about 1 -3 times as long as broad, fairly deeply

constricted, sinus open; basal and apical angles of semicells truncate and furnished

603

with a few spines. Semicells more or less triangular in vertical view; each angle with

three spines (it seems to be 9-angular). Length 24 /x; breadth 18-5-19 /x; breadth of

isthmus 10-10-5 /x. (Tab. 30, figs. 12, 13.) Sample 3.

Named after Messrs. W. & G. S. West.

Sphaerozosma Corda

229. Sph. aubertianum W. West. Cells somewhat smaller than in the type (W. West,

Journ. of Bot., Vol. XXVII, 1889, Tab. 291, fig. 17). Length 10-12 /x; breadth

13-16 /x; breadth of isthmus 5-6 /x. (Tab. 32, fig. 9.) Sample 74.

Geographical Distribution. — Maine; First record for South Africa.

Onychonema G. C. Wallich

230. Onychonema species ad O. laeve Nordst. var. micracanthum Nordst. The cells

correspond with those in Nordstedt’s variety except for the somewhat undulated

apical margin in some specimens; this may be due to the fixative as other specimens

from the fresh material have smooth apical margins. Unfortunately the measure-

ments of Nordstedt’s variety are not available. Length 16-18 breadth 20-24 /x

(with spines 24-30 /x); breadth of isthmus 4-4-5 /x. (Tab. 32, fig. 10.) Sample 74.

TRIBUS: BACILLARIOPHYTA

The different Diatom taxa have been determined by Dr. B. J. Cholnoky and the

descriptions and figures have been published by him. (Beitrage zur Kenntnis der

Sudafrikanischen Diatomeenflora II. Einige Gewasser im Waterberg-Gebiet, Transvaal.

Portugaliae Acta Biologica (Serie B) Vol. 6, No. 2, 1958, pp. 99-160, Taf. I- VII. The

diatoms of the following samples have not been determined as yet;

30, 36, 37, 38, 39, 41, 42, 43, 44, 66, 67, 68, 69, 70, 71, 72, 73, 74.

Class: Diatomeae.

Order: Centrales.

Suborder: Discineae.

Fam.: Coscinodiscaceae.

Subfam.: Melosiroideae.

Melosira Agardh.

231. M. granulata (E.) Ralfs. Samples 47, 48, 49, 50, 57.

232. M. nyassensis O.M. Sample 12.

233. M. roeseana Rabh. Sample 51.

Subfam.: Coscinodiscoideae.

Cyclotella KutQ.

234. C. kiitzingiana Thw. Sample 50.

604

Order: Pennales.

Suborder: Araphidineae.

Fain.: Fragilariaceae.

Subfam.: Fragilarioideae.

Fragilaria Lyngbye

235. F. capucina Desm. var. acuta Grun. Sample 63.

236. F. fonticola Hust. Sample 1.

237. F. ungeriana Grun. Sample 63.

Asterionella Hassa/l

238. A. africana Chy. Sample 49.

Synedra Ehrenberg

239. S. acus Kg. var. radians (Kg.) Hust. Samples 7, 59, 60.

240. 5. rumpens Kg. Samples 1, 4, 5, 7, 8, 10, 11, 12, 60, 63, 64.

241. S. rumpens var. fragilarioides Grun. Samples 9, 11, 18.

242. S. rumpens var. meneghiniana Grun. Samples 5, 7, 8, 10, 11, 12, 23, 47, 63, 64.

243. S. rumpens var. scotica Grun. Sample 50.

244. 5. ulna (Nitzsch) E. Samples 1, 4, 5, 6, 8, 9, 11, 15, 18, 19, 20, 23, 32, 46, 47, 48.

49, 50, 51, 52, 53, 55, 57, 58, 60, 63, 64.

245. S. ulna var. biceps (Kg.) Hust. Samples 1, 5, 6, 7, 8, 12, 13, 47, 48, 50, 60, 63.

246. S. ulna var. danica (Kg.) Grun. Samples 7, 63, 64.

247. S. vauciteriae Kg. Sample 6.

Suborder: Raphidioidineae.

Fam.: Eunotiaceae.

Subfam.: Eunotioideae.

Eunotia Ehrenberg

248. E. alpina (Nag.) Hust. Sample 24.

249. E. exigua (Breb.) Grun. Samples 2, 3, 24, 25, 26, 27, 28, 29, 33, 35, 45, 46. 51,

52, 53, 54, 55, 56, 57, 58, 63, 64.

250. E. garusica Chy. Samples 47, 60.

251. E. lunaris (E.) Grun. Sample 60.

252. E. lunaris var. subarcuata (Nag.) Grun. Sample 51.

253. E. mogolensis Chy. Samples 25, 26, 27, 28, 45, 52.

254. E. montana Hust. Sample 45.

255. E. pectinalis (Kg.) Rabh. Samples 26, 28, 46, 52, 58, 64.

256. E. pectinalis var. minor (Kg.) Rabh. Samples 46, 52, 54, 55, 56, 57, 58.

257. E. porcellus Chy. Sample 25.

258. E. praerupta E. var. inflata Grun. Sample 54.

259. E. pseudofiexuosa Hust. Samples 45, 54, 57.

260. E. subaequalis Hust. Samples 23, 25, 26, 27, 28, 29, 35, 45, 46, 47, 48, 51, 52, 53,

54, 55, 56, 57, 58, 59. 63, 64.

261. E. tenella (Grun.) Hust. Samples 3, 24, 25, 26, 27, 28, 29, 34, 35, 45, 46, 51, 52,

53, 54, 55, 56, 57, 58, 63.

262. E. tschirchiana O.M. Sample 7.

263. E. zygodon E. Samples 25, 26, 27, 28, 45.

264. E. zygodon var. elongata Hust. Samples 26, 27, 28, 29.

605

Suborder: Monoraphidineae.

Fam.: Achnanthaceae.

Suhfam.: Achnanthoideae.

Achnanthes Bory

265. A. exigua Grun. Samples 6, 63.

266. A. lanceolata (Breb.) Grun. var. rostrata Must. Sample 51.

267. A. linearis W. Sm. Samples 13, 14, 15, 32.

268. A. microcephala Kg. Samples 1, 4, 5, 6, 7, 8, 9, 10, 11. 12, 15, 20, 22, 23, 29,

32, 33, 45, 47, 49, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 63, 64.

269. A. minulissima Kg. Samples 1, 2, 4, 5. 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 18, 19,

20, 22, 23, 29, 32, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 63, 64.

Suborder: Biraphidineae.

Fam.: Naviculaceae.

Subfam.: Naviculoideae.

Frustulia Agardh.

270. F. rhomboides (E.) de Toni. Samples 23, 45, 46, 51, 52, 53, 54, 55, 57, 58.

271. F. rhomboides var. saxonica (Rabh.) de Toni. Samples 12, 14, 20, 29, 32, 45, 47,

50, 51, 52, 53, 54, 55, 56, 57, 58, 63, 64.

272. F. vulgaris (Thw.) de Toni var. angusta Chy. Samples 1, 7, 9, 11, 12, 13, 14, 15,

47, 50, 53, 54, 55, 57, 60, 63, 64.

Gyrosigma Hassall

273. G. kiitzingii (Grun.) Cl. Samples 1, 7, 10, 12, 13, 15, 18, 20, 23, 47, 48, 49, 50, 62.

274. G. spenceri (W. Sm.) Cl. var. nodifera Grun. Sample 6.

Caloneis Cleve

275. C. aequatorialis Must. var. tugelae Chy. Samples 14, 15, 62.

276. C. bacillum (Grun.) Cl. Samples 13, 15, 20, 63.

277. C. cliasei Chy. Sample 22.

278. C. incognita Must. Samples 14, 15, 16, 47.

279. C. silicula (E.) Cl. Samples 12, 14, 64.

280. C. silicula var. peisonis Hust. Samples 47, 50.

281. C. silicula var. truncatula Grun. Samples 14, 47, 48, 49, 50, 53, 56.

Neidiura Pfitzer

282. N. affine (E.) Cl. Samples 6, 7, 19, 25, 26, 33, 46, 54, 60.

283. N. affine var. amphirhynchus (E.) Cl. Samples 28, 46, 48, 52, 53, 54, 55, 56, 59.

284. N. affine var. longiceps (Greg.) Cl. Samples 12, 52, 53, 56, 57.

285. TV. gracile Hust. Samples 46, 52.

286. TV. gracile forma aequale Hust. Samples 29, 45, 46.

287. TV. iridis (E.) Cl. Samples 26, 33, 47, 48, 49, 53, 54, 55, 57, 59, 63, 64.

288. TV. iridis var. amphigomphus (E.) van Heurck. Samples 25, 26, 28.

289. TV. iridis var. ampliata (E.) Cl. Sample 56.

290. TV. javanicum Hust. Sample 58.

291. N. productum (W. Sm.) Cl. Samples 14, 15, 26, 46, 54, 56, 58, 63, 64.

606

Diploneis Ehrenberg

292. D. ovalis (Hilse) Cl. Sample 59.

293. D. subovalis Cl. Sample 52.

Stauroneis Ehrenberg

294. S. anceps E. Samples 6, 19, 20, 23, 47, 50, 51, 53, 54, 56, 63.

295. S. anceps forma gracilis (E.) Cl. Samples 46, 52, 53, 54, 55, 56, 58, 64.

296. S. borrichi (Petersen) Lund. Samples 53, 56.

297. S. claasseniae Chy. Sample 59.

298. S. crucicula Grun. Sample 1.

299. 5. phoenicenteron E. Samples 24, 25, 26, 28, 29, 33, 46, 48, 51, 52, 54, 55, 56, 57,

59, 60.

300. S. phoenicenteron forma gracilis (Dippel) Hust. Samples 45, 56.

Anomoeoneis Pfitzer

301. A. exilis (Kg.) Cl. Samples 1, 5, 7, 8, 9, 10, 11, 12, 14, 20, 23, 63.

302. A. sphaerophora (Kg.) Pfitzer. Samples 59, 60.

Navicula Bory

303. N. acidophila Chy. Sample 24.

304. N. anglica Ralfs. Samples 12, 13, 14, 16, 20, 22, 47, 48, 49, 50.

305. N. bryophila Petersen. Samples 5, 6, 9, 10, 11, 25, 26, 27, 45.

306. N. cari E. Samples 13, 17, 46, 59, 61, 62.

307. N. cari var. angusta Grun. Samples 20, 21, 22, 26, 46, 51, 52, 53, 54, 55, 56, 57,

58, 63, 64.

308. N. cincta (E.) Kg. Samples 14, 15, 20, 46, 49, 61, 62.

309. N. cinctaefonnis Elust. Sample 61.

310. N. confervacea Kg. Samples 47, 48.

311. N. cryptocephala Kg. Samples 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17,

18, 19, 24, 25, 26, 28, 29, 32, 33, 34, 35, 45, 46, 47, 48, 49. 50, 51, 52, 53, 54, 55, 56,

57, 58, 59, 60, 61, 62, 63, 64.

312. N. cryptocephala var. intermedia Grun. Samples 6, 63.

313. N. cryptocephala var. veneta (Kg.) Grun. Sample 57.

314. N. cuspidata Kg. Samples 48, 49, 53, 54, 58, 59, 63.

315. N. cuspidata var. ambigua (E.) Cl. Samples 15, 20, 21, 22, 46, 49, 51, 55, 56, 57,

59, 60, 63, 64.

316. N. dicephala (E.) W. Sm. Samples 15, 48.

317. N. dicephala var. neglecta (Krasske) Hust. Sample 50.

318. N. elaborata Hust. Sample 35.

319. N. exigua (Greg.) O.M. Samples 1, 13, 20, 47, 48, 50.

320. N. exiguiformis Hust. Samples 51, 63.

321. N. gastrum E. Samples 51, 52, 53, 54, 55, 57, 58.

322. N. graciloides A. Mayer. Sample 7.

323. N. gregaria Donk. Samples 2, 32, 47, 50, 51, 59.

324. N. gregarioides Chy. Samples 21, 23, 26, 46, 51, 52, 53, 54, 55, 56, 57, 58, 60, 63.

325. N. grinmiei Krasske. Samples 1, 4, 13, 35, 47, 48, 59, 63, 64.

326. N. hungarica Grun. var. capitata (E.) Cl. Samples 1, 20, 48.

327. N. krasskei Hust. Samples 25, 29.

328. N. lanceolata (Ag.) Kg. Samples 15, 47, 48, 49, 50, 59, 60.

329. N. longicephala Hust. Samples 55, 56.

330. N. mengeae Chy. Sample 57.

607

331. N. menisculus Schum. Samples 6, 11, 13.

332. N. niicrocephala Grun. Sample 50.

333. N. minima Grun. var. atomoides (Grun.) Cl. Sample 46.

334. N. minuscula Grun. Sample 47.

335. TV. minusculoides Hust. Samples 60, 61.

336. TV. muralis Grun. Samples 2, 24, 49.

337. TV. mutica Kg. Samples 7, 8, 25, 26, 27, 29, 31, 35, 51, 56, 57, 63.

338. TV. mutica forma cohnii (Hilse) Hust. Samples 3, 4, 8, 19, 20, 27, 29, 34.

339. TV. mutica var. nivalis (E.) Hust. Samples 1, 7, 10, 19, 20, 21, 33, 35, 51, 53, 55, 56.

340. TV. mutica var. pseudolagerheimii Chy. Samples 11, 28, 45, 48, 55, 58, 63.

341. TV. nyassensis O.M. Samples 12, 13, 54, 56, 57, 58, 63.

342. TV. nyassensis var. capitata O.M. Sample 63.

343. TV. nyassensis var. minor Chy. Samples 13, 14, 15, 16, 20, 23, 52, 59, 61, 62.

344. TV. perpusiUa Grun. Samples 25, 27, 28.

345. TV. pupula Kg. Samples 46, 54, 56, 60, 61, 62, 63, 64.

346. TV. pupula var. capitata Hust. Samples 47, 48, 49, 50, 55, 64.

347. N. pupula var. rectangularis (Greg.) Grun. Sample 63.

348. TV. pygmaea Kg. Samples 59, 60.

349. TV. radiosa Kg. Samples 7, 8, 11, 12, 46, 47, 48, 52, 53, 54, 55, 57, 58, 63, 64.

350. TV. radiosa var. tenella (Breb.) Grun. Samples 9, 11, 25, 32, 33, 35, 45, 47, 48,

49, 50, 52, 54, 55, 57, 58, 59, 60, 63, 64.

351. TV. rhynchocephala Kg. Samples 7, 13, 14, 15, 48, 60.

352. TV. ruttneri Hust. Sample 33.

353. TV. schroeteri Meister. Samples 1, 7, 9, 10, 12, 14, 47, 48, 50.

354. TV. seminuloides Hust. Sample 7.

355. TV. seminuloides var. sumatrana Hust. Samples 29, 45.

356. TV. seminulum Grun. Samples 29, 32, 33, 49, 59, 60.

357. TV. simplex Krasske var. minor Chy. Samples 20, 22, 23, 52, 53, 54, 55, 56, 58.

358. TV. subtilissima Cl. Samples 3, 52, 56.

359. TV. suecorum Carlson. Samples 19, 64.

360. TV. tenelloides Hust. Samples 19, 21, 29, 31, 32, 33, 35, 45, 46, 62.

361. TV. terrestris Petersen. Samples 63, 64.

362. TV. tridentula Krasske. Sample 60.

363. TV. viridula Kg. Samples 1, 60.

364. TV. viridula var. linearis Hust. Sample 1.

365. TV. viridula var. rostellata (Kg.) Cl. Samples 5, 6, 7, 9, 10, 11, 12, 13, 14, 15, 16,

18, 19, 20, 23, 28, 47, 48, 49, 50, 51, 60, 62, 63, 64.

366. TV. vulpina Kg. Sample 59.

367. TV. zanoni Hust. Samples 1, 5, 7, 8, 9, 10, 11, 12, 13, 47, 50.

Pinnularia Ehrenberg

368. P. acoricola Hust. Samples 25, 29, 45.

369. P. acrosphaeria Breb. Samples 5, 14, 15, 59.

370. P. acrosphaeria var. controversa Chy. Sample 59.

371. P. amaniensis Hust. Samples 25, 26, 28.

372. P. borealis E. Sample 23.

373. P. braunii (Grun.) Cl. Samples 54, 58.

374. P. braunii var. amphicephala (A. Mayer) Hust. Sample 45.

375. P. claasseniae Chy. Sample 35.

376. P. dactylus E. Sample 45.

377. P. divergens W. Sm. Samples 14, 15, 20, 47, 50, 54.

378. P. divergens var. undulala Herib. & Perag. Sample 13.

379. P. dubitabilis Hust. Samples 2, 3, 12, 20, 24, 25, 27, 28, 29, 33, 35, 45, 47, 48, 53,

55, 57, 60, 63, 64.

608

380. P. gibba E. Samples 6, 7, 11, 14, 15, 20, 25, 26, 27, 28, 29, 33, 45, 46, 47, 48, 49,

50, 51, 52, 54, 56, 57, 58, 59, 61, 63, 64.

381. P. gibba forma subundulata Mayer. Samples 16, 31, 33, 47, 50, 51, 52, 53, 54, 55,

56, 57, 58, 59, 63, 64.

382. P. gibba var. parva (E.) Grun. Samples 4, 7, 9, 11, 13, 14, 15, 19, 25, 26, 27, 28,

29, 31, 33, 35, 45, 51, 63, 64.

383. P. gibba var. sancta Grun. Samples 25, 28, 33, 50, 51, 54, 55, 56, 59, 63.

384. P. graciloides Must. Samples 4, 23.

385. P. interrupta W. Sm. Samples 23, 25, 26, 27, 28, 29, 32, 33, 63.

386. P. interrupta \ar. minutissima Hust. Samples 7, 24.

387. P. maior (Kg.) Cl. Samples 13, 14, 15, 58.

388. P. niesolepta (E.) W. Sm. Sample 19.

389. P. microstauron (E.) Cl. Sample 64.

390. P. microstauron forma biundulata O.M. Sample 48.

391. P. microstauron var. brebissonii (Kg.) Hust. Sample 19.

392. P. ruttneri Hust. Samples 25, 28.

393. P. subcapitata Greg. Samples 2, 3, 9, 12, 19, 20, 22, 23, 25, 27, 28, 29, 31, 32. 33,

34, 35, 45, 50, 51, 52, 53, 54, 55, 60, 64.

394. P. subcapitata var. hilseana (Janisch) O.M. Samples 2, 4, 7, 10, 11, 12, 25, 28. 31,

32, 33, 34, 35, 45, 46, 48, 50, 51, 52, 53, 54, 55, 56, 57.

395. P. viridis (Nitzsch) E. Samples 4, 12, 13, 14, 15, 16, 20, 27, 28, 33, 47, 48, 49, 50.

52, 53, 54, 56, 57, 58, 59, 63, 64.

396. P. viridis var. sudetica (Hilse) Hust. Sample 33.

Subfam.: Amphiproroideae.

Amphora Ehrenberg

397. A. ovalis Kg. Samples 12, 14, 59.

398. A. ovalis var. libyca (E.) Cl. Samples 47, 48, 49, 50.

399. A. submontana Hust. Samples 11, 13, 14, 15, 16, 23, 46, 47, 48, 49, 50, 62.

Cymbella Agardh.

400. C. amphicephala Nag. Sample 29.

401. C. amphicephala var. hercynica (N.S.) Cl. Sample 26.

402. C. bengalensis Grun. Sample 63.

403. C. cistula (Hempr.) Grun. Samples 5, 6, 7, 8, 9, 14.

404. C. cistula var. africat.j Chy. Samples 11, 12.

405. C. claasseniae Chy. Samples 25, 26, 28, 51, 54, 56, 58.

406. C. gracilis (Rabh.) Cl. Samples 7, 14, 15, 52.

407. C. javanica Hust. Sample 32.

408. C. kappii Chy. Samples 1, 5, 7, 8, 9, 10, 11, 12.

409. C. kolbeilivi^i.. Samples 1, 5, 6, 7, 8, 9, 10, 1 1, 12, 13, 14, 15, 18, 20, 47, 50, 63, 64.

410. C. microcephala Grun. Samples 1, 2, 3, 5, 7, 8, 9, 10, 11, 12, 13, 20, 23, 45, 51.

52, 53, 54, 55, 56, 57, 58, 63, 64.

411. C. miilleri (O.M.) Hust. Sample 7.

412. C. naviculiformis Auersw. Samples 13, 14, 18, 51, 52, 53, 54, 55, 56, 57, 58, 63, 64.

413. C. nylstroomensis Chy. Sample 63.

414. C. pretoriensis Chy. Samples 21, 24, 25, 26, 27, 28, 29.

415. C. pseudoincerta Chy. Sample 29.

416. C. raytonensis Chy. Samples 22, 25, 26, 27, 28, 29. 52.

417. C. raytonensis var. debegenica Chy. Sample 21.

418. C. spicula Hust. Sample 64.

419. C. spuria Cl. Samples 8, 46.

609

420. C. turgida (Greg.) Cl. Samples 1, 2, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13. 14, 15, 16, 18,

20, 23, 34, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 63, 64.

421. C. turgida forma minor Chy. Samples 7, 22, 23, 50, 52, 53, 55, 57, 58, 59, 60, 63.

422. C. turgida var. pseudogracilis Chy. Sample 63.

423. C. ventricosa Kg. Samples 1, 5, 7, 8, 9, 10, 11, 12, 13, 18, 19, 20, 22, 23, 29, 32,

33, 34, 35, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64.

Gomphonema Agardh.

424. G. acuminatum E. var. turris (E.) Cl. Sample 7.

425. G. augur E. Sample 52.

426. G. clevei Fricke. Samples 9, 50, 63, 64.

427. G. constrictum E. Samples 57, 60, 63, 64.

428. G. gracile E. Samples 1, 4, 6, 7, 8, 9, 11, 12, 13, 14, 15, 18, 23, 24, 25, 26, 28, 29,

32, 46, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 63, 64.

429. G. gracile var. lanceolatum (Kg.) Cl. Samples 7, 9, 12, 51, 59, 60, 63, 64.

430. G. lanceolatum E. Samples 4, 5, 6, 9, 10. 11.

431. G.parvulum (Kg.) Grun. Samples 2. 4, 5, 6, 7, 8, 9, 10, 11, 13, 14, 15, 16, 20, 23,

26, 29, 32, 33, 35, 45, 46, 47, 48, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 63, 64.

432. G. parvulum var. lagenula (Grun.) Hust. Samples 1, 2, 4, 5, 9, 10, 11, 12, 13, 14.

18, 19, 20, 21, 22, 23, 24, 25, 28, 29, 31, 32, 33, 34, 46, 47, 48, 49, 50, 51, 52, 53, 54,

55, 56, 57, 58, 59, 60, 61, 63, 64.

433. G. subtile E. Samples 7, 26, 28, 33.

Fam.: Epithemiaceae.

Subfam.: Epithemioideae.

Epithemia Brebisson

434. E. zebra (E.) Kg. Sample 7.

Subfam.: Rhopalodioideae.

Rhopalodia O. M idler

435. R. gibba (E.) O.M. Samples 4, 6. 9, 1 1, 12, 13, 14, 23, 47, 48, 49, 50, 62.

Fam.: Nitzschiaceae.

Subfam.: Nitzschioideae.

Hantzschia Grunow

436. H. amphioxys (E.) Grun. Samples 11, 12, 17, 20, 23, 33, 56, 59, 60, 62, 64.

437. H. amphioxys var. africana Hust. forma minuta Chy. Samples 1, 3, 4, 21, 25, 26,

27, 28, 29, 31, 34, 35, 48, 49, 51, 52, 55, 57, 58, 63.

438. H. amphioxys var. intermedia Grun. Samples 20, 57, 58, 59, 63.

439. H. amphioxys var. vivax (Hantzsch) Grun. Samples 20, 26, 27, 29, 47, 48, 49,

53, 55, 56, 57, 59.

Nitzschia Hassal

440. TV. accommodata Hust. Samples 1, 3, 4, 5, 6, 7, 8, 9, 10, 12, 13, 14, 15, 16, 17, 20,

22, 23, 25, 26, 27, 28, 32, 33, 34, 46, 47, 48, 49, 51, 53, 54, 58, 59, 60, 61, 62, 63.

441. N. amphibia Grun. Samples 48, 50.

442. N. capitellata Hust. Samples 1, 33, 46, 47, 48, 51, 52, 53, 54, 55, 56, 57, 58, 62, 64.

610

443. N. claasseniae Chy. Samples 1, 5, 7, 12, 20, 23, 57, 58, 59, 60.

444. N. clausii Hantzsch. Samples 1, 13, 14, 15, 18, 47, 50, 63.

445. N. confinis Hust. Samples 7, 13, 48.

446. N. debilis (Arnott) Grun. Sample 63.

447. N. denticula Grun. Sample 22.

448. N. diserta Hust. Sample 5.

449. N. dissipata (Kg.) Grun. Samples 1, 16.

450. N.fonticola Grun. Samples 9, 13, 17, 19, 21, 23, 28, 29, 31, 32, 33, 34, 35, 45, 46, 49.

451. N. frustulum (Kg.) Grun. var. perpusilla (Rabh.) Grun. Samples 5, 12, 18, 19,

20, 22, 60.

452. N. goetzeana O.M. Samples 13, 14, 59.

453. N. kUtzingiana Hilse. Samples 46, 47, 49, 50, 51. 52, 53, 54, 55, 56, 57, 58, 59,

60, 61, 62.

454. N. linearis (Ag.) W. Sm. Samples 1, 5, 7, 9, 11, 13, 14, 15, 16, 47, 48, 49, 50, 51,

59, 60, 62, 64.

455. N. palea (Kg.) W. Sm. Samples 2, 5, 11, 12, 24.

456. N. palea var. tropica Host. Sample 49.

457. N. purvuloides Chy. Samples 21, 26, 32, 45. 53.

458. N. perminuta Grun. Samples 46, 53, 55, 56, 58, 59, 60. 61, 62.

459. N. pseudobacata Chy. Sample 47.

460. N. recta Hantzsch. Sample 1.

461. N. romana Grun. Sample 59.

462. N. sigma (Kg.) W. Sm. Samples 1, 20, 23, 47, 48, 50.

463. N. subadapta Chy. Sample 12.

464. TV. subcommunis Hust. Sample 50.

465. TV. subvitrea Hust. Samples 14, 47, 49.

466. N. tarda Hust. Samples 8, 20, 23, 47, 49, 55, 56, 59, 60.

467. TV. transvaalensis Chy. Sample 59.

468. N. tropica Hust. Samples 1, 4, 19, 20, 22, 23.

469. TV. tryblionella Hantzsch var. levidensis (W. Sm.) Grun. Samples 1, 7, 59.

470. TV. von-hauseniae Chy. Samples 20, 25, 26, 32, 33, 63, 64.

Fam.: Surirellaceae.

Subfam.: Surirelloideae.

Cymatopleura W. Smith

471. C. solea (Breb.) W. Sm. Samples 12, 14.

Stenopterobia Brebisson

All. S. intermedia (Lewis) Hust. Samples 25, 26, 27, 28, 46.

473. S. rautenbachiae Chy. Sample 64.

Surirella Turpin

474. 5. angusta Kg. Samples 1, 7, 13, 14, 15, 20, 47, 48, 50, 59, 60, 61, 62, 63, 64.

475. S. delicatissima Lewis. Samples 22, 24, 26, 27, 28, 32, 34, 35, 45, 51, 52, 53, 54,

55, 56, 57, 58, 63, 64.

476. S. linearis W. Sm. Samples 1, 46, 56, 58, 64.

All. S. robusta E. var. splendida (E.) van Heurck. Sample 48.

478. S. tenera Greg. Samples 1, 7, 13, 14, 20, 23, 29, 45, 47, 48, 49, 50, 51, 52, 53, 54,

55, 56, 57, 58.

479. S. tenera forma minor Chy. Sample 64.

611

TRIBUS: EUTHALLOPHYTA

Class: Chlorophyceae.

Order: Protococcales.

Earn.: Hydrodictyaceae.

Pediastrum Meyen.

480. P. boryanum (Turpin) Menegh. var. granulatwn (KUtz.) A1 Braun. (Pascher

Susswasser-Flora, Heft 5, 1915, p. 101, fig. 61e.) This agrees in shape and size

with var. granulatwn A1 Braun, but no granules were perciptible. (Tab. 33, fig. 4.)

Sample 20.

481. P. boryanum var. longicorne Reinsch. (Pascher Susswasser-Flora, Heft 5, 1915,

p. 101, fig. 61d.) Sample 20.

482. P. boryanum var. minor var. nova, P. boryano var. forcipato Racib. affinis, sed

membrana cellulae neque granulata neque punctata facile distinguenda.

Near var. forcipatum Racib. (Pascher Siisswasser-Flora, Heft 5, 1915, p. 101,

fig. 61f); but the membrane is not granular or punctate. Colonies with 16-31 cells,

margin-cells 6-8 ,u, in diameter; cells 8-11 p long and 5-6 p broad. (Tab. 33, fig. 5.)

Samples 18, 20.

483. P. duplex Meyen var. reticulatum Lagerheim. (Pascher Susswasser-Flora, Heft 5,

1915, p. 95, fig. 57h.) Sample 47.

484. P. duplex var. reticulatum forma cohaerens Bohlin. (Pascher Siisswasser-Flora,

Heft 5, 1915, p. 95, fig. 57f.) (Tab. 33, figs. 1-3.) Sample 74.

485. P. tetras (Ehrenb.) Ralfs forma evoluta West. Samples 47, 48.

486. P. tetras var. excisum Rabenh. (Pascher Siisswasser-Flora, Heft 5, 1915, p. 104,

fig. 64b.) (Tab. 33, fig. 6.)

Earn.: Coelastraceae.

Scenedesmus Meyen

487. Sc. bijugatus (Turpin) KLitz. (Pascher Susswasser-Flora, Heft 5, 1915, p. 167.)

Samples 19, 23, 48.

488. Sc. quadricauda (Turpin) Breb. (Pascher Siisswasser-Flora, Heft 5, 1915, p. 165,

fig. 223.) Samples 12, 20, 23, 35, 47, 48.

Order-. Ulotrichales.

Faw. Oedogoniaceae.

Oedogonium Link.

489. Oe. boyanum spec, nova, Oe. crasso Wittr. sec. Him affinis, sed plantae bisexuales,

antheridia 3-16-cellulata, oogonia et oosporae latiores sunt.

This species is near Oe. crassum Wittr. sec. Him. (Pascher Siisswasserflora, Heft 6,

1914, p. 200, fig. 286); but the plants are monoecious, the antheridia have 3-16 cells

and the oogonia and oospores are broader. Oogonia solitary, elliptical, with a pore

in the upper part, 79-82 p broad and 112-130 p long. Oospores elliptical, not filling

the oogonium, membrane smooth and thick, 74 p broad and 91 p long. Vegetative

cells 52-54 p broad and 96-192 p long. Antheridium cells 44-48 p broad and 6-12 p

long. (Tab. 34, figs. 9, 10.) Sample 73.

612

490. Oe. mattiei spec, nova, Oe. zig-zag Cleve sec. Him affinis, sed antheridiis 1-3-

cellulatis, cellulis sterilibus latioribus, oogoniis brevioribus, zygotis globosis et

oogonium baud complentibus, cellulis antheridii latioribus brevioribusque valde

differt.

Near Oe. zig-zag Cleve sec. Him (Pascher Susswasserflora, Heft 6, 1914, p. 209,

fig. 304) ; but differs in that the oospore does not fill the oogonium, and in having broader

vegetative cells, shorter oogonia, spherical oospores, and shorter, broader, 1 -celled to

3-celled antheridia. Monoecious; oogonium solitary, more or less spherical, with a

pore in the upper part, 44-50 /x broad and 43-52 /ix long. Oospore membrane smooth

and thick, 42 jtx in diameter. Vegetative cells 28-30 jj. broad and 40-72 /x long. Antheri-

dium cells 26-28 broad and 6-8 ^ long. (Tab. 34, fig. 8.) Sample 73.

491. Oe. mesianum spec, nova, Oe. varians Wittr. et Lund. sec. Him affinis, sed ternis

vel quaternis antheridiorum cellulis, sterilibus cellulis paulo latioribus et zygotis

multo minoribus facile distinguenda.

This species closely resembles Oe. varians Wittr. & Lund. sec. Him (Pascher

Susswasserflora, Heft 6, 1914, p. 206, fig. 298); but the antheridia consist of 3-4 cells,

the vegetative cells are slightly broader and the oospores are considerably smaller.

Monoecious; oogonia solitary, subovate, with a pore in the upper part, oogonia

36 /LX broad and 38 /x long. Oospores spherical, not filling the oogonia, membrane

smooth and fairly thick, 27 jx in diameter. Vegetative cells 16-20 /x broad and 44-80 /x

long. Antheridium cells 15 /x broad and 3-8 jx long. (Tab. 34, fig. 5.) Sample 20.

Named after the late Prof. Dr. M. G. Mes, formerly Head of the Department of

Plant Physiology and Biochemistry, University of Pretoria.

492. Oe. pseudofragile spec, nova, ex affinitate Oe. fragile Wittr. sec. Him. Differt

tamen eo quod antheridiis ternae, quaternae vel quinae sunt cellulae, oogonia et

oosporae minora sunt et antheridii cellulae breviores.

Near Oe. fragile Wittr. sec. Him (Pascher Susswasserflora, Heft 6, 1914, p. 207,

fig. 302); but the antheridia consist of 3-5 cells, the oogonia and oospores are smaller

and the antheridium cells are shorter. Monoecious; oogonia solitary, more or less

spherical, with a pore in the upper part, 36-38 /x broad and 32-36 p. Oospores not

filling the oogonia, more or less spherical, membrane smooth and fairly thick, 32 p

broad and 30-32 p long. Vegetative cells 12-20 p broad and 44-80 p long. Antheridium

cells 12-16 p broad and 4-6 p long. (Tab. 34, figs. 1-4.) Sample 20.

493. Oe. subintermedium spec, nova, affinis Oe. intermedia Wittr. sec. Him, sed differt

quod oogonium in parte superiore angustatum, cellulae steriles paulo latiores,

cellulae antheridii paulo et latiores et longiores sunt quam in Oe. intermedia.

This is near Oe. intermedium Wittr. sec. Him (Pascher Susswasserflora, Heft 6,

1914, p. 207, fig. 301); but the upper part of the oogonium is attenuated, the vegetative

and antheridium cells are slightly broader and the antheridium cells are slightly longer.

Monoecious; oogonia solitary, more or less spherical, with a pore in the upper part,

34 p broad and 33 p long. Oospore more or less spherical, not filling the oogonium,

membrane smooth, 30 p broad and 28 p long. Vegetative cells 20 p broad and 60-80 p

long. Antheridia 2-celled, 19 p broad and 10-12 p long. (Tab. 34, fig. 6.) Sample 73.

494. Oe. subvaucherii spec, nova, Oe. vaucheria A. Braun sec. Him affinis, sed oogoniis

globosis differt.

This differs from Oe. vaucherii A. Braun sec. Him (Pascher SUsswasserflora, Heft 6,

1914, p. 207, fig. 303) in the oogonia being spherical. Monoecious; oogonium solitary,

spherical, with a pore in the upper part, 46 p in diameter. Oospore spherical, not

filling the oogonium, membrane smooth and thick, 40 p in diameter. Vegetative cells

22 p broad and 60-80 p long. Antheridia 2-4-celled, 19 p broad and 6-8 p long. (Tab.

34, fig. 7.) Sample 73.

613

III.— VEGETATIVE REPRODUCTION IN DESMIDS

In Desmids vegetative reproduction takes place by cell division. In the genera

possessing a median constriction the two semicells move apart during the elongation

of the isthmus which is accompanied or followed by nuclear division. Normally the

next step is the formation of a cylindrical girdle band within the wall in the median

region of the elongated isthmus. In exceptional cases the girdle band is not formed

during every cell division; in some species the girdle band is consistently absent. From

this girdle band a plate-like projection grows inwards and eventually becomes the

cross-partition. Simultaneously the elongated isthmus-halves begin to swell and

develop into new semicells. During this process of swelling the chloroplasts of the

parent semicells elongate, divide and take up their position in the newly formed semicells.

Finally, the cross-partition splits and the newly formed individuals move apart.

Cell division does not always take place in the normal manner in vitro: it was

observed that an individual in which one of the semicells had not reached full maturity

was capable of dividing again, so that the daughter individuals differed markedly in

shape and size from the normal type. (See Tab. 13, fig. 3; Tab. 21, figs. 1,4; Tab. 22,

figs. 5, 6; Tab. 23, fig. 1.)

During August, 1955, cell division was studied in two species of Micrasterias.

It was observed that the chloroplasts of individuals about to divide were a deeper green

colour than those of resting individuals.

The rate and manner of cell division in Micrasterias crux-melitensis (Ehrenb.)

Hass, are illustrated in Tab. 35, figs. A-P. Similar observations on Micrasterias denti-

culata Breb. are presented in Tab. 36, figs. A-G and Tab. 37, figs. H-K.

Cell division usually begins during the afternoon. The first divisional stages are

reached fairly rapidly; but it may take up to several days before the new semicells

reach their mature size, normal shape and effiguration.

Not infrequently however, patterns deviating from the normal are met with. For

instance, the cross septum may not be completed, and the swelling isthmus develops

curiously shaped monstrosities between the two adjacent parent semicells.

Rosenberg (1940) studied the effect of various sets of conditions on zygospore

formation in desmids. He worked mainly with Micrasterias americana (Ehrenb.)

Raffs. He found that a low rate of division was obtained by growing this species in

Benecke’s nutrient solution (250 p.p.m.) in the dark at a low temperature of 8°C. After

10 days, he examined the cultures and found that about 10 per cent of the cells were

abnormally large and contained an additional section between the two halves which

normally join at the isthmus. When these giant cells were isolated and placed under

conditions favouring rapid division, each cell divided at both isthmuses simultaneously,

giving rise to one giant cell and two normal cells. The results of several series of

experiments showed that giant cell formation takes place when conditions are unfavour-

able for cell division, for instance during the winter.

In the material collected in the districts of Warmbaths, Nylstroom and Naboom-

spruit giant cells were observed in: Euastrum insulare (Wittr.) Roy (Tab. 9, fig. 12);

Micrasterias crux-melitensis (Ehrenb.) Hass. (Tab. 21, fig. 4; Tab. 22, figs. 1-3);

Cosmarium sexnotatum Gutw. var. simplex Claassen (Tab. 24, fig. 9); C. waterbergense

Claassen (Tab. 26, figs. 4-7) and Staurastrum dilatatum Ehrenb. (Tab. 30, fig. 6). The

material of M. crux-melitensis was collected during August, 1955 and that of the other

four species during April, 1954.

As only mature giant cells were observed, and not their actual formation, it was

not possible to determine whether those in M. crux-melitensis were the result of a very

low division rate or not. It should be mentioned that numerous other variations in

this species were observed to arise during very rapid division. The material containing

giant cells of the other four species was fixed at the time of collecting.

614

Numerous variations were observed among the specimens of M. crux-melitensis

(Ehrenb.) Hass. The normal type is illustrated in Tab. 18, figs. 1, 2 and the following

figures depict the majority of the variations: Tab. 18, figs. 3-14; Tab. 19, figs. 1-11;

Tab. 20, figs. 1-6; Tab. 21, figs. 1-4; Tab. 22, figs. 1-6; Tab. 23, fig. 1.

G. S. West (1899) in discussing variation in the Desmidieae, states:

“ The ordinary method of increase in this family (Desmidiaceae) of unicellular

plants is by division of the mother-cell into two exactly similar daughter-cells,

each constituted of one of the half-cells of the mother and a newly developed half.

The latter is sometimes markedly different from the parent half, but in cases where

this difference is extreme the new halves of the next division generally conform to

the original type, though more rarely this may not occur for several generations.

The great diversity of form and wonderfully varied character of these plants

are to be associated with their confinement to small ponds or the quiet margins of

lakes, &c., localities suitable for their existence in large numbers. In these restricted

areas the unceasing effect of the struggle for existence will result in a gradually

increasing diversity of form, and this is to be correlated with the immense numbers

of individuals that are sometimes found in these situations.

As regards the conditions favourable to variation among the members of this

group of plants, it may be said that the occurrence in large quantity of a particular

species is most conducive to the production of deviations from the normal form.

It may happen that in some localised spot an immense quantity of some particular

species is occasionally produced by very rapid division, and in such a case some

variation is always met with”.

The foregoing facts were found to be applicable to fresh material of Micrasterias

crux-melitensis (Ehrenb.) Hass, kept in glass containers and studied from \2jll55

onwards. It was observed in numerous individuals that where there were differences

between the new semicells and the parent semicells, the semicells formed during the

next division reverted to the original type (Tab. 18, fig. 7; Tab. 19, fig. 7; Tab. 21,

figs. 1, 3).

West (1899) also states:

“ The most numerous variations are to be found amongst the commoner and

more widely distributed species. The majority of these variations appear to affect

only the superficial characters — the warts, spines, striolations, granules, scrobicu-

lations, &c., on the external or internal surfaces of the cell-wall. Some of them,

however, are more important modifications, being changes in the external form

of symmetry of the plant; and yet others are variations in the structure and arrange-

ment of the cell-contents.

That variation in a species has a tendency to be reproduced there can be no

doubt, especially when the reproduction takes place, as in these plants, by simple

cell-division; yet extreme modifications, which are obvious abnormalities, are

never (or very rarely) repeated in succeeding generations and may be regarded

merely as accidental occurrences in the history of species.

The five statements which follow represent the result of the direct observations

on variation in this group of plants, and may, owing to our insufficient knowledge

of the question, be subject to further alterations.

1 . The structure of the cell-contents is one of the most constant features exhibited

by a species; but this fact can be of little classificatory value owing to the

very large number of species which possess the same structure and arrangement

of the chromatophores.

615

2. The outward form of the cell, as seen in front view, varies within certain

limits, which are usually very small, but which may in exceptional cases be

considerable. The form of the vertical view is, as a rule, a more constant

feature than the form of the front view.

3. The ornamentation (scrobiculations, granulations, spinulations, &c.) of the

cell-wall is relatively constant, being always arranged according to a definite

law, which is only transgressed by variations in one or more of the individual

eomponent groups which constitute the pattern of arrangement.

4. The prolific growth and rapid division of immense numbers of Desmids

have a tendency to produce variations from the typical forms.

5. Changes in the conditions of environment cannot affect the characters of a

species unless they act for long periods of time”.

Abnormal cell division was also observed in two species of Cosmarium viz. Tab. 28,

fig. 3 and Tab. 29, fig. 4.

IV.— SUMMARY

The freshwater algal flora of the districts of Warmbaths, Nylstroom and Naboom-

spruit in South Africa has been investigated. A total of 494 species, varieties and forms

have been identified. These include 17 Cyanophyceae with seven genera, two Flagel-

latae with two genera, four Mesotaeniaceae with two genera, 207 Desmidiaceae with

12 genera, 249 Diatomeae with 26 genera and 15 Chlorophyceae with three genera.

Thus in the material studied, the Desmidiaceae and Diatomeae are best represented.

Several of the taxa in the present communication are new records for South Africa.

The diatom taxa have been studied in detail by Dr. B. J. Cholnoky and are only

mentioned in the taxonomic part for the sake of completeness as regards the survey

of the freshwater algal flora of the region under consideration. Of the 245 other taxa,

157 are described as new species, varieties or forms, as indicated in Table I.

Table I. — The Numbers of Species, Subspecies, Varieties and Forms belonging to the

Different Groups

From the investigation of this relatively small area in South Africa, it is clear that —

(1) new taxa are abundant in this area;

(2) additional taxa from this area await description since several of these were in

a condition that did not permit identification;

(3) further investigation is of the utmost importance if a relatively complete survey

of the freshwater microflora of this area is aimed at. In this respect a closer

study of the Cyanophyceae, Flagellatae, Chlorophyceae, Zygnemales and the

smaller forms of the Desmidiaceae should be undertaken.

616

ACKNOWLEDGEMENTS

It is my pleasant duty to offer my cordial thanks to the following persons and

institutions to whom I am indebted for help and advice during my studies: —

The Council for Scientific and Industrial Research, Pretoria, for liberal financial

assistance during the course of these studies;

The Librarian, Division of Botany and Plant Pathology, Department of

Agriculture, Pretoria, for the generous loan of literature;

Prof. Dr. H. G. Schweickerdt, Department of General Botany, University

of Pretoria, for his encouragement and guidance throughout this investigation;

Dr. B. J. Cholnoky, C.S.I.R., Pretoria, for collecting samples 47-74 and

determining the Diatom taxa;

Dr. W. B. Turrill, Royal Botanic Gardens, Kew, for supplying certain literature

on Desmids not available in South Africa;

Prof. Dr. H. G. Schweickerdt, Department of General Botany, and Prof.

Dr. H. L. Gonin, Department of Latin, University of Pretoria, for their assistance

with the Latin diagnoses of new taxa.

A word of special thanks is due in memory of my parents for their encouragement

and financial assistance throughout this study.

The author is most grateful to these persons and institutions for their generous

help, without which this study would not have been possible.

V. — Literature Cited

Borge, O and Erdtman, G. (1954). . On the Occurrence of Pediastrum in Tertiary Strata in the

Isle of Wight. SartryckurBotaniskaNotiserl954. Hafte2.

Lund. pp. 112-113.

Borgesen, F. (1890) Ft Idle Bidrag til Bornholm Desmidie-Flora. Botanisk

Tiddskrift. Bind 17, pp. 141-152.

Carter, Nellie (1935) Alpine Desmids from British Columbia. Journ. Linn. Soc.

London, Vol. L, No. 333, pp. 151-174.

Cholnoky, B. J. (1952) Beitrage zur Kenntnis der Algenflora von Portugiesisch-Ost-

Afrika. (Mozambique.) 1. Separata do „Boletim da

Sociedada Portuguesa de Ciencias Naturais”, Vol. IV,

2.0 Scrie (Vol. XIX), Fasc. 1, Pags. 89a 135. Lisboa/1952.

(1954) Fin Beitrag zur Kenntnis der Algenflora des Mogolflusses in

Nordost-Transvaal.* Sonderabdruck aus Band 101, Heft

1/2, 1954, der Osterreichischen Botanischen Zeitschrift,

pp. 118.

(1954) Diatomeen und einige andere Algen aus dem ,, de Hoek”-

Reservat in Nord-Transvaal. Sartryck ur Botaniska

Notiser 1954, Hafte 3. Lund. pp. 269.

(1955) Hydrobiologische Untersuchungen in Transvaal I. Ver-

gleichung der herbstlichen Algengemeinschaften in Rayton-

vlei und Leeufontein. Hydrobiologia, Vol. VII, No. 3,

pp. 137-209.

Cushman, J. A Various papers on North American Desmids in the Bulletin

of the Torrey Botanical Club.

(1903) Vol. XXX, pp. 513-514.

(1904) Vol. XXXI, pp. 161-164, PI. 7; pp. 393-397; pp. 581-584

PI. 26.

(1905) Vol. XXXll, pp. 223-229, PI. 7-8; pp. 549-553.

(1906) Vol. XXXIII, pp. 343-351.

(1907) Vol. XXXIV, pp. 599-601.

— (1908) Vol. XXXV, pp. 109-134, PI. 3-5.

* This area is situated in N.W. Transvaal.

617

618

(1895)

Jorgensen, E. G. (1947)

Joshua, W. (1882)

(1883)

(1885)

Kopetzky-rechtperg, O. (1952),

Kufferath, H. (1932)

Loub, W. (1953)

Massee, G. (1887)

Messikommer, E. (1942)

Some New and Rare Desmids of the United States, 11. Bull

Torrey Botan. Club, Vol. XXII, pp. 289-298, PI. 232-233.

Algevegetation i Madum So. Botanisk Tiddskrift, Bind 48,

Hefte 2, pp. 141-155.

Notes on British Desmidieae. Journ. of Bot., Vol. XX

pp. 300-302.

Notes on British Desmidieae — No. 2. Journ. of Bot.

Vol. XXI, pp. 290-292, 349.

On some new and rare Desmidieae — No. 3. Journ. of Bot.,

Vol. XXIII, p. 33, Tab. 254.

Burmese Desmidieae, with descriptions of new species occur-

ring in the neighbourhood of Rangoon. Journ. Linn

Soc. Bot., Vol. XXI, 1884-1886, pp. 634-655, PI. 22-25.

Artenliste von Desmidiales aus den osterreichischen Alpen.

Smn 161-24. Aus den Sitzungsberichten der Osterr.

Akademie der Wissenschaften, Mathem.-Naturw. Kl.,

Abt. I, 161 Bd., 4. und 5. Heft. Wien. pp. 239-261.

Quelques Desmidiees du Congo Beige. Annales de Crypto-

gamie exotique. Decembre 1932. Tome V, Ease. 3-4,

pp. 276-281.

Zur Algenflora der Lungauer_ Moore. Smn 162-36. Aus

den Sitzungsberichten der Osterr. Akademie der Wissen-

schaften, Mathem.-Naturw. Kl., Abt. I, 162 Bd., 7. und 8.

Heft. Wien 1953, pp. 545-569.

On causes influencing the direction of growth, and the origin

of multicellular plants. Journ. of Bot., Vol. XXV, pp.

257-267, PI. 277 (Types of sheath in Algae).

Beitrag zur Kenntnis der Algenflora und Algenvegetation des

Hochgebirges um Davos. Pflanzen geographische Korn-

mission der Schweizerischen Naturforschenden Gesellschaft.

Beitrage zur geobotanischen Landesaufnahme der Schweiz,

Heft 24.

Migula, W. (1924) Die Desmidiazeen. 2. Auflage. Handbiicher fur die

praktische naturwissenschaftliche Arbeit V.

Kryptogamen-Flora von Deutschland, Deutsch-Osterreich

und der Schweiz im Anschluss an Thome’s Flora von

Deutschland. Band II. Algen.

Nageli, C. (1849) Gattungen einzelliger Algen. [Micrasterias decemdentata

(Nag.) Arch., p. 123, Tab. VI, H, Fig. 2.]

Nordstedt, O. (1887) The Figures in Cooke’s “British Desmids’’. Journ. of Bot.,

Vol. XXV.

Nygaard, G. (1932) Contributions to our knowledge of the freshwater Algae of

Africa. 9. Freshwater Algae and Phytoplankton from

the Transvaal. (Communicated by Miss E. L. Stephens.)

Trans. Roy. Soc. of S. Afr., Vol. XX — Part 2, pp. 101-148.

Oltmanns, F. (1922) Morphologic und Biologic der Algen. I Bd.

Pascher, a. and Lemmermann, E. Pascher, Siisswasserflora. Heft 2. Flagellatae II.

(1913)

Pascher, A., Lemmermann, E. and Pascher, Siisswasserflora. Heft 5. Chlorophyceae 2.

Brunthaler, J. (1915)

Pascher, A. (1927) Pascher, Siisswasserflora. Heft 4. Volvocales.

Petersen J. B. (1936) On some Algae from Grimsey. Botanisk Tiddskrift, Bind 43,

pp. 269-277.

(1934) The Algal Vegetation of Hammer Bakker. Botanisk Tidd-

skrift, Bind 42, pp. 1-48.

Prescott, G. W. (1948) Desmids. Botan. Review, Vol. XIV, pp. 644-676.

Printz, H. (1927) Chlorophyceae (nebst Conjugatae, Heterokontae und

CharophytaX Die Natiirlichen Pflanzenfamilien. Heraus-

gegeben von A. Engler. Band 3.

619

Reinsch, P. (1878) Contributiones ad fioram Algarum aquae dulcis Promotorii

Bonae Spei. Journ. Linn. Soc. Bot., Vol. XVI, pp. 232-248.

PI. VI.

Contributions to our knowledge of the freshwater Algae of

Africa. 10. Phytoplankton from South African pans and

vleis. (Communicated by Miss E. L. Stephens.) Trans

Roy. Soc. of S. Afr., Vol. XX— Part 2, pp. 149-188.

Formation and division of binucleate giant cells in Micrasterias

americana (Ehrenb.) Ralfs. New Phytologist, Vol. 39,

No. 1, pp. 80-85.

Roy, J. (1890) Freshwater Algae of Enbridge Lake and Vicinity, Hampshire.

Journ. of Bot., Vol. XXVIII, pp. 334-338.

Roy, J. and Bissett, J. P. (1886).... Notes on Japanese Desmids. Journ. of Bot., Vol. XXIV,

pp. 193-196, 237-242, Tab. 268.

Notes on Afghanistan Algae. Journ. Linn. Soc. Bot., Vol.

XXI, 1884-1886, pp. 241-250, PI. 5.

Die von Prof. Dr. Volkens und Dr. Stuhlmann in Ost-Afrika

gesammelten Desmidiaceen, bearbeitet unter Beniitzung

der Vorarbeiten von Prof. G. Hieronymus. A. Engler,

Beitrage zur Flora von Afrika. XVI. Botanische Jahrb.,

Band 26, pp. 1-59.

Berichte iiber die botanischen Ergebnisse der Nyassa-see-

und Kinga- Gebirgs-Expedition der Hermann- und Elise-

geb. Heckmann-Wentzel-Stiftung. A. Engler, Beitrage

zur Flora von Afrika XVIII.

(1900) II. W. Schmidle, (Jber Planktonalgen und Flagellaten aus

dem Nyassasee. Botan. Jahrb., Band 27, pp. 226-237.

(1903) V. Algen, insbesondere solche des Plankton, aus dem Nyassa-

See und seiner Uingebung, gesammelt von Dr. Fiillebom.

Botan. Jahrb., Band 32, pp. 56-88.

(1904) VI. W. Schmidle, Das Chloro- und Cyanophyceenplankton

des Nyassa- und einiger anderer innerafrikanischer Seen.

A. Engler, Beitrage zur Flora von Afrika XXIV. Botan.

Jahrb., Band 33, pp. 1-33.

(1902) Beitrage zur Algenflora Afrikas. A. Engler, Beitrage zur

Flora von Afrika XXL I. Algen aus Ost-Afrika. Botan.

Jahrb., Band 30, pp. 58-68.

Schmidt,!. (1898-1899) Danmarks blaagronne Alger (Cyanophyceae Daniae). Bota-

nisk Tiddskrift, Bind 22, pp. 283-418.

Schulze, G. M. (1954) Internationaler Code der Botanischen Nomenklatur. Ange-

nommen vom Siebenten Internationalen Botanischen

Kongress Stockholm, Juli 1950.

Taylor, Wm. R. (1935) Alphine Algae from the Santa Marta Mountains, Colombia.

Amer. Journ. of Bot., Vol. 22, pp. 763-781, PI. 1-3, Taylor:

Algae.

Taylor, Wm. R. and Colton, H. S. The Phytoplankton of some Arizona pools and lakes. Amer.

(1928) Journ. of Bot., Vol. 15, pp. 596-614, PI. XLVI-XLVII.

Van Oye, P. (1949) Nouvelles Donnees sur les Desmidiees des Environs de

Matadi (Congo Beige). Hydrobiologia. Acta Hydro-

biologica, Limnologica et Protistologica. Separatum.

Vol. I, No. 3, pp. 282-308.

(1953) Etude sur les Desmidiees du Grand-Duche de Luxembourg.

Un Probleme Biogeographique. Bulletin de la Societe

Royale de Botanique de Belgique, Tome 85, pp. 157-194

(juin 1953) — ^Communication presentee a la seance du

19 octobre 1952.

Desmidiacees de la Vallee du Trient (Valais, Suisse). Bull.

de la Soc. bot. Geneve, Vol. I, pp. 251-268, PI. III.

On variation in the Desmidieae, and its Bearings on their

classification. Journ. Linn. Soc. Bot., Vol. XXXIV,

pp. 366-416, PI. 8-11.

A Treatise on the British Freshwater Algae.

ViRET, L. (1909)...

West, G. S. (1899)

(1904). . . .

Schaarschmidt, J. (1884)

Schmidle, W. (1899)

Rich, F. (1932)

Rosenberg, M. (1940)

620

621

EXPLANATION OF THE ILLUSTRATIONS.

Figure 1.

Topographical Map of the districts of Warmbaths, Nylstroom and Naboomspruit

showing the principal localities where collections were made.

Figure 2.

Localities of collections in the Doornfontein and Leeupoort area.

Figure 3.

Localities of collections in the Moddernek area.

Tabula 1

Figs.

1, 2. Stigonema pseudominutuni sp. nov.

3-7. Scytonema myochrous (Dillw.) Ag.

Tabula 2

Figs.

1. Nostoc pseudogelatinosum sp. nov.

2. N. nylstroniicum sp. nov.

3. Anabaena galpinii sp. nov.

4. A. mesiana sp. nov.

5. A. pseudocatenula sp. nov.

6. Oscillatoria subpristleyi sp. nov.

7. O. nylstromica sp. nov.

8. O. tenuis Ag.

9. O. waterbergensis sp. nov.

10. Lyngbya uliginosa sp. nov.

Figs. 3, 4 X a; 5, 6, 7, 9 x b; 1, 2, 8, 10 x c.

Tabula 3

Figs.

1. Cylindrocystis barbarica sp. nov.

2. C. cajfra sp. nov.

3. Penium margaritacewn (Ehrenb.) Breb.

4. P. margaritacewn var. incognitum var. nov.

5. P. margaritacewn var. brevior var. nov.

6. P. barbaricwn sp. nov.

7. P. libellula (Focke) Nordst. var. schweickerdtii var. nov.

8. P. subcucurbitinum sp. nov.

9. P. curtum Breb. var. waterbergense var. nov.

10-13. P. mesianwn sp. nov.

14. P. pseudorufescens sp. nov.

15. P. curcurbitinum Biss, forma botesii forma nov.

16, 17. Penium species ad P. cruciferum (De Bary) Wittr. Fig. 17 = zygospore.

Figs. 4, 5, 6, 7, 10, 11, 13 X a; 1, 2, 3, 8, 9, 12, 14, 15, 16, 17 X b.

a = 40 /x; h = 40 p.

622

Figs.

1.

2.

3.

4-5.

6-7.

8-11.

Figs.

1.

2.

3.

4-5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

Figs.

1.

2.

3.

4-6.

7.

8.

9-10.

11.

12.

13.

14.

15.

16-18.

19.

20.

21.

22.

23.

24.

Tabula 4

Closterium pseudolunula sp. nov.

C. spetsbergense Borge var. subafricanum van. nov.

C. intermedium Ralfs var. mesianum var. nov.

C. kranskopense sp. nov.

C. acerosum (Schrank) Ehrenb. var. waterbergense var. nov.

C. malinvernianum De Not.

Figs. 1, 3, 4, 8, 9, 10, 11 x a; 6 X b; 5, 7 x c; 2 X d.

Tabula 5

Closterium ralfsii Breb. var. subralfsii var. nov.

C. ralfsii var. glentigianum var. nov.

C. ralfsii var. nodosum var. nov.

C. ralfsii var. minor var. nov.

C. truncatum sp. nov.

C. subsiliqua sp. nov.

C. pseudoleibleinii sp. nov.

C. striolatum Ehrenb. var. nylstromicum var. nov.

C. striolatum var. subnylstromicum var. nov.

C. subdecorum sp. nov.

C. decorum Breb. var. minor var. nov.

C. boyanum sp. nov.

C. moniliferum (Bory) Ehrenb. var. epithemioides var. nov.

C. cymbellaeformis sp. nov.

C. atomicum sp. nov.

C. sublagoense sp. nov.

C. schweickerdtii sp. nov.

Figs. 1, 2, 3 X a; 4, 5, 6, 7, 8, 9, 10, 13, 14, 15, 18 X b; 11, 12, 16, 17 x c.

Tabula 6

Closterium kutzingii Breb. var. transvaalense var. nov.

C. warmbadianum sp. nov.

C. warmbadianum var. porulosum var. nov.

C. barbaricum sp. nov.

C. jenneri Ralfs var. dubitabilis var. nov.

C. parvulum Nag. var. minor var. nov.

C. abruptum West var. westiorum var. nov.

C. peracerosum Gay var. elegans G. S. West.

C. gracile Breb. var. brevius var. nov.

C. tumidum Johnson var. angustum var. nov.

C. ceratium Perty var. angustum var. nov.

C. tumidum Johnson.

C. Venus Kiitz. var. inflatum var. nov.

C. incurvum Breb. var. elaboratum var. nov.

C. calosporum Wittr. var. minor var. nov.

C. ehrenbergii Menegh. var. minutissimum var. nov.

C. acutum (Lyngb.) Breb. var. linea (Perty) W. & G. S. West forma minor f. nov.

C. cornu Ehrenb. var. minor var. nov.

C. cornu var. angustum var. nov.

Figs. 4, 5, 6, 7, 8, 9, 10, 11, 13, 14, 17, 18, 20, 24 X a; 2, 3, 12, 15, 16, 19,

21, 22, 23 X b; 1 X c.

a = 40 /X ; b = 40 jti.

623

Tabula 7

Figs.

1. Pleurotaenium trabecula (Ehrenb.) Nag. var. barbaricum var. nov.

2. PL pseudoehrenbergii var. nov.

3-5. PL trabecula (Ehrenb.) Nag. var. ungustum var. nov.

6. PI. caffrorum sp. nov.

7-8. PI. ehrenbergii (Breb.) De Bary var. waterbergense var. nov.

9. PL westiorum sp. nov.

10. PL trabecula var. brevis var. nov.

Figs. 7, 8 X a; 1, 2, 3, 4, 5, 10 X b; 9 X c; 6 x d.

Figs.

1, 2.

3.

4.

5.

6.

7.

Tabula 8

Pleurotaenium species ad PL ovatum Nordst.

Closterium cynthia De Not var. waterbergense var. nov.

C. caffrorum sp. nov.

C. insolitum sp. nov.

Pleurotaenium truncatum (Breb.) Nag. var. mattiei var. nov.

PL trochiscum W. & G. S. West var. galpinii var. nov.

Figs. 1, 2 X a; 3, 4, 5 x b; 6, 7 x c.

Tabula 9

Figs.

1 . Euastrum spinulosum Delp. subsp. africanum Nordst.

2. E. spinulosum subsp. africanum var. transvaalense var. nov.

3. E. divergens Josh. var. subbifidum var. nov.

4-5. E. rostratum Ralfs subsp. umbonatum W. & G. S. West var. kranskopense

var. nov.

6-7. E. brasiliense Borge var. theronii var. nov.

8. E. ansatum Ralfs.

9. E. dubitabilis sp. nov.

10-12. E. insulare (Wittr.) Roy.

13-14. E. cuneatum Jenner var. minor var. nov.

15-17. E. pseudovalidum sp. nov.

18. E. elegans (Breb.) Kiitz. var. transvaalense var. nov.

19-21. E. schweickerdtii sp. nov.

Figs. 1, 2, 3, 10, 11, 12, 13, 14 x a; 4, 5, 6, 7, 8, 9, 15, 16, 17, 18, 19, 20,

21 X b.

a = 40 /x; h = 40 p.

Figs. 5, 7, 14, 16, 20 = lateral view of cell and figs. 17, 21 = vertical view of

semicell.

Tabula 10

Figs.

1-4. Micrasterias apiculata (Ehrenb.) Menegh.

Figs. 1, 2 X a; 3, 4 x b.

Tabula 11

Figs.

1-4. Micrasterias denticulata Breb. var. africana var. nov.

Figs. 1, 2 X a; 3, 4 x b.

624

Figs.

1.

2.

3.

Figs.

1-3.

4.

Figs.

1-4.

Figs.

1-3.

4, 5.

6.

Figs.

1, 2.

3.

4.

5-7.

Figs.

1

2

3

4-7

8

Figs.

1-14.

Figs.

1-11.

Figs.

1-6.

Tabula 12

Micrasterias denticulata Breb. var. africana var. nov.

M. denticulata var. subnotata West forma cornuta forma nov.

M. nylstromica sp. nov.

Tabula 13

Micrcisterias denticulata Breb. var. subnotata West forma cornuta forma nov.

M. nylstromica sp. nov.

Tabula 14

Micrasterias groenewaldii sp. nov.

Fig. 2 = vertical view of semicell.

Tabula 15

Micrasterias truncata (Corda) Breb.

M. truncata var. africana Fritsch & Rich.

M. sol (Ehrenb.) Kiitz.

Figs. 1, 2, 3 X a; 4, 5, 6 x b.

Tabula 16

Micrasterias crux-melitensis (Ehrenb.) Hass. var. transvaalensis var. nov.

M. crux-melitensis var. evoluta Turn.

M. americana (Ehrenb.) Ralfs var. transvaalensis var. nov.

M. decemdentata (Nag.) Arch.

Tabula 17

Micrasterias crux-melitensis (Ehrenb.) Hass. var. evoluta Turn.

M. mahabuleshwarensis Hobson var. transvaalensis var. nov.

M. tropica Nordst. var. elongatissima var. nov.

M. decemdentata (Nag.) Arch. var. galpinii var. nov.

M. truncata (Corda) Breb. var. minor var. nov.

Tabula 18

Micrasterias crux-melitensis (Ehrenb.) Hass.

Fig. 2 = vertical view of semicell.

Tabula 19

Micrasterias crux-melitensis (Ehrenb.) Hass.

Tabula 20

Micrasterias crux-melitensis (Ehrenb.) Hass.

Tabula 21

Figs.

1-4. Micrasterias crux-melitensis (Ehrenb.) Hass.

625

Tabula 22

Figs.

1-6. Micrasterias cru.x-melitensis (Ehrenb.) Hass.

Tabula 23

Figs.

1. Micrasterias crux-melitensis (Ehrenb.) Hass.

2, 3. M. pumatifida (KUtz.) Ralfs var. transvaalensis var. nov.

Figs.

1-3.

4-5.

6.

7.

8, 9.

10.

11.

12.

13.

14, 15.

Figs.

1, 2.

3.

4-6.

7, 8.

9.

10.

12-14.

15, 16.

17, 18.

Figs.

1-7.

9. 10.

11, 12.

13.

14, 15.

16.

Tabula 24

Euastrum divaricatum Lund. var. transvaalense var. nov.

Cosmarium pyramidatum Breb. var. majus var. nov.

C. contractum Kirchn. var. pseudogartanense var. nov.

C. laeve Rabenh. var. majus var. nov.

C. sexnotatum Gutw. vai. simplex var. nov.

C. quadratum Ralfs var. africanum Fritsch.

C. decoratum W. & G. S. West var. waterbergense var. nov.

C. nanum sp. nov.

C. hammeri Reinsch var. minor var. nov.

C. impressidum Elfv.

Figs. 14 X a; 1-13, 15 X b.

Fig. 2 = lateral view and fig. 3 = vertical view of cell,

a = 40 b = 40 /u.

Tabula 25

Cosmarium pseudotaxichondrum Nordst. var. atomicum var. nov.

C. orthostichum Lund. var. pseudopumilum var. nov.

C. pseudoprotractum sp. nov.

C. cajfrorum sp. nov.

C. sexangulare Lund.

C. trachypleurum Lund. var. subspinosum sp. nov.

C. transvaalense sp. nov.

C. subcostatum Nordst. var. warmbadianum var. nov.

C. debaryi Arch. var. minor var. nov.

C. granatum Breb. var. africanum Fritsch.

Figs. 1, 2, 3, 9, 11, 18 X a; 4, 5, 6, 7, 8, 10, 12, 13, 14, 15, 16, 17 x b.

a = 40 /x; b = 40 /X.

Figs. 2, 6, 14, 16 = vertical view and fig. 5 = lateral view of cells.

Tabula 26

Cosmarium naterbergense sp. nov.

C. orthostichum Lund. var. transvaalense var. nov.

C. speciosum Lund. var. simplex Nordst. forma africanum forma nov.

C. tetragonum (Nag.) Arch. var. transvaalense var. nov.

C. nylstromicum sp. nov.

C. subconnatum sp. nov.

Cosmarium barbaricum sp. nov.

Figs. 8, 14, 15 X a; 1-7, 9-13, 16 x b.

Fig. 15 = lateral view of cell.

626

Figs.

1, 2.

3, 4.

5.

6.

7.

8.

9.

10.

Tabula 27

Cosmariwn binwn Nordst.

C. botesii sp. nov.

C. maximum (Borg.) W. & G. S. West var. minor West.

C. trachypleurum Lund. var. pseudonatalensis var. nov. forma iiregulare f. nov.

C. trachypleurum var. pseudonatalensis var. nov.

C. connatum Breb. var. subellipticum var. nov.

C. subtumidum Nordst. var. theronii var. nov.

C. pyramidatum Breb. var. elongatum var. nov.

Figs. 1, 2, 4, 6 X a; 3, 5, 7-10 x b.

Fig. 2 = vertical view of semicell.

Tabula 28

Figs.

1-3. Cosmariwn obsoletum (Hantzsch) Reinsch var. transvaalense var. nov.

4. C. pachydermum Lund. var. waterbergense var. nov.

5, 6. C. reniforme (Ralfs) Arch.

7-10. Xanthidium cristatum Breb. var. transvaalense var. nov.

Figs. 1, 5, 6, 9, 10 X a; 2, 3, 4, 7, 8 x b.

Fig. 10 = vertical view of semicell.

Figs.

2.

3.

4.

5, 6.

7, 8.

9, 10.

11.

12.

13, 14.

15-17.

18.

19, 20.

Tabula 29

Euastrum galpinii sp. nov.

E. spinulosum Delp. subsp. africanum Nordst.

E. divergens Josh. var. galpinii var. nov.

Cosmarium granatum Breb. var. africanum Fritsch.

C. galpinii sp. nov.

C. schweickerdtii sp. nov.

C. pseudamoenum Wille.

C. decor atum W. & G. S. West var. galpinii var. nov.

Xanthidium cristatum Breb. var. delpontei Roy & Biss, forma laevis forma nov.

Staurastrum setigerum Cleve var. pectinatum W. & G. S. West forma australe

forma nov.

St. galpinii sp. nov.

St. quadrangulare Breb. var. subarmatum var. nov.

St. brevispinum Breb. var. masoganum var. nov.

Figs. 6, 14, 15, 20 = vertical view of semicells.

Tabula 30

Figs.

1-6. Staurastrum dilatatum Ehrenb.

7-9. St. caffrorum sp. nov.

10, 11. St. barbaricum sp. nov.

12, 13. St. xiphidiophorum Wolle var. westiorum var. nov.

14, 15. St. orbiculare Ralfs var. ralfsii W. & G. S. West.

16-18. St. subtrifurcatum West forma major W. & G. S. West.

Figs. 16 X a; 14, 15, 17, 18 X b; 1-13 x c.

Figs. 3, 4, 8, 11, 12, 14, 18 = vertical view of semicells.

627

Figs.

1, 2.

3, 4.

5, 6.

7, 8.

9, 10.

11-13.

14-16.

17, 18.

19, 20.

21, 22.

23, 24.

25-27.

28, 29.

Figs.

1, 2.

3-6.

7.

8.

9.

10.

Figs.

1-3.

4.

5.

6.

Figs.

1-4.

5.

6.

7.

8.

9, 10.

Figs.

a-p.

Figs.

A-G.

Figs.

H-K.

Tabula 31

Staurastrum subgemmulatum W. & G. S. West var. mattiei var. nov.

St. mesianum sp. nov.

St. orbiculare Ralfs var. waterbergense var. nov.

St. dickiei Ralfs.

St. pseudogemmulatum sp. nov.

St. pseudogemmulatum var. warmbadianum var. nov.

St. dilatatum Ehrenb.

St. trihedrale Wolle var. australe var. nov.

St. teliferum Ralfs var. transvaalense var. nov.

St. crenatum var. nov.

St. connatum Roy & Biss. var. wannbadianum var. nov.

St. mattiei sp. nov.

St. pygmaeum Breb. var. botesii var. nov.

Figs. 2, 4, 5, 7, 10, 11, 13, 15, 16, 17, 20, 22, 24, 26, 29 = vertical view of

semicells.

Tabula 32

Staurastrum gemelliparum Nordst. var. africanum var. nov.

St. furcatum (Ehrenb.) Breb.

St. excavatum W. & G. S. West.

St. leptocladum Nordst. var. cornutum Wille.

Sphaerozosma aubertianum W. West.

Onychonema species ad. On. laeve Nordst. var. micracanthum Nordst.

Euglena spirogyra Ehrenb. var. major var. nov.

Figs. 2, 4, 5 = vertical view of cells.

Tabula 33

Pediastrum duplex Meyen var. reticulatum Lagerh. forma cohaerens Bohlin.

P. boryanum (Turpin) Menegh. var. granulatum (Kiitz.) A. Br.

P. boryanum var. minor var. nov.

P. tetras (Ehrenb.) Ralfs var excisum Rabenh.

Figs. 1, 4, 5, 6 X a; 2, 3 X b.

Tabula 34

Oedogonium pseudofragile sp. nov.

Oe. mesianum sp. nov.

Oe. subintermedium sp. nov.

Oe. subvaucherii sp. nov.

Oe. mattiei sp. nov.

Oe. boyanum sp. nov.

Tabula 35

Micrasterias crux-melitensis (Ehrenb.) Hass.

Tabula 36

Micrasterias denticulata Breb.

Tabula 37

Micrasterias denticulata Breb.

629

TOPOGRAPHICAL MAP.

202909

630

631

IL

632

633

634

635

636

637

638

639

640

641

642

643

644

645

20290')

646

80//.

647

TABULA 18.

I.

648

649

650

651

652

653

4.

654

655

656

9.

657

658

659

TABULA 30.

17

660

661

202909

662

663

TABULA 34.

202909

664

3-40 pjn.

4 42 p.m.

5-05

5 40p.m.

6 20p.m.

TABULA

665

666

TABULA 37.

667

Review

F. ora Zamhesiaca, Volume 1, Part 1, published by the Crown Agents, London; 366 pp.

including illustrations. Price 25s.

The title Flora Zanibesiaca is an appropriate one for an international project of

this kind dealing, as it does, with the flowering plants of territories linked by one of

Africa’s greatest rivers. The project is sponsored by the governments of Britain,

Portugal and the Federation and deals with the territories of the Federation of Rhodesia

and Nyasaland, Mocambique and the Bechuanaland Protectorate. The Flora is

produced by an editorial committee consisting of Dr. H. Wild of the Federal Herbarium,

Salisbury, Mr. A. W. Exell of the British Museum (Natural History), Mr. J. P. M.

Brenan of Kew and Dr. F. A. Mendonca of Portugal. The whole production is of a

very high standard with an attractive format which could well serve as a model of its

kind.

Included in this first part is a brief history of botanical exploration in the area,

a selected bibliography, a glossary of botanical terms, a key to plant families, a map

of the area and 58 excellent line drawings illustrating most of the genera described.

The arrangement follows the Bentham and Hooker system and the families now dealt

with extend from the Cycadaceae to the Polygalaceae, covering some 90 genera and

306 species. The text results from the collaboration of the following authors:

G. Cufodontis (Pittosporaceae), A. W. Exell (Papaveraceae, Fumariaceae, Cruciferae,

Polygalaceae), A. W. Exell and E. Milne-Redhead (Ranunculaceae), John Lewis

(Gymnospermae), F. A. Mendonca (Nymphaceae), N. K. B. Robson (Annonaceae,

Violaceae), P. Taylor (Resedaceae), G. Troupin (Menispermaceae) and H. Wild

(Dilleniaceae, Berberidaceae, Cabombaceae, Capparidaceae, Bixaceae and

Flacourtiaceae).

It is clear that a single publication dealing with the flowering plants of the whole

of tropical Africa on the scale of the old F,ora of Tropical Africa is too great a task for

a single institution or government to undertake. Thus regional Floras have been

embarked upon, based mainly on political subdivisions, and the Flora Zambesiaca

is the most recent of these. To the north of the region covered, there is in progress

the Flora of Tropical East Africa; to the north-west, the Flore du Congo Beige et Ruanda

Urundi; and to the west, the Conspectus Florae Angolensis. To the south, work has

started on a Flora of Southern Africa, the first part of which is nearly completed.

So little revisionary work has been done on African plants that it becomes necessary

to carry out a small revision on practically every family dealt with. This is time-

consuming work because, in order to arrive at the correct name to use for a species,

variation must be studied throughout its entire range. The more thoroughly this is

done, the more valuable will the publication be. Much of the earlier monographic

work on African plants is unrealistic in that too many species are maintained on slender

distinctions that are not supported by a study of the living plants in the field, or when

a wider range of material has become available to the herbarium botanist. Even with

the more extensive material now available, the delimitation of species limits in accordance

with reality calls for good judgement. In both respects of thoroughness and judgement

regarding species delimitation, the Flora Zambesiaca inspires confidence. Not only

will botanists and field workers in the territories concerned find this a most valuable

publication, but also those in adjoining territories.

One of the very few criticisms that can be made relates to the treatment of

Encephalartos vil osus (Gaertn.) Lem., because the plants classified under this heading

were separated some years ago as E. umbeluziensis R. A. Dyer.

L. E. CODD

202909—7

668

Addendum

Cladophora

Cladophora isaacii Simons, sp. nov.

In Bothalia 7: 194 (1960) the above species was described, with a Latin diagnosis,

and figured, but a type specimen was not designated. This omission is now rectified

in order to validate the species.

Cape Province. — Malmesbury District: Paternoster Bay, in mid- and upper-level

pools, Simons in National Herbarium, Pretoria, algal collection No. 2407 (PRE, holo-

type).

R. H. Simons

669

A Revision of the South African Species of Anthericum,

Chlorophytum and Trachyandra

By

A. Amelia Obermeyer

INTRODUCTION

The genus Anthericum was described by Linnaeus in his Species Plantarum, Ed. 1,

p. 310 (1753), to receive two well known European species, A. ramosum and A. liliago

[previously known as Phalangium, Tournefort, Inst. 368, t.l93 (1700)], four South

African species now classified under Bulbine, a tumble-weed from the Cape {Trachyandra

revoluta) and two European species at present placed under Nartheciwn. The concept

was a wide one, with the result that in time many and varied species were added to it

on the one hand, while on the other, groups of species were withdrawn to form separate

genera. Bulbine was again raised to generic rank by Willdenow in 1800.* The genus

Chlorophytum was segregated in 1808 by Ker Gawler and Caesia by R. Brown in 1810.

In 1843 Kunth discarded the genus Anthericum altogether and divided what was left

into three genera, Phalangium, Bulbinella and Trachyandra, whilst he upheld Bulbine,

Chlorophytum and Caesia and some other genera outside Africa, which do not concern

us here. His genus Phalangium was the equivalent of what is now accepted as

Anthericum L. The European species A. ramosum and A. liliago, which were known to

Linnaeus and which fit his generic description, especially that of his later works, form

the basis of this genus. In 1936 Green proposed that A. ramosum L. should be regarded

as the type species.

Baker in his monograph on the Liliaceae [Trans. Linn. Soc. 15 : 253 (1876)] went

back to Anthericum in a much broader sense (but not quite in the Linnaean circum-

scription for he recognized Chlorophytum, Caesia and Bulbine) and sank Kunth’s

genera Phalangium, Bulbinella and Trachyandra. In the Elora Capensis in 1896, he

raised Bulbinella again to generic rank. Thus Anthericum in the Flora Capensis consisted

of the sections, Phalangium and Trachyandra, whilst at the same time Baker placed

certain species in a new section Dilanthes, which will be discussed later.

When examining the South African species in and related to Anthericum, it was

found that species of Trachyandra Kunth were so different from typical Anthericum L.

that it was decided to restore the genus. Far more difficulty was experienced in

separating Chlorophytum from Anthericum. These two genera are rich in species and

have a world-wide distributio.i. It is outside the scope of this revision to decide w hether

Chlorophytum should be made a section of Anthericum as suggested by Duthie [Ann.

Stell. Univ. 4 : 1 (1926)], Perrier [Notulae Syst. Vol. V, p. 33 (1955)] and others. The

seeds of Anthericum are small and angular, usually numerous and this distinguishes

it from Chlorophytum, which has large flat seeds, very similar in appearance in all

species. Further, the trigonous capsules, the rosulate leaves, the bracteate scape and

the spongy roots usually combine to distinguish Chlorophytum from Anthericum, but,

with the exception of the seed, one or two of these features may be absent. An example

of this is for instance Chlorophytum rigidum Kunth, W'hich has the leaves somewhat

distichous and the pedicels articulated near the base, both features of many species

of Anthericum. But its seeds are flat, for which reason it is best kept under Chloro-

phytum for the present.

* In Hort. Cliff. 122 (1735) Linnaeus had created the genus Bulbine for B. caukscens but in his

Sp. PI. (1753), he placed it under Anthericum. Jussieu’s conception on the nomenclature of these genera

differs from the modern accepted view. In his work, Genera Plantarum 1789, he regards Bulbine

Willd. to be Anthericum whilst the true Anthericum L. (1753) was identified by him as Phalangium.

Poiret in Lam. Encycl. (1804) and Persoon, Synopsis (1805), both follow Jussieu’s classification.

670

A small number of species from the south western Cape regarded as Anthericum

by Baker but which show a closer affinity to Chlorophytum, are here transferred to

this genus. Kunth adopted this view with those species that were known to him, and

Adamson and Salter in their Flora of the Cape Peninsula regarded Anthericum triflorum

as a Chlorophytum. This had been suggested by Duthie. In notes published posthu-

mously (Gen. PI. Fragm. 70, 1866), Salisbury had also noticed the difference between

species now placed in the genus Trachyandra (he probably was not aware of Kunth’s

genus at the time) and the true Anthericum species. He described a new genus Dilanthes

and mentioned that A. revolutum and other closely related species should go into that

genus. It must be regarded as a synonym of Trachyandra. Baker apparently did

not recognize it as a synonym of Trachyandra and made it a section for the south

western Cape species of Chlorophytum and those of Anthericum he thought possessed

rough filaments. However, A. polyphyllum, A. galpinii, A. multisetosum and A. robus-

tum, mentioned by Baker in this section, have glabrous filaments. This heterogeneous

section cannot therefore be upheld.

Duthie’s article on the species of Anthericum and Chlorophytum of the Stellenbosch

Flats [Ann. Stell. Univ. 4, A, 1 : 1-23 (1926)], giving detailed descriptions of 13 species

(9 species of Trachyandra, 3 of Chlorophytum and 1 of Anthericum) occurring in this

area is most informative. Similarly the 20 species enumerated in the Flora of the Cape

Peninsula by Adamson and Salter must be mentioned as a valuable contribution to

our knowledge of these genera. While at Kew in 1950, Pauline Kies began a revision

of these genera. After her marriage I continued with this work and as I have introduced

many modifications I feel I must take full responsibility for the present account.

Key to Genera

Inflorescence central; flowers one to many-nate (if solitary, supported by 2 bracts);

pedicels articulated; perianth persistent; anthers basifixed, large; seeds with

a black, shiny, granular testa; leaf base folded, not tubular;

Seeds small, globose, irregularly compressed with many folds 1 . Anthericum (p. 670)

Seeds large r und a/'d flat with a poimed ilum. 2. Chlorophytum (p. 690)

Inflorescence axillary; flowers single, each supported by one bract; pedicels not

articulated; perianth deciduous, only the base persisting as a small cup or

rim if a capsule is formed; anthers versatile, small; seeds angular, grey or

brown, smooth or verrucose, often glutinous when ripe: leaf base tubular

3. Trachyandra (p. 711)

1. ANTHERICUM

Linn. Sp. PI. ed. 1 : 310 (1753); Gen. 422 (1754). Baker in J. Linn. Soc. 15 ; 290

(1876); Fl. Cap. 6 : 378 (1897); FI. Trop. Afr. 7 : 477 (1898) sensu lato. Benth. &

Hook. Gen. PI. 3, 2 : 788 (1893) sensu lato. Krause in Engl. & Prantl, Pflanzenfam.

15, A ; 282 (1930) sensu lato. Phillips, Genera of S.A. Flow. PI. 2nd ed. 183 (1951)

sensu lato.

Phalangium Juss. Gen. Pi. 52 (1789). Poir. Encyc. Meth. 5 : 242 (1804). Kunth,

Enum. 4 : 593 (1843). Salisb. Gen. Pi. Fragm. 70 (1866).

Herbaceous perennials, the parts above ground dying down in winter. Roots,

many, long, thin, fibrous often with scattered watery tubers near the tips, sometimes

fairly sturdy and woody, rarely swollen, cylindrical and fairly short (A. calyptrocarpum).

Rhizome usually creeping, knobby, often covered with fibres (the remains of old leaf

bases). Leaves distichous or rosulate; in the distichous species there are usually about

8, the primary leaves small, sometimes hairy, the later leaves becoming progressively

larger; in the rosulate species the leaves are numerous; bases sometimes dilated,

membranous, folded (never tubular); lamina flat or folded, rolled or terete, glabrous

671

or hairy, rarely glandular. Inflorescence central, simple or branched, usually over-

topping the leaves; scape flattened, sometimes narrowly winged and usually naked

in distichous-leaved species or terete and bracteate in rosulate species; lower bracts

in compound inflorescences usually large, leaf-like, the fertile bracts much smaller;

in a few species 2-keeled bracteoles present; pedicels articulated near the base or near

the middle, lengthening at anthesis. Flowers in axillary fascicles each surrounded by

bracts; in a few species flowers solitary, subtended by a bract and bracteole; flowers

in each axillary fascicle opening consecutively; open all day, seldom opening in the

afternoon {A. calyptrocarpum). Perianth rotate, white, rarely greenish, shiny, consisting

of watery, translucent cells, the dark median keel usually appearing when the flower

fades; the 3 outer segments slightly narrower than the 3 inner, marcescent, covering

the capsule when it ripens. Stamens 6, adnate to the very base of, and slightly shorter

than the perianth segments; filaments glabrous or papillate in the upper half, usually

flattened below; anthers large, basifixed, the base at the back of the anther forming

a rimmed pit where the filament is inserted; anthers introrse and curling backwards

when fading. Ovary sessile, globose with about 10-30 biseriate ovules; with septal

glands; style filiform often slightly dilated in upper half, smooth, often declinate;

stigma minutely penicillate or capitate. Capsule loculicidally 3 valved, globose, obtuse,

apiculate or beaked, smooth or with transverse ridges (the number of ridges roughly

corresponding with the number of ovules). Seeds small, irregularly angled, minutely

granular, black; embryo cylindric, endosperm firm.

Distribution: A few species in southern Europe and northern Africa; many

found in tropical Africa extending southwards, one species being recorded from Cape

Town. A few are also recorded from central America.

Type Species: A. ramosum L.

Typical European species bear racemes with solitary flowers, supported by one

bract in each axil, whereas our tropical and subtropical species have several bractiferous

buds clustered in each axil, or, if reduced to one flower, with at least two semi-opposing

bracts. A. ramosum L. and A. liliago L., the typical species, moreover have 8 chromo-

somes whereas our South African species have 7 as far as is known. Kunth placed

the species with the fascicled buds in a separate second section [Enum. Plant. 4 : 595

(1843) under Phalangium] and added here the South American species, some from India

and two doubtful ones.

The South African species of Anthericum may be regarded as southern oflf-shoots

of tropical members which migrated south; on reaching the drier, colder parts of the

highveld, they became small and stunted, the Karroo effectively barring further expan-

sion. These small impoverished highveld plants lose many of their usual characteristics

and are often difficult to distinguish. It is likely that the effect of the frost, fires and

drought affected some features i.e. hairiness, size, etc. An example is A. galpinii which

grows up to 3 feet high in the eastern Transvaal lowveld and produces divaricately

branched racemes, whereas some specimens, including the type, are only 8 inches

high and have nearly simple racemes. Similarly I have come to the conclusion that

A. multisetosum, A. rubrovittatum and A. robustum are merely small or tall forms of

one species, A. angulicaule; the shape of the flowers, bracts, pedicels and capsules

remains constant.

Several species e.g. A. cooperi and A. fasciculatum, have adapted themselves very

well to the eastern highveld where they are common although usually small. C. E. Moss

studied the genus for many years and I had the privilege of reading his manuscript

notes which are preserved at the Moss herbarium of the University of the Witwatersrand.

The variability of these eastern highveld species led Moss to suggest that “ although

no actual experiments appear to have been performed in connection with the matter,

specimens occur which are difficult to account for on any hypothesis other than that

they are hybrids or hybrid segregates.”* I found that the numerous species described

* Duthie’s statement in Ann. Stell. Univ. 4, A. 1 : 3 (1926) that “ no indication of hybridisation

has been met with ” is not contradictory for she dealt mainly with Trachyandra species.

672

by Baker and others were often forms of one variable species and consequently reduced

the number of species considerably. Even so when using the key, it will not be found

easy to determine a stunted specimen which has lost many of its characteristics.

I agree with Moss that it is likely that A. coopen, A. saundersiae, A. fasciculatum and

A. galpinii hybridize. It would for instance, account for the variability of the filaments

in some of the species from the Drakensberge near van Reenen. A. cooperi and

A. capitatwn were described from plants collected in that area and the only difference

seems to have been that A. cooperi had smooth filaments while in A. capitatum they

were rough in the upper half. Similarly A. nudicaule was also separated from A. cooperi

because of its rough filaments but N. E. Brown pointed out that even some plants

on the type sheet of A. nudicaule had smooth filaments. In Anthericum the filaments

are constant for each species, smooth or rough. So their variability points to a crossing

of two species, one with smooth and the other with rough filaments. The filaments

of the specimens of A. cooperi found on the Transvaal highveld are smooth, so possible

parents of the mixture found near van Reenen could be A. cooperi on the one hand

and either A. saundersiae (from the lower regions of Natal) or A. fasciculatum (a

highveld species) on the other, these both with rough filaments. A. fasciculatum and

A. galpinii probably also hybridize at times. It would be interesting to investigate

these species to see if they do hybridize easily and also to study their polymorphism

with regard to factors inhibiting growth, like frost, fires and drought. It was felt that

the systematic work on these closely related species could not be solved satisfactorily

in the herbarium and that theie is need for research work in the field.

Geographical Notes: Seventeen species of Anthericum were found to occur in

southern Africa. Of these three are confined to the Transvaal; A. trichophlebium

has been found mainly around Pretoria and Johannesburg; it is related to A. whytei

but segregated geographically and it is much smaller than its northern relative. The

second, an interesting new species, A. cyperaceum is at present only known from the

Transvaal bushveld. Its congested inflorescence resembling a cyperaceous plant is

unusual. A third species, A. radula from the Woodbush in the Transvaal, found by

Rehmann, has not been collected again. All the other species show a wide range,

some being recorded from localities over a thousand miles apart. Anthericum calyptro-

carpum for instance was described from Angola but it is also found in South West Africa,

eastern Southern Rhodesia and near Pretoria. Anthericum whytei was described from

Nyasaland but shows a continuous distribution through the Zambesi country to South

West Africa. A. anceps is found in Southern Rhodesia and northern South West

Africa. It is very closely related to a polymorphous widespread species from the

eastern highveld, A. cooperi. Here too probably, geographical isolation caused them

to become somewhat different. One other species worth mentioning is A. rangei Engl.

& Krause; it was described from South West Africa but has been found south as

far as the Cape Peninsula (where incidentally it flowers in summer). It is very unob-

trusive and was described as A. scariosum by Duthie in 1926. Although the type of

A. rangei from South West Africa was probably destroyed during the Second World

War, the description and the record of intermediate localities between the Cape and

South West Africa, made me decide to consider them one species. It is the only true

Anthericum species found so far south. All the other species found in the winter

rainfall area usually described as Anthericum species, belong either to Trachyandra or

Chlorophytum.

Superficially the South African species could be grouped into two sections, but

several species form links between these two groups. The members of the first group

have distichous leaves and a usually compressed, naked scape, the flowers are arranged

in fascicles in the axils of large bracts and the capsules are round and obtuse. In the

second group the leaves are rosulate and the terete scape beais bracts. In A. trans-

vaalense, A. radula, A. longistylum and A. haygarthii, which come under this latter

673

group, the flowers are solitary, with an outer bract and an inner 2-keeled bracteole and

the capsule is beaked. Intermediate forms are A. krauseanum, A. rangei and A. calyp-

trocarpum, which have the leaves rosulate and the terete scape bracteate, but the

flowers are not always solitary nor is the capsule always beaked (for in A. krauseanum

it is globose and with transverse ridges similar to those of the first section). A further

link is made by A. acutum which has the leaves more or less distichously arranged and

the flowers clustered but its capsule is beaked.

Morphology

Roots: They are usually very abundant, nearly always thin, and often with watery

tubers near the tips. (These are usually left behind when the plant is pulled out of the

ground). I presume that these tubers are produced during the summer months and

are used for water storage. A. angulicaule and A. cyperaceum have the roots more

sturdy and no tubers were seen on the herbarium specimens. A. calyptrocarpum with

its swollen roots is an exception. No plants showed the production of the long root

hairs so often found in Chlorophytum and Trachyandra.

Rhizome: It is horizontal and knobby, each shoot arising from a nodular, woody

base. It is nearly always covered with fibres from old leaf bases.

Leaves: In the distichous species there are about 8, the outer very small, the inner

ones becoming progressively larger. They clasp at the base but are never tubular.

The veins are usually prominent. Sometimes the two outer veins fuse giving the leaf

a prominent, usually yellow margin.

Inflorescence: In the distichous-leaved species the flowers and their bracts are

arranged in small clusters in the axils of larger bracts. Actually these clusters, spirally

arranged on the rhachis, represent repressed side branches. Occasionally in some

species, e.g. A. anceps, A. saundersiae, A. fasciculatum, etc., a basal cluster may

elongate; the flowers on this branch are then usually distichously arranged.

In the four closely related species A. longistylum, A. radula, A. transvaalense and

A. haygarthii the flowers are always solitary with an outer bract and an inner 2-keeled

bracteole. The inner bracteole in the 2 last mentioned species is not situated exactly

opposite the outer bract, but to one side, probably as a result of the spiral arrangement

of the raceme; the inner bracteole may have two horns or in haygarthii an oblique

one-sided horn. In this rosulate group the scape is terete, with the lower bracts leaf-like,

the upper becoming progressively smaller. In the distichous species the scape is usually

compressed and naked and the lowest bract of the inflorescence is large, the following

floral bracts being much smaller.

The species all have articulated pedicels. The articulation is found usually below

the centre in our species. A specimen of A. acutum collected by Werdermann showed

two small bracts produced at the articulation. An articulated pedicel may therefore

be interpreted as a peduncle bearing a pedicel on top. It is usually this upper part,

the pedicel proper, which elongates during the flowering and the fruiting stage. The

inflorescences of these African species therefore appear to be the result of the reduction

of a very complex inflorescence. As in Chlorophytum, the flowering period of an

inflorescence lasts a long time as the flowers open consecutively. It is the only

inflorescence produced on that shoot. Beside it a new shoot will develop Lorn a

lateral bud of the rhizome.

Flowers: These are very similar in structure; the perianth being white, dark

keeled, delicate in texture with the watery cells translucent as in Chlorophytum. The

stamens are often declinate. The filaments are smooth, or in some species rough in

the upper part. The anthers are large and basifixed, with the filament tip inserted in

a rimmed pit at the base of the anther. When the anthers fade they curl backwards

in a spiral. Often the margins of the open locules are wavy. The ovary is oblong

or ovoid with 8-24 biseriate, axillary ovules in each loculus. It was seen that in some

674

specimens one or both rows of ovules would double themselves — the basic number

of 16 becoming thus 24 or even 32. The funicle is slender and short.

Capsule: In the distichous species (and also in A. krauseanum) the capsule is oblong,

with many transverse ridges (the number of ridges corresponding roughly with the

number of ovules). In the other species it is ovoid and apiculate or beaked and the

transverse ridges may be absent.

Seed: It is small and angular with a black and granular testa. During development

inside the rounded capsule the globular, young seeds are pressed into irregular, angular

shapes.

Key to Species

Leaves distichous, ca. 8; scape usually naked seldom bracteate, often compressed, sometimes

narrowly winged; capsule rounded, obtuse, rarely apiculate, with transverse ridges; flowers

usually clustered:

Robust plants up to 180 cm high (usually smaller); leaf margin often raised; roots woody,

sturdy; pedicels articulated near the base; filaments smooth, shorter than anthers:

Flowers in a divaricately branched raceme (simple in small or starved plants); floral bracts

small, usually dark, firm; style not exserted from closed flower 1. A. angulicaule

Flowers in dense fascicles supported by long basal bracts ; floral bracts membranous, whitish,

large; style exserted from closed flower 2. A. cyperaceum

Smaller, more slender plants, 10-50 cm high; leaf margin not raised; roots wiry, numerous,

with scattered tubers; pedicels articulated near the base or the middle; filaments smooth

or rough in upper half, more or less equalling anthers in length:

Leaves linear, not over 2 cm wide, usually much narrower:

Filaments smooth (occasionally rough in A. cooperi):

Raceme divaricate with bare wiry branches and the axils congested with subulate bracts

many flower buds and an occasional, accessory branch; scape slender, subterete,

not winged, naked 3. A. galpinii

Raceme more or less spicate, occasionally with a basal ascending branch ; pedicels short,

flowers closely arranged on rhachis; bracts enclosing the buds; scape flat, winged,

stout, naked or with some sterile bracts near spike 4. A. anceps

Raceme congested in a pseudo-capitate inflorescence, always simple, rarely somewhat

elongated; bracts membranous, usually whitish and large, occasionally small; scape

flat, fairly stout, naked ; filaments smooth, occasionally rough 5. A. cooperi

Filaments rough:

Raceme divaricately branched, flowers often congested at first; ovules ca. 10; leaves

flaccid; gregarious plants, found in the lower parts of Natal 6. A. saundersiae

Raceme seldom divaricately branched, usually spicate, or with many interrupted spikes

on rhachis; ovules ca. 14; plants variable, found on the highveld of the Transvaal

and Orange Free State to South West Africa 7. A. fasciculatum

Leaves lanceolate-ovate, over 2 cm wide, plants densely and shortly hairy:

Leaf not “ petioled ; filaments scabrid; Pretoria and surrounding districts 8. A. trichophlebium

Leaf “petioled ”; filaments smooth; northern South West Africa, Southern Rhodesia to

Nyasaland 9. A. whytei

Leaves rosulate, many (subdistichous in A. acutuni) scape terete, bracteate; capsule ovoid (cylin-

drical in A. acutum) beaked or apiculate, smooth or with some transverse ridges; flowers

single or 1-3-nate:

Plants grasslike, xerophytic;

Roots wiry, numerous, with scattered tubers:

Capsule ovoid, beaked, glabrous; leaf margin not thickened; filaments smooth; southern

South West Africa, south-western Cape 10. A. rangei

Capsule small, globose, obtuse or apiculate, with transverse ridges; leaf margin thickened;

flowers very small, pedicels filiform; filaments rough; northern South West Africa

11. A. krauseanum

Roots swollen, cylindrical; capsule smooth, acute; Angola, northern South West Africa,

Southern Rhodesia, Transvaal 12. A. calyptrocarpum

Plants not grasslike (if with linear leaves then flowers solitary with an inner two-keeled bracteole):

Leaves sub-distichous; flowers 1-3-nate on simple or sub-simple racemes, congested at first

elongating during anthesis; capsule cylindrical, apiculate 13. ^4. acutum

Leaves rosulate, numerous; flowers single with an ovate outer bract and a two-keeled inner

bracteole; inflorescence branched or simple; capsule beaked:

Inflorescence divaricately branched (simple in young plants); bracts small:

Scape glabrous 14. A. longistylwn

Scape scabrid 15. A. radula

Inflorescence simple, seldom with a basal, ascending branch; bracts large:

Leaves hairy; inner bracteole with two horns 16. A. iransvaalense

Leaves glabrous; inner bracteole oblique with one horn 17. A. haygarthii

675

1. A. angulicaule Bak. in J. Linn. Soc. 15 : 305 (1876) and in FI. Cap. 6 : 382

(1896). Type: Cape, Keiskamma, Hutton (K, lecto.!, PRE, photo.).

A. robustum Bak. in FI. Cap. 6 : 386 (1897). Type: Zululand, Wood 3972 (K, holo.,

NH. iso!, PRE, photo.). A. multisetosum Bak. in FI. Cap. 6 : 386 (1897). Type:

Swaziland, Galpin 1013 (K, holo., PRE, iso!). A. mbrovittatum Poelln. in Bol. Soc.

Brot. 16, 2 : 49 (1942). Type: Transvaal, Barberton, Galpin 539 (B, holo!, PRE,

GRA, iso!).

Plants variable in size, 20-150 cm high. Roots hard, woody, fairly stout, ca.

3 mm in diam. Rhizome with irregular woody knobs, horizontal, covered with fibres

from old leaf bases. Leaves linear to linear-lanceolate, up to 75 cm long and 2-5 cm

broad, coriaceous, ribbed, clasping at the base, tapering in upper half, apex acuminate,

margin raised or flat; lamina glabrous, seldom minutely horizontally pubescent on

the prominent ribs above (glabrous to the naked eye). Inflorescence branched (simple

in starved plants); scape flattened, often narrowly winged, occasionally bracteate,

rhachis often zigzagging between axils; bracts small ca. 3 mm long, deltoid, dark,

apiculate, clustered ; pedicels articulated near the base, up to 1 cm long in fruit, angular.

F/oweri' 3-4-nate, white with dark keeled perianth segments ca. 1 cm long; tips ciliate,

3 outer often with dark tips; filaments glabrous, short; anthers large; ovary with

ca. 16 ovules per cell. Capsule 6 mm long. Seeds typical.

Flowering Period: From June onwards through the summer.

Distribution: Eastern Cape, Natal, Orange Free State, Transvaal, Bechuanaland,

Swaziland, Portuguese East Africa. Usually found on grassy mountain slopes.

Cape. — Peddie: Keiskamma River Mouth, edge of lagoon, Galpin 7652 (PRE). [In

the FI. Cap. 6 : 382 Baker quotes Zeyher, from Zwartkops River, but this is probably

Chlorophytum capense (L.) Voss].

Natal. — Pinetown: Galpin 12097 (PRE); Nkandhla: Qudeni, Gerstner 634 (PRE).

Swaziland. — Havelock Mine, Miller 2999 (PRE).

Transvaal. — Lydenburg: Kemp’s Heights near Lydenburg, Marais 14 (PRE). Bar-

berton: Saddleback Mountain slopes, Galpin 539 (PRE). Belfast: Waterval Onder,

Young 1465 (PRE). Waterberg: farm Roodepoort No. 15, Palala Road, Galpin

11611 (PRE). Potchefstroom: Losberg, Theron 800 (PRE).

Orange Free State. — Kroonstad: Bothaville, Goossens 1206 (PRE).

Bechuanaland. — Taungs, Brueckner 591 (PRE).

Portuguese East Africa. — Mocuba, Faulkner 242 (PRE).

A. angulicaule varies a good deal in size. It grows in grassveld which is regularly

ravaged by fires. Of the synonyms, A. robustum seems to be its optimal form and

A. multisetosum the small stunted form appearing early in spring after cold, drought

and fires; A. rubrovittatum is intermediate. The type from the eastern Cape has no

prominent leaf margin but this I consider to be merely a variation. The raised margin

results from the fusion of the two outer ribs. The specimens from the western Free

State and western Transvaal are more slender in habit. All the above specimens have

the same essential characters; the hard, woody roots, knobby rhizome, flattened scape,

small dark bracts, pedicels articulated near the base and very short filaments,

2. A. cyperaceum Kies sp. nov. distincta.

Radices crassae lignosae. Folia linearia conduplicata rigida glabra 80 cm longa,

6-12 mm lata, marginibus prominentibus. Scapus teres rectus costatus, 45-180 cm

longus. Flores axillares fasciculati congest!. Bracteae inferiores 15 cm longae foliosae

superiores 5 mm longae membranaceae basi rubrae brunneae carinatae marginibus

laceratae. Pedicelli 15 mm longi, prope basin articulati. Perianthium 1 cm diam.

Filamenta laevia, antheris breviora; stylus longe exsertus. Capsula ovoidea apice

acute transverse rugosa.

676

Hard, glabrous, rush-like plants up to 180 cm tall. Foots hard, woody, fairly

stout, 2 mm in diam. spreading. Rhizome with knobs, wcody, horizontal, covered

with fibres from old leaf bases. Leaves erect, linear, up to 80 cm long, ca. 14 mm

broad, folded, glabrous, ribbed, hard, clarping at the base, long acuminate in upper

half, margin raised, yellow. Inflorescence with flowers massed in 3-4 dense clusters

at the apex, supported by a long, patent, basal bract; scape round, compressed, up

to 180 cm tall, hard; lower basal bract leaf-like, long acuminate, patent, up to 17 cm

long; floral bracts ovate, 8 mm long, white, thin; pedicels patent, articulated near

the base, elongating during anthesis, up to 15 mm long in fruit. Flowers in several

dense clusters; perianth segments ca. 14 mm long, tips of outer segments dark on

outside; filaments glabrous, short; anthers large, longer than filaments; style declinate,

crook-shaped, ultimately longer than perianth; ovary with ca. 10 ovules per loculus.

Capsule oblong, 1 cm long, transversely rugose, apiculate. Seeds typical.

Flowering Period: February-March.

Distribution: Transvaal, collected in the Waterberg, Middelburg and Lydenburg

districts; (“in mixed open bushveld on hillside” Story).

Transvaal. — Lydenburg: near Steelpoort, Story 4071 (PRE). Middelburg: Zoete-

velden, van der Merwe, 1306 (PRE). Waterberg: Foie Evans in PRE 28768 (PRE,

holo.).

A very distinct and conspicuous species, nearest A. angulicaule Bak. The congested

inflorescence subtended by large, boatshaped bracts, separates it from all the other

species.

3. A. galpinii Bak. FI. Cap. 6: 385 (1897); aggregate species.

Glabrous or hairy plants up to 75 cm high. Roots many, thin, wiry with scattered

tubers near the root-tips. Rhizome small, horizontal, sometimes covered with fibres

from old leaf bases. Leaves distichous, about 8 per shoot, primary small, the folllowing

becoming progressively larger, linear to linear-lanceolate, 6-60 cm long, 2-10 mm

broad, soft or firm, closely ribbed, glabrous or shortly, patently setose on the ribs

especially on the primary leaves, the margin fimbriate or glabrous, base somewhat

dilated, clasping. Inflorescence a divaricately branched raceme with patent, bare side

branches emerging from the axillary flower fascicles, rarely simple in starved or young

plants; scape terete or flattened and narrowly winged, ribbed; lower axillary bracts

often with a long, soft awn, up to 3 cm long; floral bracts small, usually dark, apiculate;

pedicels up to 1 cm in fruit, articulated below the middle. Flowers with perianth

variable in size, segments 10-15 mm long; stamens declinate, 4 situated posteriorly,

2 anteriorly with the style; filaments smooth; ovary oblong with 10-30 ovules per cell.

Capsule rounded, ca. 6 mm high, closely transversely ribbed. Seeds typical.

Flowering Period: November-March.

Distribution: Warmer parts of the Transvaal, Southern Rhodesia to northern

South West Africa and Angola; also in Portuguese East Africa. Usually in grassland

or along river courses in sand.

Key to Varieties

Primary leaves short, hairy, later leaves glabrous; bracts aristate; plants wiry; in grassveld

a. var. galpinii

Leaves glabrous, coriaceous; bracts small, apiculate; usually along river courses, in sand

b. var. matabelense

Leaves soft, margin fimbriate with red or white cilia up to 2 mm long; on rocky grass slopes

c. var. norlindii

{a) var. galpinii

A. galpinii Bak. in FI. Cap. 6 : 385 (1897). Type: Transvaal, Barberton, Galpin 1160

(K, holo., PRE, NH, iso!).

677

A. patulum Bak. in FI. Cap. 6 : 386 (1897). Type: Transvaal, Barberton, Saddleback

Mountain, Galpin 1232 (K, holo., PRE, GRA, iso!). A rehmannii Bak. in Bull. Herb.

Boiss. Ser. 2:8: 781 (1901). Type: Transvaal, Potgietersrus, Klippan, Rehmann

5432 (Z, holo!, PRE, photo.). A. divaricatum Bak. ex Schinz in Viert. Nat. Ges. Ziir.

49 : 174 (1904), non Jacq. Type: Transvaal, Pietersburg, Shilouvane, Junod 665

(Z, holo! PRE, photo.). A. bakerianum Poelln., in Fedde, Rep. 50 : 232 (1941) nom.

nov. for A. divaricatum Bak. ex Schinz. A. junodii Bak. in Viert. Nat. Ges. Ziir. 49 : 175

(1904). Type: Transvaal, Pietersburg, Shilouvane, Junod 1735 (G, holo!, PRE, photo.).

A. delagoense Poelln. in Fedde, Rep. 53, 6 : 134 (1944). Type: Portuguese East Africa,

Lourengo Marques, Schlechter 11709 (B, holo., probably destroyed, PRE, iso.!). Fig. 1.

Flowering Period : October-April.

Distribution: Transvaal, Potuguese East Africa and Southern Rhodesia, grass-

veld.

Transvaal. — Barberton: Thorncroft 24 (PRE). Nelspruit: Pretorius Kop, van der

Schyjf 1106 (PRE). Pilgrims Rest: Calais, Killick & Strey 2556 (PRE). Pietersburg:

between Duiwelskloof and Munnik, Schweickerdt 1038 (PRE). Sibasa: Kruger National

Park, Baiandbai, Lang (TM 32158, 32149, PRE). Waterberg: near Vaalwater, Meeuse

& Strey 10424 (PRE).

Portuguese East Africa. — Delagoa Bay, Forbes (B). Lourengo Marques, Sul do

Save, between Boane and Gova, Myre & Carvalho, 1367 (PRE).

Southern Rhodesia. — Matobo, West 2443 (SRGH). Bulawayo, Feiertag (SRGH).

The variety galpinii is a xerophytic form, fairly uniform in appearance and is

common around Barberton, the type locality, and elsewhere in the Lowveld. A. patulum

also from Barberton, must have been collected in more shady, moist surroundings.

It is glabrous and has few-flowered fascicles but no good differences could be detected.

It agrees with specimens collected in Southern Rhodesia found near rivers.

{b) var. matabelense {Bak.) Oberm., stat. nov.

A. matabelense Bak. in FI. Trop. Afr. 7 : 484 (1898). Type: Southern Rhodesia,

Matabeleland, banks of the Matengwe River, Holub (K, holo.). A. volkii Soelch,

in Mitt. Bot. Staatssamm. Muenchen II: 186 (1956). Type: South West Africa,

Grootfontein, Blockfontein on dunes, Volk 1685 (M, holo.!, PRE, photo.).

Flowering Period : January-May.

Distribution. — Angola, South West Africa, Southern and Northern Rhodesia,

Transvaal.

Angola. — Between Kembo and Kutile Rivers, Pocock 457 (PRE).

South West Africa. — Okavango Native Territory: Omuramba Khaudum, North of

Tamso, de Winter & Marais 4735 (PRE, K).

Northern Rhodesia. — Mkushi, Fiwila, Robinson 2574 (SRGH).

Southern Rhodesia. — Matoba, farm Besna kobila. Miller 1941, 2523 (SRGH); Enter-

prise, Wild 3698, 3232 (SRGH); Bikita, Devuli Bridge near Birchenough Bridge,

Obermeyer 2482 (PRE).

Transvaal. — Soutpansberg : Dongola Reserve, Sandveld, Pole Evans 4586 (PRE,

SRGH); 4307 (PRE); Pretoria: Hammanskraal, Codd 5622; 3468 (PRE).

(c) var. norlindii (Weim.) Oberm., stat. et comb. nov.

Chlorophytum norlindii Weim. in Bot. Not. 1937, 434, photo., p. 435. Type: Southern

Rhodesia, Makoni, near Maidstone, Norlindii & Weimarck 4128 (LD, holo., PRE,

SRGH, iso.!).

Flowering Period: January-March.

Distribution: Southern Rhodesia, Transvaal.

678

Fig. 1 . — Anthericum galpinii Bak. a, habit showing distichous leaves and tuberous root swellings, X i.

b, capsule, X i. c, stamen, X 2. d, seed, X 6.

679

Southern Rhodesia. — Umtali, Quagga’s Hoek, Chase 4874 (SRGH, PRE). Inyanga,

Chase 559 (PRE). Salisbury: Brain 8327 (SRGH); Hunyani, Eyies 4613 (SRGH);

Makabusi, Wild 2275 (SRGH); Marandellas, Dehn 34 (SRGH). Urungwe: near

Msukwe River, Davies (SRGH 67829).

Transvaal. — Waterberg: Visgat (near Dorset) near Vaalwater, Meeuse 10565 a

(PRE).

A. limosum Bak. in Trans. Linn. Soc. ser. 2, Bot. 1: 257 (1878) from Angola,

Barro do Bengo, in swampy ground, Welwitsch 3803, 3804 (BM, holo.?) may be this

species but as I have neither seen the type nor any other material from that region,

I hesitate to sink A. galpinii and its varieties under this species. The type at BM

(PRE, photo.) does not agree with Baker’s description.

4. A. anceps Bak. in J. Linn. Soc. 15: 305 (1876); Fl. Cap. 6: 382 (1896); FI.

Trop. Afr. 7: 482 (1898). Type: Southern Rhodesia, South African Gold Fields,

Baines s.n. (K, holo.).

A. rautanenii Schinz in Bull. Herb. Boiss. Ser. 2, 8: 625 (1908); Poelln. in Fedde,

Rep. 52: 260 (1943). Type: South West Africa, Amboland, Olukonda, 603

(Z, holo.!, Pre, photo.). A. otavense Engl. & IGause in Engl. Bot. Jahrb. 45: 129

(1911); Poelln. in Fedde, Rep. 52: 260 (1943). Type: South West Africa, Otavi,

Dinter 622 (B, holo., probably destroyed: NBG, iso.!, PRE, photo.).

Plants up to 40 cm high, solitary or gregarious. Roots many, thin, spreading,

occasionally with tubers near the tips. Rhizome small, horizontal, knobby, covered

with fibres from old leaf-bases. Leaves linear, up to 40 cm long, 1 cm broad, not much

attenuated until near the apex, base clasping, slightly dilated, folded, glabrous, ribbed.

Infllorescence a simple dense and closely flowered raceme, occasionally with a short,

ascending basal branch which may be sterile; about as long as the leaves; scape stout,

flattened, narrowly winged; lower sterile bracts clasping scape, upper floral bracts

ascending with a patent, soft awn, glumaceous, up to 2 cm long, nerved; pedicels

short, hidden by the bracts, up to 6 mm long in fruit, articulated below the middle.

Flowers many in each axillary fascicle, the lowest “ fascicle ” sometimes elongating

to form a short distichous-flowered side branch; perianth white, with the segments

II mm long, the outer with dark tips; filaments glabrous, short; anthers typical;

ovary with ca. 12 ovules per cell. Capsule globose, 7 mm in diam., closely, transversely

ribbed. Seeds typical.

Flowering Period: October-March.

Distribution: Southern Rhodesia, northern South West Africa, apparently in

sandy mopani grassveld.

Southern Rhodesia. — Bulawayo, Eyies & Johnston 1180 (GRA). Wankie, Levy, 38

(PRE). Plumtree, McLeod 6 (PRE).

South West Africa. — Grootfontein : Otavi Valley, Dinter 5509 (B); Arisotavi, Boss

(TM 35683, PRE); Gautscha Pan, Maguire 21966 (PRE); Rotenfels, Rehm (M).

Outjo: Feigenwasser, Fb/A: 1795 (M, PRE). Okavango Native Territory : near Nzinzi,

de Winter 3997 (PRE).

5. A. saundersiae Bak. in Fl. Cap. 6: 384 (1896). Type: Natal, Saunders (K,

holo.).

A. pulchellum Bak. in J. Bot. Lond. 1872: 140 nom. nud.; Fl. Cap. 6: 386 (1897).

Type: Natal, Durban, Gerrard 554 (K, holo.). A. rudatisii Poelln. in Fedde, Rep.

53: 129 (1944). Type: Natal, Umzinto, Ifafa, Rudatis 1235 (B, holo.!, PRE, photo.),

var. angustum Poelln. l.c. Type of var. : Natal, Marianhill, collector unknown (B,

holo., probably destroyed).

Plants up to 40 cm high, gregarious. Roots thin, long (no tubers seen). Rhizome

creeping, small, knobby, covered with sparse, short fibres from old leaf bases. Leaves

680

erect, linear, up to 30 cm long, 1 cm broad, gradually tapered to the subulate apex,

clasping at the base, soft, usually flat. Inflorescence a branched or sub-simple, contracted

and congested raceme with the lowest branch often patent; scape flat, winged, bare;

lowest bract much longer than the following ones, leaflike; floral bracts small, 8 mm

ovate, subulate, dark; pedicels up to 8 mm long, articulated near the base. Flowers

very many in each axillary fascicle; perianth with segments 1 cm long; filaments

minutely papillate in upper part, slightly longer than anthers; ovary with 7-10 ovules

per cell; style declinate. Capsule globose, 6 mm in diam. with lax, transverse ridges.

Seeds typical.

Flowering Peiiod: October-March.

Distribution; Natal, usually in low grasslands near river mouths which become

inundated at times.

Natal. — Port St. Johns; West Gate, Galpin 3404 (PRE); Schonland 4200 (GRA)

Port Shepstone; Port Edward, Moss 19177 (J). Umzinto; Scottburgh, Mauve 4013

(PRE, G); Umlazi; Isipingo, Salter 381/34 (BOL). Durban; Westville, Johnston

(NH 20172); Berea, Forbes 479 (NH). Hlabisa; Ward 1929; Hluhluwe Game Reserve,

Ward 2806 (NH).

In the FI. Cap. 6; 387 (1897), Baker cites Nelson 185 from the Vaal River, near

rapids, as this species. Although the specimen is poor, it is recognizable as A. fascicu-

latum, the type of which was collected in the same area.

6. A. cooper! fiaA". in J. Linn. Soc. 15; 304 (1876); FI. Cap. 6; 382 (1896). Type;

Natal, Cooper 1004 (K, syn.); Basutoland, Cooper 3302 (K, syn.).

A. pachyphyllum Bale, in J. Linn. Soc. 15; 304 (1876); FI. Cap. 6; 381 (1896). Type;

Cape, Grahamstown, collector unknown (K, holo.!, PRE, photo.). A. triflorum var.

minor Bak. in FI. Cap. 6; 383, pro parte, as to Galpin 1025 (PRE, GRA) from Barberton.

A. nudicaule Bak. in FI. Cap. 6; 384 (1896). Type; Griqualand East, Tyson 1054

(K, holo, PRE, iso.!). A. capitatum Bak. in FI. Cap. 6; 384 (1896). Type; Natal,

Van Reenen’s Pass, Wood 4795 (K. holo., NH, iso.!). A. adscendens Poelln. in Bol.

Soc. Brot. 16, 2; 70 (1942). Type; Lydenburg, Wilms 1503 (B, holo.!, PRE, photo.).

A. pascuorum Poelln. in Bol. Soc. Bret. 16, 2; 46 (1942). Type; Natal, Rudatis 456

(B, holo.!, PRE, photo.).

Small plants 10-40 cm high, usually gregarious. Roots thin, many, producing

some scattered tubers near the tips. Rhizome horizontal, knobby, woody, covered

with fibres fiom old leaf bases. Leaves erect or falcate, linear to lanceolate, variable

in length, 5-30 cm long, 3-10 mm broad, attenuated near the obtuse, apiculate apex,

clasping at the base, flat or usually folded, glabrous or minutely pubescent on the

prominent ribs and margin with minute, horizontal hairs or minutely ciliate. Inflore-

scence a simple raceme, usually congested near the apex; scape compressed, narrowly

winged, usually naked, biacts variable, small or large, green or white and membranous;

pedicels short, up to 8 mm in fruit, articulated below the middle. Flowers 1-3-nate,

congested or occasionally somewhat lax, with the rhacliis visible between the flower

fascicles; perianth with segments 8-12 mm long; filaments smooth or papillate in

upper half, longer than anthers; ovary with ca. 16-24-32 ovules per cell (one or both

rows may sometimes double the number of ovules usually produced). Capsule globose,

8 mm in diam., with many transverse ridges. Seeds typical.

Flowering Period; October-March.

Distribution; Eastern Cape, Natal, eastern Free State, Transvaal; in grassveld.

The size of the bracts varies from small and green to large, white and membranous;

the type (form a) according to the description, has the large bracts, a common form

in the Drakensberg area. It is very similar in appearance to A. capitatum which Baker

separated because of the papillate (not smooth) filaments. I suspect that this variation

in the filaments may be the result of hybridization, possibly with A. saundersiae. The

681

Transvaal and eastern Cape specimens show a form with small, green bracts, somewhat

more lax inflorescence and fairly short, falcate leaves (form b). This form agrees with

A. pachyphyllum Bak. It was not practicable however, to maintain these forms as

separate species.

(o) Typical form with large, white, membranous bracts and a congested inflore-

scence; filaments smooth or papillate.

Natal. — Klip Rivier; near van Reenen, Wood 6158, 12098 (PRE). Alfred: near

Umzimkulu, Killick & Marais 2010 (PRE). Utrecht: Kaffir Drift, Thode A367 (PRE).

Swaziland. — Stewart (TM 10122, PRE).

Transvaal. — Carolina: Galpin 12516 (PRE). Middelburg: Dullstroom, Noome

(TM 20800, PRE). Belfast: Codd 5632 (PRE); Pilgrims Rest: Graskop, Galpin

14530 (PRE).

{b) Form with small greenish bracts, a lax raceme and leaves usually fairly short,

falcate {A. pachyphyllum Bak.).

Cape. — Albany: Rockcliffe near Sidbury, Daly 816 (PRE); Alexandria: Bushman’s

River Bridge, on Port Elizabeth Road, Archibald 4015 (PRE); Idutywa, Galpin 10923

(PRE).

Transvaal. — Potchefstroom: Bur tt Davy \Q62 {'PRE). Pretoria: \vi\h, Leendertz Al?>

(PRE). Benoni: Bradfield 291 (PRE). Johannesburg: Witpoortjie, Gilmore 663

(PRE). Waterberg: near Nylstroom, Burtt Davy 2072 (PRE).

7. A. fasciculatum Bak. in J. Linn. Soc. 15: 316 (1876); FI. Cap. 6: 383 (1896).

Type: Vaal River Plains, Barber (K, holo.).

A. subulatum Bak. in Bull. Herb. Boiss. Ser. 2, 1: 781 (1901). Type: Transvaal,

Pretoria, Donkerhoek, Rehmann 6550 (Z, holo.!, PRE, photo.). A. hereroense Schinz

in Bull. Herb. Boiss. Ser. 2, 1: 857 (1901); Poelln. in Fedde, Rep. 52: 237 (1943).

Type: South West Africa, Hereroland, Orumbo, Dinter 1306 (Z, holo.!, PRE, photo.),

var. longibracteatum Poelln., l.c. Type of var. : Aitzas, Dinter 823 (B, holo.?). A.

co«rat/7?7 Bak. in Bull. Herb. Boiss. Ser. 2,4: 997 (1904). Type: Transvaal, Germiston,

Modderfontein, Conrath 731 (GZU, holo.!, PRE, photo.). A. lydenburgense Poelln. in

Bol. Soc. Brot. 15, 2: 73 (1942). Type: Transvaal, Lydenburg, Wilms 1538 (B, holo.!,

PRE, photo.). A. wilmsii Diels ex Poelln. in Fedde, Rep. 53: 131 (1944). Type:

Transvaal, Lydenburg, Wilms 1502 (B, holo. ! L, iso., PRE, photo.). A. wilmsii E>'\q\s

ex Burtt Davy and Pott-Leendertz in a First Check list of the Flowering Plants and

Ferns of the Transvaal and Swaziland, Ann. Transv. Mus. 3: 134 (1912) nom. nud

Plants up to 60 cm high, gregarious or solitary. Roots many, thin, spreading,

with some tubers near the tips. Rhizome horizontal, woody, knobby, covered with

fibres from old leaf bases. Leaves linear to filiform, 5^0 cm long and 1-5 mm broad,

flat or usually folded, soft or coriaceous, ribbed, glabrous or with short, stiff hairs, on

the ribs; base vaginate, dilated and with a ciliate margin; outer leaves very small.

Inflorescence a simple raceme or with some divaricate basal branches or an interrupted,

spike-like, branched inflorescence, overtopping the leaves; scape terete, seldom

compressed, often ribbed, lowest bract slightly larger than the upper floral bracts,

glumaceous, subulate, closely folded, amplexicaul, varying in length from 6-12 mm,

ribbed, light or dark; pedicel up to 1 cm in fruit, articulated below the middle. Flowers

congested or laxly arranged, perianth with segments ca. 14 mm long; filaments rough

in upper half; ovary with ca. 14 ovules per cell. oblong, 9 mm. typical.

Flowering Period: October-March.

Distribution: Eastern and south-western Transvaal, Bechuanaland, South West

Africa, Orange Free State, northern Cape, Basutoland; a common grassveld species.

Apparently a variable, adaptable species with a wide distribution. The following

forms were distinguished but it was not possible to maintain them as distinct varieties.

682

(a) Typical form. Found near the Vaal River from Vereeniging to Barkly West,

apparently in moist surroundings. It is a soft leaved, gregarious species with usually

a simple inflorescence and small, soft, whitish bracts.

Transvaal. — Benoni: Benoni, in water or on its edge, Bradfield 349 (PRE).

Orange Free State. — Fauresmith, Kies 334 (PRE). Boshoff : Smitskraal, Bunt Davy

(PRE, 12904b).

Cape. — Kimberley: sandveld near Riverton, Acocks 2260 (PRE). Barkly West:

Schietpan Kalk, Acocks 1645 (PRE). Vryburg: Armoedsvhkte, Mogg 8023 (PRE).

(b) Highveld form. Probably as a result of drought and cold the plants remain

small and bear few flowers. They equal A. subulatum Bak. and A. conrathii Bak. the

types of which were collected near Johannesburg.

Transvaal. — Heidelberg: Leendertz (TM 4687, PRE). Middelburg: Jenkins (TM

10181, PRE). Wakkerstroom: near Volksrust, Schweickerdt 621 (PRE).

Swaziland. — Mbabane: Ukutula, Compton 25252; near Mbabane, Compton 27548

(PRE).

(c) Bushveld form. Solitary, large, coriaceous plants with compound, interrupted,

spike-like inflorescences; it is found in the warmer parts of the Pretoria district to

Lydenburg. It is extremely common in grassveld. Often recorded from limestone

areas. They equal A. lydenburgensis Poelln. and A. wilmsii Poelln.

Transvaal. — Benoni: Bradfield 350 (PRE). Pretoria: University Farm, Pretoria,

Kies 364 (PRE); near Delmas, Codd, (PRE). Johannesburg: Milner Park, Moss

6040 (PRE J.).

(d) Western form. Gregarious, leaves narrow, shortly pubescent; flowers evenly

spaced on a simple rhachis. This equals A. hereroensis Schinz.

South West Africa. — Grootfontein: Rehm (M), Schoenfelder S 432 (PRE). Otji-

warongo: Volk 463 (M). Gibeon: Asis, Volk 721 (M).

Cape. — Mafeking, Brueckner 553 (PRE), Leistner 580 (PRE).

(p) Crisped form. Gregarious, leaves soft, fairly broad, margin often crisped,

ciliate; inflorescence often divaiicately blanched.

Cape. — Mafeking: Acocks 18778 (PRE); near Setlagoli, Kalahari Thornveld, Leistner

567 (PRE). Vryburg: Armoedsvlakte, Sharpe (PRE 7399), Foley (PRE 2760). Herbert:

Campbell, Acocks 1421 (PRE).

Transvaal. — Christiana: Kameelpan, Theron 433 (PRE), 612 (PRE); “ Kaffraria ”

near Christiana, Burn Davy (PRE, 12774). Pretoria: near Pienaar’s River Station,

Codd 6222 (PRE).

8. A. trichophlebium Bak. in FI. Cap. 6: 382 (1896). Type: Transvaal, Pretoria,

hills above Apies River, Rehmann 4314 (K, holo.).

A. pretoriense Bak. in Bull. Herb. Boiss. Ser. 2, 1: 780 (1901). Type: Transvaal,

Pretoria, hills above Apies River, Rehmann 4314 (Z holo. ! PRE, photo).

A. vaginatum Bak. in Bull. Herb. Boiss. II, 4: 997 (1904). Type: Transvaal, Irene,

Conrath 1243 (GZU, holo! PRE, photo.). In the publication no number is mentioned

but on the type sheet this is given as 1243.

Small, pubescent plants up to 20 cm high. Roots many, hard, wiry, 1-2 mm in

diam. Rhizome woody, horizontal, covered with fibres fiom persistent leaf bases.

Leaves ca. 8, distichous, primary small, ovate, becoming progressively larger and

ovate-lanceolate to oblong-oblanceolate, ca. 1 1 cm long, 3 ■ 5 cm wide, acute at the

apex, narrowing towards the clasping base, lamina firm, ribbed, setose with wlrite

patent, velvet hairs. Inflorescence a simple, congested, many flowered raceme, rarely

with a short basal side-branch; scape flattened, sturdy, ribbed, pubescent, 10-20 cm

long; bracts lanceolate-acuminate, up to 3 cm long; pedicels articulated below the

683

middle, up to 8 mm in fruit. Flowers congested, with the perianth greenish-white,

segments ca. 1 cm long; filaments with unequal short papillae in upper half; ovary

with ca. 14 ovules per cell. Capsule typical, 8 mm long. Seeds typical.

Flowering Period : September-March.

Distribution; Transvaal; found in the Pretoria, Johannesburg, Rustenburg

and Waterberg districts, in grassveld, not very common.

Transvaal. — Pretoria: Mogg 16323A (PRE); Meintjies Kop, Leendertz 912 (PRE).

Johannesburg; Bryanston, GillUand (J 28093, PRE). Rustenburg: Swartruggens,

Sutton 793 (PRE). Waterberg: near Nylstroom, Hdfstrom & Acocks 207 (PRE).

9. A. whytei Bak. in FI. Trop. Afr. 7: 493 (1898). Type: Nyasaland, Mount

Zomba, Whyte (K, holo.).

A. friesii Weimarck in Bot. Not. Lund 1937; 422, photo, p. 423. Type: Southern

Rhodesia, Inyanga near Cheshire, Norlindh & Weimarck, 4330 (ED, holo., PRE,

SRGH, iso.).

Plants up to 1 m high. Roots many, thin. Rhizome small, creeping, covered

with fibres from old leaf bases. Leaves about 8, becoming progressively larger, the

largest pair with a lanceolate lamina, 30 cm long, 3-5 cm broad, tapering to an acute

apex and narrowing at the base into a long “ petiole ” up to 18 cm long, closely ribbed

with the raised ribs densely and minutely pubescent, margin narrow, yellow. Inflore-

scence a simple, congested raceme, seldom with 1-2, short, basal ascending branches;

scape flattened, narrowly winged, glabrous, but for the ciliate wings, often with a small,

sterile, adpressed bract near the apex; fertile bracts small, ovate, acute, up to 8 mm

long, membranous; pedicels up to 1 cm in fruit, articulated below the middle. Flowers

congested with the white perianth segments 12 mm long, the apex of the outer segments

dark and papillate; filaments glabrous, 4 mm; anthers 10 mm; ovary with ca. 18-24

ovules per cell; style ultimately longer than perianth. Capsule typical, closely ribbed.

Seeds typical.

Flowering Period : November-March.

Distribution: Northern South West Africa, Southern and Northern Rhodesia,

Portuguese East Africa, Nyasaland. Usually in moist grassland in Mopani woodland.

South West Africa. — Okavango Native Territory: Okavango River near Masari

Camp, de Winter 4089 (PRE).

Southern Rhodesia. — Victoria Falls, Rogers 5624 (GRA). Que Que, McLeod 30

(PRE). Salisbury, Brain 9784 (SRGH).

Northern Rhodesia. — Mazabuka, near Chirundu Bridge, Drummond 5410 (SRGH);

Mazabuka, Vet. Officer, C.R.S. 587 (PRE); Livingstone-Maramba Road, Young

1091 (PRE, SRGH).

10. A. rangei Engler & Krause in Engl. Bot. Jahrb. 45: 125 (1910); Poelln. in

Fedde, Rep. 52: 260 (1943). Type: South West Africa, Diamond area II, Aus, Range

157 (B, holo., probably destroyed).

A. scariosum Duthie in Ann. Stell. Univ. 4: 15, t5 (1926); Ann. Bol. Herb. 4: 136

(1928). Type: Cape, Stellenbosch, Duthie 1818 (STE. holo., J. iso.!). A. tubiferum

Dinter in Fedde, Rep. 29: 263 (1931); Fedde, Rep. App. 23: 57 (1923) nom. nud. ;

Poelln. in Fedde, Rep. 52; 242 (1943). Type: South West Africa, Klinghardt Moun-

tains, Dinter 4001 (B, lecto.l, PRE, iso.! photo.). A. rigidifolium Poelln. in Bol. Soc.

Brot. 16, 2: 48 (1942). Type: Cape Town, Claremont, Schlechter 524 (B, holo.!

PRE, photo.).

Grasslike, wiry plants up to 40 cm high. Roots numerous, wiry, often with many

scattered watery tubers. Rhizome lateral, thickly covered with fibres from old leaf

bases. Leaves many, rosulate, erect, straight or spirally curled, linear or filiform 5-20

cm long, 1-2 mm wide, dilated and membranous at the base, glabrous, wiry, margin

684

mirutely cartilagino-dentate. Inflorescence simple or with some ascending branches,

laxly flowered; scape terete, bracteate, glabrous; floral bracts fascicled, membranous;

pedicels up to 10 mm long in fruit, articulated below the middle. Flowers 1-4-nate,

perianth spreading, segments ca. 10 mm long; filaments smooth; ovary ovoid with

ca. 10 ovules per cell, style glabrous, 6 mm long. Capsule ca. 8 mm long, ovoid, beaked,

transversely rugose. Seeds ca. 2 mm in diam. tetrahedral.

Flowering Period: September-March (in the Cape from January to March,

in leaf May-November; those from the southern Namib were in flower in September).

Distribution : Southern South West Africa, the south-western Cape to the Cape

Peninsula. “ In gravelly soil, rare ” Duthie.

South West Africa. — Diamond Area I: Buchu Mountains, Peilberg, on covered

dunes, Dinter 6506 (B).

Cape. — Namaqualand: Richtersveld, Kawarass, Marloth 12424 (PRE). Calvinia:

Lokenburg, Acocks 17553 (PRE). Clanwilliam: Smith 2617 (PRE). Caledon: Rivier

Zondereinde, Schlechter 9884 (B, PRE). Hills west of Baths, Purcell 57, 58 (NBG).

Bredasdorp: between Bredasdorp and Malagas, Lewis 3304 (NBG). Duthie also

cites: Cape, Rondebosch, Wolley Dod 2484 (BOL.) and specimens collected by Guthrie.

As the type of A. rangei was not at Berlin, I had to rely on the description and

distribution to identify the species. Specimens collected by Dinter near the type

locality of A. rangei and agreeing with the description, made it reasorably certain that

it must have been this species. In their description Engler & Krause give the measure-

ments of the pedicel and the perianth segments as 4 mm. This is much smaller than

those of A. rangei and near to the small flowers of A. krauseanum, but they may have

measured immature buds. A. krauseanum has not been recorded so far south.

11. A, krauseanum Dinter in Engl. Bot. Jahrb. 48: 354 (1912); Poelln. in Fedde,

Rep. 52: 238 (1943). Type: South West Africa, Damaraland, Windhoek, Skaap

River, Dinter 1913 (B, holo., probably destroyed; NBG, iso.!, PRE, photo.).

A. rigidiim Krause in Engl. Bot. Jahrb. 48: 353 (1913) e descr. ; non Baker (1872);

nec de Wild. (1913). A. rigens Poelln. in Fedde, Rep. 50: 232 (1941), nom. nov. for

A. rigidum Krause', Poelln. in Fedde, Rep. 52: 256 (1943). Type: South West Africa,

Auas Mountains, Dinter 1894 (B, holo., probably destroyed). A. kyllingioides Krause

in Engl. Bot. Jahrb. 51: 441 (1914); Poelln. in Fedde, Rep. 52: 239 (1943) e descr.

Type: South West Africa, Hereroland, Kaistfeld, Seiner 661 (B, holo., probably

destroyed). A. brachyphyllum Suess. in Mitt. Bot. Staatssam. Muenchen, I: 49 (1950).

Type: South West Africa, Rehoboth, Buellspoort, Rehm, (M, holo.!, PRE, photo.).

A. durum Suess. in Mitt. Bot. Staatssam. Muenchen, I: 49 (1950); non Poelln. 1941.

A. suessenguthii Solch in Mitt. Bot. Staatssam. Muenchen, II: 176 (1956), nom. nov.

for A. durum Suess. A. pimgens PoeWn. in Fedde, Rep. 52: 241 (1943) e descr. Type:

South West Africa, Hereroland, Brakwater, Dinter 1510 (B, holo., probably destroyed).

Plants up to 55 cm, hard, glabrous, grasslike. Roots many, thin, 1-2 mm in diam.

Rhizome horizontal, small, covered with flbres of old leaf bases. Leaves stiff, rigid,

erect or falcate, flat, linear, 20-30 cm long, 1-3 mm wide, margin and midrib raised.

Inflorescence a branched, laxly flowered panicle; scape terete, arcuate at the base,

bracteate with small bracts; floral bracts minute, fascicled; pedicels filiform, up to

6 mm in fruit, articulated near the base. Flowers 1-3-nate, with perianth segments

5 mm long; filaments rough; ovary ovoid. Crr/ww/c small, 2 mm high, usually broader

than long, tipped by the persistent style-base, transversely rugose. Seeds typical.

Flowering Period : December-January.

Distribution: Known from the northern part of South West Africa, usually

in sandy scil.

South West Africa. — Windhoek: Khomas Plateau, Friedenau, Gassner 209

(M); Voigtland, Volk (M 11335); Auas Mountains, farm Kromhoek, Merxmiiller

685

793 (M). Rehoboth: Gams Mountains, Merxmiiller 955 (M). Grootfontein : Otavi

Mountains, Rehm (M); Grootfontein, Dinter 806 (NBG); Tsumeb, Naegelsbach 66

{M); Auros, 5975 (M); Dinter 5609 (B). Ovamboland: Rehoboth, Fo/^ 2519 (M).

Seiner noted that the Herero na.me was ehosuo-tshisumba, i.e. “ make-up grass ”,

as the Hereros grind the roots and then mix it with fat to make an ointment which

they rub on their bodies. But in the description of the type plant {Seiner 661), Krause

describes the roots of A. kyllingioides as thin. {A. kyllingioides is here considered

to be a synonym of A. krauseanum). It is possible that this species, like A. rangei,

mentioned earlier, has tubers near the root tips, but so far these have not been seen

on specimens in the herbarium. It is also possible that there is a mistake somewhere.

The following species, A. calyptrocarpum, with its swollen roots would seem a more

likely source for an ointment ingredient.

12. A. calyptrocarpum Bak. in Trans. Linn. Soc. Ser. 2: Bot. 1, 258 (1878); Fl.

Trop. Afr. 7: 480 (1898). Type: Angola, Huilla, between Mampulla and Lopollo,

Welwitsch 3786 (BM, holo., PRE, photo.).

A. curvifolium Krause in Engl. Bot. Jahrb. 48: 354 (1913) e descr; Poelln. in Fedde,

Rep. 52: 258 (1943). Type: South West Africa; Auas Mountains, Dinter 1878 (B,

holo., probably destroyed). A. caespitosum Dinter in Fedde, Rep. 29: 267 (1931);

Poelln. in Fedde, Rep. 52: 234 (1943). Type: South West Africa, Grootfontein,

Otavi, Dinter 5284 (B, holo.!, PRE, iso.!).

Grasslike plants variable in size, 10-60 cm high. Roots fairly short, uniform,

fleshy, many. Rhizome small, horizontal, covered with fibres from old leaf ba.ses.

Leaves many, filiform, 10-30 cm long base, dilated and membranous, fascicled; pedicels

up to 8 mm long in fruit, articulated below the middle. Flowers 1-4-nate, with perianth

segments rotate; filaments smooth; ovary with ca. 8 ovules per cell. Capsule XmhmaXt,

5 mm long, glabrous, smooth, often capped with the withered perianth. Seeds typical,

0-75 mm long.

Flowering Period: December-March. Flowers open in the afternoon.

Distribution: Angola, Southern Rhodesia, South West Africa, Transvaal; usually

in dry grasslands, also in vleis.

South West Africa. — Okavango Native Territory: Andara, Merxmiiller 2080 (M,

PRE). Kaokoveld: Ondongua, Barnard (SAM 44167, NBG); Namakunde, Volk

997 (M), Barnard 583 (NBG). Grootfontein: Omaue, Volk 2849 (PRE). Rehoboth:

Goellschau, Volk (M 11496). Windhoek: Binsenheim, Volk (M 11034); Voigtland,

Volk (M 11399).

Southern Rhodesia. — Inyanga: near Inyanga town, Norlindh & Weimarck 4140

(PRE, SRGH). Salisbury: Marandellas, Wild 3292 (SRGH). Umvukwes: Darwin,

Umsengedzi River, Wild 3984 (SRGH). Umtali: Dora Ranch, Chase 6362 (SRGH).

Transvaal. — Pretoria: Bronkhorstspruit, Repton 1214 (PRE). Witbank: Gilifillan

7263 (PRE). Waterberg: near Vaalwater, Meeuse & Strey 10410 (PRE, BM).*

The species is apparently very adaptable and this could be the reason for its wide

distribution. Usually it is found in dry grassveld but Prof. Merxmiiller collected his

specimens in a vlei, the flowers floating on the surface of the water.

Baker describes the perianth with a circumscissile area at the base; it breaks off

irregularly however, wherever pressure is greatest. The perianth may persist on top

of the capsule, which is unusual for this genus.

A species closely related to A. calyptrocarpum Bak., is the Angolan species, A.

tenellum Welw. ex Baker in J. Linn. Soc. Ser. 2, Bot. 1: 256 (1878); type: Angola,

Morro de Monino, Welwitsch 3796 (BM, holo.). Mr. John Lewis of the British Museum

who compared the types, found A. tenellum to be a more slender species.

* TANGANYIKA. — Seronera to Kleins Camp, mile 57, Greenway & Turner 9997 (PRE).

686

13. A. acutum Wright in Kew Bull. 1914, 170 (1914). Type: Natal, Richmond,

Ensikeni, Haygarth in Hb. Wood 12063 (K, holo., NH, PRE, iso.!).

Glabrous plants up to 80 cm high. Roots many firm, ca. 4 mm in diam. Rhizome

small, usually covered with fibres of old leaf bases. Leaves subdistichous, ca. 8, becoming

progressively longer, linear-lanceolate, up to 50 cm long, 1-2 cm broad, tapering

gradually to the apex, clasping at the base, thin, flat, ribbed. Inflorescence a sub-simple,

lax raceme or with 1-2 short, basal branches; scape terete, bracteate with 1-3 sterile,

leaf-like bracts, fertile bracts resembling the sterile bracts but becoming smaller;

pedicels very short in bud, up to 8 mm in fruit, articulated near the middle; flowers

1-2-nate, the secondary flower (or flowers) sometimes borne on contracted, bracteate

side branches. Flowers with perianth segments up to 17 mm long, with a broad 5-ribbed

green keel; filaments glabrous, slightly dilated at the base, about 8 mm long; ovary

narrow, oblong, 4 mm high, ca. 20 ovules per cell. Capsule (immature) cylindrical,

3 grooved, 15 mm long, 5 mm in diam., green, coriaceous, with transverse wrinkles,

the persistent style-base forming an apiculus. Seeds (immature) black, angular with

a well developed funicle.

Flowering Period: December, probably short.

Distribution: Southern Natal, north eastern Cape.

Natal. — Richmond: Hutchinson 1829 (PRE, GRA); Underberg: Drakensberg

Garden, Werdermann 1421 (B). Giant’s Castle: Symons (TM 25296, PRE).

Cape. — Mount Currie: Kokstad, Hutchinson 1829 (PRE).

This species is of interest as it shows a link between the truly distichous species

with several fiowers in the axil of an outer bract and those that are rosulate and have

only one flower in the axil of each outer bract. The arrangement of the bi-nate flowers

shows that it is the vestige of a contracted raceme, for the second flower is sometimes

situated on a peduncle bearing small bracts. Fruiting material was collected for the

first time by Werdermann 1421 (B) at Underberg. Although the capsules were not

yet quite ripe, they showed a resemblance to those of A. haygarthii. The seeds although

shrunken and immature, showed many folds and would probably become angular

when ripe.

14. A. longistylum Bak. in J. Linn. Soc. 15: 305 (1876); FI. Cap. 6: 381 (1896).

Type: Transvaal, Baines (K, holo.).

A. polyphyllum Bak. in FI. Cap. 6: 383 (1896). Type: Transvaal, Barberton, Galpin

1149 (K, holo., PRE, GRA, iso.!). A. recurvifolium Bak. in Kew Bull. 1906: 28.

Type: Southern Rhodesia, near Salisbury, Cecil 143 (K, holo.). Fig. 2.

Glabrous plants up to 100 cm high. Roots numerous, fibrous, firm, up to 3 mm in

diam. Rhizome woody, usually covered with fibres from old leaf-bases. Leaves many,

rosulate, erect, or sometimes curled back, linear, up to 30 cm long, 2-8 mm broad,

long tapered in upper half, clasping below, folded, ribbed, margin minutely papillate.

Inflorescence a lax ascending panicle, sometimes simple, many flowered with the solitary

flowers subtended by a bract and bracteole; scape terete, bracteate; basal sterile

bracts leaf-like; fertile bract small, ovate-acuminate, auriculate, bracteole smaller,

situated obliquely opposite bract, 2-cleft to emarginate with the apices usually unequal :

pedicels short, up to 4 mm, stout and erect in fruit, articulated below the middle.

Flowers with the perianth segments lanceolate, 13-16 mm long, reflexed when open;

stamens declinate, 4 erect, 2 decumbent with style, filaments short, glabrous, flattened;

anthers large, 7 mm; ovary ovate, acute with 9-30 ovules per loculus, style exceeding

the perianth. Capsule erect, ovate, beaked, 13 mm long, 6 mm in diam., glabrous,

transversely rugose, apiculate. Seeds typical.

Flowering Period: November-March.

Distribution: Natal, Transvaal, Bechuanaland, Southern Rhodesia; diy bushveld,

rocky slopes or sandy flats.

687

Fig. 2. — Anthericum longistylwn Bak. a, habit showing rosulate leaves x -J. b, capsule, X 2. c, outer

auriculate bract and inner emarginate bracteole, X 2. d, flower.

688

Natal. — Bergville: Killick 1043 (PRE).

Transvaal. — Barberton: Galpin 408 (PRE). Nelspruit: Kruger National Park,

Pretorius Kop, Codd 5664 (PRE, SRGH). Pilgrim’s Rest: Rogers 18238 (PRE);

Killick & Strey 2558 (PRE). Pretoria: Premier Mine, Rogers 25033; Hammanskraal,

Kies 367 (PRE), Codd 3493 (PRE). Letaba: Tzaneen, Phillips 3294 (PRE). Waterberg:

near Maraheki, farm Waterval 6, Meeuse & Strey 10381, 10383 (PRE).

Cape. — Mafeking: Brueckner 453 (PRE).

Bechuanaland. — Near Derdepoort, Codd 8893 (PRE), 8862 (PRE, SRGH). Gabe-

rones, Van Son (TM 28665, PRE). Mahalapye, de Beer 555 (SRGH).

Southern Rhodesia. — Rusape, Dehn R 59 (SRGH). Bulalima-Mangwe: Embakwe,

Feiertag 45411 (SRGH). Salisbury: Beatrice, Drewe A1 (SRGH). Matobo, West

2447 (SRGH).

The types of A. longistylum and A. polyphyllum were compared at Kew by Mr.

W. Marais who found them to be conspecific. The filaments of A. polyphyllum are

glabrous and not scabrous as described. Baker describes the flowers of A. longistylum

as 2-3-nate, but this is somewhat misleading for the flowers, although sometimes found

close together, do not emerge from the axil of one bract. The part of the scape present

on the type of A. longistylum is bare but this is unusual. The leaves of some speci-

mens are very narrow.

15. A. radula Bak. in Bull. Herb. Boiss. Ser. 2, 1: 781 (1901). Type: Transvaal,

Pietersburg, Houtbosch, Rehmann 5805 (Z, holo.!, PRE, photo.).

Plants resembling A. longistylum, but with the inflorescence glandular-scabrid.

Roots many, long, firm. Rhizome short, covered with fibres from old leaf bases. Leaves

rosulate, many, linear, ca. 20 cm long, 4 mm wide, folded, glabrous, rigid, ribbed.

Inflorescence branched, glandular-scabrid, 35 cm high; scape firm, terete, bracteate,

glabrous at the base, scabrid with stipitate glands above; lower sterile bracts large,

leaf-like; each solitary flower subtended by a bract and bracteole; bract ovate,

membranous, bracteole smaller, unequally bifid; pedicels 3 mm long. Flowers With

perianth segments 16 mm long, 2 mm broad; filaments smooth, 4 mm long; anthers

4 mm; ovary ovoid, apex acute, style exserted, declinate. Capsule unknown.

Flowering Period: Not stated, probably December.

Distribution: Transvaal, Houtbosch, only known from the type locality.

Transvaal. — Pietersburg: Houtbosch, Rehmann 5805 (Z, holo.!, PRE, photo.).

The species very closely resembles A. longistylum and may eventually prove to be

a glandular variety cf this species.

16. A. transvaalense in FI. Cap. 6: 384 (1896). Type: Transvaal, Barberton,

Galpin 1035 (K, holo., PRE, NH, GRA, iso.!).

A. indutum Poelln. in Fedde, Rep. 50: 232 (1941), nom. nov. for A. vestitum Bak. ex

Schinz in Viert. Nat. Ges. Zurich, 49: 174 (1904); non Bak. in J. Linn. Soc. 15: 307

(1876). Type: Transvaal, Pietersburg, Shiluvane, Junod 862, 1441 (K, holo.).

Plants up to 80 cm high. Roots numerous, firm, ca. 2 mm in diam. Rhizome

woody, horizontal, covered with fibres of old leaf bases. Leaves rosulate or semi-

distichous, straight or in a loose spiral, erect, 5-35 cm long, up to 6 mm broad, closely

ribbed (midrib absent) flat, setose with long, white, patent hairs. Inflorescence glutinous,

simple, seldom with a short basal branch, many flowered, elongating during anthesis;

scape terete, pubescent, bracteate above, sterile bracts pubescent, leaf-like but smaller

and narrower; each solitary flower supported by a large glabrous ovate, acute, auriculate

bract and a bracteole with 2 long subulate horns; pedicels up to 3 mm in fruit, articulated

in the middle. Flowers closely arranged on rhachis with bracts imbricate; perianth

glutinous, segments 13 mm long, filaments short, anthers 5 mm long; ovary ovoid,

each cell with ca. 15 ovules. Capsule ovoid, beaked, transversely rugose. Seeds typical.

689

Flowering Period: November- April.

Distribution: Orange Free State, Swaziland, Transvaal, Southern Rhodesia,

usually in grassveld on rocky outcrops.

Orange Free State. — Senekal: Ferrara, iron stone hills, Goossens 984 (PRE).

Swaziland. — Stegi, Isateki Beacon, Compton 27297 (PRE).

Transvaal. — Belfast: .6o/m5 12385 (PRE). Benoni: Bradfield hllh {?KE). Krugers-

dorp: Muldersdrift, Gilliland (J, 26103, PRE). Middelburg: Mogg 17314 (PRE).

Lydenburg: Steenkampsberg, Codd 8222 (PRE, SRGH). Pietersburg: Woodbush

Hill, Pott 4730 (PRE); Haffenden Heights, Junod 4120 (PRE).

Southern Rhodesia. — Inyanga, Hopkins (SRGH 9448); W/iellan 603 (SRGH).

Like A. longistylum Bak. and A. haygarthii, this species also has narrow-leaved

forms.

17. A. haygarthii (Wood & Evans) Kies comb. nov.

CMorophytum haygarthii Wood & Evans in J. Bot. Lond. 37: 254 (1899). Type:

Zululand, ’Nkandla, Haygarth in Hb. Wood 7448 (NH, holo.!, PRE, iso.).

Anthericum bracteatum Thode ex Poelln. in Fedde, Rep. 53, 6: 126 (1944). e descr.

Typj: Natal? Transvaal? “ On stony kopje and plateau, 2100 M ” Thode 3500,

flow. January 1921 (B, holo., probably destroyed).

Plants up to 80 cm high. Roots many, firm, stout, 3 mm in diam. Rhizome small,

woody, obliquely horizontal, covered with fibres of old, persistent leaf-bases. Leaves

semi-distichous, linear to lanceolate, 20-80 cm long, up to 14 mm broad, tapered in

upper half, base clasping, ribbed, glabrous, firm, flat. Inflorescence a simple raceme,

many and closely flowered, with bracts imbricate; scape terete, stout, glabrous, with

several leaf-like sterile bracts; rhachis pubescent and viscous above; each solitary

flower with a membranous ovate bract and an inner oblique boatshaped bracteole,

which covers the bud; pedicels erect, stout, 5 mm long in fruit, articulated in the

middle. Flowers with perianth segments 1 5 mm long; stamens with glabrous filaments;

ovary ovoid, with ca. 12 ovules per cell. Capsule ovoid, transversely rugose, sutures

prominent, apex beaked. Seeds typical.

Flowering Period: November-February.

Distribution: Natal, eastern Transvaal, stony slopes and swamps.

Zululand. — Ubombo: Gerstner (NH 22874).

Natal. — Vryheid: Galpin 10198 (PRE). Paulpietersburg: Dumbe Mountain, Galpin

9654 (PRE).

Transvaal. — Ermelo: Mavieriestad, Pott 5173 (PRE). Wakkerstroom : van Dam

(TM 24339, PRE); Thode (NH 16488). Barberton: Schagen, slopes of Amajuba

Mountain, Liebenherg 3105 (PRE). Pilgrim’s Rest: Mauchsberg, Sabie, Smuts &

Gillet 2384 (PRE); Graskop, Codd 6736 (PRE). Belfast: Dullstroom, Galpin 13045

(PRE).

A. transvaalense and A. haygarthii are closely related.

Species Excluded *

A. brevicaule Bak. in Journ. Linn. Soc. XV: 298 (1876); FI. Cap. 6: 391 (1897)'

Type: South Africa without locality, Thunberg (UPS?). Caesia brevicaulis (Bak.)

Dur. & Schinz, Consp. FI. Afr. V: 353 (1895). Identified as Caesia contorta (L.f)

Dur. & Schinz.

A. brevifolium Thunb. Prodr. 62 (1794). Bak. in Fl. Cap. 6: 388 (1897). Type:

South Africa, without locality, Thunberg (UPS, holo., PRE, photo.). Identified as

Caesia contorta (L.f.) Dur. & Schinz.

* See also p. 759.

690

A. longipedicellatum Poelln. in Fedde, Rep. 53; 128 (1944). Type: Cape, Rivers-

dale, Rust 14 (B, holo.!, PRE photo.). Identified as Ornithogalum zeyheri Bak.

Bak. in Viert. Nat. Ges. Zurich 49: 175 (1904). Type: Transvaal,

Letaba, Spelonke, top of Mamotsuiri, Junod 1448 (G, holo.!, PRE, photo.). Identified

as an Albuca sp.

A. muhiceps Poelln. in Bol. Soc. Brot. 16, 2; 68 (1942). Type: Cape, Riversdale,

Rust 237 (B, holo.!, PRE, photo.). Identified as Bulbine frutescens Willd. var.

A. nonscriptum Poelln. in Bol. Soc. Brot. 16, 2: 45 (1943). Type: Ex hortus

Berlin-Dahlem, 1826-7 (B, holo., probably destroyed in 1943). Poellnitz thought it

originated from the Cape. Identified by Kies as A. ramosum L. from Europe.

A. pauper Poelln. in Bol. Soc. Brot. 16, 2: 47 (1942). Type: Transvaal: Heidelberg.

Wilms 1542 (B, holo.!, PRE, photo.). Identified as Ornithogalum leptophyllum Bak.

A. quadrifidum Poelln. In Bol. Soc. Brot. 16, 2: 47 (1942). Type: Cape: Pondo-

land. Port Grosvenor, Bachmann 276 (B, holo.!, PRE, photo.). Identified as Ornitho-

galum leptophyllum Bak.

A. seineri (Engl. & Krause) Poelln. in Fedde, Rep. 53: 136 (1944). Identified as

Ornithogalum seineri {Engl. & Krause) Oberm. cf. Bothalia 7: 401 (1961).

A. unilaterale Thunb. nom. tant. ex Juel, Plant. Thunb., 121 (1918), (UPS, PRE,

photo.). Possibly Eriospermum cernuum Bak.

For species excluded from Anthericum before 1897 the Flora Capensis vol. 6 should

be consulted.

2. CHLOROPHYTUM

Ker Gawler in Bot. Mag. t.l071 (1808); Kunth, Enum. 4: 602 (1843); Baker in J.

Linn. Soc. 15; 321 (1876); Benth & Hook. Gen. PI. 3, 2: 788 (1883); Krause in Engl.

& Prantl, Pflanzenfam. 15, A; 284 (1930); Phillips, Gen. S.A. Flow. PI. ed. 2: 183

(1951).

Asphodelopsis Steud. ex Bak. in J. Linn. Soc. 15: 321 (1876).

Hartwegia Nees in Nov. Act. Nat. Cur. 15, 11: 372 (1831).

Herbaceous perennials the parts above ground persistent or dying down in winter.

Roots variable, usually long, many, swollen, rarely wiry and with scattered tubers,

in some south western Cape species, with sessile “ tubers ” (arrested roots) on the

rhizome. Rhizome creeping, sometimes covered with fibres (the remains of old persistent

leafbases). Leaves rosulate, rarely subdistichous, a few or many, rarely 1 ; lamina

usually flat, glabrous or pubescent, rarely glandular, base often tapered into a “ petiole ”,

margin usually fimbriate. Inflorescence a central, simple or branched raceme, rarely

glandular-pubescent; scape terete, bracteate ; bracts large or small; pedicels articulated

near or below the middle, rarely at the apex. Flowers 1-6-nate, opening consecutively,

open all day. Perianth rotate or reflexed, rarely cup-shaped or urceolate, white, rarely

pinkish, translucent, keel becoming dark with age; the 3 outer segments slightly

narrower than 3 inner; marcescent, covering capsule, rarely reflexed in fruit. Stamens 6,

hypogynous; filaments glabrous or papillate, shorter, or in certain species, longer

than the segments; anthers basifixed, usually large, introrse. Ovary sessile or shortly

stipitate, trigonous with 6-30 biseriate ovules: style simple, smooth; stigma minute,

penicillate. Capsule trigonous, oblong, obcordate, or globose in outline, smooth or

with transverse ridges or tuberculate. Seeds large, flat, round in outline with a notch

where the pointed hilum is situated, testa black shiny, minutely granulate.

Distribution: Africa, Madagascar, Asia.

Type Species: Chlorophytum inornatum Ker Gawler.

691

Like Anthericum, Chlorophytum is a widespread genus containing nearly 300

species most of which are found in tropical Africa, with some in Madagascar and a few

in Asia. In South Africa the twenty species could be divided into two sections which

coincide with their geographical distribution, namely nine from the winter rainfall

area and therefore confined to the south-western Cape, and eleven from the summer

rainfall region, these latter being more closely related to the tropical species. Chloro-

phytum inornatum is the type species of the genus published by Ker Gawler in 1804.

It comes from Sierra Leone, but is not found in Namaqualand as Baker led us to believe

in the Flora Capensis. On the whole our South African species are fairly uniform

and fit in well with the genus. There remains, however, the problem whether to accept

Dasystachys as a separate genus or as a section of Chlorophytum. For the present

I regard only those species which have single flowers, each supported by a bract and

bracteole, a simple raceme, non-articulated, very short pedicels and exserted stamens

as true Dasystachys. But Chlorophytum papillosum Rendle which reaches our northern

borders seems to form a link between these two genera. It possesses the exserted stamens

and the non-articulated pedicels but its flowers are arranged in clusters. Dasystachys

has the typical Chlorophytum type of capsule and seed and was considered to be a

section of it by Engler and von Poellnitz. Baker referred it to Dasystachys but I have

left it under Chlorophytum.

Geographical Notes. — On the whole the species are found to inhabit fairly

small areas. C. comosum has a wide distribution but it has the same range as the eastern

forests which it inhabits. With C. modestum and the subtropical species C. gazense

and C. laxum, etc., it forms a group of closely related, shade and moisture loving species.

C. papillosum also has a wide distribution in the Zambesi lowlands. C. capense,

C. bowkeri, C. krookianum and C. aridum are closely related and are found in more

open country where they occur in clumps in the shade of small thickets. Their leaves

are firmer and usually slightly fleshy; the more favourable the situation the broader

the leaf. The south-western Cape species are xerophytic. Why they do not dry black

and why the others from the summer rainfall area do, could not be explained. Dr.

P. H. B. Talbot examined the root of a specimen of C. bowkeri but he found no signs

of mycorrhiza.

Morphology

Roots: They are typical for each species or group of closely related species. The

roots are numerous, long, thin or evenly or unevenly swollen. Those of C. triflorum

are unique; they are hard and dark-coloured, swollen near the insertion to the rhizome

and tapering gradually to the tip. C. viscosum has pink roots which are narrowly

cylindrical; they become hollow when dry through shrinkage of the spongy mesoderm.

C. rigidum has thin, wiry roots which show affinity to certain species of Anthericum.

C. trachycarpum from northern South West Africa and Southern Rhodesia also has

thin roots but these bear tubers near the tips also reminiscent of Anthericum. C.

undulatum and C. crassinerve often possess sessile “ tubers ” on the rhizome. These

“ tubers ” are roots arrested in their development, an adaptation to dry, unstable

conditions. If the conditions become favourable, they will elongate. If they were

favourable from the start no “ tubers ” are formed, the roots then merely showing a

spindleshaped thickening near the tip. In most species, especially those from the

summer rainfall area, the mesoderm is spongy and there is a thin, inner, woody core.

The root hairs are extremely well developed attaining up to 2 mm in length; they

intertwine and form a matted surface. With age the root hairs and mesoderm disappear

leaving only the thin, hard core.

Rhizome: The perennial rhizome is small and compact forming new plants hori-

zontally.

692

Leaves: These show much variation especially in the south-western Cape species,

where they have adapted themselves to dry conditions. Here we get such xerophytic

adaptations as a reduction in the number of leaves, narrow lamina, pubescence, folding

or twisting and a hardening of the epidermis. The margin of the leaf is ciliate in nearly

all species. In some of the south-western Cape species the primary leaves remain small

and resemble ligulate squamae, often purple spotted. The hard sclerenchymatous

nerves of the leaf-bases persist as fibres in these species.

Inflorescence: It is a simple or branched raceme. Its flowering time is extended

over a long period, as the many buds in each axil develop consecutively. Each axillary

fascicle of flowers may be regarded as a reduced flowering branch. Occasionally in

C. crispum and C. capense, v/e still find a minute peduncle on which these axillary

flowers are arranged spirally. If flowering conditions are favourable, many flowers

are produced; if it is too cold or too dry, many buds remain latent or are retarded

so that in one axil a capsule, a flower and small buds may be found side by side. In

this way the inflorescence will flower for months and incidentally it is the only one

produced (centrally) during the season. This is in sharp contrast to Trachyandra and

Biilbine w'here the (axillary) inflorescence appears, flowers and fades within a short

period, perhaps a fortnight, whilst subsequent young racemes are to be seen near the

base, developing and flowering soon after a rainy spell. As the bracts may be regarded

as the equivalent of the leaves, the oldest and sterile ones found on the scape are the

largest and still very leaf-like; the floral ones are small. If the leaf shows a special

characteristic, such as long cilia on the margin, the floral bracts will also show this.

The pedicels are usually articulated near the middle, rarely near the base (C. rigiduin

and C. monophyllum) or near the apex (C. macrosporum and C. trachycarpum). If

not fertilized the flower drops off at the articulation. In those with the articulation

near the base, the upper part of the pedicel has elongated during anthesis; in those

with articulation near the apex, the lower part has developed. The location of the

articulations is constant for each species as in Anthericwn. The pedicels do not change

their position while the capsule ripens. This they have in common with Anthericwn

but it is in contrast to the genus Trachyandra and some species of Bidbine where the

position of the pedicel changes after the flower has dropped off if not fertilized, or when

a capsule is formed.

Flowers: They are white, consisting of delicate, many cornered, tr-anspaxent,

watery cells which reflect the light in all directions, giving it a crystalline, translucent

appearance. The keel is not visible at first, but appears when the tepals fade. The

perianth is usually rotate, sometimes reflexed straight back; in C. papilloswn it is

cupshaped; in C. macrosporwn the perianth is urceolate at the base with the segments

ligulate above the cup, not at all like the usual flower form. The flowers are open

all day and are apparently scentless. The perianth is persistent, either fitting closely

around the capsule or reflexed. The segments do not fuse at the top in fading as in

Trachyandra. The stamens have very delicate filaments which shrivel up unevenly

so that they sometimes have a rough surface when dry. In some species they are

scabrid but in the majority the surface appears smooth or, when magnified, with

retrorse, short papillae. They are usually shorter than the perianth except in C.

papilloswn where during anthesis, they elongate beyond the perianth cup. Apparently

the rate of growth varies in the filaments, some developing more slowly. The large,

introrse anthers are basifixed and curl backwards when fading. The ovary is usually

oblong to obcordate or globose in outline. The ovules are biseriate and vary from

6-11 in certain species whereas in others we find 20-30 per loculus. It may happen

that a whole row doubles itself so that for instance, 30 instead of the usual 20 ovules

may be found in a loculus. The style is simple and also elongates rapidly during

anthesis. The stigma is minute, penicillate.

693

Capsule: It is trigonous, the seeds are flattened perpendicularly in each compressed

compartment. The walls are usually smooth with a raised margin. In C. rigidum,

C. monophyllum and C. viscosum transverse ridges are present, like those found on most

Anthericum capsules. In C. trachycarpum the ovary is papillate, the papillae afterwards

developing into irregular tuberculate ridges on the capsule.

Seed: It is flat, large (2-4 mm in diam.) round in outline and notched where the

hard pointed hilum is situated; the testa is black, shiny, granula".

Vegetative reproduction occurs in all species through the formation of lateral

buds on the rhizome. C. comosum is of special interest for its propagation by leaf

tufcs found at the apex of the inflorescence. These sink down to earth and become

rooted.

Chromosome Number: The basic number of chromosomes was found to be seven

in C. krookianum and C. comosum and in the tropical species C. laxum and C. inor-

natum.

Key to Species

Species from summer rainfall area; filaments smooth (except C. crispuni)', plants usually drying

black:

Filaments shorter than perianth segments which are spreading or reflcxed:

Inflorescence patently branched, floral bracts shorter than the flowers;

Ovary sessile:

Scape compressed, oval in cross-section, glaucous, coriaceous plants; S.E. Cape 1. capense

Scape terete:

Small plants 20-50 cm high, leaves flat up to 10 cm long, margin distinctly crisped

fimbriate; filaments papillate; S.E. Cape 2. crispuni

Larger plants. Laves more than 10 cm long, usually half-folded, margin smooth or

minutely crisped fimbriate, filaments smooth; eastern Cape, Transvaal, Natal:

Leaves ca. 2 cm wide, usually half-folded, undulate; ovary with ca. 10 ovules in each

cell, capsule globose; Transvaal bushveld i.aridwn

Leaves ca. 6 cm wide, flat or somewhat folded; ovary with ca. 16 ovules per cell,

capsule oblong; large Aloe-like plants; eastern Cape to eastern Transvaal

4. krookianum

Ovary stipitate, perianth urceolate, pedicels very slender, up to 2-5 cm long; north-eastern

Transvaal, Southern Rhodesia 5. macrosporum

Inflorescence simple or with 1-2 long, basal, ascending branches:

Cansule smooth, roots swollen;

Capsule obcordate in outline; raceme simple, scape arcuate at the base thrusting the few

flowered raceme outside leaf-rosette; sm.all plants found in the warm eastern coastal

area 6. mode stum

Capsule globose:

Leaves linear 7. pulchellum

Leaves lanceolate, attenuated towards the base; raceme often with a leafy apical

tuft 8. comosum

Capsule oblong; raceme dense with flowers close together and the bracts usually larger than

the flowers 9. bowkeri

Capsule tuberculate; roots thin with small tubers near the tips; northern South West

Africa, Southern and Northern Rhodesia 10. trachycarpum

Filaments ultimately exserted from campanulate perianth; perianth glandular-papillate on

outside. Northern South West Africa and southern tropical Africa.... II. papillosum

Species from winter rainfall area; filaments rough; plants not drying black; rhizeme often with

fibres :

Inflorescence divaricately branched, seldom simple; roots all alike, thin or slightly thickened;

leaves somewhat distichous, margin not fimbriate; fertile bracts small, clustered; pedicels

articulated near the base; capsule transversely rugose:

Plants glabrous, roots thin:

Leaves numerous 12. rigidum

Leaf solitary J3. monophyllum

Plants glandular, roots pink, slightly swollen 14. viscosum

694

Inflorescence simple, rarely with 1-2 basal, ascending branches; roots various; leaves in a

rosette; margin minutely fimbriate; fertile bracts usually large, often resembling the

perianth; pedicels articulated near the middle; capsule smooth with a raised margin:

Roots all alike, hard, thickest at place of insertion to rhizome, tapering gradually, with some

transverse wrinkles; Cape Peninsula and surrounding districts 15. triflorum

Roots various but not as above:

Leaves glabrous (except minutely ciliate margin):

Plants with many leaves:

Leaf margin not prominent, apex usually long tapering:

Leaves in an elongated rosette; roots hard, not much thickened; ovules ca. 18

per cell; Namaqualand 16. namaqueme

Leaves usually in a flat rosette; roots all alike or dimorphous i.e. with anchoring

roots which at first have a lanate, spongy skin which later sloughs off leaving a

thin, hard, woody core, and often with small sessile tubers on rootstock between

the old fibres 17. undulatum

Leaf margin raised, apex usually obtuse 18. crassinerve

Plants with one old and one young, green leaf; 1-2 short, membranous bract-like primary

leaves present at the base 19. pauciphyllum

Leaves pubescent, rhizome small, bulbous 20. lewisae

1. C. capense (L.) Voss in Vilmorin’s Blumeng. 1; 1086 (1895); O. Kuntze, Rev.

Gen. PI. 3: 316 (1898); Van Oostroom in Blumea 4: 495 (1941); Adamson & Salter

in Flora of the Cape Peninsula, 185 (1950), as “ C. capense (L.) Druce ”.

Asphodelus capensis L. Syst. Nat. ed. 10, 982 (1758). Type: Miller’s Ic. t.56 (1760).

Linnaeus describes the plant, “A scapo nudo, ramoso, fol. lanceolatis planis ” and

refers to this plate. The specimen No. 431-6, preserved in the Linnaean herbarium,

with a simple raceme and named Asphodelus capensis L. Syst. Nat. ed. lOA does not

represent this species. It was impossible to identify it however, from the indistinct

photo at PRE.

Anthericum rouwenortii de Gorter, Cat. Plant. Hort. Ulenp. 51 (1783); van Oostroom

in Blumea 4: 493 (1941). Type: probably Cape (L, holo.! PRE, photo.). It was

said to have come from Ceylon but this was already doubted at the time. It was in

cultivation at the Hortus Ulenpas in Holland round about the same time that

Anthericum datum Aiton was cultivated at Kew. They could have been raised from

the same batch of seed which must have been collected in the south-eastern Cape.

A. datum Ait. in Hort. Kew, 1: 448 (1789). Type: Cape, probably (BM, holo., PRE

photo.). Aiton refers to Miller’s plate 56.

Phalangium datum (Ait.) Redoute, Lil. t.l91 (1807).? P. fastigiatum Poir. in Lam..

Encycl. 5: 246 (1804). No type preserved. Redoute asserted that Poiret’s species

was not the same as Aiton’s A. datum but as no specimen exists, this cannot be verified.

P. “ fasciculatum ” Bak. in J. Linn. Soc. 15: 331 (1876) in error for P. fastigiatum

Poir.

Chlorophytum datum (Ait.) R.Br. ex Sprengel, Syst. II: 88 (1825); Roem. & Schult.,

Syst. 7: 454 (1829); Kunth, Enum. 4: 604 (1843); Saunders in Ref. Bot. 3, t.216

(1870); Bak. in J. Linn. Soc. 15: 330 (1876), et in FI. Cap. 6: 399 (1897).

Herbaceous, glaucous plants up to 1 m tall. Roots many, long fleshy. Rhizome

small, vertical. Leaves many, in an elongated rosette, linear-lanceolate 30-45 cm

long, 25 mm broad, gradually narrowed towards the apex, base clasped firmly around

rhizome and other leaves, closely ribbed, glaucous, coriaceous, glabrous. Inflorescence

a tall, much branched raceme with several to many ascending, lateral branches; scape

compressed, firm, with some leaf-like bracts, covered with a white bloom; fertile

bracts small, deltoid; pedicels short, up to 4 mm, articulated above the middle when

in fruit. Flowers laxly arranged, 3-many nate, if many, arranged in short, spiral,

bracteate fascicles; perianth rotate with segments 9 mm long; filaments glabrous

(minutely papillate when magnified); ovary with ca. 10 ovules per cell. Capsule

695

8 mm, broadly obovate in outline, apex deeply emarginate. Seeds typical, 3 mm

in diam.

Flowering Period : October-May.

Distribution; Known only from the eastern Cape, Humansdorp to Albany

districts.

Cape. — Humansdorp: Jeffrey’s Bay, Lewis 3608 (NBG), Fourcade 4321 (BOL). Uiten-

hage: Sandhills, Ecklon & Zeyher 111 (NBG). Port Elizabeth: Cradock Place, on

downs, Galpin 6448 (PRE, GRA); Redhouse, Paterson 1960 (GRA); Mackay Bridge,

Sundays River, Urton 240 (GRA); Addo, Drege 8719a (L), 8719 (K). Albany: bank

of Fish River, Penrock farm. Dyer 1188 (PRE). King William’s Town: Pirie, Sim

1 104 (PRE).

Miller’s plate, t.56, was made from a plant raised from seed in the Chelsea Garden

in 1751 ; it flowered in March the following year. Miller stated that part of the seed

parcel went to Holland. He goes on to say that although he could not find a reference

to it in any botanical books, it seems likely to have been cultivated formerly in the

Dutch gardens for it can be seen on flower paintings, which were painted up to sixty

years previously. Whether the flowers painted were really those of Chlorophytum

capense we do not know. The shape of the flower is so elementary that several other

species might also have been mistaken for it, e.g. Trachyandra hirsuta. After the 1751

consignment of seeds, he mentions that several more packets arrived and he therefore

supposed it to be quite common at the Cape. But he describes the seeds as angular,

which is not correct. The capsule on the plate is deeply grooved, therefore the seeds

inside must have been compressed, which is typical for Chlorophytum. Once again

he may have confused the seeds with those of a Trachyandra species.

The species is endemic in the eastern Cape where it has been collected in dry,

shady vegetation on river banks, often near the sea. Redoute’s plate (Lil. t.l91) and

that of the Ref. Bot. 3 : t.216 are more life-like. It is doubtful whether it ever occurred

on the Peninsula. Adamson says “ doubtfully native, rare under trees or by streams

on the Groote Schuur and Newlands Housing Estates ”. I think these Cape Peninsula

plants are better placed under C. comosum.

2. C. crispum (Thunb.) Bak. in J. Linn. Soc. 15: 331 (1876); FI. Cap. 6: 398

(1897).

Anthericum crispum Thunb. in Prod. 63 (1794); Thunb., FI. Cap. ed. Schultes, 324

(1823). Type: Cape, Thunberg (UPS, holo., PRE, photo.). Anthericum rustii Poelln.

in Fedde, Rep. 53: 127 (1944), e descr. Type: Cape, Riversdale, Rust 298 (B, holo.,

probably destroyed). Poellnitz says the type was leafless. It is almost certainly this

species. Anthericum rosenbrockii Poelln. in Bol. Soc. Brot. 16; 69 (1942). Type:

Cape, Port Elizabeth, Rosenbrock 82 (B, holo.! PRE, photo.). Bulbine crispa (Thunb.)

Roem. & Schult., Syst. Veg. 7: 448 (1829); Kunth, Enum. 4: 568 (1843).

Small plants variable in size. Roots many, unevenly swollen, spongy below,

older parts thinner and harder. Rhizome compact, vertical. Leaves forming a flat

rosette, linear-lanceolate to ovate-lanceolate, 3-10 cm long, 5-15 mm broad, herbaceous,

striate, glabrous but for the prominent, fimbriate, crisped margin, base cuneate, apex

acuminate. Inflorescence a much branched raceme, up to 50 cm high, usually smaller,

sometimes simple in starved plants; scape terete, glabrous, bracteate, fertile bracts

resembling the sterile ones, small, 3 mm long, deltoid; pedicels up to 6 mm in fruit,

articulated in the middle. Flowers 1-15 nate, if many, spirally arranged on a minute

rhachis; perianth rotate, segments 8 mm long, white, green keeled; filaments muricate;

anthers basifixed, revolute; ovary with ca. 8 ovules per cell; style declinate. Capsule

5 mm long, emarginate, with faint, transverse ridges. Seeds typical.

Flowering Period ; September-May.

Distribution: South-eastern Cape, Riversdale to Alexandria districts.

696

Cape. — Riversdale: Hills near Riversdale, Muir 2757 (BOL). George: Keurboom’s

River, Longkloof, Fourcade 3393 (BOL). Oudtshoorn: Moeras Rivier, Compton

23158 (NBG). Uniondale: Compton 10540 (NBG); Misgund, Esterhuysen 6959

(BOL). Uitenhage: on sandhills near Zwartkops River, Zeyher 1069 (BOL, NBG,

PRE). Alexandria; Addo National Park, Korhaan Vlakte, Archibald 5266 (PRE).

3. C. aridum Oberm. sp. nov. C. krookiano Zahlbr. affine sed minore differt.

Planta glabra ad 1 m alta. Radices tumescentes. Folia lanceolata ad 30 cm

longa 2 cm lata glabra striata marginibus crispulatis fimbriatis. Racemus erectus

ramosus; bracteae inferiores 3 cm longae caducae superiores fertiles minutae deltoides;

pedicelli in fructu filiformes 7 mm longi ad centrum articulati. Perianthii segmenta

9 mm longa; ovula in quoque loculo circ. 10. Capsula parva 5 mm longa globosa

vel obcordata.

Plants glabrous, up to 1 m high. Roots typical, swollen, with many thin rootlets

and the swollen parts covered with long, interwoven roothairs. Rhizome horizontal

or vertical, small, woody. Leaves many, in a basal rosette, erect, linear-lanceolate,

up to 30 cm long, 2 cm broad, semi-folded, glabrous, striate, margin white with a

narrow red rim, crisped, fimbriate. Inflorescence an erect, branched, spreading raceme

ca. 60 cm high; scape terete, bracteate, lower bracts up to 3 cm, caducous, upper fertile,

minute, deltoid; pedicels filiform up to 7 mm in fruit, articulated in the middle. Flowers

ca. 4-nate; perianth white, delicate, shiny, reflexed when fully open, free from the

base, segments lanceolate, 9 mm long, 3 outer with a brown spot at the apex, the inner

wider than outer, 5 mm broad; filaments glabrous, 3 mm, erect, spreading; ovary

with ca. 10 ovules; stigma apical, minutely papillate. Capsule globose to obcordate

in outline, 5 mm in length, emarginate, sulcate. Seeds 2 mm in diam.

Flowering Period: December-February. Bees collect much pollen from them.

Distribution: Transvaal, drier areas of the Bushveld.

Transvaal. — Lydenburg: Koffiehoogte in dry Bushveld, J. M. J. de Wet (Prinshof

001403) in flower Dec. 1956 (PRE). Pretoria; near Rust der Winter, Mauve 4067

(PRE, holo.), Codd 2225 (PRE); Rooikop, Smuts & Gillett 2526 (PRE). Potgietersrus:

Leendertz TM 10172 (PRE). Waterberg: Towoomba Pasture Research Station,

Codd 3620 (PRE); Soutpansberg; Smuts & Gillett 4115 (PRE). Kruger National

Park, Mtile-kop, Klokwene, van der Schyff & Marais 3742 (PRE).

The plants increase in size in cultivation or in more moist and warm surroundings

and the leaves then become wider and more flabby, reclining in the upper half. It

then resembles C. k'rookianum although still smaller in all respects and with a more

slender scape. After moie research it might eventually be considered a variety of

C. krookianum. It is also closely related to C. capense but the softer leaves with their

crisped, fimbriate margin, the terete scape and the reflexed perianth distinguish it from

the Cape species.

4. C. krookianum Zahlbr. in Ann. Nat. Hofmus. Wien, 15, I-: 19 (1900) e descr.

Type: Cape, East Griqualand, Newmarket-Umzinkulu, Krook (PI. Penth. 784, W,

holo., destroyed); Capeu Port St. Johns, Isnuka, Galpin 3495 (PRE, neo.).

C. longipedunculatum Forbes in Bothalia 4: 37 (1941) et in FI. PI. S.A. 22: t.861

(1942). Type: Transvaal, Nelspruit, Sabie, Pole Evans 4247 (PRE, holo.!).

Anthericum magnificum Poelln. in Eol. Soc. Brot. Ser. 2, 16: 44 (1942). Type: Natal,

Umzinto, Campbellton-Dumisa, Rudatis 1845 (B, holo.!, PRE, photo.).

Large, glabrous plants up to 2 m tall when in flower. Roots typical. Rhizome

small. Leaves many, in a basal rosette, erect or upper half reclining, lanceolate, up

to 136 cm long, 7 cm wide, half folded, glabrous, shining, somewhat succulent, slightly

narrowed to the clasping base, attenuate in upper half, maigin narrow, membranous,

yellow, sometimes ciisped, fimbriate. Inflorescence a tall, branched raceme with

4-8 ascending branches, up to 2 m; scape woody, terete, bracteate; sterile bracts

697

up to 16 cm long, fertile usually small, deltoid, acuminate, scarious, caducous; pedicels

short during anthesis, up to 22 mm long in fruit, articulated in the centre. Flowers

in compact fascicles, 3-6 nate; perianth rotate with white, delicate, shiny segments

ca. 16 mm long, 4 mm wide, 3 exterior ones with a green dot at the apex, becoming

red-streaked when fading: filaments 5 mm, glabrous, declinate; ovary with 15-20

ovules per cell: style declinate, stigma apical, minutely penicillate. Capsule oblong,

ca. 12 mm. long, attenuated to the base, coriaceous. Seeds typical.

Flowering Period: January-April.

Distribution: Eastern Cape, Natal, Swaziland to eastern Transvaal, usually in

moist situations.

Cape. — Port St. Johns: Isnuka, Galpin 3495 (PRE, neo, BOL).

Natal. — Alfred: near Stafford’s Post on Harding Road, Marais (PRE, GRA). Pieter-

maritzurg: Dyer (PRE). Lion’s River: Nottingham Road, McClean 892 (PRE).

Weenen: Estcourt Research Station, laboratory kloof. West 1783 (PRE). Estcourt:

Loskop, Cathkin Park, Hewlett 115 (NH). Zululand, Hewlett (NH 40426).

Swaziland. — Hlatikulu, Stewart (BOL 10072, PRE, TM 9640); Compton 26398

(PRE, Swaziland Herb.).

Transvaal. — Barberton: Liebenberg 2462 (PRE); van Dam (PRE. TM 2114T),

I This is the largest Chlorcphytum in South Africa. Although the type was destroyed,

the description and the locality indicated this species.

5. C. macrosporum Bak. in J. Linn. Soc. 15: 330 (1876) et in FI. Cap. 6: 400

(1897). Type: “ South African Goldfields ” Baines s.n. (K, holo.). The type locality

is most probably in Southern Rhodesia where the species occurs.

C. rhodesianwn Phill. in FI. PI. S.A. 14, pi. 540 (1943). Type: Southern Rhodesia,

Bulawayo, Pole Evans (PRE 8821, holo.!).

Plants glabrous, up to 150 cm high. Roots thick, numerous, lanate. Rhizome

compact, woody. Leaves in an ascending, basal rosette, linear-lanceolate, up tc 40

cm long, 15 mm broad, tapering towards base and apex, usually half-folded, glabrous,

somewhat glaucous, margin wavy. Inflorescence a tall, branched, sp eading raceme,

many flowered, much taller than the leaves; scape terete, bracteate; bracts lanceolate-

acuminate, upper fertile bracts deltoid, acute; pedicels 1 cm in bud, up to 2-5 cm in

fruit, articulated above the middle. Flowers 2-6 nate; perianth urceolate at the base,

the free part of the segments narrowed at the base, ligulate, 15 mm long, reflexed in

the open flower and in fruit, greenish; filaments smooth, spreading, swollen in upper

part, 3 much shorter than the others, anthers typical; ovary stipitate, protiuding from

the perianth cup, with ca. 20 ovirles per cell. Capsule oblong, 10-15 mm long, stipe

very short or absent. Seeds typical, 3 mm in diam.

Flowering Period: November-March.

Distribution: Southern and Northern Rhodesia, northern Transvaal.

Transvaal. — Sibasa: Kruger National Park, Punda Maria, Lang (TM 32253, PRE);

Baiandbai. Lang (TM 32147, PRE, NH); between Mabase and Baiandbai, Lang

(TM 31105, PRE).

Southern Rhodesia. — Bulawayo, Kolbe 4026 (BOL), Rogers 13801 (SRGH). Salisbury:

Eyles 593 (SRGH) ; Wild 2292 (SRGH). Matobo, Miller 2042 (SRGH, PRE). Umtali,

Chase 4288 (SRGH). Manica: Odzani River Valley, Teague 352 (BOL). Nuanetsi,

1712 (SRGH, PRE). Hartley: //orrrZry 3387 (PRE, SRGH). NDanga: Lundi,

Sabie River, Hall (NBG 540/50).

Northern Rhodesia. — Lusaka, Noak 294 (SRGH).

The species is unusual because of the cup-shaped base of the perianth and the

stipitate ovary, which is not noticeable in the fruit however. It shares this charac-

1517525-2

698

teristic with C. andongense Bak. from Angola, which must be very closely related to it

or perhaps conspecific. The leaves of C. andongense Bak. are broader.

6. C. modestum Bak. in J. Linn. Soc. 15: 329 (1876) et in FI. Cap. 6: 397 (1897).

Type: Natal, Krauss 177 (K, holo.).

C. dwbanense O. Kuntze, Rev. Gen. 3, 2: 316 (1898). Type: Natal, Durban Bluff,

Kuntze (NYS, holo., PRE, photo.).

Small plants up to 30 cm high. Roots typical, many, long, swollen. Rhizome

compact, vertical. Leaves linear-lanceolate, 15-23 cm long, thin, soft, flat, glabrous,

apex acute, tapered towards the base into a long “ petiole ”, membranaceously dilated

at the clasping base. Inflorescence simple, very seldom with a short, basal side branch,

few flowered ; scape arcuate at the base, so that the raceme is exserted outside the leafy

rosette, as long as or somewhat longer than the leaves; lowest bract linear, acuminate,

up to 2 cm long, upper smaller, subulate; pedicels short, 4 mm in fruit, articulated

below the middle. Flowers 1-4 nate; perianth with segments 1 cm long; filaments

glabrous; ovary with 4-6 ovules per cell. Capsule obcordate, 1 cm long, 8 mm wide

at the emarginate apex. Seeds typical, 3 mm in diam.

Flowering Period: November-March.

Distribution. — A forest and coastal bush species found in Pondoland, Natal

and Portuguese East Africa, perhaps extending further northwards.

Cape. — Port St. Johns: Moss 3489 (PRE); Schonland 3982 (PRE). Lusikisiki: Blen-

kiron 16042 (BOL).

Natal. — Durban: Berea, IFocxi 9187 (PRE), 9406 (NH). Estcourt: Loskop, Cathkin

Park, Howlett 8 (PRE). Umzinto : Scottburgh, common on the pavements and in

clearings, usually in shady situations. Mauve 4049 (PRE).

Portuguese East Africa. — Lourengo Marques, Polana, Gomes Pedro 48 (PRE).

Related to the tropical African species C. gazense Rendle, C. laxum R. Br. and

C. petiolatum Bak.

7. C. pulchellum Kunth, Enum. 4: 605 (1843); Bak. in FI. Cap. 6: 398 (1897).

Type: South Africa, Lalande (B, holo., probably destroyed). As Lalande collected

in the eastern Cape, it is probable that the species came from that area. Queenstown,

Gwatyn Farm, mountain sides, Galpin 8333 (PRE, neo.).

Small plants up to 70 cm high. Roots many, swollen near the tips. Rhizome

small, vertical, with some fibres from old leaf bases. Leaves forming a loose rosette,

spreading from near the base, ca. 9 per shoot, linear, up to 30 cm long, 4 mm wide,

folded ribbed, hard, glabrous. Inflorescence a long simple or few branched raceme,

up to 70 cm high; scape longer than the leaves, firm, terete, bracteate; sterile and

fertile bracts small, deltoid, apiculate; pedicels up to 6 mm long, articulated in the

middle. Flowers many, laxly arranged on rhachis, 2-4 nate; perianth segments 1 cm

long; filaments probably glabrous, shrivelling unevenly when dry, anthers typical;

ovary with 10-12 ovules per cell. Capsule not seen.

Flowering Period; December.

Distribution: Probably from eastern Cape.

Cape. — Queenstown: Gwatyn farm, mountain sides, Galpin 8333 (PRE, neo.).

The species is closely related to C. comosum but the leaves are narrower and firmer

and the ovary has 10-12 ovules per cell. The leaves do not make an elongated rosette

as in C. capense. The filaments are described as retrorsely papillate; those on the

Galpin specimen appear to have been glabrous, but have shrivelled irregularly. Our

specimen is taller, but otherwise agrees with the description of the type.

8. C. comosum (Thunb.) Jacques, Journ. de la Soc. Imp. et Centr. d’Hortic. 8: 345

(1862); Bak. in J. Linn. Soc. 15: 329 (1876) and in Fl. Cap. 5: 400 (1897); Wood

699

in Natal Plants 3: 279 (1902); Adamson & Salter in Flora of the Cape Peninsula,

185 (1950); Watt & van Brandwyk in the Medicinal and Poisonous Plants of Southern

Africa, 13 (1932).

Anthericum comosum Thunb. Prod. 63 (1794), and in FI. Cap. ed. Schultes 323 (1823);

Roem. & Schult., Syst. Veg. 7: 475 (1829). Type: Cape, Uniondale, Langekloof

Thunberg (UPS, holo., PRE, photo.). ? A. sternbergianum Roem. & Schult., Syst.

Veg. 7: 1693 (1829). Type: apparently described from a plate and description made

by Sternberg who named it C. comosum. Neither the plant nor the plate could be

found. A. nlanifolium Thunb. a and c, nom. tant. ex Juel in Plant. Thunb. 121 (1918),

(UPS, PRE, photo.). *

Phalangium comosum (Thunb.) Poir. in Lam. Encyc. 5: 252 (1804). P. viviparum

Hort., Bak. in FI. Cap. 6: 400 (1897).

Chlorophytum burchellii Bak. in J. Linn. Soc. 15: 330 (1876). C. datum var. burchellii

Bak. in FI. Cap. 6: 399 (1897). Type: Cape: Albany, Blue Krantz, Burchell 3650

(K, holo.). C. delagoense Bak. in Fl. Cap. 6: 399 (1897). Type: Portuguese East

Africa, Delagoa Bay, Monteiro (K, holo.). Baker thought that the raceme was part

of a panicle but both Killick and Marais, who studied the type at Kew, thought it to

be simple and rot a branch. They could not distinguish it from C. comosum. C.

longituberosum Poelln. in Bol. Soc. Brot. 2, 16: 56 (1942). Type: Natal, Scottspoort

(alt. 4,000 fl.) Thode 3495 (B, holo.! PRE, photo.). 1 could not trace this locality.

C. vallistraopH Poelln. in Bol. Soc. Brot. 2, 16: 79 (1942). Type: Cape, Albany,

Trapp’s Valley, Daly 625 (B, holo.!, GRA, iso.!).

Plants up to 80 cm high. Roots long, many, swollen near the tips. Rhizome

small, vertical. Leaves variable, 10-30 cm long, 1-2 cm broad, tapering at the base

and apex, soft, usually flaL arranged in a loose rosette. Inflorescence much longer

than the leaves, a simple long, lax raceme, sometimes with 1-2 basal ascending branches;

often the apical flowers suppressed and the bracts enlarged to form a leafy tuft; scape

long, terete, bracteate; sterile bracts narrow lanceolate, 2-4 cm long, acuminate,

fertile small, deltoid, subulate; pedicels thin, up to 8 mm in fruit, articulated near

the middle. Flowers in small, spiral, bracteate, axillary fascicles, 1-6 nate; perianth

with narrow segments, 1 cm long, sometimes reflexed; filaments glabrous, slightly

swollen above the middle, shrinking unevenly when dry; ovary with ca. 6 ovules per

cell. Capsule globose in outline, emarginate, 5 mm in length. Seeds typical, slightly

convex on one side.

Flowering Period: During the summer months.

Distribution: Found along the eastern forest areas from Swellendam to the

Soutpansberg; an important constituent of the forest undergrowth; also found near

streams and in moist shady places.

Cape. — Humansdorp: T/ioi/e A 1053 (PRE). Knysna: Ashkop, Fowrcai/p 3709 (BOL) ;

Belvidere, Duthie 994 (BOL). George: Compton 10689, 14342 (NBG); Wilderness,

Compton 15546 (NBG); Grootvadersbosch, Marloth 3493 (PRE), Zeyher 4236 (NBG).

Port Elizabeth: Loerie, Dix 183 (GRA). Queenstown: Gwatyn, Galpin 8192 (PRE);

Katberg, near Sanatorium, Schonland 4323 (GRA). Komgha: Flanagan 2248 (PRE,

BOL, NBG, GRA). Kentani: Pegler 1171 (PRE, BOL).

Natal. — Bergville: Mont aux Sources, Bayer & McLean 109 (PRE). Estcourt:

Cathkin Park, Ndema Forest, Galpin (PRE, 11760). Durban: Berea, Forbes 819

(NH). Zululand: Eshowe Bush, Gerstner (NH 22596), Lawn 110 (NH).

Transvaal. — Pietersburg: Woodbush, Pott 4790 (PRE), Codd 9415 (PRE). Sout-

pansberg: Entabeni, Obermeyer 901 (PRE).

Portuguese East Africa. — Delagoa Bay, Monteiro (K).

* Add to synonymy: Hartwegia comosa Nees in Nova Acta 15, 2, 373 (1831); Kunth, Enum. 4:

607 (1843).

Caesia comosa (Thunb.) Spreng. Syst. 2: 88 (1825); Kunth l.c. 610.

700

Characteristic of this species are the leafy tufts at the apices of the racemes. These

bend down to earth because of their weight and take root. In this way they multiply

very rapidly and cover vast stretches in the undergrowth of the eastern Cape forests.

The plants however do not always produce these leafy rosettes. Those seen from the

northern Transvaal never had them. Adamson records it from the Cape Peninsula

as “ not more than a garden escape ”. It is used as a purgative by the Xhosas.

This plant and the variety mentioned below are well known cultivated stoep plants

in South Africa, known as “ Hen and Chickens ”. It is also well known in Europe

where for many years the variety with a broad yellow band on each side of the midrib

of the leaf has been a popular pot plant called C. variegatum (FI. Mag. 1875, t.l52).

It is often mistakenly called C. capense or C. elatum [cf. Gerome, Nouvelles Precisions

sur les Varietes Panachees et sur les Caracteres Distinctifs entre C. datum et C. comosum,

in J. Soc. Nat. Hort. France, 28: 98 (1927)]. In America it is called the Bracket

plant or Spider plant. In Baileya 9; 29 (1961) Dress has written an article on the

species of Chlorophytum in cultivation in America, their descriptions and a key. He

doubts whether the cultivated plants which there bear the names C. capense and C.

comosum are really diflFerent. As I have not seen the larger and broader leaved plant

which they call “ C. capense ”, I cannot give an opinion cn it. In South Africa C.

comosum may bear leafy tufts or not and the leaves of the wild plants are usually longer

and broader than those of the cultivated specimens. True C. capense is rather rare and

never cultivated. Lawrence in Gentes Herbarium 8: 11 (1949) wrongly applied the

name C. capense (L) Kuntze to C. comosum.

9. C. bowkeri Bak. in Ref. Bot. t. 352 (1873); in J. Linn. Soc. 15: 332 (1876)

et in FI. Cap. 6: 398 (1897). Type: Probably from the eastern Cape, Bowker (K,

holo.). Described and drawn from a living plant.

C. stamineum Zahlbr. in Ann. Nat. Hofmus. Wien 15, 1 : 18 (1900), with figure in tex^

e descr. Type: Cape: Griqualand East, Krook (PI. Penth. 510; W, holo., destroyed).

C. strictum Bak. in Bull. Herb. Boiss. Ser. II, 1 : 782 (1901). Type: Transvaal, Middel-

burg, Elandspruitberge, Schlechter 3831 (Z, holo.!, PRE, photo.). C. nigricans Bak.

Bull. Herb. Boiss. Ser. 2, 4: 997 (1904). Type: Transvaal, Johannesburg, Modder-

fontein, Conratli 666 (Z, holo.!, PRE, photo, GZU, iso.!). C. wilmsii Engl. & Krause

in Engl. Bot. Jahrb. 45: 138 (1910). Type: Transvaal, Lydenburg, near the town,

Wilms 1522 (B, holo., probably destroyed, K, iso.!, PRE, photo.). C. peiiongibrac-

teatum Poelln. in Port. Acta Biol. 1: 229 (1945) e descr. Type: Natal, Scottspoort,

Thode, 3496 (B, holo., probably destroyed). This locality could not be traced.

Plants glabrous, gregarious, up to 1 m high. Roofs typical. Rhizome small,

woody. Leaves many, rosulate, linear to lanceolate, 30-60 cm long, 2-6 cm broad,

semi-folded, somewhat undulate, slightly narrowed to the clasping base, attenuate in

upper half. Inflorescence simple, seldom with a basal branch, up to 100 cm high,

with many flowers closely arranged and supported by large bracts, hiding the rhachis;

scape terete, firm, ca. 40 cm long with some lai'ge, sterile bracts; fertile bracts lanceolate,

subulate, up to 2 cm long, upper usually smaller, caducous or persistent; pedicels

2-4 nate, up to 6 mm long, articulated near the base, erect. Flowers 2-4 nate; perianth

with segments 1 cm long, rotate; filaments smooth; ovary with ca. 15 ovules per cell,

style declinate. Capsule oblong, 1 1 mm long, 7 mm broad, firm. Seeds typical.

Flowering Period: November-March, usually in the spring.

Distribution: Eastern Cape, Natal, Swaziland, Transvaal, Bechuanaland, Southern

Rhodesia; often in damp situations or in grassveld.

Cape. — King William’s Town: Hogsback Mountain, Rattray 363, Sim 1099, 20293

(PRE, BOL). Stutteiheim: Evelyn Valley, Leighton 2678 (BOL); Stutterheim com-

monage, Acocks 9540 (PRE). East London: Wormald 21 (GRA). Kentani: Pegler

1172 (BOL, PRE). Komgha: near Komgha, grassy hills, Flanagan 568 (PRE).

701

Natal. — Weenen: Estcourt Research Station, Acocks 10753 (PRE). Tnanda: Wood

1228 (NH). Klipriver: van Reenen’s Pass, Schweickerdt 951 (PRE).

Swaziland. — Near Forbes Reef, Codd and Mul/er 316 (PRE).

Transvaal. — Barberton: Rimer’s Creek, Ga/p/n 1204 (PRE). Pilgrim’s Rest: Kowyn’s

Pass, Codd 6449; 7040 (PRE). Belfast: Leendertz (TM 10132, PRE); Machadcdorp,

Galpin 13006 (PRE); Lydenburg, Smuts & GiUett 2491 (PRE). Bronkhorstspruit:

Rhenosterkop, Young 2133 (PRE). Pretoria: Mogg 15911 (PRE). Potchefstroom:

Z.OM1V 1301 (PRE). Waterberg: below Krantzberg, .Djrr c6 Frews' 4205 (PRE). Marico:

Lekkerlach, Louw 806 (PRE). Potgietersrus: Kwarriehoek, Steyn 83 (PRE). Pieters-

burg: Chuniespoort, Mogg & Barnard 1165 (PRE). Soutpansberg: Elim, Obermeyer

TM 29327 (PRE).

Bechuanaland. — Lobatsi, Rogers (BOL).

Southern Rhodesia. — Salisbury, Eyies 2081 (PRE).

The typical plant (cultivated) had wide leaves; those from moist areas also have

them but some from around Pretoria and several other localities have a much narrower

lamina, a xerophytic adaptation. The veiy long, subulate bracts and the flowers

congested on the simple raceme are typical for the species. Although there are usually

about 15 ovules per cell, the type plant was said to have 20-30 ovules. The doubling

of one or both rows of ovules occasionally occurs.

C. niagnificum Weimarck from Southern Rhodesia seems to be a wide-leaved

luxuriant form of C. bowkeri.

10. C. trachycarpum Oberm. sp. nov. C. brachystachyo Bak. affine sed ita dififert:

capsulae tuberculatae minores. Fig. 3.

Radices tenues apice tuberiferae. Folia oblongo-lanceolata plana tenuia marginibus

fimbriatis. Racemas simplex circ. folds aequalis; scapus teres saepe nudus; flores

congest!; bracteae ovatae abrupte aristatae; pedicelli breves sub fructu ad 5 mm long!

ad apicem articulati. Perianthii segmenta 1 cm longa; filamenta glabra; ovarium

papillatum sulcatum. Capsula parva, lata irregulariter tuberculata.

Plants up to 40 cm high, gregarious. Roots many, thin, wiry, with tuberous swellings

near the tips. Rhizome compact, small. Leaves oblong-lanceolate, up to 40 cm long,

5 cm broad, attenuate below and above, flat, thin, striate, margin crisped, fimbriate:

outer primary leaves much smaller. Inflorescence simple, shorter or somewhat longer

than the leaves; scape terete, erect, usually bare, flowers closely arranged on rhachis;

bracts broad, suddenly narrowed into a soft, long, ciliate awn; pedicels short, up to

5 mm in fruit, articulated below apex. Flowers 3-5 nate; perianth-segments ca. 1 cm

long; stamens with smooth filaments; ovary globose, with ca. 7 deep, transverse

and 6 vertical grooves, raised areas pulvinate, papillate; 6 ovules per cell, style,

exserted. Capsule trigonous, 3 mm high, 5 mm broad, with rough, tuberculate,

transverse ridges, seldom nearly smooth. Seeds typical, 14 mm in diam. ca. 2 per cell.

Flowering Period : December-January.

Distribution: Northern South West Africa, Southern and Northern Rhodesia

in the West; a shade plant usually found near rivers.

South West Africa. — Okavango Native Territory: between Sambiu and Masari,

de Winter 4081 (PRE, holo.).

Southern Rhodesia. — Lomagundi: Urungwe, Mensa Pan near Chirundu bridge,

Drummond 5372 (SRGH). Darwin: Umvukwes, Umsengedzi River, Wild 3980

(SRGH). Wankie: 40 miles south of Wankie, Wild 4748 (SRGH).

Northern Rhodesia. — Mazabuka, Kandabwe River Dam, Robinson 1790 (SRGH);

near Chirundu bridge, Drummond 5415 (SRGH).

L

702

703

The species is related to C. hrachystacliyum Bak. but it has a papillate ovary, a

small broad tuberculate capsule and thin roots with scattered tubers. The type of

C. brachystachyum, unfortunately, is incomplete consisting only of 2 small, fimbriate

leaves ca. 1 1 cm long, 1 • 5-2 cm wide. For the rest we have Baker’s description made

from a living plant sent by Buchanan from Nyasaland. He describes the roots as

cylindrical, fleshy and the raceme branched at the base but he does not mention the

ovary or capsule. I therefore think that the ovary must have been smooth, otherwise

he would have remarked on it. In the Salisbury Herbarium is a specimen from Zomba

in Nyasaland {Banda 360) which I think is true C. brachystachyum; it has firm roots,

fimbriate leaves and lanceolate, ciliated bracts; it has a smooth, oblong ovary. West

3776 from Inyanga, Southern Rhodesia could also belong to Baker’s species. An

unusual feature in C. trachycarpum is that the articulation of the pedicel is just below

the perianth.

11. C. papillosum Rendle in J. Linn. Soc. 30: 422 (1895). Type: East Africa,

Tanganyika, between the coast and Uyui, Taylor (BM, holo., PRE, photo.).

C. dolomiticum Dinter in Neue Pfi. Deutsch-Siidwest Afr. 23 (1914). Type: South

West Africa, Tsumeb, Dinter 2703 (NBG, iso.!). C. tsiimebensis Dinter in Bot. Jahrb.

57: 237 (1921). Type: South West Africa, Tsumeb, Dinter 2703 (B, holo.!, PRE,

photo., NBG, iso.!).

Dasystachys papillosa (Rendle) Bak. in FI. Trop. Afr. 7: 514 (1898).

Anthericiim lunatum FoeWn. in Fedde, Rep. 52: 259 (1943). Type: South West Africa?

Outjo, Dinter 1398 (B, holo.!, PRE, photo.). Fig. 4.

Plants up to 80 cm high. Roots long, many, hard and fibrous when old, younger

parts soft and fleshy. Rhizome small, compact, horizontal, fibrous. Leaves many,

in an elongated rosette, linear to lanceolate, ca. 40 cm long, 2-5 cm wide, soft, glabrous,

shiny, half-folded, margin undulate, ciliate. Inflorescence simple or seldom with an

ascending branch below, exceeding the leaves; scape terete, glabrous below, glandular-

papillate above, bracteate; bracts 5-2 cm, linear-subulate, ciliate; pedicels short up

to 5 mm in fruit, not articulated. Flowers ca. 5-nate, congested on the raceme, perianth

white, glandular-papillate, globose at the base, constricted near the middle, with the

segments slightly spreading, 1 cm long, apex of segments green tipped; filaments

glabrous, lengthening when the flower opens and ultimately much exserted, up to

12 mm long; ovary trisulcate, placed on a small disk with 10-14 ovules per cell; style

exserted dedinate; stigma capitate, minute. Capsule oblong, trigonous, green, shiny,

1 cm long, shortly stipitate. Seeds 4 mm in diam.

Flowering Period; January-April.

Distribution; Tanganyika, Southern Rhodesia, northern South West Africa; a

shade plant.

South West Africa. — Outjo: Narebis, Barnard 182 (NBG). Okavango: Mohembo,

de Winter 4404 (PRE). Kaokoveld: near Kaoko Otavi, de Winter & Leistner 5545

(PRE).

Bechuanaland. — Ngamiland, Curson 555 (PRE).

Southern Rhodesia. — Bulawayo: Brain 5048 (SRGH). Ndanga: Triangle, Wild

3722 (SRGH).

The wild specimens are glanduliferous but in cultivation in Pretoria this character-

disappeared, the plants becoming very luxuriant and growing to twice their usual size.

12. C. rigidutn Kunth, Enum. 4: 604 (1843). Type: Cape, Tulbagh, Great

Winterhoek, Drege 8738 (K, holo.!, PRE, photo.).

704

Fig. 4. — C/ilorophytuni papillosum Rendle. a, habit, x b, outer and inner bracts, c, cross-section

of capsule, d, capsules, e, papillate flower with globose base, exserted stamens and style

X 2. f, flat black seed with cuneate funicle, x 5.

705

Anthericum rigidum (Kunth) Bak. in J. Bot. 10: 141 (1872) et in FI. Cap. 6: 381 (1896).

A. schultesii sensu Bak. in J. Linn. Soc. 15: 315 (1876) et in FI. Cap. 385 (1897);

sensu Dutliie in Ann. Univ. of Stell. 4, 1: 17, t. 6 (1926). Baker, who had rot seen

the type of Chlorophytum dubiwn Roem. & Schult. (which was given the name Trachy-

andra schultesii by Kunth, when he transferred it to his new genus), quoted specimens

under A. schultesii in the FI. Cap., which should have come under C. rigidum. [Chloro-

phytum dubiwn is a synonym of Chlorophytum triflorum (Ait.) Kunth.].

Plants up to 50 cm high. Roots wiry with age, soft and fleshy when young. Rhizome

small, horizontal or inclined, often covered with fibres of old leaf bases. Leaves more

or less distichous, variable, up to 30 cm long, 8 mm wide, glabrous, folded or flat,

somewhat rigid; primary leaves short, falcate. Inflorescence with a few branches up

to 50 cm high; scape firm, terete with a few small bracts; fertile bracts in small losulate

clusters, dark ovate; pedicels up to 1 cm in fruit, articulated near the base. Flowers

3-6 nate; perianth variable in size, 25-35 mm in diam. white with a 5-nerved brown

keel; filaments slender, scabrid; ovary with 6-15 ovules per cell.* Capsule trigonous

with transverse, rugose ridges. Seeds 3 mm in diam.

Flowering Period : May-October.

Distribution: Cape: Caledon, Somerset West, Stellenbosch, Paarl and Tulbagh

districts.

Cape. — Caledon: lower S. slopes of French Hoek Mts. (Gem Pk), Esterhuysen 11426

(BOL). Somerset West: Sir Lowry’s Pass, Bolus 5560 (BOL); Villiersdorp, Bolus

5268 (BOL); Gordon’s Bay, Guthrie (BOL 25718). Stellenbosch: Jonkershoek,

Compton 15291 (NBG). Paarl: Wemmershoek, Compton 20196 (NBG); Hercules

Pillar, Barker 4854 (NBG); Bainskloof, Bolus 4071 (BOL). Tulbagh: Ontongsberg,

Leighton (BOL 24289).

The specimens from Sir Lowry’s Pass, Bolus 5560 and Gordon’s Bay, Guthrie

in herb. Bolus 25718, are larger than the type specimen but otherwise agree.

13. C. monophyllum Oberm. sp. nov. C. rigido Kunth affine, sed monophyllo

differt.

Planta ad 50 cm alta. Rhizoma non visa. Folium solitarium angustissimum

ad 40 cm longum 2 mm latum complanatum rigidum glabrum basi dilatatum. Racemus

ramosus divaricatus; scapus teres glabrus; bracteae minores fuscae, bracteae fertiles

fasciculatae; pedicelli sub fructu ad 1 cm longi prope basin articulati. Segmenta

perianthii rotata vel reflexa 1 cm longa; filamenta scabra; ovula in quoque loculo

circ. 10. Capsula 1 mm alta transverse costata.

Plants up to 50 cm. Roots not seen. Rhizome not seen. Leaf 1, linear, 40 cm

long, 2 mm broad, folded, rigid, glabrous, broadened at the very base. Inflorescence

a much branched divaricate panicle; scape terete, glabrous, about as long as the raceme,

with a small dark bract; branches elongate, congested above; fertile bracts minute,

deltoid, clustered; pedicels short in bud, up to 1 cm in fruit, articulating near the base.

Flowers laxly arranged in upper part; perianth with spreading or reflexed segments,

1 cm long; filaments slender, scabrid; ovary with ca. 10 ovules per cell. Capsule

typical, 7 mm high, with transverse ridges. Seeds immature.

Flowering Period: December.

Distribution: Only known from the type locality.

Cape. — Ceres: Cold Bokkeveld, sandflats, Adamson D9 (PRE, holo., BOL, iso).

The species is closely allied to C. rigidum but Adamson noted that it had only

one leaf, which moreover is much narrower. It will be interesting to see more material

of this species.

* Kunth found 6 ovules per cell, while I counted up to 15.

706

14. C. viscosum Kunth, Enum. 4: 605 (1843). Type; Cape, Namaqualand, near

the mouth of the Orange River, Drege 2673 (B, holo.!, L, K, iso., PRE, photo.).

Anthericum viscosum (Kunth) Bak. in J. Bot. 10; 141 (1872) in J. Linn. Soc. Bot. 15;

316 (1876) et in FI. Cap. 6; 387 (1897). Dur. & Schinz, Consp. FI. Afr. 5; 387 (1893).

Poelln. in Fedde, Rep, 52; 243 (1943). A. kovismontanum Dint, in Fedde, Rep. 29;

268 (1931). Type; South West Africa, Kovis mountains, Dinter 6645 (B, holo.!, PRE,

photo.). A. longibracteatum Dinter in Fedde, Rep. 29; 267 (1931); Poelln. in Fedde,

Rep. 52; 240 (1943) including var. brevibracteatum l.c. and including forma submem-

Poelln. in Fedde, Rep. 52; 244(1943). Type; South West Africa; Diamond

Area II; Pomona, Dinter s.n. 2nd June 1929 (B, holo.!, M, iso.!, PRE, photo.). Type

of var.; South West Africa; Diamond Area 1; Halenberg, dunes, Dinter 4066 (B

holo! PRE, photo.). ,

Gregarious plants hard in texture, up to 60 cm high (usually smaller in the Namib)

with all parts glandular; glands shortly stipitate. Roots many, uniform, pinkish,

slightly swollen, 3 mm in diam. hollow when dry, the soft mesoderm having shrunken.

Rhizome short, horizontal, covered with old, hard leaf-bases which may eventually

disintegrate into fibres. Leaves distichous or rosulate, variable in size, 15-35 cm long,

2-10 mm broad, dilated at the clasping base, flat, closely ribbed, glandular, hard;

primary outer leaves ca. 2 cm long, pale, membranous. Inflorescence simple or laxly

branched, shorter or longer than the leaves; scape glandular with a few small bracts;

fertile bracts small, membranous; pedicels short at first, up to 1 1 mm in fruit, articulated

near the base. Flowers laxly arranged, ca. 4-nate; perianth spreading, white, dark

keeled, segments ca. 8 mm long; filaments scabrous; ovary with ca. 19 ovules per

cell. Capsule ovoid, apiculate, trigonous, closely transversely ribbed. Seeds flat.

Flowering Period; June-October.

Distribution; South western Cape, recorded from the Piketberg, Vanrhynsdorp,

Calvinia and Namaqualand districts, to the southern part of South West Africa.

Usually in sand; noted to be fairly frequent where found.

Cape. — Piketberg; sandy slope between Yerloren Vlei and Rooikransberg, Pillans

7849 (BOL). Clanwilliam; near Clanwilliam, sandy arid fynbos, Acocks 19762 (PRE).

Vanrhynsdorp; Zandkraal, marginal strandveld, Acocks 15165 (PRE). Calvinia;

Lokenburg, Acocks 17391 (PRE). Namaqualand; Witbank, Pillans 5145 (B).

South West Africa. — Diamond Area I; Klinghardt Mountains, Dinter 4048 (PRE,

B). Luederitz; Rheinpfalz, Dinter 6376 (B); Peilberg, Buchu Mountains, Dinter

6507 (B); Kovis Mts. Dinter 6291 (B).

The plants are variable in size, due no doubt to surroundings; those from more

favourable localities grow up to 60 cm high, whereas those from the Namib remain

stunted, seldom over 15 cm high. The pink, uniform roots and glandular covering

allow for easy recognition of the species.

15. C. triflorum {Ait.) Kunth, Enum. 4; 606 (1843); Adamson & Salter, Flora

of the Cape Peninsula, 185 (1950).

Anthericum triflorum Ait. in Hort. Kew 1; 448 (1789); Willd. Spec. Plant. 2; 140

(1799); Roem. & Schult., Syst. Veg. 7; 466 (1829); Bak. in FI. Cap. 6; 383 (1896);

Duthie in Ann. Univ. Stell. 4; 18 (1926). Type; ex hortus Kew (BM, holo., PRE,

photo.). Introduced in 1782 by George Wench. A. pauciflorum Thunb., Prod. 63

(1794); and in FI. Cap. ed. Schultes 320 (1823). Type; Cape, Lion Mountain, and in

the Swartland, Thunberg a (UPS, holo., PRE, photo.). A.bipedunculatum Jacq., Coll.

Suppl. 88; 10; leones 2; 18, t. 410 (1795). Type; Plate 410, l.c. No specimen

preserved. A. dubium (Roem. & Schult.) Poelln. in Fedde, Rep. 50; 232 (1941).

Phalangium bipedunculatum Poir. in Lam. Encyc. 5; 244 (1804). P. pedunculatum

Bak. in error for P. bipedunculatum Poir. in J. Linn. Soc. 15; 315 (1876). P. triflorum

(Ait.) Pers., Synops. 1; 368 (1805).

707

Chlorophytum dubium Roem. & Schult., Syst. Veg. 7: 455 (1829). Type: Cape (M,

holo.!, PRE, photo.). There is no locality or collector noted on the label, only a number

1624. C. brehmeanum Roem. & Schult., Syst. Veg. 7: 454 (1829) e descr. Type:

Cape, Brehm (the type could not be located).

Trachyandra? brehmeana (Roem. & Schult.) Kunth, Enum. 4: 586 (1843). * T. pauciflora

(Thunb.) Kunth, Enum. 4: 584 (1843). T. schuUesii Kunth, Enum. 4: 586 (1843).

[Kunth changed the epithet of Chlorophytum dubium Roem. & Schult., to schultesii

when he transferred it. The specimens quoted by Baker under Anthericum schultesii

in the FI. Cap. belong to Chlorophytum rigidum].

Plants sturdy, up to 1 m high. Roots many thick, hard, dark, swollen above,

tapering gradually to the tip. Rhizome small, compact, horizontal, with a few fibres

from old leaf bases. Leaves rosulate, variable in size, linear-lanceolate, up to 25 cm

long, 2 cm wide, long tapered in upper half, clasping at the base, folded or flat, soft,

striate, glabrous. Inflorescence a simple raceme, seldom with a short, basal, ascending

branch; scape firm, terete, bracteate; bracts diminishing in size, upper subulate, dark

keeled, with a membranous edge; pedicels short at first, up to 14 mm in fruit, articulated

just above the middle. Flowers 2-4 nate; perianth white, thin, reddish brown on the

outside, segments 10-12 mm, spreading or reflexed; filaments scabrid; ovary with

11-17 ovules per cell. Capsule typical, ca. 2 cm long, 12 mm wide. Seeds typical,

4 mm in diam.

Flowering Period: July-October.

Distribution: South western Cape; in crevices on mountain slopes or in sand.

Cape. — Wynberg: Sea Point, above Ban try Bay, Smith 2883, 2932 (PRE); Green

Point, Pappe (NBG 22905); Cape Town, Rogers 2469 (GRA), Wolley-Dod 2778 (BOL);

Salt River, Zeyher 4659 (NBG); Melkbosch Strand, Esterhuysen 2842 (BOL). Malmes-

bury: Mamre Hills, 450 (NBG, PRE), Compton 21663 (NBG); near Darling,

Esterhuysen 3875 (NBG, BOL). Piketberg: between Verloren Vlei and Rooikransberg,

Pillans 8080 (BOL).

It is one of the oldest Chlorophytum species from the Cape known to be cultivated

in Europe. The roots are very typical and distinguish it from all other species.

16. C. namaquense Schltr ex Poelln. in Ber. Deutsche Bot. Ges. 61: 207 (1943).

Type: Cape, Namaqualand, Vogelklip, Schlechter 11295 (B, holo., probably destroyed,

PRE, iso.!).

C. marlothii Poelln. in Port Acta Biol. 1 : 229 (1945). Type: Namaqualand, Anenous,

Marloth 12218 (B, holo., probably destroyed, PRE, iso.!). C. inornatum sensu Bak.

in FI. Cap. 6: 399 (1897); non Gawler.

Fairly large plants, up to 80 cm high. Roots many, probably soft and spongy

when young, woody and ca. 2 mm thick when old. Rhizome small, covered with fibres

from old leaf bases. Leaves in an elongated rosette, lanceolate, long tapered in upper

half, clasping at the base, up to 40 cm long, 25 mm broad, flat, glabrous, margin

minutely ciliate. Inflorescence simple or with 1-2 ascending branches, 40-80 cm high,

much exceeding the leaves; scape glabrous, stout, up to 5 mm in diam. bracteate,

the bracts diminishing in size, fertile bracts ca. 5 mm acuminate; pedicels short at

first, 1 cm long in fruit, articulated in the middle. Flowers 2-4 nate; perianth white,

with a dark keel, segments rotate, 12 mm long; filaments scabrid; ovary with ca.

20 ovules per cell. Capsule typical, sutures prominent, 15 mm long, 7 mm broad.

Seeds typical, 3 mm in diam.

Flowering Period: August-September.

Distribution: Cape: known only from Namaqualand, apparently rare.

Cape. — Namaqualand: Mesklip, Barker 1883 (NBG), Compton 5869 (NBG); Modder-

fontein. Bolus 6585 (BOL.).

* In J. Bot. Lond. 10: 139 (1872) Baker refers it to Anthericum brehmeanum.

708

Baker in the FI. Cap. 6: 399 (1897) refers Bolus 6585 from Modderfontein, Nama-

qualand, to C. inornatum Gawler from Sierra Leone but this species has a flat rosette

with soft leaves, only 2-5 seeds per loculus and a scabrid, villous scape. The difference

in localities moreover makes it unlikely that they could be conspecific.

17. C. undulatum (Jacq.) Oberni. comb. nov.

Anthericum undulatum Jacq., Coll. Suppl. 68, 10; leones 2: 18, t. 411 (1793). Bak.

in J. Linn. Soc. 15: 304 (1876); and in FI. Cap. 6: 380 (1876). Type; Plate 411,

l.c. No specimen preserved.

A. graminifolium Willd., Sp. Plant. 2; 139 (1799) nom. nov. for A. undulatum Jacq.

(1793); non Thunb. (1794); Roem. & Schult. in Syst. Veg. 7: 463 (1829). [Thunberg’s

A. undulatum, a synonym of Trachyandra hispida (L.) Kunth, was published in his

Prodromus, which appeared in 1794, so that Jacquin's name remains valid]. A. bolusii

Bak. in FI. Cap. 6: 385 (1897). Type; Cape, Namaqualand, Bolus 6601 (K, holo.,

BOL, iso. !). A. pauciflorum Thunb. var. /S, minor Bak. in J. Linn. Soc. 15: 315(1876);

A. triflorum Ait. var. minor Bak. in FI. Cap. 6; 383 (1897). Type of var. : Cape,

Paarl, between Paarl and Lady Grey Railway Bridge, Drege 8723 a (L, iso. !, PRE,

photo.). {Galpin 1025 from near Barberton, Transvaal, quoted in Flora Cap. 6; 383,

is Anthericum cooperi Bak.). A. tumidum Poclln. in Bol. Soc. Brot. 16, 2: 77 (1942).

Type: Cape Piketberg, Diels 191 (B, holo. !, PRE, photo.). A. pleiophyllum Schlechter

ex. Poelln. in Fedde, Rep. 53: 6 (1944). Type: Cape, Namaqualand, Schlechter Will

(B, holo. !, PRE, iso. !). A. dielsii Poelln. in Fedde, Rep. 53: 127 (1944). Type:

Cape, Clanwilliam, Olifantsdal, Diels 366 (B, holo. !, PRE, photo.).

Phalangium undulatum (Jacq.) Poir. in Lam. Encycl. 5; 242 (1804).

Chlorophytum graminifolium (Willd.) Kunth, Enum. 4: 606 (1843). C. tuberculatum

Duthie in Ann. Stell. Univ. 4: 15, t. 7 (1926). Type: Cape, Stellenbosch, Flats, Dm/Z/zc

992 (STE, holo., BOL, iso.!). C. piquetbergense Poelln. in Port. Act. Biol. 1: 230

(1945). Type: Cape, Piketberg, Nieuwe Kloof, Diels 194 (B, holo.!, PRE, photo.).

Plants variable in size, up to 50 cm high. Roots many, covered with a soft mesoderm

when young and then ca. 2 mm in diam., the soft parts disappearing with age, leaving

the wiry, hard, inner core, sometimes the root tips swollen; short sessile “tubers”

(arrested roots) often present on rhizome between fibres of old leaf bases. Rhizome

small, horizontal, covered with fibres of old leaf bases. Leaves forming more or less

flat rosettes, very variable in size and appearance, linear to lanceolate, 5-20 cm long,

2-10 mm wide, tapered to the apex, attenuate to the base if lamina is wide, extended

near the base if it is narrow, erect or spirally twisted, flat or folded, margin straight

or crisped, ciliate; outer primary leaves absent or present as ligulate bracts up to 4 cm

by 1 cm, spotted with red. Inflorescence a simple raceme, seldom with a basal ascending

branch ; scape terete, firm, bracteate, fertile bracts membranous, resembling the perianth ;

pedicels short at first, 14 mm long in fruit, articulated near the middle. Flowers fairly

close together, 2-4 nate; perianth white, with a dark, often wine-red keel, spreading

or reflexed, segments 8-15 mm long; filaments scabrid; ovary with ca. 12 ovules per

cell, style declinate, stigma small, penicillate. Capsule oblong, 12 mm long, 8 mm broad

emarginate, sutures prominent. Seeds typical, 3 mm in diam.

Flowering Period: July-October.

Distribution; South western Cape districts, common.

Cape. — Stellenbosch; Faure, 5r;-e>’ 516 (PRE). Bellville: north of Tygerberg, Co/zrp/on

20056 (NBG). Hopefield: Salter 3031 (BOL). Malmesbury: near Vredenburg,

Leighton 588 (BOL, PRE). Tulbagh: Zeyher 28 (NBG). Paarl: Hercules Pillar,

Barker 1822 (NBG). Piketberg; de Hoek, Esterhuysen 5518 (BOL). Clanwilliam:

near Doom River Bridge, Barker 6554 (NBG); Bidouw Berg, Schlechter 8687 (PRE).

Vanrhynsdorp: sandveld, Marloth 8270 (PRE). Calvinia: Perdefontein, Acocks

709

17310 (PRE); Lokenburg, Acocks 18908 (PRE). Namaqualand : Springbok, Acocks

19433 (PRE). Ceres: Spes Bona, Marloth 10366 (PRE). Laingsburg: Matjiesfontein,

Acocks 17145 (PRE). Prince Albert: Marchand in herb. Marloth 10510, (PRE).

Anthericum undulatum figured and described by Jacquin was said to have come

from the Cape but Baker in the Flora Capensis could not place any specimens under

this species. When, however, a plant collected at Lokenburg in the Calvinia district,

flowered at the Division of Botany, its resemblance to the Jacquin plate was very striking.

As it is a very common Cape species, although variable and therefore given many

names, it is likely that Jacquin had received seeds or plants from this species. The

flowers were painted singly by Jacquin but it is seen occasionally that the subsequent

buds can be absent or very small. He also described the filaments as glabrous but those

of A. bipedunculatwn on the previous plate and A. "\flexuosum ” on the following,

were also called glabrous by him which in fact they are not but the small papillae on the

filaments are easily overlooked. It was suggested that the plate represented A. liliago

L. from North Africa, but this species has an ovoid, pointed ovary whereas Jacquin

described it in his species as “ subrotundum, triquetrum ”. Jacquin’s plant more-

over flowered in October, which is the flowering time for the Cape species, whereas

A. liliago from North Africa flowers in May. Acocks commented on the variability

of C. undulatum as follows: “ My feeling is that all these Chlorophy turns are a single

species no matter whether they have short, narrow straight leaves with straight, undulate

or crenulate margins, or broad, folded, arcuate leaves, long or short, straight edged,

undulate or crisped; no tubers, few or many, thin or fat; short inflorescence or long;

so much depends on where they grow, whether in deep, sandy soil, near a water course,

or on heavy soil, or on next to no soil in the crevices of shale and on whether they

have had lots of well distributed rain, no rain, or intermittent rain with droughts in

between; and on how many times they were eaten off, before they managed to come

into flower; not to mention of course whether they were growing in shady places,

in dense scrub or in bare exposed places. If a plant is obstinate enough to be a perennial

in those parts it has to be tough and adaptable ”.

It was seen that the root-tubers were young roots arrested in their development;

under moist conditions they elongate to become normal roots. In optimal surroundings

they are absent. Jacquin described the three inner segments as undulate, and these

suggested the name, but it is a rather obscure character, difficult to see in dried flowers.

18. C. crassinerve (Bak.) Ohenn., comb. nov.

Anthericum crassinerve Bak. in J. Bot. 29, 71 (1891) et in FI. Cap. 6: 384 (1896). Type:

Cape: Namaqualand, near Ookiep, Bolus 6600 (BOL, syn. !, K, iso.); Scully 114

(BOL, syn. !).

Glabrous plants up to 40 cm high. Roots spongy, when young, thin and hard

when old; short, bulbous, erect or spreading root-tubers present on rootstock between

fib'^es. Rhizome compact, covered with fibres from old leaf bases. Leaves in a rosette,

ca. 5, lanceolate, ca. 12 cm long, 14 mm broad, coriaceous, glabrous, closely ribbed,

margin prominent, apex obtuse, apiculate, base attenuate; primary leaves rudimentary,

membranous, mottled with purple. Inflorescence simple, up to 40 cm high; scape

terete, firm, glabrous, spotted, bracteate; bracts membranous resembling the perianth

segments, ovate, auriculate, acuminate; pedicels short at first, up to 1 cm in fruit,

articulated in the middle. Flowers laxly arranged, 2-4 nate; perianth spreading or

reflexed, segments 15 mm long, white, red keeled; filaments scabrid; ovary with ca.

15 ovules per cell. Capsule typical. Seeds typical.

Flowering Period: August-October.

Distribution: Known only from Namaqualand.

Cape. — Namaqualand: between Garies and Khamieskroon Pillans in herb. BOL.

19174; Brakdam, Esterhuysen 5680 (BOL); Scully 14 (BOL); Morris 'm herb. BOL,

19173; 6 miles N. of Garies, Leighton 1396 (BOL).

710

The species is closely allied to C. undulatum; the obtuse, broad leaves with the

thick red margin distinguish it.

19. C. pauciphyllum Obenn., sp. nov. C. undulata (Jacq.) Oberm. affine sed foliis

paucis pubescentibus differt.

Planta ad 15 cm alta. Radices plures spongiosa. Folia primaria brevia ligulata

maculata; folium maturum unum linearis ad 45 cm longum 5 mm latum planum

rigidum costatum pubescente vel glabrescente ; folium novum immaturum sub anthesi

pubescente. Racemus simplex vel basi ramo uno ad 26 cm alto; scapus gracilis teres

glabrus bracteatus; pedicelli ffiiformes ad centrum articulati. Segmenta perianthii

rotata 6 mm longa, filamento scabro.

Plants up to 15 cm high. Roots many, ca. 10 cm long, covered with a soft, spongy,

lanate tissue which disappears with age leaving a hard, thin, inner core: younger roots

usually spindle shaped and with several fascicled thin rootlets appearing from the tips

of the swellings. Leaves few; primary short, ligulate, clasping the scape, purple spotted;

old mature leaf (from previous season) linear, up to 45 cm long, 5 mm wide, flat, rigid,

ribbed, pubescent to glabrous, margin prominent; new leaf 20 cm long at time of

flowering, pubescent, clasping scape at the base. Inflorescence simple, 26 cm tall;

scape slender, terete, glabrous with a small sterile bract; fertile bracts small, 5 mm long,

membranous, pedicels filiform, articulated near the middle. Flowers laxly arranged,

1-2 nate; perianth rotate, 12 mm in diam. white with a green keel; filaments densely

scabrous; ovary typical. Capsule not seen.

Flowering Period : March-September.

Distribution: Only known from Clanwilliam in the Cape; apparently a mountain

species.

Cape. — Clanwilham: N. of Bulshoek Barrage, Barker 7301 (NBG, holo., PRE, photo.).

Citadelskop near Wupperthal, fairly frequent, Leipoldt 1071 (BOL); Nieuwoudt

Pass, Esterhuysen 8150 (BOL).

Easily distinguished from the related species C. undulatum because of its solitary,

long leaf, and from C. lewisae which has several short leaves and a bulbous rhizome.

20. C. lewisae Oberm., sp. nov. C. undulato (Jacq.) Oberm. affine sed rhizomate

bulboso foliis setcsis differt.

Planta ad 20 cm alta. Radices longae lanatae. Rhizoma bulbosa. Folia 2-4

ad basin et marginem setosa. Racemus simplex vel basi ramulo solitario; scapus

pubescens bracteatus; pedicelli tenues 9 mm longi ad basin articulati. Segmenta

perianthii reflexa 6 mm longa; filamenta scabra; ovula in loculis 12.

Small pubescent plants up to 20 cm tall. Roots fairly thin, the spongy tissue not

much swollen, lanate. Rhizome bulbous, ca. 1 cm in diam. horizontal, covered with

fibres of old leaf bases. Leaves 2-4, linear-lanceolate, up to 16 cm long, 5 mm broad,

attenuate in upper half, terminating in a tiny, black, bulbous point, base slightly dilated,

clasping the bulbous rhizome, upper surface glabrous, lower surface, especially the

margins, setaceous. Inflorescence up to 25 cm tall, simple or with one short, ascending

side branch; scape terete, thin, pubescent, with 1-3 small, sterile bracts; fertile bracts

ovate, auriculate, 4 mm long; pedicels thin, 9 mm long, articulated below middle.

Flowers 6-10, laxly arranged on rhachis, 2-3 nate; perianth reflexed, white with 3

dark-green nerves, segments obovate, 12 mm; filaments scabrous; ovary oblong

with ca. 12 ovules. Capsule and seed not seen.

Flowering Period: September.

Distribution : Cape, Calvinia.

Cape. — Calvinia: top of Botterkloof Pass, Lewis (SAM 62038, holo., NBG, PRE,

photo.); same locality Johnson 561 (NBG).

The globose rhizome and the hairy leaves distinguish it from the other species.

711

Species Excluded

C. Bak. in FI. Cap. 6: 397 (June 1897), non Hua. Type: Natal, Weenen,

Wood 4425 (K, holo., NH, iso.!).

This species was found to be identical with Ornithogalum longiscapum Bak. in

Bull. Herb. Boiss. Ser. 2, 1: 854 (1901). There is also C. vaginatum Hua [Contr. FI.

Congo Fran?. Lil. 22 (1897)] from the French Cameroons which was published in

the same year as C. vaginatum Bak. Professor Aubrevilles in Paris and Mr. W. Marais

at Kew, were unable to discover in which month this volume appeared. As it was

therefore impossible to discover which name was the older of the two, it was decided

to regard Hua’s name as the one published first. This makes Baker’s name a synonym

and so the name Ornithogalum longiscapum Bak. remains valid.

C. drepanophyUum Bak. in FI. Cap. 6: 398 (1897). Type: Namaqualand, near

Nababeep, Bolus 6584 (K, holo.). This is Trachyandra falcata {L.f.) Kunth.

C. haygarthii Wood & Evans in J. Bot. Lond. 37: 254 (1899) and in Wood’s Natal

Plants 1 : 79 t. 98 (1899). Type: Zululand, Haygarth (Wood 7448, NH, holo.!, PRE,

iso). This is Anthericum haygarthii (Wood et Evans) Kies, comb. nov.

Chlorophytum norlindii Weim. in Bot. Not. 1937: 434, photo, p. 435. Type:

Southern Rhodesia, Makoni district, Maidstone, Norlindh and Weimarck 4128 (LD,

holo. !, PRE, iso. !). This is Anthericum galpinii Bak. var. norlindii (Weim.) Oberm.

comb, et stat. nov.

Species Insufficiently Known

Chlorophytum leipoldtii Poelln. in Ber. Deutsche Bot. Ges. 61 : 207 (1943) by error

“C. leipoldii.” Type: Cape, Leipoldt s.n. (B, holo., probably destroyed in 1943). If

this specimen was the same as Leipoldt 493 from Clanwilliam (GRA), and from the

description it seems to be, it is a synonym of C. undulatum (Jacq.) Oberm.

C. dregei Poelln. in Port. Acta Biol. 1: 228 (1945). Type: Cape, Drege 1827

(B, holo., probably destroyed in 1943). It may be found in other herbaria. As the

specimen had no roots it cannot be determined with certainty whether it was C. triflorum

or C. undulatum.

C. schlechterianum Poelln. in Ber. Deutsche Bot. Ges. 61 : 208 (1943). Type: Cape,

Schlechter s.n. ex hortus Berlin-Dahlem (pressed for herb. 1900; B, holo., probably

destroyed in 1943). The roots unknown. It could be C. undulatum or C. namaquense.

Species Erroneously Referred to South Africa

C. blepharophyllum Schweinf. ex Bak. in J. Linn. Soc. 15: 111 (1876). The type

comes from Central Africa, Gallabat. Baker in the above publication mentions:

“Africa australis, in ditione Transvaal, Baines', perianthium non vidi Mr. W.

Marais looked up the Baines specimen at Kew and found it had three labels: 1. S.A.

Goldfields. 2. Seqetse, Quae Quae River. 3. Matabele Land, source of the Gwailo

River, therefore all from Southern Rhodesia. It has not been recorded from the

Transvaal.

C. inornatum sensu Bak. in FI. Cap. 6: 399 as to Bolus 6585 (1897); non Gawler.

Baker places Bolus 6585 from Namaqualand under this tropical species. It is C.

namaquense Poelln.

3. TRACHYANDRA

Kunth, Enum. 4: 573 (1843). Dilanthes Salisb. Fragm. 70 (1866). Liriothamnus

Schlechter in Notizbl. Bot. Gart. & Mus. Berlin 9: 145 (1924).

712

Perennials, suffrutescent or herbaceous arid in this case the parts above ground

dying down in winter; with glandular pubescence in some species. Roots various;

fibrous or spindleshaped, sometimes swollen only near the tips; in some fused to form

a “ toed ”, contracted tuber. Stems upright, woody, naked or covered with old leaf

bases, or more usually developed as a rhizome. Rhizome vertical, seldom horizontal.

Leaves uniform or dimorphous; base tubular, often persistent as a sheath of fibres.

In dimorphous-leaved species, tubular, membranous, basal squamae are present,

surrounding the leaves and scape and sometimes the shoots; leaves arranged in a

rosette, rarely somewhat distichous, or in congested spirals on aerial stems; lamina

flat, triquetrous, terete or canaliculate, hairy or glabrous or glandular-pubescent;

sometimes spirally contorted, undulate or plicately folded. Inflorescence an axillary,

single or branched raceme rarely sub-umbellate; scape terete, naked or with seme

sterile bracts (vestiges of side-branches); bracts one per flower; pedicels single,

not articulated, changing their position during anthesis and when the capsule ripens,

erect, patent or recurved, sometimes making a complete locp. Flowers single, scentless

or with a strong scent, usually opening in the afternoon, closing in the evening.

Perianth rotate or recurved, erect or pendulous; usually white, rarely yellow, pink

or mauve, dark keeled; 3 outer segments slightly narrower than 3 inner; often with

yellow, green or dark didymous maculae near the base, caducous, the flower falling

off completely if not fertilised; if fruit is formed, the perianth base persists below the

ovary forming a small, sometimes stipitate rim or cup. Stamens 6, adnate to the very

base of the perianth, slightly shorter than the segments, spreading or theirner cornivent;

filaments similar or dimoiphous, retrorsely scabrid, occasionally declinate; anthers

versatile, introrse. Ovary sessile, globose with 2-16 bi-seriate ovules in each chamber;

with septal glands; style filiform, ultimately exserted, often declinate; stigma small,

minutely penicillate. Capsule loculicidally 3-valved, globose or trisulcate, coriaceous,

seldom fleshy; apex obtuse or apiculate, stipitate through abortion of lower ovules;

smooth or tuberculate with a few or many, smalt or large gland-tipped tubercles, rarely

with 3 lateral horns. Seeds angled, usually brown or grey, smooth or verrucose, with

or without prominent ridges; with immersed yellow glands when immature, becoming

glutinous when ripe, probably as a result of the glands erupting.

Distribution: A predominantly South African genus with the majority of the

species in the south western Cape; found all over South Africa and South West Africa,

a few extending to Southern Rhodesia, Angola, Nyasaland, Kenya and Abyssinia.

Suggested type species; T. hispida (L.) Kunth.

Species of Trachyandra are found throughout southern Africa, but the majority

are endemic to the winter rainfall region of the south western Cape. A few species

extend further northwards, one as far as Abyssinia. Linnaeus and others after him

classed this group of plants with Anthericum but Kunth recognized it as constituting

a separate genus which he named Trachyandra, giving a good description of it in

Enum. 4: 573 (1843). Baker in the Journ. Linn. Soc. 15: 307 (1876) reverted to the

old classification and sank the genus as a section under Anthericum where it remained

until now. On close examination, however, it will be seen that the species belonging

to this section are very different from true Anthericum ard more closely related to

Bulbine. The axillary inflorescence, the single, non-articulated pedicel and the deciduous

perianth are but a few obvious characters that distinguish them from Anthericum at

a first glance. As in Bulbine, the perianth, if the flower is not fertilized, drops off

completely; if seed is set, however, its base remains and forms a cup or rim below

the capsule. Both have axillary inflorescences, pedicels that change their position

when the capsule is formed and leaf-bases that are tubular. Furthermore a parallel

development is found in the formation of tuberous roots and the production of aerial

stems in some of the species. The caulescent habit of T. involucrata induced Schlechter

to put it in a separate genus, Liriothamnus, with the remark that it showed affinity to

Bulbine caulescens. But separation solely on its caulescent habit, is not sufficient.

713

When Kunth published the name Trachyandra, referring to its scabrid stamens,

he treated the name as feminine as can be seen from his epithets.

Resume of the Sections

The genus can be divided into three sections. The first, § Liriothamnus, could be

regarded as the most primitive. The roots are fibrous and the outer leaves are similar

to the later ones. The inflorescence is simple in the majority of species, the perianth

(with very few exceptions) immaculate ar.d rotate and the filaments aie uniformly

scabrid and spreading. The number of ovules varies from 10-2 per loculus. There is

a tendency to reduce the number of ovules.

The second section, § Trachyandra, shows a much greater diversity in the root

system, in the special rudimentary outer leaves or squamae and in the more elaborate

floral structure. Amongst its species are so called “ tumble weeds ”. The flowers

are often strongly scented, sweet or musklike. The pendulous flower has an ingenious

use for its retrorsely scabrid filaments. These form a tube around the ovary, and

the nectar, copiously exuded frcm the septal glands, is collected between the retrorse

excrescences of the filaments and attracts insects.

The third section, § Glandulifera, is distinguished from the other two sections in

the development of glandular hairs. In some varieties of T. asperata these may be

restricted to a few microscopic stipitate glands on the ovary which afterwards crown

the tubercles that emerge wuh the development of the capsule. In T. gerrardii and

T. sabidosa, however, the whole raceme may be densely glandular while the furry

capsule somewhat resembles the fruit of a plane tree, being densely covered with

pectinate glard-tipped tubercles. The degree of pubescence ard glardulosity is very

variable. The inhorescerce is always branched (sometimes reduced to a simple raceme,

i.e. in some specimens of T. asperata var. stenopliylla), the perianth maculate and the

lower ovules absent or abortive giving the capsule a stipitate appearance. This may

not be obvious at first in capsules which are densely covered wuh tubercles, the stipe

being hidden by them. The leaves are usually triquetrous. Archibald anoi Phillips

noted that the plants exuded a strong unpleasant odour.

Morphology:

Root system: The roots of the § Liriothamnus are fibrous, and numerous. In

the § Trachyandra they have become storage organs and vary a good deal. In the

tumble-weed grorps there are many arranged in a congested circle; they are often

swollen, elongate-: pirdle-shaped but not contracted or fused. The thickness depends

on the season and whether growth ard flower-prcduction have used up much of their

contents. The next step, as seen in T. jacquiniana ard related species, seems to have

been the contraction into tubers ard a reduction in their number. This develcpment

is confined to species of the winter rainfall region. Ultimately the roots and rootstock

beccme fused forming a solid “ toed ” foot as in some Bidbines. How the new parts

deveb p is not yet prc perly ur derstccd. The many hard outer skins suggest that new

roots are formed irside the old skins which may remain in situ for a long time. In

the § Glandulifera they are long, spreading, often branched and the lower half of the

root near the tip is swollen. In many species frcm sardy areas the root is covered

by a der se felt ce mposed c f long interlacing root-hairs. These are apparently permanent

ar.d seem able to absorb water very quickly.

Stem-rhizome: One of the most striking features of this genus is the development

of a woody stem which may reach a height of up to 6 ft. in T. adamsoni. A close ally,

T. involucrata, also produces a woody, rather gnarled, naked stem while T. acocksii

and T. burkei have short ones that are covered with persistent leaf-bases. In the

§ Glandulifera the Cape species T. scabra ard T. sabulosa also possess short branched

stems. The anatomy of the wood was studied by Adamson [J. Bot. Lond. 69: 10

(1931)] who found it to be of typical monocotyledonous development as in Dracaena.

714

The cambium is external and complete amphivasal vascular strands are added, forming

separated, secondary ground tissues. Most species however, have a hard, irregularly

shaped, woody rootstock. Growth usually takes place in a vertical direction, but in

certain species, e.g. T. brachypoda and T. chlamydophyJla the rhizome grows horizontally.

Leaves: The bases of all leaves are tubular. A remarkable development has taken

place in those species here put in the second section, § Trachyandra, where the leaves

have become dimorphous, the first or outer leaves having been changed into short

tubular membranous prophylls functioning as a protective covering for the vital inner

parts. I decided to use the term “ squamae ” for these organs as “ basal rudimentary

leaves ” is so cumbersome and “ prophylls ” vague. Jackson’s definition of squama,

“ usually the homologue of a leaf ”, fits these organs very well. If the monocotyledonous

leaf is interpreted as a leaf-base and petiole, we may regard the squamae as the sheathing

leaf-bases which have lost their petioles. Their reduction from leaf to squama can be

followed in various species. T. tabularis, which may be regarded as a link, shows

outer leaves that occasionally lose their lamina, thus becoming short tubular prophylls,

something often seen in the Liliaceae. In the next phase the lamina has disappeared

permanently and the squamae now become very different from the leaves. They are

now truly dimorphous, brown or white and membranous. One to three narrow tubular

squamae can be seen surrounding each leaf- and each scape-base. A further change

now takes place in that each shoot becomes surrounded with squamae in addition to

those fitted around the leaf- and scape-bases. With the expansion of the shoot these

outer squamae often burst and become tattered. In some other species the squamae

are wide and loose-fitting, and remain whole. In some, e.g. T. bulbinifolia and T.

dissecta the thin, inner squamae differ from the more sturdy outer squamae.

The leaves sometimes show a slight twist or are strongly wavy or plicately folded

or form a stiff spiral. This spiral twisting, waving and folding, is met with in other

Monocotyledons from Namaqualand, e.g. in Ornithogalum, Moraea, Dipcadi, Albuca,

Babiana, etc. It could be an adaption to the dry conditions. Certain specimens of

one species however show it markedly whereas others growing near it have straight

leaves. Pubescence varies from short raised points to long hairs even in one species.

Arber in her classic work on the Monocotyledons remarks on p. 130 that in this

phylum there is a greater variety in the width of the leaves amongst the different species

than in the Dicotyledons, which could be expected if the “ blade ” is interpreted as

an expansion of the petiolar region. This variation in leaf-width is very marked even

in individuals of one species.

Of cytological interest are the spiral thickenings of the vessels in the lamina. They

draw out into fine spiral threads when the leaf is broken (cf. Duthie l.c.). It was also

observed in T. laxa var. erratica.

Inflorescence: A number of species, all from the first section, have simple racemes

i.e. T. sa/tii, T. esterhuysenae, T. reflexipilosa, etc., but in the majority of species,

including the whole § Glandu/ifera, the inflorescence is branched. In some of the

branched species, we meet specimens which occasionally bear simple racemes probably

because the plants were immature or depauperate. On the other hand there is a definite

tendency in some species to suppress side branches. In some species like T. falcata

and T. ciliata, side-branches, although small and suppressed, can always be detected

near the base of the raceme. In T. longepedunculata these side branches have often

disappeared altogether, only the empty bracts remaining. In the above species and some

others however, the raceme may elongate to 2 or 3 times its original length during

anthesis, becoming prostrate in the process.

In the “ tumble-weeds ” the raceme is much branched and divaricate. At first

the branches, branchlets and pedicels are erect but after flowering they spread out and

downwards forming a “ ball ” which after detachment rolls around in the wind. It

is extremely light, the scape consisting mainly of pith. An interesting development

715

are the accessory branches. It is found in the majority of the species but often one

or both side-branches are suppressed. We meet it for instance in T. laxa var. erratica,

where the 3 lowest branches are trichotomous arising from the clavate apex of the scape;

the upper branches are alternate. In other species, e.g. in T. jacquiniana these accessory

branches are found in nearly every axil but one of the branches remains small. If the

branch is completely suppressed we still find its bract. In the § GlanduUfera, T.

asperata, and T. gerrardii also have accessory branches.

Bracts: These do not vary very much. Some become nearly amplexicaul, e.g.

in T. falcata. They show a resemblance to the leaves. From a phylogenetic point

of view they are of interest in that they are the last surviving organs in cases where

floral side-branches are suppressed. The rhachis, pedicels and flowers may have

disappeared but the bracts remain as a relic. This also occurs in Anthericum and

Chlorophytum.

Pedicels: They are not articulated. In some cases it would appear as if there

could be an articulation near the apex. This occurs in some species, e.g. in T. tabularis,

after fertilisation when the perianth-base elongates into a short stipe below the persistent

rim. If the flower is not fertilized, it drops off completely at the apex of the pedicel.

In this case the pedicel remains thin and does not change its position. If a capsule

is formed, however, the pedicels move into certain positions, upwards or downwards,

patent or with the apex recurved, sometimes forming a complete loop. The ultimate

position is constant and makes a useful character for recognizing the species or

varieties.

Flowers: When the perianth closes after flowering, it fits snugly around the ovary,

the apical part protruding above it as a rod. It does not twist. Through pressure

of the enlarging ovary, it tears off below and falls off like a pointed cap, leaving its

saucershaped base below the capsule. The perianth is either rotate or revolate. The

first species Linnaeus described from the Cape was named Anthericum revolutum by

him because of its recurved perianth. It forms a graceful “ ball ” around its pedicel.

It is pendulous. A number of related species in this section have the same type of

recurved, pendulous perianth but in the majority of the species it is rotate and the

flowers face upwards. They are usually white, but in T. arvensis they are yellow, in

T. thyrsoidea mauve, and in T. tortilis and T. hirsutiflora they are pink, while in many

a faint pink tinge may be seen. In those species where the segments curve backwards,

we find a pair of yellow, green or dark spots near their base, probably serving as honey

guides. Duthie, Adamson, Marloth, Jacquin and others noted a strong smell, sweet

in some species, musk-like in others. The flowers usually open in the afternoon and

close during the night. In the branched racemes a few flowers open daily but on the

simple racemes, e.g., in T. saltii, they flower profusely from the base upwards after

good rains. Within a few days the flowering period is over and the unobtrusive plants

sink back into obscurity. The plants however, have young inflorescences near the base,

ready to emerge when conditions become favourable once more. In T. ciliata and some

others the raceme continues to lengthen if conditions remain favourable.

Stamens: In the rotate perianths the filaments are uniform, shortly scabrid and

spreading. At the base, where they surround the ovary in bud, the filaments are always

smooth. In the revolute perianths however, the filaments are dimorphous. The outer

spread and are shortly scabrid but the inner are connivent around the ovary; they are

flattened and smooth in the lower half where they surround the ovary but have lateral

and dorsal fringes. Above the ovary they bend outwards and there have long, retrorse

excrescences. As the flowers hang upside down the copious nectar is caught amongst

these outgrowths. Miss Duthie records that bees are the chief pollinating insects.

The anthers are small and versatile.

Ovary and capsule: The number of ovules per cell is fairly constant for each species,

occasionally 2 more or 2 less. In the § Trachyandra the number of ovules is usually

about 10 but in one plant of T. jacquiniana 16 ovules were counted in a cell, the largest

i

716

number for the genus. However, there is a tendency, especially in the § Glandulifera,

to reduce the number of ovules or to abort the lowest. The capsule then becomes

obovoid or shows a stipe at the base above the persistent perianth rim. The widespread

species T. saltii has about 8 ovules per loculus, only the upper usually being fertile. In

the variety seciinda with reflexed pedicels only 4 were counted in some of the specimens

examined. In the related Angolan species, T. pyrenicarpa, each loculus produced

only one seed, a large one for the genus. The ovules are axial and biseriate. The

placenta is sometimes swollen, perhaps the result of insect irritation. The septal glands

produce much nectar. A curious feature of T. margaretae is the presence of a dense

covering of simple, erect hairs on the young ovary. In T. zebrina the ovary is glabrous

when young but some reflexed setae appear later on the capsule. The § Glandulifera

(apart from the two non-glandular species just mentioned) is the only section which

has the ovary not glabrous; here it is sparsely to densely covered with stipitate glands.

When the capsule matures, simple or complex pectinate tubercles or horns are formed

on which the glandular hairs are mounted. The glands erupt and disappear but the

excrescences persist in covering the capsules, partly or entirely. The capsules open

loculicidally, the walls sometimes recurving. The style is always glabrous. [Baker in

the Flora Capensis 6, 379 (1897) erroneously mentions in the key “ filaments and style

scabrous ”] . At first the style is short but it eventually becomes longer than the stamens.

The stigma is small, penicillate.

Seed: When immature, the testa in most species shows golden flecks which appear

to be glands which burst when the seed is ripe and thus give it a glutinous coat. The

seed is angled, usually grey, smooth or verrucose in certain species. The seeds of the

Glandulifera species have a different but definite verrucose pattern for nearly each

species; it may be globose or flattened or in some with a crenulate wing and with

large and small tubercles near the perimeter, the thick central area being fairly smooth.

The species have mostly adapted themselves to very definite, special habitats. The

tumbleweeds for instance are always found growing in sandy soils. Others are grassveld

species, i.e. the widespread. T. saltii, while others, e.g., T. erythrorrhiza, prefer marshy

surroundings. Adamson in the Flora of the Cape Peninsula, records a number of

species that flower after veld-fires e.g. T. muricata, T. hirsuta, T. hirsutiflora, etc.

Apparently growth is stimulated but this problem has not yet been fully investigated.

It occurs in many other genera. The bare ground will become warmer after the vegetative

covering has been removed and this may also be a reason for more rapid flowering.

Because of their capacity to develop so rapidly they will have completed their annual

life cycle by the time their neighbours crowd around them once more.

The plants have no economic importance. Karsten (The old Company’s Garden

at the Cape, 1951) mentions that shoots of a wild Anthericum were served as asparagus

on Van Riebeeck’s table. It could have been T. diraricata which is plentiful around

the Cape, producing the fattest and most succulent shoots, or perhaps T. falcata, but

our recent records of this species show its southernmost habitat to be Saldanha Bay.

The plants, especially the tumbleweeds, are eaten by animals, records showing that pigs

even dig up the roots. They are not poisonous, although they were often suspected.

In this review 45 species are enumerated from the Union and South West Africa;

10 of these are new species. A few extend beyond our borders, i.e. T. saltii which is

found in Southern Rhodesia to Kenya and Abyssinia. T. arvensis occurs in Northern

Rhodesia. T. reflexipilosa has also been discovered in Southern Rhrdesia. In Angola

occurs T. pyrenicarpa, a species closely related to T. saltii but with one large seed per

locule. There is one glandular species, T. malosana, recorded from Nyasaland and

Southern Rhodesia. It is close to T. asperata and may be conspecific. Baker in the

Flora of Trop. Africa 7: 491 (1898) refers specimens collected in Angola and Beira

to “Anthericum elongatum Willd.” It may prove to be a mixture of species, some perhaps

belonging to the tumbleweed group and others probably to T. saltii. Von Poellnitz

717

described a large number of Anthericum species from tropical Africa in various publi-

cations. Many of his type-specimens were destroyed during the second World War.

Some of these may prove to belong to Trachyandra.

Key to Sectfons

Plants not glandular-pubescent;

Plants with the outer leaves occasionally smaller but not changed into squamae; roots fibrous;

inflorescence simple, seldom branched; perianth immaculate, seldom maculate; filaments

shortly scabrid § 1. Liriothanimis

Plants with the outer leaves changed into squamae; roots usually in a circle, often swollen,

sometimes contracted or fused into a few tubeis and also fused to the rhizome; inflorescence

branched, seldom simple through reduction; perianth usually maculate; filaments uniform

or dimorphous, the outer spreading the inner connivent around the ovary and then curving

outwards, there with long retrorse bristles § 2. Trachyandra

Plants glandular-pubescent with the stipitate glands few, minute, confined to the ovary and afterwards

the capsule, to densely scabrid-glandular on all parts of the raceme § 3. Glandulifera

Key to Species

Section I. LIRIOTHAMNUS. [Liriothamnus Schlechter in Not. Bot. Gart.

& Mus. Berlin 9: 145 (1924)].

Roots wiry, thin, seldom lanate. Aerial stems sometimes developed or usually

with a rhizome which is often covered with fibrous leaf-base remains. Leaves uniform

(in T. brachypoda and T. tabular is lamina of primary leaves often reduced). Inflorescence

simple or branched: scape often arcuate near the base. Perianth erect, rotate, imma-

culate (maculate in T. adamsonii and T. biirkei)', filaments spreading, uniform, shortly

scabrid; ovules 10-2 per cell, the lower sometimes abortive. Capsule glabrous, often

stipitate. Seeds smooth or verrucose.

Distribution: Widely distributed.

Plants producing naked, aerial, woody stems: (two species, found in the Vanrhynsdorp and

Namaqualand districts; Liriothamnus Schlechter):

Shrubs with bare stems up to 6 ft. high; leaves flat, over 2 cm wide, perianth maculate; capsule

erect 1 . T. adamsonii

Shrubs usually about 2 ft. high; leaves terete ca. 6 mm in diam.: perianth not maeulate;

capsule recurved 2. T. involucrata

Plants producing short aerial stems, which are covered with leaf-bases:

Leaf-bases hard, thick, quill-like; leaves many; inflorescence simple, perianth immaculate 3. T. acocksii

Leaf-bases long with reticulated thickenings; leaves few; inflorescence usually branched;

perianth maculate 4. T. biirkei

Plants without aerial stems; with rhizomes; raceme simple or branched; nerves of leaf-bases often

persistent as fibres;

Inflorescence simple;

Ovary glabrous;

Leaves usually less than 5 mm wide, variously pubescent but not with long reflexed setae

only:

Leaves flat or terete, glabrous or with soft long hairs and short curly ones; scape sharply

arcuate below pushing inflorescence outside leaf rosette; 4-8 ovules per loculus

(Summer rainfall region) 5. T. saltii

Leaves filiform, glabrous; flowers congested near the apex; pedicels short; 2 ovules

per loculus (Great Winterhoek Mountains, Cape) 6. T. esierhuysenae

Leaves filiform with a uniform, white, short curly pubescence; inflorescence lax; pedicels

ca. 15 mm long; 6 ovules per loculus (Calvinia) 7. T. gracilenta

Leaves usually 8 mm wide, flat; upper surface glabrous, lower surface and scape evenly

setose with long, straight, retrorse hairs ; immature scape densely reflexi-pilose ; pedicels

untidy, straggling in all directions, patent or erect in fruit, 15-25 mm long, thin; raceme

long, lax, flowers in irregular groups along rhachis 8. J. reflexipUosa

Ovary densely pubescent with tawnish, erect hairs 9. T. margaretae

718

Inflorescence branched:

Plants rosulate:

Mouth of leaf-base fimbriate (eastern Cape) 10. T. affinis

Mouth of leaf-base not fimbriate (Cape Peninsula):

Sclerotic plants; leaf margin raised, smooth; flowers laxly arranged on rhachis; perianth

segments ca. 8 mm long; capsule globose, dry; seeds smooth, usually 1 per

loculus 1 1 . r. brachypoda

Soft plants; leaf margin not raised, minutely denticulate; flowers fairly close together;

perianth segments ca. 12 mm long; capsule trisulcate, slightly fleshy; seeds with

3 prominent, crenulate ridges, tubercular, usually several per loculus 12. T. tabularis

Plants sub-distichous, softly hairy; leaves soft, erect; seeds with small ridges, somewhat

tuberculate 13. T. hirsiita

Section 2. TRACHYANDRA. [Dilanthes Salisb. Fragm. 70 (1866)].

Roots many, swollen, arranged in a congested circle, fusiform or contracted,

sometimes fused together and with the rhizome. Outer leaves changed to squamae.

Inflorescence branched, often with accessory branches at the nodes, seldom simple

through reduction. Perianth erect or pendulous, rotate or recurved, maculate (except

in T. chlamydophylla, T. hispida, T. pecidiaris and T. hirsutiflora where it is immaculate) ;

filaments subequal, shortly scabrid or dimorphous with the three outer spreading,

shortly scabrid, the three inner connivent around the ovary, with dorsal and lateral

fringes, curving outwards above and there with long retrorse bristles; ovules 16-4 per

cell. Capsule glabrous (except in T. zebrina and T. hirsutiflora) never stipitate. Seeds

smooth or vermcose.

Distribution: Found mainly in the south western Cape; a few species belonging

to the tumbleweed group found northwards as far as Southern and Northern Rhodesia

and Angola.

Inflorescence a simple raceme or rarely with a few short basal, ascending branches, elongating

during anthesis; scape often with a few sterile bracts (vestiges of undeveloped side branches);

plants glabrous or hairy; perianth immaculate (maculate in T. falcata and T. ciliata):

Each leaf-base wrapped up in separate tubular, brown, firm squamae but no squamae surrounding

shoots; inflorescence simple; rhizome horizontal; roots all alike, hard, cylindrically

swollen, not bulbous; perianth immaculate, glabrous 14. T. chlamydophylla

Each shoot as well as the leaf-bases and scapes wrapped up in tubular, membranous, loose

squamae, the outer bursting when plants expand; inflorescence simple or with a few short,

basal branches (T. falcata)-, pedicel usually recurved in fruit (except in T. hirsutiflora)-,

rhizome vertical; roots often bulbous; perianth immaculate or maculate, hispid on the

outside (glabrous in T. longepedunculata):

Inflorescence simple, scape naked:

Racemes capitate, dense, short, the lower pedicels long, overtopping the apex of the

inflorescence; pedicels recurved in fruit; leaves usually longer than the raceme:

Bracts long, linear-acuminate, ending in a soft awn, hispid; (Cape Peninsula) 15. T. hispida

Bracts deltoid, glabrous, white, membranous, edges fimbriate (South West Africa)

16. T. pecidiaris

Racemes spicate, white- or fawn-tomentose ; pedicels erect in fruit 17. E. hirsutiflora

Inflorescence with a few, short side-branches or rarely simple through reduction, if simple

then with 1-3 sterile bracts below inflorescence (vestiges of side branches), when old

often prostrate:

Leaves flat, ca. 2-5 cm wide; perianth softly hairy on outside, sometimes glabrous or

becoming glabrous: young inflorescences with bracts closely imbricate, resembling an

ear of corn; large plants:

Lower bracts not amplexicaul; inflorescences ultimately prostrate; pedicels recurved

in fruit 18. T. ciliata

Lower bracts amplexicaul; inflorescence erect; pedicels erect in fruit \9. T. falcata

Leaves terete, 2 mm in diam. scapes bracteate, often mottled at the base; perianth glabrous

outside; slender, glabrous plants 20. T. longepedunculata

Inflorescence a divaricate raceme, seldom simple in immature or starved plants; scape naked,

erect; perianth maculate:

Squamae surrounding shoots absent. Leaves and scapes arranged horizontally on the discoid

rhizome; each leaf- and scape-base separately surrounded by several close-fitting tubular

719

squamae; roots many, firm, often spindle-shaped but not fused or contracted into tubers,

the new ring of spreading roots formed above the old one; perianth recurved; filaments

dimorphous, 3 inner with long, retrorse bristles in upper half; tumbleweeds:

Flowers white, pedicels up to 1 cm long:

Stout, glabrous, littoral plants; panicle branches usually short, dichotomous or trichotomous,

patent, many-flowered; capsule often somewhat fleshy, ca. 1 cm in diam., perianth

recurved from the middle 21. T. divaricata

Smaller plants, glabrous or with the base of scape hairy and leaves rough; not littoral;

racemes laxly flowered; capsule small, dry; perianth recurved from near the base:

Scape-base minutely, sparsely hairy, seldom glabrous; leaves rough, flat or rolled; lowest

branches not trichotomous; perianth segments ca. 8 mm long; filaments yellow at the

base (southern and south eastern Cape) 22. T. revoluta

Scape-base glabrous; leaves terete, smooth, sometimes glutinous; branches of raceme

alternate or the 3 lowest trichotomous with the apex of the scape clavate; perianth

segments ca. 12 mm long; filaments yellow in the middle (Kalahari sandveld) 23. T. laxa

Flowers yellow, plants slender, laxly flowered; pedicels up to 15 mm long 24. F. arvensis

Squamae surrounding shoots present, bursting when plants develop (often worn away in older

specimens); leaves and scapes arranged vertically on the narrow rhizome; leaf- and scape-

bases surrounded by thin, smaller inner squamae, which are usually different to the outer

ones; roots various; perianth spreading or recurved ; filaments uniform or slightly dimor-

phous:

Roots many, not fused or contracted:

Leaves 2 or few, opposite, flat, up to 5 cm wide, smooth or usually muricate ; roots many,

usually fairly thin and long 25. T. niuricata

Leaves many (if few, roots spindleshaped), linear, hairy or glabrous; squamae numerous,

white, entire or inner fimbriate:

Roots somewhat spindleshaped ; inner squamae often fimbriate ; leaves minutely setaceous

or glabrous; filaments minutely papillate 26. T. bulbinifolia

Roots not spindleshaped, usually long, felted:

Plants lanate; inflorescence umbellate (Namib) 27. T. lanata

Plants with retrorse, coarse hairs; inflorescence not umbellate, compact; flowers

mauve or pinkish (Karroo) 28. T. thyrsoidea

Roots few, contracted, bulbous, often fused into one or a few hard tubers or tuber split at

the base into several points:

Pedicel recurved in fruit; small plants; leaves few, flat glabrous, linear-lanceolate, usually

transversely plicate 29. T. tortilis

Pedicels patent or erect in fruit:

Ovules 12-16 per cell; ovary oblong; leaves linear, up to 5 mm wide, glabrous or with

reflexed, long, silky hairs, wavy when young; large plants up to 40 cm high; panicles

divaricate with accessory branches in nearly all axils; pedicels short ascending;

capsule ultimately touching rhachis; roots usually fused, often hard. . 30. T. jacquiniana

Ovules 6-1 1 per cell; ovary ovoid; leaves filiform or (in T. paniculata) linear-lanceolate,

variously hairy or glabrous; plants usually up to 30 cm high; panicles with or

without accessory branches:

Leaves few, flat, linear-lanceolate, ca. 1 cm broad, glabrous above, shortly pubescent

below: panicle with accessory branches; roots soft, bulbous, not fused J^l.T.panicidata

Leaves several, filiform or linear, less than 1 cm broad, glabrous or variously pubescent:

Capsule, pedicel and rhachis shortly setose with sparse, reflexed bristles; squamae

forming a long neck, often transversely striped with dark bands. ... 32. T. zebrina

Capsule, pedicel and rhachis glabrous :

Pedicels 1-3 cm long, patent:

Inflorescence about as long as the leaves, pedicels up to 3 cm long, side-branches

of inflorescence about as long as main branch; roots fused into a hard

tuber split below 33. T. kanooica

Inflorescence overtopping the leaves, divaricately branched with many accessory

branches; pedicels 10-15 mm long; squamae usually forming a long.

sleek, brown neck; root-tubers not fused 34. T. patens

Pedicels short, less than 1 cm, erect, capsule ultimately touching rhachis:

Leaves 3-15, soft, more or less flat above, convex below, glabrous or with a

few to many, retrorse bristles along margin; bracts aristate; raceme simple

or few-branched ; small plants 35. F. oligotricha

Leaves 2-6, usually wavy, wiry, light green, ribbed, muricate or glabrous; scape

shortly hairy or glabrous; bracts aristate; plants variable in size 36. T. fle.xi folia

Leaves ca. 3, filiform, straight; plants glabrous, racemes divaricate, laxly

flowered ; bracts minute, mucronate ; outer squamae hard, brown, forming

a neck, inner thin, breaking up into shreds 37. T. dissecta

720

Section 3. GLANDULIFERA.

Roots wiry or stout and woody, often with some fusiform swellings near the tips.

Rhizome woody, sometimes produced into short branches. Leaves uniform, triquetrous

or flat and keeled, glabrous or hairy, bases often persisting as fibres. Inflorescences

branched, often with accessory branches, rarely simple through reduction. Perianth

erect, rotate, maculate; filaments spreading, uniform, shortly scabrid; ovules 6-2

per cell, the lower sometimes abortive. Capsule covered with a few to many gland-

tipped tubercles, usually obovoid or with a short stipe. Seeds verrucose. Glands

few to many, either confined to a few on the ovary, which become stipitate when the

capsule develops, or the whole raceme glandular; in some species the raceme is scabrid

with truncate, gland-tipped tubercles; the glands disappear with age.

Distribution: Two species are endemic in or near the Cape Peninsula in sandy

habitats, a third is found only in South West Africa, the majority occur in the eastern

Cape, northwards to the eastern Transvaal, on mountain grasslands or in marshy

places, extending as far as Nyasaland.

Leaves in a terminal fascicle at the end of very short, woody branches; ovules 2 per cell:

Pedicels patent, recurved at the apex in fruit (South Western Cape):

Capsule ob-triangular, contracted at the base, with a few gland-tipped tubercles and some

transverse ridges 38. T. scabra

Capsule globose (stipe hidden) densely covered with dendroid, gland-tipped tubercles 39. T. sabidosa

Pedicels erecto-patent in fruit; plants densely glandular, glands shortly stipitate (South West

Africa) 40. T. glandulosa

Leaves rosulate or somewhat distichous from a basal rhizome; pedicels erect or twisted; ovules

6-2 per cell (eastern Cape to eastern Transvaal):

Ovules 6-A per cell; base of plant fibrous; capsule not horned:

Capsule ca. 5 mm in diam. when fully developed (unknown in T. capdlata) with few to many

short gland-tipped tubercles and some transverse ridges on the skin:

Roots thin fibrous, fusiform near the tip:

Leaves filiform to linear, usually 15-20 cm long, tapering to the apex, grasslike 41. T. aspercita

Leaves few, triquetrous, over 40 cm long, each face ca. 8 mm broad, soft, with soft golden

hairs 42. T. capdlata

Roots red, stout, ca. 4 mm in diam. cylindrical; leaves flat, somewhat distichous; pedicels

erect 43. T. erythrorrhiza

Capsule ca. 10 mm in diam. when fully developed, hurry, being covered with pectinate, gland-

tipped tubercles; robust plants with hairy leaves and scabrid-glandular, branched

racemes 44. T. gerrardii

Ovules 2 per cell; base of plant not fibrous; capsule with 3 lateral horns 45. T. gijfetdi

Section 1. LIRIOTHAMNUS (Schltr.) Oberm., stat. nov.

Liriothamnus Schltr. in Not. Bot. Gart & Mus., Berlin 9: 145 (1924).

1. T. adamsonii {Compton) Oberm., comb. nov.

Liriothamnus adamsonii Compion in LBoi-Lond. 69 \ 10(1931). Type: southwestern

Cape, Vanrhynsdorp, mouth of the Doom River near Klaver, Compton (NBG. 318/22,

holo.!).

Shrubs with a vertical, woody stem, up to 180 cm high, branched near the apex.

Roots about 5 mm in diam. lanate. Stems ca. 2 cm in diam. covered with hard leaf-

bases which may eventually disappear. Leaves in tufts near the apex, somewhat fleshy

glaucous, glabrous, 16-30 cm long, 15-35 mm wide, with about 25 nerves, margin

minutely denticulate. Inflorescence axillary, simple or with 1-2 small basal branches,

30-50 cm long, bracts widely ovate, long acuminate, 10 mm long, 3 mm wide,

membranous, margin minutely denticulate; pedicels erect, stout, up to 10 mm long

in fruiting stage. Flower with perianth white or faintly flushed with pink, maculate

and probably recurved; segments up to 14 mm long; filaments scabrid; ovary with

ca. 9 ovules per loculus, style terete, stigma minute. Capsule erect, ovate, 12 mm long,

5 mm broad, apex obtuse. Seeds 3 mm long, tuberculate.

Flowering Period : August.

Distribution: Only known from the type area.

721

Cape. — Clanwilliam; Doom River Bridge, Compton 22779 (NBG); Doornpoort,

Hall 810 (NBG).

Live plants of this interesting species were collected by Compton in 1922 and

cultivated at Kirstenbosch where they flowered in August, 1926. Adamson collected

fruiting material at the type locality in September 1923 and examined the secondary

thickening of the stem which was like that of Dracaena.

2. T, involucrata (Bak.) Obenn., comb. nov.

Anthericum involucratum Bak. in J. Linn. Soc. 15; 311 (1876). Type: Cape, Nama-

qualand near Mierenkasteel, Drege 2681 (K, holo., L, iso.! PRE, photo.).

Liriotliamnus involucratus (Bak.) Schltr. in Notizbl. Bot. Gart. & Mus. Berlin 9: 145

(1924).

Small gnarled, woody shrubs up to 60 cm high. Boots many, slightly thickened,

lanate. Stems ca. 1 cm in diam. at the base, 3 mm in diam. near the tips. Leaves in

tufts near the apex usually on short young branches; leaf-bases forming a membranous,

wide, conspicuous tube sometimes produced into a denticulate point opposite the

lamina; lamina terete, succulent, ca. 17 cm long, ca. 6 mm in diam. glaucous, glabrous,

canaliculate, long acuminate in upper half. Inflorescence simple or branched, up to

26 cm long; bracts small, 7 mm long, acuminate, scarious; pedicels up to 2 cm long

in fruit, recurved in an S-shaped loop. Flower with perianth rotate, immaculate, white,

segments 10 mm long, 2 mm broad; filaments spreading, scabrid; ovary globose with

about 9 biseriate ovules per cell. Capsule pendulous narrowly ovoid, 18 mm long,

tapered below and above; opening at the apex where the points then recurve giving

it a 3 horned appearance. Seeds (immature) verrucose?

Flowering Pei icd: July. A faint scent was noted at times when a plant flowered

at the Division of Botany.

Distribution: Namaqualand.

Cape. — Namaqualand: Richtersveld, Twee Rivieren, Marloth 12268 (PRE); Kubus

Mt., Dyer & Verdoorn 1841 (PRE); Nigramoep, Acocks 19355 (PRE); Karee Mt.,

Schlechter 8190 (PRE).

3. T. acocksii Oberm., sp. nov., distincta, nullis e specibus notis propinqua.

Plantae ad 45 cm altae gregariae, caulibus brevibus lignosis. Folia glabra glauca

dura, 12-30 cm longa, basi reliquiis penniformibus breves erecti. Flores basi cupuli-

formes; ovula in loculis 4. Capsida globosa parva stipitata. Semina glabra.

Small glabrous, glaucous shrublets up to 45 cm high, with ha^d leaves, growing

in dense clumps. Roots hard, woody. Stems short, branched, woody, densely coveted

with the congested, hard quill-like leaf-bases. Leaves 12-30 cm long, triangular in cross-

section, hard, glaucous, glabrous, straight or with a lax spiral twist when young.

Inflorescence a simple raceme up to 50 cm long; bracts minute, subulate, membranous,

white; pedicels erect, up to 15 mm long in fruit. Flower with perianth forming a short

cup at the base, segments white, spreading, keel b'oad and dark; filaments uniform,

spreading; ovary with ca. 4 ovules per cell. Capsule globose, dry, 5 mm in diam.,

contracted at the base. Seeds about 2 per cell, smooth, grey.

Flowering Period: January-May, apparently after rains.

Distribution: Recorded only from the Great Karroo, where the plants are

confined to the eastern slopes growing high up on rugged hills, amongst boulders.

Cape. — Britstown: Brewershoek, Loots (PRE, 7609). Beaufort West: Watt (PRE,

26454); near Nelspoort, Acocks 15877 (PRE). Richmond: 7 miles W. N.W. of

Richmond, Acocks 16338 (PRE, holo.!).

4. T. burkei {Bak.) Oberm., comb. nov.

Anthericum burkei Bak. in J. Bot. Lond. 1872, 140; in J. Linn. Soc. 15: 298 (1876).

722

Type; Transvaal, Apies River, Burke (K, holo.). The locality is doubtful as the species

occurs only in the eastern Kalahari sandveld.

Bulbinella biirkei (Bak.) Benth. in Gen. Plant. 3: 784 (1883); Bak. in FI. Cap. 6: 358

(1896).

Hard, grasslike plants forming short, woody stems at the base which are densely

covered with reticulate fibres. Roots many, covered by a short, felted covering of

roothairs. Stems short, woody below, the new shoots formed laterally, covered with

the tubular, reticulated leaf-bases which afterwards remain as coarse, long fibres. Leaves

few, the long, reticulated leaf-bases tightly clasped around the young leaves and bases

of scapes, venation consisting of parallel sclerenchymatous veins connected by ascending

oblique lateral veins; lamina semi-terete, flattened above, up to 40 cm long, 2 mm

wide, hard. Inflorescence a divaricate raceme with di- and trichotomous branching,

ca. 30 cm tall; scape short, arcuate at the base, terete, glabrous; pedicels, patent,

up to 15 mm long; bracts minute, subulate. Flower with perianth spreading or slightly

recurved, segments ca. 6 mm long, maculate with 2 green or dark spots at the base;

filaments erect shortly scabrid; ovary with 2 ovules per cell. Capsule small, erect,

ca. 5 mm, somewhat 3 lobed, constricted at the base. Seeds apparently 1 per cell (no

ripe seed seen).

Flowering Period: September-March. Flowers sweetly scented {Acocks).

Distribution: A psammophyte found in the south eastern Kalahari sandveld*

Recorded from the Kimberley district, north western Free State, south western Trans'

vaal and Bechuanaland.

Cape. — Kimberley: near Riverton, Acocks 742 (PRE).

Orange Free State. — Boshof: Smitskraal, Burtt Davy (PRE 11312, 11356).

Transvaal. — Christiana: Kameelpan, Theron 518 (PRE). Wolmaransstad: Boskuil,

Sutton 53 (PRE). Lichtenburg: Kinges 1890 (PRE). Mafeking: Brueckner 451

(PRE), Leistner 568 (PRE).

Bechuanaland. — 14 miles N.W. of Molepolole, Codd 8943 (PRE).

The reticulate structure of the persistent long leaf-base, resembling some Bulbinella

species is unique in the genus. This character no doubt induced Bentham to place it

in that genus but the structure of the flower and capsule make it a Trachyandra.

5. T. saltii (Bak.) Oberm., comb, nov., aggregate species.

Anthericiim saltii Bak. in J. Linn. Soc. 15 : 309 (1876); in FI. Trop. Afr. 7; 492 (1898).

Fig. 5.

Grasslike plants 10-50 cm high. Roots many, wiry, fairly stout. Rhizome vertical.

Leaves variable in size, filiform to linear, 5-50 cm long, 1-25 mm wide; lamina flat

or rolled, tapering to the apex, below gradually dilated into a tubular, membranous

base, mouth fimbriate; glabrous or pubescent with long straight hairs sometimes

intermixed with short curly hairs; nerves in basal part of leaf sometimes persistent

as fibres (if sclerenchymatous because of dry conditions). Inflorescence a simple

raceme, usually many-flowered; several young racemes usually present near the base

of the plant; scape arcuate near the base, protruding outside the leaf-rosette, 6-40 cm

long; bracts small, narrow, cuspidate, margin fimbriate or smooth; pedicels 8-15

mm long, ascending, sometimes recurved in fruit. Flower with perianth segments

ca. 1 cm long, typical: stamens typical; ovary turbinate, with ca. 8 ovules per cell,

the lower usually aborting. Capsule globose 5 mm in diam. glabrous, constricted at

the base. Seeds grey, smooth.

Flowering Period: November-March after good rains. A number of flowers

open in the afternoon and fade during the night, the inflorescence finishing its flowering

period in 2 or 3 days. Young inflorescences are present below and emerge at the next

favourable opportunity.

723

724

Distribution: Common on the highveld and eastern grassveld of the Cape,

Natal, Orange Free State and Transvaal, Bechuanaland, South West Africa, Southern

Rhodesia and extending northwards to Abyssinia. Absent from the karroo and winter-

rainfall districts.

This species seems very adaptable; it has been found in dry surroundings growing

only a few inches high or near the river banks, 20 inches high. It has been

described several times throughout its range but no sound differences could be found

between the synonyms cited. Mr. John Lang of the British Museum was kind enough

to compare specimens from South Africa with the type of Anthericum saltii Bak. and he

found them to be conspecific.

Baker and others after him, attached some importance to the absence or presence

of basal fibres, referring the fibrous ones to Anthericum micranthum Bak. and those

without to A. elongatum var. holostachyum Bak. This sclerenchymatous tissue is

developed more profusely under dry conditions, but has no systematic value.

From the ample material seen, I could group the specimens roughly into 3 forms.

These were not restricted to geographical areas. The distribution for instance of the

var. secunda with pedicels recurved in fruit, agrees with that of var. saltii in South

Africa. The var. oatesii with its long, erect, pedicels was found in Southern Rhodesia

and in the Kimberley area; side by side, however, the var. saltii with short, erect pedicels

also appeared.

Key to Varieties

Pedicels erect in fruit:

Pedicels, ca. 1 cm long a. var. saltii

Pedicels ca. 2-3 cm long b. var. oatesii

Pedicels recurved and often pseudo-secund in fruit, ca. 15 mm long c. var. secunda

{a) var. saltii.

Anthericum saltii Bak. in J. Linn. Soc. 15: 309 (1876). Type: Abyssinia, Salt (BM,

holo.). A. micranthum Bak. in J. Bot. 1891, 71; et in FI. Cap. 6: 388 (1897). Type:

Cape, Griqualand West, du Toit’s Pan near Kimberley, Elliott 1220 (K, holo., incomplete,

according to Miss Kies who examined the type). A. elongatum var. holostachyum

Bak. in Flor. Cap. 6: 389 (1897). Type of var.: Cape, Griqualand East, Tyson 2122

(K, holo.). .4. Dinter in Fedde, Rep. 29: 266 (1931). Poelln. in Fedde,

Rep. 52: 235 (1943). Type: South West Africa, Otavi, Dinter 5275 (B, holo.! PRE,

photo.). A. aristatum Poelln. in Fedde, Rep. 50: 321 (1941); et in Fedde, Rep. 52:

234 (1943). Type: South West Africa, Grootfontein, Schoenfelder 919 (B, holo.!,

PRE, iso.!). A. hrevitepalum Poelln. in Bol. Soc. Brot. 16, 2: 71 (1942). Type:

Transvaal. Komatipoort, Schlechter 11846 (B, holo.!, PRE, photo.). A. gracilitepalum

Poelln. in Bol. Soc. Brot. 16, 2: 53 (1942). Type: Transvaal, Middelburg, Nazareth,

Schlechter 4480 (B, holo.!, PRE, BOL, iso.!). A. amboense Poelln. in Fedde, Rep.

52: 233 (1943). Type: South West Africa, Amboland, Umkuanyama, Rautanen 439

(B, holo.? probably destroyed; K, iso.). A. cepaefolium Dinter apud Poelln. in Fedde,

Rep. 52: 235 (1943). Type: South West Africa, Windhoek, Auas Mts., Dinter 1885

(B, syn. ? probably destroyed, NBG. iso.!) Gaub, Dinter 2458 (NBG, iso.!). A. lanzae

Cuf. in Miss. Biol. Borana, Race. Bot. Angiosp. IV, 306, photo, p. 307 (1939), e descr.

Type: Eritrea; Arero, Miss. Biol. 321 (RO, syn.); Javello, Miss. Biol. 544 (RO, syn.)

A. kdssneri Poelln. in Fedde, Rep. 50: 322 (1941) e descr. Type: Kenya, Kdssner

(B, holo.?).

South West Africa. — Grootfontein: Auros, Volk 801, 8368 (M). Keetmanshoop:

We hern Karasberg, Pearson 7970 (BOL, distributed under the name A. pilosum Bak.).

Cape. — Barkley West: Newlands, Ferrar 24 (NH); Warrenton: near Warrenton,

Hafstrom H950 (PRE). Kimberley : Acoeks 11 (PRE), Esterhuysen 1 1 59 (BOL).

725

Griqualand East: near Clydesdale, Tyson 2122 (GRA). Middelburg: Bangor farm,

Bolus 14047 (BOL), Gill 109 (PRE).

Basutoland. — Leribe, Dieterlen, 322, 348 (PRE).

Orange Free State. — Fauresmith: Smith 960 (PRE). Kroonstad: Pont 625 (PRE).

Vredefort; Parys, Bruce 145 (PRE). Bloemfontein: Lam & Meeuse 4787 (L.).

Natal. — Estcourt: Acocks 10610 (PRE). Polela: Killick & Marais 2089 (PRE).

Transvaal. — Sibasa: Kruger National Park, Baiandbai Lang (TM 32150, PRE, NH);

near Punda Maria, Lang (TM 31107, PRE). Warmbaths: Sidey 1159, Collins 2409

(PRE). Carolina: Galpin 12211 (PRE). Germiston: Modderfontein, McLean (PRE).

Heidelberg: Suikerbosrand, Schlechter 3498 (PRE, BOL). Christiana: Theron 437

(PRE).

Southern Rhodesia. — Urungwe Reserve, Mgunje, Wild 4155 (SRGH, PRE); Mt.

Selinda, Hack 54 (SRGH). Umtali, Chase 5938 (SRGH, PRE). Salisbury, Wild

3685 (SRGH).

Northern Rhodesia. — Mapanza, Choma, Robinson 2948 (SRGH); Medinilunga,

Milne-Redhead 1078 (PRE).

Kenya. — Ngong Hills, Greenway 8475 (PRE); western slopes of Kilimanjaro, Greenway

4394 (PRE); Moyale, Gillett 13980 (PRE; named Anthericiim lanzae Cufodontis).

Uganda. — Ruizi River, Jarrett 341 (PRE; named A. kdssneri Poelln.).

{b) var. oatesii (Bak.) Oberm., stat. et comb. nov.

Anthericum oatesii Bak. in J. Bot. Lond. 1878; 324; et in Oates, Matab. ed. 2: 411,

t.l3 (1889). Type: Southern Rhodesia, Matabeleland, Oates (BM, holo.). A. betschu-

anicum Poelln. in Bol. Soc. Brot. 16, 2: 53 (1942). Type: Cape, Bechuanaland,

Kuruman, Marloth 1014 (B, holo.!, PRE, iso.!).

Southern Rhodesia. — Umvukwes, Mazoe, Wild 3985 (SRGH). Matobo, West 2506

(SRGH). Farm Besnakobila, Miller 4867 (SRGH, PRE); Embakwe, Feiertag (SRGH,

45422). Makoni, Maidstone: Norlindh & Weimarck 4132 (SRGH).

Cape. — Kimberley: Macfarlane, Acocks 1410 (PRE, BOL).

Orange Free State. — Fauresmith: Henrici 4247 (PRE).

(c) var. secunda (Krause & Dinter) Oberm., stat. et comb. nov.

Anthericum secundum Krause & Dinter in Engl. Bot. Jahrb. 45: 127 (1911). Poelln.

in Fedde, Rep. 52: 260 (1943). Type: South West Africa; Grootfontein, Dinter

855 (B, holo.? probably destroyed; NBG, iso.!). The description noted that the

pedicels were erect after the flowers had dropped off; probably the flowers were not

fertilised. The type number in NBG shows recurved, fruiting pedicels. In the original

description the length of the pedicel was stated to be 1 dm, but this must have been

1 cm. In some specimens only four ovules were counted in a cell.

Transvaal. — Pretoria: Magaliesberg, Witfontein, Smith 12 (PRE); Wonderboom,

Repton 2762 (PRE). Germiston: Modderfontein, Conrath 660, 665 (GZU). Nelspruit:

Pretorius Kop, van der Schijff (PYCE). Petersburg: Moss 15665 (PRE, J.). Pilgrim’s

Rest: Calais, Killick & Strey 2557 (PRE).

Orange Free State. — Senekal: Doornkop, Goossens 684 (PRE).

Cape. — Albert: Burghersdorp, Pocock 103 (GRA).

Natal.— Hlabisa: Ward 1557 (PRE). Richard’s Bay, Lawn 1728 (NH).

Portuguese East Africa. — Delagoa Bay, Junod 399 (SRGH); Inyamosan, Schlechter

12072 (PRE).

Nyasaland. — Zomba, Jackson 2082 (SRGH).

Southern Rhodesia. — Inyanga, Chase 3683 (SRGH); Fries, Norlindh & Weimarck

(SRGH).

L

726

6. T. esterhuysenae Oberm., sp. nov. T. saltii afiinis sed plantis glabrescentibus

ovulis duobus differt.

Plantae graminiformes. Folia subteretia glabra, basi reliquiis duris longis fibrosis.

Racemi simplices ad apicem congesti; scapus bracteis minutis; pedicelli erecti. 0\ula

in loculis 2 pendula. Capsula erecta globosa.

Grass-like plants up to 50 cm long. Roots unknown. Rhizome unknown. Leaves

semiterete to narrowly linear, up to 50 cm long, 1-2 mm broad, glabrous; leaf-bases

persisting as hard, long fibres. Inflorescence simple, flowers aggregated near the apex;

scape erect, straight, up to 50 cm long, with some empty bracts below raceme; bracts

small; pedicels short ca. 5 mm long, ascending to erect. Flower with perianth segments

ca. 5 mm long; filaments papillate (nearly smooth in one flower); ovary with 2

pendulous, collateral ovules in each loculus. Capsule (immature) globose, erect.

Flowering Period: February.

Distribution. — Apparently confined to high altitudes on mountains in the south

western Cape districts.

Cape.^ — ^Worcester: Slanghoek Mts., Observation Peak, shale band. Ester Imysen 5613

(BOL, holo.!, PRE, photo.). Caledon: Kogelberg, Esterhuysen 9961 (BOL). Stellen-

bosch: Hottentots Holland Mts., above Diepgat, Esterhuysen 16728 (BOL); Banhoek

Kloof, Esterhuysen 19902 (BOL). Piketberg: Twenty-four Rivers Mts., above Porter-

ville, sandy, swampy flats, Esterhuysen 16616 (BOL).

The 2 pendulous collateral ovules show affinity to Bulbmella, which it does resemble

superficially. In Bulbinella however, the periantih is star-shaped and persistent where-

as our species has a typical Trachyandra perianth, for in fading the segments fuse above

the ovary, and it then drops off, where the pressure of the expanding capsule tears

it apart. The species resembles T. tabularis but this species has a branched inflorescence

and 6 ovules per cell to mention but a few differences.

I have named it after Miss E. Esterhuysen who collected it in four different localities.

So far there is no record of anybody else having collected it.

7. T. gracilenta Oberm., sp. nov. T. esterhuysenae Oberm. alfinis, sed ita differt:

plantae pilis minutis albis crispis indutae, pedicelli longiores, loculi 6 non 2-ovulati.

Plantae graminiformes. Radices multae lanatae ad acumen tumescentes. Folia

linearia tomentosa, pilis minutis albis crispis induta. Racemi simplices laxi, scapo

tomentoso, pedicellis sub anthesi 15 mm longis. Ovula in Kculis 6.

Grass-like plants up to 40 cm high. Roots many, covered with felted roothairs;

with some swellings near the tip. Rhizome small, woody. Leaves linear, rolled, 40 cm

long, 1-2 mm broad, ribbed, evenly pubescent with small, white curly hairs, bases

persistent but usually not breaking up into fibres. Inflorescence a simple, lax raceme,

20-30 flowered: scape slender terete, minutely pubescent; bracts minute, white,

subulate, fimbriate, folded around pedicel: pedicels 15 mm in flower, spreading.

Flowers with perianth probably spreading, thin, dark keeled, segments 1 cm long;

stamens uniform, shortly scabrid; ovary with ca 6 ovules per cell. Capsule unknown.

Flowering Period: September.

Distribution: Only known from type locality.

Cape. — Calvinia, Lokenburg, Stinkfontein hills, on flat top; arid fynbos vegetation,

Acocks 18550 (PRE, holo.!).

A graceful, grass-like plant distinguished from T. esterhuysenae by its pubescence,

and the longer pedicels. Moreover the ovary contains 6 ovules per cell, not 2.

8. T. reflexipilosa {O. Ktze.) Oberm., comb, et stat. nov.

A. flliforme Thunb. var. reflexipilosum O. Kuntze, Rev. Gen. 3, 2: 315 (1898). Type:

Natal, Charlestown, Otto Kuntze s.n. (K, holo.).

727

Setose, robust, many-leaved plants up to 70 cm high. Roots many, wiry, thin.

Rhizome small woody. Leaves linear-lanceolate, up to 45 cm long, 8 mm broad, flat

or slightly rolled, upper surface glabrous, lower ribbed, retrorsely fulvous setose on

margin and ribs, seldom nearly glabrous. Inflorescence a simple, many-flowered, tall

raceme, up to 1 m high; scape terete, retrorsely setose especially near the base and when

immature; rhachis glabrous; bracts small, subulate, surrounding pedicel, flmbriate;

pedicels slender up to 25 mm long in fruit, glabrous, irregularly spaced along rhachis

and spreading haphazardly in all directions. Flowers with perianth segments spreading,

1 cm long; filaments scabrid, uniform; ovary with 6 ovules per cell. Capsule gXohosQ,

4 mm in diam. contracted at the base, glabrous. Seeds (immature) with oblong white

raphides (?) on skin.

Flowering Period: October-March.

Distribution: Northern Natal, Swaziland, eastern Transvaal, eastern Southern

Rhodesia, on moist grassland, swamps or black turf.

Natal. — Vryheid: Burtt Davy (PRE, 11470).

Swaziland. — Mbabane, Forbes Reef Road, swamp, Compton 2151 A. (PRE); Hull’s

Farm, Compton 25431 (PRE).

Transvaal. — Lydenburg: Wilms 1507 (PRE, L). Pilgrims Rest: Graskop, Galpin

14480 (PRE). Nelspruit: Mauchsberg, Sabie, C/7/c/t 2289 (PRE). Barberton:

Shiya-lo-ngubo Dam, Codd 6426 (PRE). Witbank: Springbok Colliery, Kies 388

(PRE).

Southern Rhodesia. — Melsetter: Chimanimani Mts., Mt. Poza, grasslands, Goodier

509 (SRGH, PRE).

A mountain grassland species resembling T. saltii but distinguished by the long,

reflexed tawny hairs, which are characteristic especially on the young racemes emerging

from between the leaves. The thin, untidily straggling pedicels are also typical. In

drying the plants usually turn black.

9. T. margaretae Oberm., sp. nov., T. saltii, afflnis, sed ovario piloso differt.

Plantae graminiformes gregariae. Radices fibrosae durae. Folia filiformia leviter

hirsuta pilis patentibus. Racemi simplices, primum prope apicem floribus congest!,

scapo sub anthesi extendo. Ovarium hispidum pilis longis erectis indutum; ovula in

locuiis 4-6.

Hard, grasslike plants growing in clumps up to 60 cm high. Roots hard, thin,

woody. Rhizome small, woody. Leaves terete or linear, up to 40 cm long, 2 mm broad,

canaliculate or rolled, ribbed, hard, sparsely hairy with small patent hairs or nearly

glabrous, leaf-bases persistent as fibres. Inflorescence simple, glabrous, flowers at first

congested near the apex, rhachis elongating during anthesis; scape terete; bracts

small, clasping the pedicel, subulate, fimbriate; pedicels ascending, up to 12 mm long.

Flowers with white, glabrous perianth segments 1 cm long; filaments very shortly

scabrid; ovary densely covered with long erect hairs; ovules 4-6 per cell. Capsule

not seen.

Flowering Period: August-November.

Distribution: Natal, eastern Transvaal to Woodbush Mts., apparently in damp

grassveld.

Transvaal. — Barberton: Saddleback Mountain, summit, damp hollows, Galpin 538

(PRE, holo.). Nelspruit: Sabie, Pole Evans (PRE, 28725, SRGH). Pietersburg:

Woodbush Mts., grassveld. Moss 15434. (PRE, J).

Natal. — Utrecht: Tweekloof, Thode A388 (PRE). Estcourt: Cathedral Peak, Ester-

huysen 15464 (BOL).

The species is closely related to T. saltii but the hairy ovary is a very unusual

feature. Professor C. E. Moss named the plant margaretae after his wife, but he did

not publish the name.

728

10. T. affinis Kunth, Enum. 4: 579 (1843). Type: Cape, Port Elizabeth, Addo,

Drege 8727 (P, iso., PRE, photo.).

Anthericum affine Bak. in J. Bot. Lond. 10: 138 (1872).*

A. pubescens Bak. in J. Linn. Soc. 15: 309 (1876) et in El. Cap. 6: 390 (1897). Type:

Cape, Mountain grassland near Somerset East, MacOwan 1589 (K, holo.). A. pudicum

Bak, in FI. Cap. 6: 388 (1897). Type: Cape, Uitenhage, amongst shrubs near the

Zwartkops River, Zeyher 1070 (K, holo., BOL, PRE, iso.!). A. longiciliatum Poelln.

in Bol. Soc. Brot. 16, 2: 54 (1942). Type: Cape, Bathurst, Trapp’s Valley, Daly

561 (B, holo.! PRE, photo., GRA, iso.!). A. pseudofalcatum Poelln. in Bol. Soc.

Brot. 16. 2: 63 (1942). Type: Cape, Port Elizabeth, Zwartkops River, Ecklon &

Zeyher, Aspod. 113 (B, holo.!, PRE, photo.). Poellnitz in his description mentions

that the collector and locality were unknown; apparently he was unaware of the

Ecklon & Zeyher system of numbering places. A. longifolium var. burchelli Bak. in

J. Linn. Soc. 15: 312 (1876) et in FI. Cap. 6: 394 (1897). Type of var.: Cape, Port

Alfred, Burchell 3795 (K, holo.).

Slender or fairly robust plants up to 1 m high, solitary or a few together. Roots

wiry. Rhizome small. Leaves erect, somewhat fleshy, semiterete to linear, tapering

towards the apex, canaliculate above, up to 40 cm long, 8-16 mm broad, usually glabrous,

occasionally margin ciliate or lamina pubescent; mouth of tubular base fimbriate.

Inflorescence erect, branched, laxly flowered up to 1 m high; scape somewhat com-

pressed, pubescent; bracts small, ovate, acuminate; pedicels usually up to 16 mm

long in fruit (up to 4 cm in one variety), erect. Elowers white with spreading perianth

segments 12 mm long, immaculate; filaments scabrid; ovary with 6 ovules per cell.

Capsule globose, 5 mm in diam. Seeds smooth.

Fljwering Period: September-June. Flowers sometimes opening fairly early

in the day, sweet smelling.

Distribution: Humansdorp to Durban, in clearings or on grassy slopes near the

coast.

Cape. — Humansdorp: Eerste River, Eourcade 1186 (BOL); Hankey, Eoiircade 3328

(PRE). George: Sinksa Bridge, Wasserfall (PRE). Uniondale: Joubertina, Ester-

huysen 7099 (BOL); Haarlem, Esterhuysen 6034 (BOL). Bathurst: Port Alfred,

Rogers 28057 (GRA); Tyson (BOL 25743, GRA 10, PRE, in Herb. Marloth 8510);

Acocks 17692 (PRE). King William’s Town: Tyson 3072; Sim 1513 (BOL); Galpin

5955 (PRE). Albany: Grahamstown, Daly & Sole 100 (GRA); 319 (PRE); Britten

5169, 5137 (GRA); near Grahamstown, Dyer 2108 (GRA); Botha’s Hill, Lotsy &

Goddyn 149, 171 (L). Kentani: Nobongubo, Pegler 1397 (GRA); Qolora Mouth,

grassy slopes, Pegler 1349 (BOL).

Natal. — Umzinto: Scottburgh, in clearings along railway line. Mauve 4050 (PRE).

Claremont: Wood 7727 (NH).

The species is variable in size, pubescence and width of leaf, no doubt due to the

variation in climate. The specimens from George are usually smaller and more

pubescent than those from the warmer Natal South Coast. Some collectors record

that the plants appear after veld fires which stimulate early development. As a result

of injuries, usually from veldfires, the side branches of the raceme are often mutilated

and remain as small bracteate buds near the base of the scape. The apical bud however,

probably originally protected by the tissues of the side-branches, is stimulated to develop

into a long, densely flowered, simple raceme becoming prostrate with age. This

traumatic form was described by von Poellnitz as a separate species, A. pseudofalcatum

and by Baker as a variety, var. burchellii, of A. longifolium. Baker thought that the

* Baker places here A. filiforme /3, Thunb. (L, holo., PRE, photo). I have not seen the actual

specimen but this appears to be correct. In J. Linn. Soc. 15: 308 (1876) Baker makes it a variety, var.

affinis, of A. jacquinianum Schult. f.

729

small bracteate buds, that is the injured, sterile side-branches, could give rise to new

plants (“ scapi elongati, decumbentes, pilosi, e nodis vivipari ”), but this is a wrong

interpretation.

Some specimens from the Cape, e.g. Archibald 5697 from Alexandria district,

Langebosch Height, in pine plantations; Acocks 17692 from Bathurst district, Port

Alfred, West Bank, flats below dunes, possess very long pedicels, reaching a length

of up to 4 cm.

11. T. brachypoda (Bak.) Oberm., comb. nov.

Anthericum brachypodum Bak. in FI. Cap. 6: 389 (1897). Duthie in Ann. Stell. Univ. 4,

A: 10, t.Il, fig. 4, 5 (1926). Adamson & Salter, Flora of the Cape Peninsula 183 (1950).

Type: Cape, sand dunes near Cape Town, Bolus 3921 (K, holo.).* A. lowryense Bak.

in Bull. Herb. Boiss. Ser. 2, 4: 996 (1904). Type: Cape, Somerset West, Sir Lowry’s

Pass, Schlechter 5364 (Z, holo!, B, GRA, iso.!). A. obtusifolium Poelln. in Bol. Soc.

Brot. 16, 2: 74 (1942). Type: Cape, Table Mountain, 1/17, Bergius (B, holo.!, PRE,

photo.). A. validum Poelln. in Bol. Soc. Brot. 16, 2: 79 (1942). Type: Cape, Malmes-

bury, near Hopefield, Bachmann 811 (B, holo.!, PRE, photo.). A. submaculatum

Poelln. in Bol. Soc. Brot. 16, 2: 78 (1942). Type: Cape, Swellendam, Rhenosterkop,

Schlechter 10575 (B, holo!, PRE, photo.). There are only two buds on the plant.

Von Poellnitz described the flower from a detached one in a capsule and noted it to

be maculate. I doubt whether it belonged to this plant for T. brachypoda is immaculate.

A. tenuifolium Adams, in J. S.A. Bot. 9: 137 (1943); Adamson & Salter, Flora of the

Cape Peninsula 183 (1950), nom. nov. for A. stenophyllum Adamson (non Bak.) in J.

S.A. Bot. 7: 189 (1941). Type: Cape Peninsula, Smitswinkel Bay, Adamson 3128

(BOL, holo.!). A. brachypodum Bak. var. caespitosum Adamson in J. S.A. Bot. 7:

189 (1941). Adamson & Salter, Flora of the Cape Peninsula, 183 (1950). Type of var. :

Cape Peninsula, between Rondebosch & Kenilworth, Adamson 2789 (BOL, holo.).

Plants up to 70 cm high, xerophytic, solitary or in clumps. Roots many, whitish,

thin or slightly swollen at times. Rhizome horizontal, unbranched or branched, forming

clumps of plants. Leaves 4-12 per tuft, often persisting for a long time, erect, linear,

20-50 cm long, 2-4 mm wide, sclerotic, ribbed, glabrous, the margins usually raised,

yellow, occasionally with a spiral twist; sometimes outer leaves consisting of a

membranous sheath only without a lamina. Inflorescence usually much taller than

the leaves, branched or reduced to a simple raceme; lowest branches with unequal,

accessory branches, scape terete, firm, shiny; bracts small, deltoid apiculate; pedicels

short, up to 4 mm long in fruit, erect. Flowers with perianth immaculate, white, rotate;

segments ca. 8 mm long; ovary with 4 ovules per cell; Capsule globose, ca. 5 mm

in diam. Seeds ca. 2 mm in diam. black, minutely verrucose (mature?).

Flowering Period: October-April.

Distribution: Cape Peninsula and surrounding districts, “damp sandy soils

with humus ” {Adamson)-, on Stellenbosch Flats in sandy stony places, locally frequent

( Duthie).

Cape. — Peninsula: Table Mountain, Esterhuysen 11680 (BOL); Constantia Corner,

Salter 7164 (BOL); Vygekraal, Wolley Dod, 5577 (BOL). Stellenbosch: S.W. of

Stellenbosch, Salter 2049 (BOL). Bredasdorp: Struys Bay, Esterhuysen 4387 (BOL).

Swellendam: Rhenoster Hills, Marloth 12034 (PRE).

* In the Flora Capensis, Ecklon d; Zeyher, Asphod. 106, from mountains near Simonstown, is

also cited. Mr. W. Marais, who examined this specimen at K found it to be correctly named. In

the Berlin Herbarium this number represents a plant of A. hirsutum Thunb. together with a detached

inflorescence of an Urginea species. This collection was described by von Poellnitz as A. praetermissum,

in Bol. Soc. Brot. 16, 2: 62, 1942, a nomen confusum, cf. p. 579.

1517525-3

730

Adamson separated those plants with persistent leaves growing in dense tufts,

and with leaf margins not thickened, into the variety caespitosum. Duthie, who studied

this species around Stellenbosch, noted that the leaves persisted long after flowering

and that they sometimes formed tufts but she did not recognize two distinct varieties.

It is the only species from the south western Cape to flower in summer.

12. T. tabularis (Bak.) Oberm., comb. nov.

A. tabulare Bak. in FI. Cap. 6: 391 (1897). Adamson & Salter, Flora of the Cape

Peninsula, 182 (1950). Type: Cape, Table Mountain, Bolus 4726 (K, holo., BOL,

iso.!). A. palustre Adamson in J. S.A. Bot. 7: 187, fig. 1 (1941); Adamson & Salter,

Flora of the Cape Peninsula, 183 (1950). Type: Cape Peninsula, Patrys Vlei, Salter

8472 (BOL, holo.). A. glabrum Adamson in J. S.A. Bot. 7: 100 (1941); Adamson

& Salter, Flora of the Cape Peninsula 182 (1950). Type: Cape Peninsula, Steenberg,

Adamson 2768 (BOL, holo.).

Solitary, glabrous plants variable in size, up to 100 cm tall but usually much smaller.

Roots wiry, thin. Rhizome small, woody. Leaves many, forming a basal, erect rosette;

outer leaves smaller, often consisting only of a tubular sheath without a lamina; lamina

of produced leaves linear to semi-terete, up to 50 cm long, 2-10 mm broad, long tapered

in upper half, ribbed, margin minutely denticulate. Inflorescence divaricately few-

branched or simple; lowest branches showing suppressed accessory branches; scape

usually about as long as the leaves, firm, terete, arcuate; side-branches curved upwards,

densely flowered near the top; bracts small, 5 mm, acuminate; a few sterile bracts

from suppressed side-branches often present on scape; pedicels erect, up to 1 cm in

fruit, those of dropped sterile flowers recurved. Flowers with perianth immaculate,

white often tinged with pink; segments 14 mm long; ovary with 6 ovules per cell.

Capsule globose, succulent, glabrous, sulcate 9 mm long. Seed rough, with 3 large

crenulate ridges or wings.

Flowering Period : August-September.

Distribution: Known only from Table Mountain and surrounding mountains

at fairly high altitudes, often on wet cliffs.

Cape. — Peninsula: Patrys Vlei, Salter 8534 (BOL); Cirkels Vlei, Leighton 667 (BOL);

Kommetjie near Simonstown, Galpin 4749 (PRE). Table Mountain, Marloth 776

(PRE), 14056 (PRE), Haagner (Conrath 778, GZU). Caledon: Berg, Kleinmond,

de Vos 231 (BOL).

In the herbarium, Adamson’s two species, A. glabrum and A. palustre could not

be separated from T. tabularis. They may be forms however that show differences in

their natural habitat.

13. T. hirsuta (Thunb.) Kunth, Enum. 4: 577 (1843).

Anthericum hirsutum Thunb., Prod. 63 (1794) et in FI. Cap. ed. Schult. 322 (1823).

Bak. in J. Bot. Lond. 1872: 139 et in FI. Cap. 6: 394 (1897); Duthie in Ann. Stell.

Univ. 4, A: 8, t2, fig. 1, 10 (1926); Adamson & Salter, Flora of the Cape Peninsula,

182 (1950). Type: Cape, Thunberg (UPS, holo., PRE, photo.).

T. corymbosa Kunth, Enum. 4: 577 (1843). Type: Cape, Table Mountain, Drege

8716 (P. iso., PRE, photo.). In E. Meyer’s Zwei Pflanzengeogr. Documente, Drege

8716 was said to have been collected in the Tulbagh district, Klein Drakenstein between

Berg River and Drakenstein Berge but the Paris specimen gives the locality as Table

Mountain.

Plants up to 60 cm high. Roots thin, wiry, many, slightly swollen in spring.

Rhizome small, compact. Leaves 4-6 per shoot, subdistichous, arranged in an erect

fanlike manner, linear to broadly linear, 7-40 cm long, 5-20 mm broad, flat, striate,

pubescent, firm, dark green, 2 inner smaller falcate; the basal tube often produced

into a prominent triangular point opposite the lamina; outer leaves often short, or

without a lamina. Inflorescence taller than the leaves, with a few ascending branches,

731

the lowest axils showing buds of suppressed, accessory branches, rarely simple; scape

arcuate, firm, woody, pubescent, 30-40 cm long; racemes laxly flowered, elongating

during anthesis; bracts small, subulate, ciliate; pedicels erect, up to 1 cm long and

curving inwards in fruit. Flowers with perianth rotate, immaculate; segments 10-12

mm long; stamens spreading; ovary with 6 ovules per cell. Capsule globose, 6 mm,

glabrous. Seed verrucose, grey, 3 mm.

Flowering Period: Septeraber-October.

Distribution: A mountain species found on the Peninsula and surrounding

districts.

Cape. — Cape TowiuDevils Peak, Bolus 3794 (BOL); Peninsula, Kuhl & van Hasselt

[L, 909, 84 (261)]; Kenilworth Race Course, Salter 7714 (BOL). Caledon: Koude

Rivier, 5c/?/cc/;tcr 9730 (B, BOL, L.). Stellenbosch: Dwt/u'c 653 (BOL). Paarl: Hercules

Pillar, flats at the North base, Leighton 558 (BOL). Malmesbury: near Darling,

Esterhuysen 3873 (BOL). Piketberg: near Goedverwacht, Bolus (BOL, 25730).

Baker refers Burchell 4118 from the Bathurst district, Theophilis, to this species

and mentions that it is a form with prostrate “ viviparous ” flowering stems. Although

I did not see this specimen, I suspect that it is T. affinis Kunth, or possibly T. ciliata.

T. hirsuta does not occur so far east.

The species was found in an unusual habitat in the Caledon district, near Bot

River Lagoon, next to running water in a marshy area, growing in clumps; plants

luxuriant, flowering profusely, September 1960, Mauve 4070 (PRE, NBG).

Section 2. TRACHYANDRA.

14. T. chlamydophylla (Bak.) Oberin., comb. nov.

Anthericum chlamydophyllum Bak. in FI. Cap. 6: 389 (1897); Duthie in Ann. Stell.

Univ. 4, A; 14, t.II, fig- 2, 11 (1926); Adamson & Salter, Flora of the Cape Peninsula

181 (1950). Type: Cape, Tulbagh Kloof, MacOwan 2603 (K, holo.).

Plants up to 70 cm high. Roots many, all alike, hard, thick, often covered with

a tomentum of thick, long, felted root hairs. Rhizome compact, hard. Squamae

brown, narrow tubular, several surrounding one leaf or scape (but none surrounding

a shoot). Leaves numerous, terete or subterete, 30-70 cm long, 3-5 mm in diam.

usually erect and straight, glabrous or slightly muricate, ribbed. Inflorescence a simple

raceme with ca. 30 flowers, compact at first and shorter than the leaves, ultimately

longer and then usually prostrate and falsely secund; scape arcuate at the base with

a few sterile bracts (vestiges of suppressed side-branches); fertile bracts subulate, up

to 7 mm long; pedicels patent, lengthening during anthesis, curved downwards near

the base then spreading, up to 28 mm in fruit; scape and pedicels thickening markedly

in fruiting stage. Flowers with perianth rotate, immaculate; segments 10-13 mm long;

filaments subequal, erect or spreading; ovary with 4 ovules per cell. Capsule globose,

ca. 8 mm in diam., persistent perianth-base wide, slightly scalloped. tetrahedral,

3-5 mm in diam.

Flowering Period. — August-November. Sweet scented.

Distribution. — Cape Peninsula and surrounding districts. Although the type

locality was stated to be Tulbagh Kloof, no subsequent collections were made from

this area. Duthie remarks that it grows most luxuriantly on well-drained, sandy soil;

when it occurs in other situations it is often much stunted.

Cape. — Wynberg: Wynberg Hill, Salter 8993 (PRE) 8786 (BOL). Kirstenbosch,

Barker 2607 (PRE). Stellenbosch: Faure, Esterhuysen 11927 (PRE, BOL). Paarl:

Klapmuts, Hdfstrom & Acocks 188 (PRE). Malmesbury: Mamre, Leighton 1822 (BOL).

Worcester: Botha’s Halt, Gillett 276 (BOL).

732

15. T. hispida (L.) Kunth, Enum. 4: 575 (1843).

Anthericum hispidum L., Spec. PI. ed. 2 (1762); Jacq. Coll. Suppl. 91; Ic. 2: 17 t.409

(1786-1793); Thunb., Prod. 63 (1794) et in FI. Cap. ed. Schult. 321 (1823); Willd.,

Spec. 2: 145 (1799); Bak. in FI. Cap. 6: 393 (1897); Duthie in Ann. Stell. Univ. 4,

A: 7, t.II, fig. 3, 7, 8 (1926); Adamson & Salter, Flora of the Cape Peninsula, 181

(1950). Type: Cape, Timnherg (Linn. H. Cat. 432-23, syn.; 432-24, syn., LINN. herb.

PRE, photo.). 1 propose making this also the type species of the genus. A. squamewn

L.f., Suppl. 202 (1781). Roem. et Schult., Syst. Veg. 7: 481 (1829). Type: Cape,

Thunberg (Linn. Herb. Cat. 432-22; LINN, holo., PRE, photo.). A. undulatum

Thunb., Prod. 63 (1794) et in FI. Cap. ed. Schult. 321 (1823); Willd., Spec. Plant

2: 140 (1799). Roem. & Schult., Syst. Veg. 7: 470 (1829). Type: Cape, Thunberg

(LINN. Cat. no. 432-24, holo., UPS, iso., PRE, photo.). A. paradoxum Roem. &

Schult., Syst. Veg. 7: 459 (1829). Type: Cape, without locality or collector (M, holo.!,

PRE, photo.). The specimen bears a label with “ 1623 ” on it. A. subpUosum Poelln.

in Bol. Soc. Brot. 16, 2: 64 (1942). Type: Cape, Piketberg, Piqueniers Kloof,

Schlechter 10748 (B, holo.! GRA, iso, PRE, photo.). A. congeslum Adams, in J. S.A.

Bot. 10: 133 (1944); Adamson & Salter, Flora of the Cape Peninsula, 181 (1950).

Type: Cape, Claremont, Salter 8762 (BOL, holo.!). The number 8767 in the text

is a misprint.

Phalangium squamewn (L.f.) Poir. in Lam. Encyc. 5: 246 (1804).

Arthropodium hispidum (L.) Spreng. Syst. 2: 87 (1825).

Bulbinellal squamea (L.f.) Kunth, Enum. 4: 573 (1843).

Trachyandra undulata (Thunb.) Kunth, l.c. 583. T. paradoxa (R. & S.) Kunth, l.c. 576.

Small plants usually solitary or occasionally in clumps up to 30 cm high. Roots

fleshy, swollen above, very irregular with numerous, filiform secondary roots. Rhizome

small. Squamae membranous, large, surrounding a whole shoot as well as the base

of the leaves and the scape individually, gaping, membranous, white. Leaves 2-5

per tuft, linear to linear-lanceolate, 6-45 cm long, 2-10 mm broad, straight or with

a lax twist or occasionally, plicately folded, thin, flat, glabrous or hispid, often with

a purple margin, glaucous, erect or prostrate. Inflorescence a simple, congested raceme;

peduncle at first short, lengthening during flowering, 2-28 cm long, hispid or glabrous;

bracts large, ovate-lanceolate, acuminate, up to 15 mm long, membranous; pedicels

up to 3 cm long, hispid, curving down in fruit. Flowers with perianth hispid on outside,

pinkish white, slightly spreading, immaculate; segments ca. 1 cm long; stamens

dimorphous, filaments pink, inner more muricate than outer, ciliate at the base; ovary

with 8 ovules per cell. Capsule ovate, glabrous, 4 mm. Seed 2 mm in diam.

Flowering Period: June-September; sweet smelling.

Distribution: Cape and neighbouring districts; occasional on seasonally damp

flats and lowest slopes on the Peninsula according to Adamson.

Cape. — Cape Town: Marloth 165 (PRE), Haagner in herb. Conrath 1241 (GZU);

Claremont Royal Observatory, Adamson 2950 (PRE); Rondebosch, Bolus 3731 (BOL),

Salter 8762 (GRA, BOL); Wynberg, Schlechter 1060 (GRA); Green Point, Tyson

(GRA). Caledon: Rivier Sonder Einde Mountains, 542 (BOL). Malmesbury:

Mamre Hills, Ro/'/rer 4606 (BOL). Stellenbosch: Burmester (GZ\J), Duthie 533 (BOL).

Worcester: Stettyn, Leipoldt 3384 (BOL). Clanwilliam: Doom Rivier, Schlechter

8055 (B, BOL, PRE, L.). Hopefield: Bolus 12871 (BOL, NH).

Adamson records that the tubers may form buds giving rise to new shoots. He

separated his species A. congestum because of its short peduncle whilst he also found

it growing in clumps. Duthie found these various forms around Stellenbosch but did

not separate them.

733

16. T. peculiaris (Dinter) Oberm., comb. nov.

Anthericum peculiare Dinter in Fedde, Rep. 29: 263 (1931); Poelln. in Fedde, Rep.

52 : 254 (1943). Type: South West Africa, Liideritz: Halenberg, 40 m. E. of Liideritz

Bay on a steep, hot slope of a mica schist mountain, Dinter 665 1 ( B, holo. ! PRE, photo.).

Small plants up to 10 cm long. Roots tuberous, fused together and with the

rhizome. Squamae short membranous white, gaping. Leaves few, linear, 10 cm long,

4 mm wide, glabrous. Inflorescence a congested raceme, many flowered; scape 6 cm

long, hirsute; bracts ovate, white, membranous, glabrous, margins fimbriate; pedicels

recurved in fruit, hirsute, up to 12 mm long. Flowers with a white perianth, segments

6 mm long, sparsely hirsute outside. Capsule glabrous, 5 mm long. Seeds black,

verrucose, 14 mm in diam., tetrahedral.

Flowering Period: In fruit, August.

Distribution: Only known from type specimen, collected at Halenberg, South

West Africa.

South West Africa. — Liideritz: Halenberg, 40 m. E. of Liideritz Bay on a steep,

hot slope of a mica schist mountain, Dinter 6651 (B, holo.! PRE, photo.).

Very near T. hispida (L.f.) Kunth but with small glabrous bracts and apparently

geographically isolated from the Cape species. Dinter found only four plants which

were in fruit. So far not collected again.

17. T. hirsutiflora [Adamson) Oberm., comb. nov.

Anthericum hirsutiflorum Adamson in Journ. S.A. Bot. 7: 98 (1941); Adamson &

Salter, Elora of the Cape Peninsula, 181 (1950). Type: Cape, Wynberg, Bolus 3448

(BOL, holo.). A. pilossisimwn Poelln. in Bol. Soc. Brot. 16, 2: 61 (1942). Type:

Cape, Muizenberg, H^ilms 3759 (B, holo! PRE, photo.). A. canaliculatum sensu Bak.

(non Aiton) in J. Linn. Soc. Bot. 15: 309 (1876) et in El. Cap. 6: 391 (1897).

Hairy plants up to 60 cm high. Roots many, swollen, spreading, yellow. Rhizome

small. Squamae surrounding shoots as well as leaf- and scape-bases, tubular, mem-

branous. Leaves 2-4 per shoot, subterete, canaliculate, up to 50 cm long, 2 mm in

diam. hard, striate, scabrid. Inflorescence a tall simple tomentose raceme, sometimes

elongating to form a second set of flowers; scape terete, purple, covered with white,

straggling hairs; bracts narrow ovate, subulate, hairy; pedicels densely hairy, erect,

up to 3 cm in fruit. Flowers with cupshaped perianth, densely hairy, pink or pale

mauve on outside, immaculate, segments 12 mm long; stamens shortly scabrid; ovary

hairy, with 10 ovules per cell. Capsule dark purple, globose, hairy, 15 mm in diam.

Elowering Period: June-October.

Distribution: Cape Peninsula and surrounding districts. According to Adamson

it is found in sand, generally amongst rocks, at low altitudes, flowering most freely

after fires.

Cape. — Wynberg: Kenilworth, Salter llhA (BOL); Tokai, Guthrie 1131 (BOL). Simons-

town: Laubner (in Herb. Conrath 662, GZU), Wright 214 (L); Red Hills, Lam &

Meeuse 4125 (L). Paarl: Franschhoek Mountains, Phillips 8549 (BOL). Caledon:

near Stanford, Acocks 15500 (PRE). Clanwilliam: lower Olifants River, Marloth

8391 (PRE).

Baker confused this species with Anthericum canaliculatum Ait., which is a synonym

of A. ciliatum L.f., a very different species. The rough woolly pubescence and the

erect fruiting pedicel easily distinguish it from T. ciliata and T. hispida.

18. T. ciliata (L.f.). Kunth, Enum. 4: 585 (1843).

Anthericum ciliatum L.f., Suppl. 202 (1781). Thunb., Prod. 63 (1794) et FI. Cap.

ed. Schult. 324 (1823); Willd. Spec. Plant. 2: 146 (1799). Bak. in J. Bot. Lond.

1872: 139 et in FI. Cap. 6: 396 (1897). Adamson & Salter, Flora of the Cape

Peninsula, 182 (1950). Type: Cape, Thunberg (UPS, holo., PRE, photo.). A.

canaliculatum Ait., Hort. Kew. 1: 448 (1789) et Hort. Kew. ed. 2: 268 (1811).

734

Wind., Spec. Plant. 2: 141 (1799). Ker Gawler in Bot. Mag. t. 1124 (1808).

Lindley in Bot. Reg. t.877 (1825) under var. rufum. Roem. & Schult., Syst. Veg.

7: 460 (1829). Kunth, Enum. 4: 578 (1843). Type: Cape, Introduced to Kew by

Francis Masson in 1774 (BM, holo., PRE, photo.). A. longifolium Jacq. Coll. Suppl.

92; Ic. 2: 18, t.413 (1786-1793). Willd. Spec. Plant. 2: 139 (1799). Roem. & Schult.,

Syst. Veg. 7: 464 (1829). Type: Iconotype, Cape, Jacquin, Ic. t.413. A. vespertinum

Jacq., Hort. Schoenbr. 1; 44, t.85 (1804). Roem. & Schult., Syst. 7: 464 (1829).

Ker Gawl. Bot. Mag. t.l040 (1807). Type: Iconotype, Jacquin, Hort. Schoen. t.85.

A. blepharophoron Roem. & Schult., Syst. Veg. 7: 461 (1829). Type: Iconotype,

Jacquin, Hort. Schoenbr. t.413. A. falcatum sensu Bak. in FI. Cap. 6: 394 (1897)

as to Bwchell 4076, non L. A. recurvatum Dinter in Fedde, Rep. 29, 264 (1931).

Poelln. in Fedde, Rep. 52: 255 (1943). Type: South West Africa, Diamond Area I:

Klinghardt Mountains, Dinter 3960 (B, holo.! PRE, photo.). A. pilosiflorum PoeUn.

in Bol. Soc. Brot. 16, 2: 60 (1942). Type: Cape, Namaqualand, Brakdam, Pearson

5948 (B, holo., probably destroyed) including var. subpapillosum l.c. Type var.: Cape

Peninsula, Bergius (B, holo! PRE, photo.). A. spongiosum Poelln. in Bol. Soc. Brot.

16, 2: 65 (1924). Type: Cape, van Rhynsdorp, Zout Rivier, Bergius (B, holo! PRE,

photo.). A. hamaturn Poelln. in Bol. Soc. Brot. 16, 2: 67 (1942). Type: Cape,

Hopefield, Bachmann 801 (B, syn. probably destroyed); Riversdale, Rust 565 (B, syn!

PRE, photo.). A. Poelln. in Bol. Soc. Brot. 16, 2: 73 (1942). Type: Cape,

Clanwilliam, Packhuisberg, Schlechter 10811 (B, holo! GRA, PRE, L, iso.!).

Phalangium canaliculatum (Ait.) Poir. in Lam. Enc. 5: 249 (1804). P. longifolium

(Jacq.) Poir. l.c. 243. P. vespertinum (Jacq.) Poir. l.c. 249.

Bulbine ciliata (L.f.) Link, Enum. 1: 329 (1821). Roem. & Schult. in Syst. Veg. 7:

450 (1829). B. canaliculata (Ait.) Spreng., Syst. 2: 86 (1825).

Trachyandra blepharophora (Roem. & Schult.) Kunth, Enum. 4: 582 (1843). T.

Kunth, Enum. 4: 582 (1843). Type: Cape, Said anha Bay, between Dassen-

berg & Groenekloof, Drege 1493 (L, iso! PRE, photo.). *

Plants variable in size, up to 50 cm high. Roots swollen and spongy but not fused»

becoming thinner and harder later in the season. Rhizome discoid, small. Squamae

membranous, thin, white, surrounding shoot as well as leaf- and scape-bases. Leaves

variable, linear, up to 100 cm long and 4 cm broad, dull glaucous, with a juicy, spongy

mesoderm, soft, flat or keeled with some scattered hairs on margin and keel or glabrous.

Inflorescence branched with 1-2 basal branches or simple (through reduction), erect

at first, sometimes elongating during anthesis and becoming prostrate ; scape pubescent

at first, glabrous with age, with some sterile bracts or vestiges of suppressed side-branches ;

bracts large, boatshaped, 1 cm long, dark, subulate, auriculate, margin minutely ciliate

the young, compact racemes with the imbricate, subulate bracts resembling an ear

of corn ; pedicels 8 mm long in flower, erect, up to 2 cm long in fruit and then recurved

and sub-secund. Flowers with a recurved, translucent, white perianth; segments

ca. 1 cm long, often hairy outside, with yellow spots near the base (these sometimes

faint); the outer stamens erect, scabrid, the inner connivent around the ovary, retrorsely

scabrid in lower half; ovary with 10 ovules per cell. Capsule ovoid to globose or cylin-

drical, glabrous, 6-14 mm long, slightly fleshy. Seeds black, minutely verrucose.

Flowering Period: August-September. Said to be scentless.

Distribution: From the southern part of South West Africa to the south eastern

Cape, usually near the coast in sand, occasionally further inland. Sometimes spreading

as a weed around Cape Town.

Cape. — Peninsula: Karbonkelberg, Leighton 704 (BOL); near Cape Town, Alberth

(GZU). Hopefield: near Hopefield, Bolus 12872 (BOL). Clanwilliam: Schlechter

8422 (GRA). Piketberg: Piquenierskloof, Schlechter 10773 (GRA, L). Caledon:

Gansbaai, Gillett 4406 (BOL). Knysna: Fourcade 1563 (GRA). Port Elizabeth:

* Add to synonymy: T. canaliculata, T. longifolia, T. vespertina, l.c.

735

Walmer, Cruden 329 (GRA); New Brighton, Paterson 2146 (GRA). Bathurst: near

Theopolis, Burchell 4076 (L.); this specimen is cited under Anthericum falcatum L.L

by Baker in Flor. Cap. 6; 394 (1897).

South West Africa. — Diamond Area I; Klinghardt Mts., Dinter 3960 (B, holo.L

PRE, photo.).

Adamson in the Journ. of S.A. Bot. 7: 97 (1941) discusses the plates in the Bot.

Magazine (t.ll24) and Bot. Register (t.877). He found that Baker’s Anthericum

canaliculatum of the Flora Capensis really represented a new species, which he described

as A. hirsutiflorum. But he could not place these two plates as he had not seen the

type of A. canaliculatum Ait., which I consider a synonym of T. ciliata. As Curtis

mentions in the text that the plant was introduced by Masson and as Aiton himself

refers to this plate, Bot. Mag. t.ll24, in the second edition [Hort. Kew. 2: 268 (1811),

under A. canaliculatum] it is likely that the plates were made from the old type plant

or its offspring. After years in unnatural surroundings it had probably degenerated

somewhat. Lindley called his plant var. rufum as the pedicel and the “ interior ” of

the flower was hairy. He must have meant the exterior of the flower. He described

the filaments as smooth but this might have been a careless observation. Both Curtis

and Lindley suggested that the species might be identical with Trachyandra hirsuta

but this species is different; it has wiry roots, broader leaves, etc.

19. T. falcata {L.f.) Kuntli, Enum. 4: 586 (1843).

Anthericum falcatum L.f., Suppl. 202 (1781). Willd., Spec. PI. 2: 138 (1799). Thunb.,

Prod. 63 (1794); et in FI. Cap. 3d. Schult. 323 (1823). Type: Cape, Thunberg (UPS,

holo., PRE, photo.). Anthericum drepanophyllum (Bak.) Schlechter ex Index Kew. 7:

(1929); cf. Schlechter in Schultze, Aus Namaland und Kalahari, p. 202 with figure

in text (1907). Schlechter refers this species to Anthericum but does not mention the

basonym. Baker described it as Chlorophytum drepanophyllum. Anthericum weissianum

Dinter in Fedde, Rep. 29: 262 (1931), Poelln. in Fedde, Rep. 52: 257 (1943). Type:

South West Africa, north bank of Orange River, near its mouth, Weiss in herb. Dinter

6608 (B, holo.! PRE, photo.).

Bulbine falcata (L.f.) Roem. & Schult., Syst. Veg. 7: 451 (1829).

Chlorophytum drepanophyllum Bak. in FI. Cap. 6: 398 (1897). Type: Cape, Nama-

qualand: near Nababeep, Bolus 6584 (K, holo.).

Fairly tall, large plants up to 60 cm high. Roots very many, long, spreading, swollen

in spring, covered with a thick tomentum of roothairs. Rhirome small. Squamae

membranous surrounding the shoot, leaves and scape-base. Leaves usually 4, up to

30 cm long and 3-5 cm wide, falcate, flat, leathery, glabrous or shortly pubescent,

margin minutely ciliate. Inflorescence an erect, branched, raceme with the branches

ascending and with accessory branches, sometimes simple through reduction, over-

topping the leaves, densely flowered; scape about as long as the raceme, terete, glabrous

at the base, or hairy at first becoming glabrous, stout; lower bracts amplexicaul, short,

forming a wide erect collar around the stem, apiculate, floral bracts widely ovate,

subulate, membranous, white with a brown patch near the tips, closely imbricate in

bud; pedicels short, 1 cm long, erect. Flower with perianth pale mauve, maculate;

segments 12 mm long; stamens scabrid; ovary with 8-10 ovules per cell. Capsule

turbinate, 12 mm long, apiculate, dry. Seeds grey, with raised hyaline margins.

Flowering Period: July-October.

Distribution: Western Cape, coastal region from Saldanha to the southern

part of South West Africa, in sandy soil. Collectors all found it to be common where

they collected it. Schulze noted that it made a nice spinach. Marloth called it the

wild cauliflower.

736

Cape. — Hopefield: Saldanha Bay, Bohnen 1053 (PRE); Vredenburg, Bohnen 1054

(PRE). Clanwilliam : Marloth 5859 (PRE). Van Rhynsdorp: Klaver, Andreae 423

(PRE). Calvinia: Driefontein, Voor Hantam, Marloth 12818 (PRE). Laingsburg:

Matjesfontein Foley 163 (PRE). Robertson: Vink River along road from Robertson

to Worcester, van Breda 983 (PRE). Namaqualand : Richtersveld, Kawarass, Marloth

12423a (PRE); between Springbok and Hondeklip, Dyer & Verdoorn 1804 (PRE);

Kamies Mountain, Garies, Leistner 742 (PRE); Lilyfontein, Rodin 1468 (PRE).

South West Africa. — Diamond Area I, north bank of Orange River near its mouth,

Weiss in herb. Dinter 6608 (B, holo.! PRE, photo.).

20. T. longepedunculata (Stead, ex Roem. & Schult.) Kunth. Enum. 4: 584

(1843).

Anthericum longepedunculatum Steud. in Roem. & Schult., Syst. Veg. 7: 457 (1829).

Bak. in J. Bot. Lond. 1872, 138 et in FI. Cap. 6: 390 (1897). Duthie in Ann. Stell. Univ.

4, a: 12, t.l (1926). Adamson & Salter, Flora of the Cape Peninsula, 180 (1950).

Type: Cape, Ludwig (in herb. Steudel originally but it could not be traced with

certainty).*

Anthericum revolutum a in Herb. Thunberg (UPS); Juel, PI. Thunb., 121 (1918) nom.

Cape, Thunberg (UPS). This is the sheet to which Baker refers in J. Bot. 1872, 138.

Soft plants 18-60 cm high, solitary or several together. Roots swollen, ca. 7,

contracted, not fused, whitish with numerous long thin rootlets. Rhizome small,

compact. Squamae membranous, whitish, 15-30 cm high, surrounding the shoot as

well as the leaf- and scape-bases. Leaves 2-8, subterete to linear, 11-40 cm long,

1-7 mm broad, tapering gradually to the apex, soft, glabrous or very minutely scabrid,

dark green, often mottled near the base. Inflorescence 1-5 per plant, occasionally

branched but usually simple through reduction, many flowered, often shorter than the

leaves, elongating and falsely secund in fruiting stage; scape usually short, glabrous,

or minutely scabrid, often reddish-brown and mottled near the base; sterile bracts

(vestiges of suppressed branches) present; bracts boatshaped, cuspidate, sub-amplexi-

caul; pedicels short in bud, elongating to 3 cm in fruit and then patent, slightly recurved.

Flower with perianth rotate, white to pinkish with reddish-brown keels, immaculate,

segments 1 cm long, inner ciliate at the base; stamens with white filaments, inner with

longer bristles, anthers orange; ovary with 9 ovules per cell. Capsule cylindrical,

ca. 8 mm long, 3 mm in diam., thin walled, apiculate. Seeds black.

Flowering Period: August- October. “Sweet smelling, bee-pollinated” Duthie.

Distribution: Cape Peninsula and surrounding areas, usually in damp marshy

places at low altitudes.

Cape. — Cape Town: Green Point, MacOwan 1986 (GRA); Claremont Sanatorium,

Wolley Dod 2968 (BOL); Sassar, Bohnen 1063 (PRE). Malmesbury: Darling Flora

Reserve, Rycroft 1995 (BOL). Stellenbosch: flats at Faure, Esterhuysen 11928 (BOL,

PRE). Caledon: Stanford, Compton 19936 (BOL). Hopefield: near Hopefield,

Bolus 12807 (BOL, PRE).

Adamson noted that the inflorescence did not lengthen in fruit but Duthie observed

that it did. The herbarium specimens seem to indicate that they do lengthen during

anthesis. The large forms seem to come very close to T. ciliata.

* Mr. W. Marais who searched for the type at the Kew herbarium wrote the following: “There

is no Ludwig specimen at Kew but there is one alleged to be from Lehmann. On this sheet is written

‘ Lehmann ’ in the hand of W. J. Hooker. Then a label ‘Anthericum longepedunculatum n. sp. C.B.S.'

in an unidentified hand. The ‘ n. sp.’ leads one to think that it is by Steudel or by a person who wrote

labels for him. Below this label ' Trachyandra longepedunculata Kunth ’ in the hand of Planchon.

Then a note by N. E. Brown: ‘ Matches the plant labelled in Herb. Thunberg Anthericum revolutum

exactly ’. The only suggestion is that this specimen is part of the Ludwig specimen which came to

Hooker via Lehmann and that Baker knew about this. That would explain the citation of Ludwig

and the omission of Lehmann by Baker in the J. Linn. Soc. 15: 138 and in the Flora Capensis 5: 390 ”.

737

21. T. divaricata {Jacq.) Kunth, Enum. 4; 580 (1843).

Anthericum divaricatum Jacq. Hort. Schoenbr. 4: 7, t.414 (1804); Roem. & Schult.,

Syst. Veg. 7: 463 (1829). Adamson & Salter, Flora of the Cape Peninsula, 184 (1950).

Type: Iconotype, Jacquin, Hort. Schoenbr. 4: 7, t.414. A. revolutum (4)

sensu Thunb., Prod. 62 (1794) et in FI. Cap. ed. Schult. 318 (1823). Juel, PI. Thunb.

121 (1918) nom. non L. A. revolutum sensu Ker Gawl. in Bot. Mag. t.l044 (1804)

exclud. syn. non L. A. revolutum sensu Bak. in J. Bot. 1872, 138; in Trans. Linn.

Soc. 15: 311 (1876) et in FI. Cap. 6: 393 (1897), non L.

Phalangium aethiopicum ramosum, floribus albis, petalis reflexis. Comm, in Hort.

Amst. 1: 67, t.34 (1697).

Plants robust up to 90 cm high. Roots many, not much thickened, occasionally

growing to a great depth. Rhizome woody, thick, irregular in shape. Squamae narrow,

tubular, surrounding each leaf- and scape-base separately. Leaves linear, up to 100

cm long, 4-12 mm wide, tapering gradually to the apex, flat, glabrous, somewhat fleshy,

flexible, erect or usually prostrate, straight or with a lax spiral twist, bright green

occasionally orange at the base. Inflorescence stout, usually with accessory branches,

divaricately branched; scape 10-50 cm high, stout glabrous; bracts small, 4 mm long,

membranous, widely ovate at the base; pedicels 4-12 mm long. Flowers erect, perianth

segments 7-12 mm long, white, green-keeled, with a yellow dot near the base, spreading,

recurved from the middle ; stamens yellow in lower half, dimorphous, 3 outer spreading,

3 inner connivent around ovary, spreading and retrorsely scabrid above; ovary with

12 atropous ovules per cell. Capsule globose, 12 mm in diam., slightly inflated, dry

or somewhat fleshy, yellowish. Seeds 2 mm in diam. smooth, tetrahedral.

Flowering Period: “August-September but also at other times; flowers opening

in the morning ”. Adamson.

Distribution: South western to south eastern Cape, common on sand and on

dunes near the sea.

Cape. — Namaqualand: Hondeklip Bay, Pillans 18254; 2 miles S. of Goodhouse,.

Lewis 64663 (NBG), Barker 6259 (NBG). Hopefield: Saldanha Bay, Clarkson 375.

Cape: Blouberg Strand, Bohnen 1052 (PRE), Compton 8923, 8924 (NBG); Robben

Island, Walgate 482 (PRE). Peninsula: Schusters Bay, Lam & Meeuse 4117 (L.);

Kommetjie, Eyles 7741 (SRGH), Fishhoek, Marloth 2844 (PRE). Heidelberg: Witsand,

Gillett 811 (BOL). Knysna: Plettenberg Bay, Fourcade 1051, 1489 (BOL), Kapp 114

(PRE). Bathurst: Port Alfred, Rogers 28052; Hutton 741.

Commelin’s plate, t.34 in Hort. Amst. 1 (1697) is the oldest drawing of a

Trachyandra. Baker referred it to Anthericum revolutum L. but its stout succulent

appearance on the plate indicates that it is better placed under T. divaricata.

23. T. revoluta (Z,.) Kunth, Enum. 4: 579 (1843).

Anthericum revolutum L., Sp. PI. 310 (1753). Roem. & Schult., Syst. Veg. 7: 462

(1829). Adamson & Salter, Flora of the Cape Peninsula, 184 (1950). Type: no type

preserved. Cape: Hermanns, Galpin 12898 (PRE, neo.). A. flliforme, (y), Thunberg,

Prod. 62 (1794) et in FI. Cap. ed. Schult. 318 (1824). non Ait. Type: Cape, on hills

below Table Mountain, Thunberg, sheet (y) (UPS, holo., PRE, photo!). A. elongatum

Wind., Sp. PI. 2: 136 (1800); Roem & Schult. Syst. Veg. 7: 456 (1829); Bak. in

FI. Cap. 6: 389 (1897). Willdenow gave Thunberg’s plant a new name as Alton had

already used the epithet flliforme for an Anthericum which is now Bulhinella flliforme

(Ait.) Kunth. Sprengel in Syst. 2: 83 (1825) in error united them. A. jacquinianum

sensu Bak. in J. Bot. Lond. 1872, 308, excl. var. et in Trans. Linn. Soc. 15: 308 (1876);

A. scabrum sensu Duthie in Ann. Univ. Stell. 4, A: 15, t4 1926, non L.f.

IPhalangium revolutum Poir. in Encycl. Meth. Bot. 5: 247 (1804).

IDilanthes revolutum Salisb. Fragm. 70 (1866).

Trachyandra elongata (Willd.) Kunth, Enum. 4: 584 (1843).

738

Adamson, Notes on Some Cape Peninsula Species of Anthericum in Journ. S.A.

Bot. 7; 93 (1941).

Plants up to 50 cm high, usually smaller. Roots many, crowded, yellow, somewhat

swollen, fusiform, firm, the circle of new roots produced above the old one. Rhizome

small, discoid. Squamae narrow, tubular, reddish-brown, surrounding each leaf-

base and also the scapes, separately (no squamae surrounding the shoots). Leaves

erect or prostrate when older, sometimes slightly undulate or with a spiral twist, 10-40

cm long, 1-4 mm in diam. linear, firm and somewhat fleshy in texture, dark green,

scabrid, at least when young, especially on the margins. Inflorescence a much branched,

lax, pseudo-dichotomous, divaricate raceme, usually overtopping the leaves; scape

slender, up to 30 cm long, 3-6 mm in diam., scabrid especially at the base when young;

bracts small, 2-5 mm long; pedicels up to 1 cm, erect in bud and in fruit, recurved

during anthesis. Flowers pendulous, perianth recurved forming a “ ball ” around

pedicel, segments 8-10 mm long, narrow, white, brown-keeled and with 2 yellow

maculae near the base; filaments yellow below, dimorphous, outer sparsely, inner

densely shaggy in lower half, and curved around the ovary; ovary with ca. 10 ovules

per cell. Capsule sub-globose, greyish brown, ca. 2 mm in diam., tetrahedral.

Flowering Period: August to November. Odour heavy, musklike (Duthie).

Adamson noted that it flowered freely after fires.

Distribution : Eastern to western Cape, near the coast or more inland occasionally.

“ Common on seasonally damp, heathy flats ”. Adamson. “ Seasonal swamps ”

Duthie.

Cape. — Cape Town: Kuhl & van Hasselt (L 909, 84/270), Bolus (BOL. 25734); Path

above Kalk Bay, Goulinis (BOL. 25733); W3mberg: Schlechter 1546 (GRA); Ronde-

bosch. Bolus 3750 (BOL); Noordhoek Flats, Esterhuysen (BOL, 25736); Sassar,

1060 (PRE); Peninsula, Klaassenbosch, Wolley DodlASA (QOL). Stellenbosch:

Duthie (BOL 25744). Hopefield: Saldanha Bay, Hoetjies Bay, Bolus (BOL. 12867).

Ceres: Bokkeveld, Marloth 8364 (PRE). Van Rhynsdorp: Bitterfontein, Schlechter

11038 (GRA, L.). Namaqualand: near Soebatsfontein, Verdoorn & Dyer 1797 (PRE);

foot of Kamies Mountains near Garies, Esterhuysen 1367 (BOL). Flumansdorp:

Kruisfontein, Galpin 4748 (PRE); Albertinia, Muir 1191 (BOL). Bathurst: Port

Alfred, Hutton 1154 (GRA), Rogers 28052 (PRE). Port Elizabeth: Paterson 166

(GRA). Alexandria: Archibald 3686 (PRE). East London: Galpin 3357 (PRE).

It is the oldest known species from the Cape. Adamson in his helpful article

(Journ. S.A. Bot. 7: 93, 1941), discusses the confusion that existed between Anthericum

revolutum L. and A. divaricatum Jacq., while later on A. flexifolium sensu Jacq. (not

of L.f.) also became involved. A. divaricatum Jacq. is a stout, glabrous bright green

plant found on the dunes and flats near the sea. A. revolutum L. has a wider distribution

and is usually found on firmer soils. It is more slender than A. divaricatum Jacq. and

the base of the dark green leaves and scape are scabrid. The description of A. revolutum

by Linnaeus although very short, does mention “ folia asperis ” which points to the

slender plant.* I therefore agree with Adamson that this smaller plant should be

regarded as A. revolutum L. [T. muricata (L.f.) Kunth also possesses asperous leaves

but it is not as common as T. revoluta (L.) Kunth].

In literature these two species were often confused. Miller, Thunberg and Baker

seem to have thought that the two names were conspecific. Ker Gawler’s plate in the

Botanical Magazine, t.l044 (1807) also figures the more succulent A. divaricatum Jacq.

In Thunberg’s herbarium, A.flliforme y is a synonym of A. revolutum L. As the epithet

filiforme was invalid, Willdenow gave it a new name, A. elongatum. Roemer & Schultes

then enumerated all 3 names in their Syst. Veg. 7: 456, 462 (1829). They also gave

* ‘'Anthericum revolutum L. was founded entirely upon a species described by Tournefort as

Asphodelus foliis compressus, asperis, caule patulo and its country is unknown.” N. E. Brown on a sheet

at J.

739

a new name, A. jacquinianum, to a plate (Ic. t.412) of an Anthericum which Jacquin had

mistakenly called A. flexifolium L.f. When Baker revised the Anthericum species for

the Flora Capensis he made this species, A. jacquinianum, a synonym of A. elongatum

Willd. As neither the roots nor the squamae of the species figured by Jacquin, corres-

pond with those of A. revolutum L. (A. elongatum Willd. is a synonym), I do not think

this was correct and have kept them separate.

23. T. laxa {N. E. Br.) Oberm., comb, nov., aggregate species.

Anthericum laxum N.E. Br. Kew Bull. 1909, 143.

Plants glabrous, up to 60 cm high. Roots many, somewhat fleshy or thin, often

with a lanate pubescence of root hairs. Rhizome small, irregular, woody. Squamae

narrow tubular, membranous, brown, several surrounding each leaf- and scape-base

separately but none surrounding a shoot. Leaves many, terete, up to 40 cm long,

3 mm in diam., erect or drooping, often glutinous from secretions of sub-epidermal

glands situated in longitudinal lines. Inflorescence a divaricate much branched raceme,

branches either alternate or the first three arising trichotomously from a clavate apex,

the following branches alternate, laxly flowered; bracts ovate, ca. 3 mm long, mem-

branous, white with a brown keel; pedicels up to 6 mm long, erect in bud, pendulous

during anthesis, erect in fruit. Flowers pendulous with perianth 1 cm in diam. recurved

from near the base, forming a “ ball ” around pedicel, white, dark keeled, with 2

yellow spots near the base, scentless, producing much nectar; 3 outer stamens scabrid,

3 inner forming a tube, curving outwards above ovary and there yellow and densely,

retrorsely scabrid; the basal area touching the ovary, smooth with lateral, and dorsal

fringes; ovary with ca. 10 ovules per cell. Capsule ca. 6 mm long, globose, dry. Seed

dark brown, angled.

Flowering Period: In late summer after rains, February to April. In South

West Africa in the southern part the flowering period could be any time after a rare,

rainy spell. The dry inflorescence starts rolling about in April.

Common Names: Rolbossie, tolbossie, sandui. Said to be eaten by animals, pigs

even digging out the roots.

Distribution: Kalahari sandveld; Bechuanaland, western Transvaal, western

Free State, northern Cape, South West Africa.

Key to Varieties

Branches of inflorescence alternate ; scape not clavate at the apex; plants rather slender {a)\ar.Taxa

Three lowest branches of inflorescence trichotomous; scape clavate at the apex; sturdy plants

{b) var. erratica

{a) var. laxa.

Anthericum laxum N. E. Br. in Kew Bull. 1909, 143. Type; Bechuanaland, Kalahari,

Bachakuru, West of Serowe, Lugard 234 (K, holo. ! PRE, photo.).

A. glutinosum Dinter in Fedde, Rep. 16: 338 (1920); and in Fedde, Rep. App.

23: 55 (1923) name only. Type: South West Africa, Diamond Area 1: Klinghardt

Mountains, Schafer 559 (B, holo., probably destroyed). Dinter apparently overlooked

his publication of the species in 1920 and republished it in 1931. Von Poellnitz now gave

the second plant i.e. Dinter 3940, a new name i.e. A. dinteri Poelln. in Fedde, Rep 50:

232 (1942), nom. nov. for A. glutinosum Dinter in Fedde, Rep. 29: 265 (1931) non

Dinter 1920. Type: South West Africa, Diamond Area 1: Buntfeldschuh, Namib,

Dinter 3940 (B, holo ! PRE, photo.). They are conspecific. A. nigrobracteatum Dinter

in Fedde, Rep. 29 : 265(1931). Type: South West Africa : Diamond Area 1 , Klinghardt

Mountains, Dinter 3952 (B, holo! PRE, photo.). This plant was referred to under

the name of A. glutinosum by Dinter in Fedde, Rep. App. 23: 55 (1923). It was

collected in the same area as A. dinteri Poelln. and the differences are too small to

merit the rank of a separate species. A. buchubergense Poelln. in Fedde, Rep. 52: 245

(1943). Type: South West Africa, Diamond Area 1, Buchu Mountains, Dinter 6571

p.p. (B, holo.! PRE, photo.). lA. pachyrrhizum Dinter in Fedde, Rep. 29: 157 (1923).^

740

Type; South West Africa; locality uncertain, probably Diamond Area 1, Klinghardt

Mountains in the S.W. Namib, Dinter 1780 (B, holo! PRE, photo.)- The specimen

is sterile but the roots and leaves are typical of T. laxa. ?A. brunneoviride Dinter ex

Poelln. in Fedde, Rep. 52: 261 (1943). Type: South West Africa, Bethany: near

Aus, Dinter 6202 (B, holo! PRE, photo.). The specimen is sterile but roots and leaves

are typical of T. laxa. A. elongatum sensu Poelln. in Fedde, Rep. 52: 249 (1943).

South West Africa, Bethany: Sandverhaar (near Aus), Pearson 4278, 4438 (BOL).

The scape of one of the specimens was scabrid showing affinity to T. revoluta.

Cape. — Prieska: Bryant B 338 (PRE). Kuruman; Batlharos, Silk 211 (PRE, SRGH),

Pole Evans 2W9 {PKE). Hay: Matsap, 2446 (KMG). Barkley West: Acocks

1557 (PRE, KMG). Kimberley: Koppies Kraal, Muir (PRE).

Transvaal. — Potgietersrus : Naboomspruit, Galpin 353 (PRE).

South West Africa. — Bethany: Sandverhaar, Pearson, 4278, 4438 (BOL).

(b) var. erratica {Oberm.) Oberm.. stat. et comb. nov.

Anthericum erraticum Oberm. Journ. S.A. Bot. 2: 187 (1936). Type: Transvaal,

Maquassi, van Niekerk (TM 35736, PRE, holo!). A. arvense Schinz var. rigidum Suess.

in Mitt. Bot. Staatssam. Muenchen 1: 49 (1950). Type var.: South West Africa;,

Grootfontein, Rehm (M, holo.!). Fig. 6, p. 000.

Cape. — Barkley West, Smith 2332 (PRE). Kimberley: Doornfontein, Adams (PRE

KMG, BOL). Hay: Padkloof, Acocks 2082 (PRE, KMG). Mafeking; Brueckner

570 (PRE). Vryburg: Burn- Davy (PRE, 14672).

Orange Free State. — Boshof: near Boshof, Schweickerdt 1115 (PRE). Kroonstad:

Goossens 1173 (PRE).

jBechuanaland. — Genesa, Rogers (BOL, 12591). Mochudi, Rogers 6317 (BOL).

Kanye, Hillary & Robertson 622 (PRE).

Transvaal. — Lichtenburg: Grasfontein, Sutton 380 (PRE). Potchefstroom, on road

to Schoemansdrift, de Wet 1850 (PRE). Warmbaths; Radium, Obermeyer TM 35737

(PRE); near Pienaars River, Codd 848 (PRE).

South West Africa. — Gobabis; farm Onreg, Merxmiiller 1117 (PRE, M); Basson

107 (PRE). Windhoek: Meyer (PRE). Otjiwarongo: Waterberg Plateau, Boss

(PRE).

T. revoluta and T. laxa are two closely related tumble weeds; the first is a Cape

species with rough, fiat or rolled leaves and a scape that is pubescent at the base. T.

laxa occurs in the Kalahari sandveld, is entirely glabrous and has terete, often glutinous

leaves. It seems wisest to consider the umbellate species {Anthericum erraticum) a variety

of T. laxa for these two forms occur side by side in some areas; but for the difference

in the arrangement of the lowest branches of the inflorescence, they agree in other

respects except that T. erratica is usually a somewhat stouter plant.

The herbarium material from South West Africa is very scanty. When more

material becomes available and with more field work we will be able to judge better

whether all Dinter’s species are conspecific with T. laxa. In tumble weeds one can of

course expect a wide and erratic distribution.

24. T. arvensis {Schinz) Oberm., comb. nov.

Anthericum arvense Schinz in Verb. Bot. Ver. Brand. 31: 216 (1889). Bak. in FI.

Trop. Afr. 7: 210 (1898). Poelln. in Fedde, Rep. 52: 245 (1943). Type; Southwest

Africa: Ovamboland, Olukonda, Schinz (Z, holo.! PRE, photo.). A. flavoviride Bak.

in FI. Trop. Afr. 7: 490 (1898). Type: Bechuanaland, Ngamiland, Botletle Valley,

Lugard, 194 (K, holo.). A. pallidiflavum Engl. & Gilg in Warburg, Kunene Sambesi

Exp. 187 (1903). Type; Angola (near S.W. border), Okavango Riverbank at Kavanga,

Baum 412 (B, holo., probably destroyed). A. gilvum Krause in Engl. Bot. Jahrb. 48:

353 (1912). Poelln. in Fedde, Rep. 52: 249 (1943). Type: South West Africa:

741

Fig. 6. — Trachyandra laxa var. erratica Oberm. a, habit, X b, basal squamae enveloping a

leaf, X 2. c, pendulous flower, d, capsule, x 3. e, seed, x 4.

742

Hereroland, Klein Namas near Hochanas Dinter 1941 (B, hole., probably destroyed);

var. brumeolum Poelln. in Fedde, Rep. 52: 250 (1943). Type of var.: South West

Africa, Hereroland, Otjispera near Epata Seiner III/357 (B, probably destroyed).

Slender, glaucous, glabrous plants up to 70 cm high. Roots many crowded,

fusiform, dark brown. Rhizome woody, compact, globose, irregular. Squamae tubular,

membranous, several enveloping each leaf- and scape-base, the outer gaping. Leaves

linear, flat, or subterete, up to 60 cm long, 2-5 mm broad, glaucous, longitudinally

grooved, long tapering to the apex. Inflorescence ca. 70 cm tall, a much branched

divaricate, many flowered raceme; peduncle slender, ca. 40 cm long, 5 mm wide;

side branches thin, lax, spreading; bracts minute, 2 mm long, white, membranous,

acute; pedicels slender 2-3-5 cm long, recurved in flower, remaining so if flower is

not fertilized, erect in fruit. Flowers with perianth pendulous, campanulate at the

base, recurved, yellow, maculate ; segments 1 cm long, 1 mm wide ; stamens dimorphous

the outer spreading, muricate, the inner connivent, curved outwards above the ovary

with the filaments flattened below and these with lateral and dorsal fringes, densely

covered with retrorse, obtuse papillae in upper half; ovary with 10-12 ovules per cell.

Capsule erect, narrow ovate, 7 mm long, glabrous. Seeds smooth, often with an orange

band (probably consisting of glands) near the hilum.

Flowering Period : August-March.

Distribution: Northern South West Africa, Bechuanaland, Southern Rhodesia,

Northern Rhodesia, Angola.

Common on the sandy banks and islands of the Okavango River and its tributaries,

also straying into surrounding country.

South West Africa. — Okavango Native Territory: between Runtu and Kapako,

Okavango River, de Winter 3730 (PRE); Andara, Merxiniiller 1962 (PRE, M). Groot-

fontein: Gautscha Pan, 157 miles E. of Grootfontein, Story 6150, 5321 (PRE).

Southern Rhodesia. — Salisbury: Hunjani River, in sand, PF/7(/4623 (SRGH); Arthur’s

Seat, Hunyani, Greatrex (SRGH, 13969).

Northern Rhodesia. — Barotseland: Shangombo, Kwando or Mushi River, Codd

7469 (PRE); Nangweshi on the Zambesi River, Codd 7201 (PRE, SRGH).

25. T. muricata (L.f.) Kunth, Enum. 4: 576 (1843).

Anthericum muricatum L.f., Suppl. 202 (1781); Thunb., Prod. 63 (1794) et in FI. Cap.

edit. Schult. 322 (1823); Willd. Sp. PI. ii: 145 (1799); Roem. et Schult., Syst. Veg.

7: 459 (1829); Bak. in Journ. Bot. Lond. 1872, 139 et in FI. Cap. 6: 395 (1897);

Adamson & Salter, Flora of the Cape Peninsula, 183 (1950). Type: Cape, Thunberg

(UPS, holo., PRE, photo.!). Anthericum flmbriatum Thunb., Prod. 63 (1794); Roem.

& Schult., Syst. Veg. 7: 465 (1829) et in FI. Cap. edit. Schult. 322 (1823). Type: Cape,

Thunberg (UPS, holo., PRE, photo.). A. bachmannii O. Ktze, Rev. Gen. 3, 2: 315

(1898). Type: Cape, Malmesbury: Hopefield, 802 (B, holo.! PRE, photo.).

A. scahromarginatum Schlechter apud Poelln. in Bol. Soc. Brot. 16: 2. 80 (1942).

Type: Cape, Calvinia: Doom Rivier, Schlechter 10873 (B, holo.! GRA, PRE, iso.!).

A. diphyllum Dinter in Fedde, Rep. 16: 338 (1920); Poelln. in Fedde, Rep. 52: 248

(1943). Type: South West Africa, Diamond Area I: Klinghardt Mountains, 5'c/tfl/cr 553

(B, hole.! PRE, photo.). A. apicicolum Krause in Engi. Bot. Jahrb. 52: 236 (1921).

Type: South West Africa: Klinghardt Mountains, Schafer 553 (B, holo., probably

destroyed). Scldfer 553 is the type for both A. diphyllum and A. apicicolum. A. longi-

folium sensu Sprengl., Syst. 1 1 : 84 (1825). A. longifolium sensu Bak. in J. Bot. Lond.

1872, 139 et in FI. Cap. 6 : 394 (1897). A. longifolium sensu Duthie in Ann. Stell.

Univ. 4, A: 9, t2, fig. 6, 9 (1926). Non Jacquin. Willdenow in Sp. PI. II, 140 (1799),

thought that A. muricatum L.f. and A. longifolium Jacq. could be conspecific and this

was accepted as a fact by Sprengel, Baker & Duthie; cf. Adamson, Journ. S.A. Bot.

7: 99 (1941).

Arthropodium muricatum (L.f.) Spreng., Syst. 2: 87 (1825).

Trachyandra flmhriata (Thunb.) Kunth, Emm. 4: 583 (1843).

743

Few leaved, muricate or glabrous plants up to 50 cm high. Roots many, orange,

thick, firm and long. Rhizome small. Squamae small, membranous, the outer loosely

surrounding shoot, the inner closely surrounding leaf- and scape-bases. Leaves 2-10,

opposite, flat, variable in size, linear to oblong, up to 80 cm long, 5-50 mm wide,

muricated, especially along the raised margin and nerves, seldom glabrous. Inflorescence

a laxly flowered, divaricate panicle up to 50 cm high, often with unequal accessory

branches; scape firm, muricate or glabrous, nearly as long as the racemes; bracts

small, 4 mm long, deltoid, auriculate; pedicels up to 12 mm in fruit, ascending.

Flowers with spreading, maculate perianth, segments ca. 10 mm long; inner stamens

connivent; ovary with 8 ovules per cell. Capsule globose, 5 mm in diam. glabrous.

Seeds black, tetrahedral, with hyaline margins.

Flowering Period: July-October. Adamson notes that it flowers most freely

after fires.

Distribution: Cape Peninsula, south western Cape to southern South West

Africa; on mountain slopes.

Cape. — Cape Town: Hills about Cape Town; Prior (PRE), Camps Bay, Strey 528

(PRE). Malmesbury: Riebeeck Kasteel Mountain, 6029 (BOL). Stellen-

bosch: Guardian Peak, Esterhuysen 11997 (BOL); Jonkershoek, Lam & Meeuse

4406 (L.). Paarl: lower clay slopes of Seven Sisters, Esterhuysen 9047 (BOL). Clan-

william: hills between Witte Els Kloof and Lamberts Hoek Berg, Pillans 9166 (BOL,

PRE). Calvinia: Nieuwoudtville, Schmidt 225 (PRE).

South West Africa. — Diamond Area 1 : Klinghardt Mts., Schafer 553 (B).

The type of A. scabromarginatum from Doom Rivier, Calvinia has numerous

leaves which are short and glabrous. It may prove to be a separate species. As the

species appears to be variable however, I have retained it here for the present.

26. T. bulbinifolia (Dinter) Oberm., comb. nov.

Anthericum bulbinifolium Dinter in Fedde, Rep. 29: 261 (1931). Type: South West

Africa: Liideritz Bay, Dinter 3836 (B, lecto.! PRE, photo.), 6693 (B, iso., probably

destroyed); Rheinpfalz, Dinter 6387 (B, probably destroyed). A. schlechteri Poelln.

in Fedde, Rep. 53: 130 (1944). Type: Cape, Namaqualand, Goechas, Schlechter

11376 (B, holo., PRE, GRA, iso.!). lA. odoratissimum Dinter, in Fedde, Rep. 29:

260 (1931). Type: South West Africa, Diamond Area 1 : coast at Rheinpfalz, Dintet

6377 (B, holo. 1 PRE, photo.); including var. brevibracteatum Dint.

Plants up to 40 cm high. Roots many, somewhat swollen near the rhizome,

tapering to the tips, often felted. Rhizome compact, vertical. Squamae surrounding

shoot forming a collar, outer firm, brownish, inner many, irregularly fimbriate, very

thin, white, membranous. Leaves many, linear, up to 18 cm long, 5 mm wide, long

tapered to the apex which is loosely curled when young, glabrous or minutely muricate

or ciliate, apparently erect, glaucous. Inflorescence a divaricately branched raceme,

laxly flowered; scape terete, slender, arcuate near the base, glabrous; bracts small,

ovate, auriculate. mucronate, edges minutely ciliate; pedicels up to 15 mm long in

fruit, erect. E lowers with a white, spreading, maculate perianth, segments 14mm long;

stamens uniform, filaments very shortly papillate, inner apparently connivent (no live

flowers seen); ovary with 12 ovules per cell. Capsule oblong, rounded, 6 mm long.

Seeds with sharp ridges.

Flowering Period: May-November, probably after good rains.

Dstribution: Southwest Africa, north western Cape; known from the coastal

sand near Liideritz Bay to Port Nolloth, Steinkopf and Goechas in Namaqualand.

South West Africa. — Luederitz: on road to Nautilus, Kinges 2597 (PRE); Luederitz

Bay 1 m S. of Lagoon, coarse sandy soil, mountainous country, Giess & van Vuuren

672 (PRE).

744

Cape. — Namaqualand : sandy plain North of Ugrabis, Marloth 2666 (PRE); Steinkopf,

Marloth 6753 (PRE); Port Nolloth, 26 miles East, Theron 1263 (PRE); Karroid hills

3 miles S. of the Orange River Mouth, Pillans 5608 (BOL); Kookfontein, Bolus 6604A

(BOL).

Anthericum odoratissimum Dinter inhabits the same area as T. hulbinifolia and flowers

at the same time. Giess & van Vuuren 867 from Aus may prove to be ^4. odoratissimum

as the squamae are large, membranous and gaping, not fimbriate, and therefore unlike

those of T. bulhinifolia. But no other differences were detected. Here further obser-

vations in the field and more collections will be necessary to settle this problem.

27. T. lanata (Dinter) Oberm., comb. nov.

Anthericum lanatum Dinter in Fedde, Rep. 29: 261 (1931). Poelln. in Fedde, Rep.

52: 251 (1943). Dinter in Fedde, Rep. App. 23 : 57 (1923) name only. Type: South

West Africa: Diamond Area 1, Klinghardt Mountains, Dinter 3901 (B, holo.! PRE,

iso.!). A. sublanatum Dinter in Fedde, Rep. 29: 262 (1931). Poellnitz in Fedde,

Rep. 52: 25 (1943). Type: South West Africa, LUderitz: Halenberg, Dinter 6628

(B, holo., PRE, photo.).

Small plants up to 30 cm. Roots many, felted, up to 22 cm long. Squamae

membranous, many, loosely arranged; outer squamae apparently absent or small.

Leaves many, linear, up to 25 cm long, 2 mm wide, tortuous, lanate with long, soft,

retrorse hairs. Inflorescence longer than the leaves, umbellate, with about 9 branches

arising out of the apex of the scape, secondary branches with unequal accessory

branches, many flowered; scape lanate or glabrous, firm, up to 14 cm long, about as

long as the umbel; bracts small, membranous, apiculate, auriculate; pedicels up to

8 mm in fruit, erect, thin. Flowers with perianth white, brown-keeled, maculate,

segments 6 mm long; filaments shortly papillate, uniform, inner erect, connivent !

(no live flower seen); ovary with ca. 10 ovules per cell. Capsule globose, black, 5 mm

n diam. Seeds (unripe) with hyaline ridges.

Flowering Period: August-September.

Distribution: South West Africa, Namib

South West Africa. — Diamond Area 1 : Pomona, S. of Liideritz Bay, Schaefer in herb.

Marloth 6597 (PRE).

The differences mentioned between T. lanata and A. sublanatum by Dinter, are not

important.

28. T. thyrsoidea (Bak.) Oberm., comb. nov.

Anthericum thyrsoideum Bak. in J. Bot. Lond. 1872, 139; FI. Cap. 6: 393 (1897).

Type: Cape, Tulbagh, Yuk River Hoogte, Burchell 1231 (K, holo.).

Small plants up to 18 cm long, hirsute. Roots many, somewhat swollen but not

bulbous. Rhizome small. Squamae membranous, broadly oblong, loosely arranged

around the shoots, leaf- and scape-bases. Leaves many, linear, up to 16 cm long,

2 mm broad, tapering gradually to the apex, covered with long, tawny, retrorse hairs.

Inflorescence a many flowered, fairly compact raceme, in some plants sub-umbellate

at the base; scape short, firm, ca. 7 cm long, setose with tawny, retrorse hairs; bracts

conspicuous, white, membranous, deltoid, cuspidate, margin irregularly dentate;

pedicels filiform, hirsute, firm in fruit up to 8 mm, about as long as the bracts. Flowers

with a mauve or pink, maculate perianth (maculae near the middle of segment), segments

8 mm, spreading (or recurved?); filaments shortly papillate, inner connivent? (no

live flower seen); ovary with 10 ovules per cell. Capsule globose, 5 mm in diam.

glabrous. Seeds not seen.

745

Flowering Period: August.

Distribution: Cape: originally collected by Burchell in the Tulbagh district;

recently found in the White Hill Karroo Garden.

Cape. — Laingsburg: White Hill Ridge, Compton 3228, 13397 (NBG); Matjiesfontein,

Beattie (NBG 2562/14, BOL).

29. T. tortilis (Bak.) Ohenn., comb. nov.

Anthericum tortile Bak. in Bull. Herb. Boiss. Ser. 2, 4: 996 (1904). Type: Cape:

Tulbagh, Saron, Schlechter 4846 (Z, holo.! PRE, photo.). A. salteri Leighton in Flow.

PI. of A. 18; t. 687 (1938). Type: Cape: Namaqualand, Springbok, Salter 966 (BOL,

holo.). A. oocarpum Schlechter ex Poelln. in Bol. Soc. Brot. 16, 2: 75 (1942). Type:

Cape, Namaqualand, Zuurfontein, Schlechter 8525 (B, holo.! PRE, photo.).

Small plants up to 15 cm. Roots fused into a few long tubers and with some

straggling thin, branched rootlets at the apex and tips. Rhizome small. Squamae

membranous, broad surrounding shoot, leaf- and scape-base. Leaves 3-6, linear,

up to 10 cm long, 6 mm broad, flat, long tapering to the apex, transversely, plicately

folded, glabrous or minutely pubescent, glaucous, margin prominent. Inflorescence a

divaricate panicle overtopping the leaves, side branches short 1-5; scape arcuate at

the base, firm; bracts broad, auriculate, apiculate, membranous, shorter than the

pedicels; pedicels up to 5 mm, recurved in fruit. Flowers with pale pink, maculate,

spreading perianth segments, 1 cm long; filaments scabrous above, fimbriate below,

declinate; ovary with 10 ovules per cell. Capsule narrow ovoid, 7 mm long. Seeds

ridged.

Flowering Period: June-August. Flowers scented.

Distribution: Cape, South-western districts, Namaqualand, Vanrhynsdorp-

Tulbagh and Hopefield.

CxPE. — Namaqualand: Mesklip, Esterhuysen 5953 (PRE, BOL). Vanrhynsdorp:

Nieuwerust, Esterhuysen 5982 (BOL). Hopefield: Bolus 12871 A (NH).

The leaves of this species are most unusual. They are folded in a compressed,

transversely plicate manner like a concertina.

30. T. jacquiniana (Roem. & Schult.) Oberm., comb. nov.

Anthericum jacquinianum Roem. & Schult. Syst. Veg. 7: 462 (1829) nom. nov. for

A. flexifolium sensu Jacq. Ic. 2, t.412 (1786-1793); non L.f. Type: Iconotype, Cape,

Jacquin, Ic. 2, 18; t.412. A. elongatum Willd. var. c, flexifol’um sensu Bak. in FI.

Cap. 6: 389 (1897).

T. jacquinii Kunth, Enum. 4; 578 (1843).

Plants up to 50 cm high. Roots 1-7, bulbous, separate or fused above, becoming

hard with age, often covered with many, thin, hard skins. Rhizome small, woody.

Squamae large, membranous, surrounding shoot, leaf- and scape-base. Leaves ca.. 12

per shoot, filiform or linear, flat, up to 15 cm long, 2 mm broad, wavy when young,

later erect, straight, with lax, long, patent silky hairs, dull blue-green. Inflorescence

a divaricate panicle with many short, accessory branches in the axils of the main branches,

many flowered; scape terete, firm, straight, laxly silky haired below, glabrous with

age; bracts short, auriculate, apiculate, membranous; pedicels short, 5 mm in fruit,

erect; if flower is not fertilized the short pedicel recurves; buds sessile. Flowers with

recurved, maculate perianth, segments 11 mm long; filaments erect, inner connivent,

hirsute below, muricate above; ovary with 12-16 ovules per cell. Capsule oblong

or, if few ovules are fertilized, obovoid. Seeds with ridges.

Flowering Period: June-September. Sweet smelling. “Anysblom ” (Mm/ot/i).

“ Flores odoratissimi, aemuli heliotropio peruviano ” {Jacquin).

Distrbution: Western Cape; Laingsburg, Carnarvon, Williston, Clanwilliam,

Calvinia, Namaqualand.

746

Cape. — Laingsburg: near Ketting Station, Acocks 17136 (PRE). Williston: Acocks

14700 (PRE). Carnarvon: 3 m. S. of Carnarvon, Acocks 16426 (PRE). Clanwilliam :

near Doom Rivier, Lewis (SAM 64668, NBG); Agtertuin, Schlechter 10865 (BOL,

PRE). Clanwilliam: Esterhuysen 5536 (BOL, PRE). Calvinia: Driefontein, Voor

Hantam, Marloth 1280 (PRE); near Lokenburg, Acocks 18485 (PRE). Namaqualand :

Brakdam, Esterhuysen 5688 (PRE); Bowesdorp, Lewis (SAM 55873, NBG); Garies,

Acocks 16436 (PRE).

Although no type is preserved, the specimens cited agree with the Jacquin plate.

They show the short, erect pedicel, the leaf has a lax pubescence of long silky patent

hairs, the root is tuberous and the outer squamae form a collar around the base of

the shoot.

The following specimen differs somewhat from typical T. jacquiniana; Acocks

19290 from near Van Rhynsdorp, succulent Karroo of hummocky flats, “ rare ” (PRE).

The leaves are rather short, wavy, terete, striate, tapered at the apex, forming a neat

flat rosette raised off the ground. This plant is glabrous and has the leaves spreading

instead of erect.

31. T. paniculata Oberm., sp. nov. T. muricatae affinis, sed radicibus bulbosis,

folds hirsutis differt.

Radices tuberosae sed non inter se adnatae. Folia circ. 4, lanceolata pubescentia

saepe undulata. Racemus multiflorus, ramis multis patentibus secundariis ferens;

pedicelli erecto-patentes ca. 1 cm longi. Flores albi maculati; ovula in loculis 10.

Capsula parva globosa.

Plants up to 30 cm high. Roots many, short, bulbous. Rhizome small. Squamae

thin, loose, membranous. Leaves 3-4, linear-lanceolate, up to 20 cm long, 1 cm broad,

flat, straight or wavy, glabrous above, pubescent below, margin shghtly raised, sparsely

fimbriate, tapered to the apex. Inflorescence a divaricate raceme with accessory

branches in many of the axils, many flowered, overtopping the leaves; scape with

a slight curve, about as long as the panicle, pubescent below; bracts small, ovate,

apiculate, auriculate; pedicels erecto-patent, ca. 1 cm long in fruit. Flowers with a

white, maculate perianth, segments 8 mm long; filaments papillate; ovary with 10

ovules per cell. Capsule globose ca. 4 mm in diam. Seeds not seen.

Flowering Period: September-October.

Distribution: Only known from the Clanwilliam district in the south western

Cape.

Cape. — Clanwilliam: Ohfants River, Rondegat, Schlechter 10792 (PRE, holo., L,

iso.). Pakhuis, MacOwan 1985 (NBG); east side of the Olifants River Valley in shady

fissures in sandstone rocks, Diels 366 pro parte (B).

The species resembles T. muricata but the roots are few, tuberous and short. It

differs from T. tortilis in that the roottubers are not fused, the leaves pubescent, the

raceme much branched and the pedicels erect in fruit. The leaves may show some

transverse folds when young.

32. T. zebrina {Schlecht. ex Poelln.) Oberm., comb. nov.

Anthericum zebrinum Schlechter ex Poelln. in Bol. Soc. Brot. 16, 2: 66 (1942). Type:

Cape, Namaqualand, Brackdam, Schlechter 11128 (B, holo.! PRE, GRA, L, iso.l). Fig. 7.

Plants up to 20 cm high. Roots many, some short and bulbous, others longer and

thinner, with swellings near the tips. Rhizome small. Squamae membranous, long,

tubular forming a neck around shoot, often up to 4 cm long, usually transversely

striped with dark brown lines; apex horizontal, irregularly toothed. Leaves about

4 per shoot, filiform to linear, ca. 18 cm long (occasionally up to 45 cm), 1-5 mm

broad, wiry, glossy, bright green, straight or wavy when young, shortly pubescent or

glabrous. Inflorescence a branched, many flowered raceme, side branches ascending,

very short in young plants; scape terete, pubescent; side branches and pedicels with

V,

747

748

short, hard, patent bristles; bracts ovate, apiculate or subulate, auriculate, white,

membranous, ciliate; pedicels patent-ascending, bristly, up to 8 mm long. Flowers

with a white or pinkish-mauve perianth with brown keels and maculae, hispid on the

outside, segments ca. 9 mm long; filaments with short papillae; ovary with 8 ovules

per cell, glabrous when young, becoming hispid with sparse, short, reflexed hairs when

capsule develops. Capsule globose, hispid. Seeds not seen.

Flowering Period: July-November.

Distribution: South-western Cape; known only from Namaqualand.

Cape. — Namaqualand: near Caries, Acocks 19314 (PRE), Theron 1265 (PRE);

Karrachas Pass, between Springbok and Kubus, Verdoorn 1823 (PRE); Brackdam,

Esterhuysen 5678 A (BOL, PRE).

Typical of this species are the usually striped squamae and the hispid raceme

and capsule.

33. T. karrooica Obenn., sp. nov., distincta, nullis e specibus notibus propinqua.

Plantae parvae ad 15 cm altae. Radices tuberosa, superne inter se adnatae. Folia

filiformia. Racemus ramis patentibus 1-3 ramo principali aequilongis, scapo brevi,

pedicellis patentibus 2-5 cm longis. Flores maculati; ovula in loculis 6.

Plants up to 15 cm high. Roots fused above with rhizome, split below into 2-3

tubers or entire, soft when young, becoming hard and brown skinned when old.

Rhizome small. Squamae small, thin, transparent, white, membranous. Leaves about

7 per shoot, filiform, 7-15 cm long, 1 mm broad, glabrous or with a few silky hairs,

apex apiculate. Inflorescence with 1-3 spreading branches nearly as long as the main

branch, about as long as the leaves; scape short, up to 5 cm, shortly pubescent; rhachis

glabrous; bracts small subulate; pedicels patent up to 2-5 cm long, apex occasionally

somewhat recurved in fruit. Flowers with a white perianth, maculate with yellow

spots, segments 8 mm long; filaments retrorsely papillate; ovary with 6 ovules per

cell. Capsule globose, 4 mm. Seeds (immature) with hyaline ridges.

Flowering Period: February-March. There is one record of it flowering in

October (Bolus 13830 from Richmond). Bryant noted that the flowering season was

short. At Vogelstruisbult in the Prieska district, he found it to be fairly common

but he could not find more than 3 specimens in flower.

Distribution: Apparently a rare, short flowering Karroo species collected only

once in each of the following districts, Calvinia, Richmond, Prieska and Kimberley.

Merxmiiller 1719, from South West Africa, Namib, 35 miles East of Henties Bay,

also seems to belong here.

Cape. — Richmond: False Karroo, 17 m west of Richmond, Acocks 16341 (PRE,

holo.!); Vlakplaats, Bolus 13830 (BOL). Calvinia 56 m. west of Eraserburg (locally

abundant) Acocks 14153 (PRE). Prieska: Vogelstruisbult, Bryant, J.256 (PRE).

Kimberley: Mauritsfontein, Acocks 5759 (KMG, with a very young inflorescence).

South West Africa. — Omaruru: 35 m east of Henties Bay, Merxmiiller 1714 (M,

PRE).

34. T. patens Oberm., sp. nov. T. flexifoliae afiinis sed pedicellis longioribus

patentibus differt, etc.

Plantae ad 40 cm altae. Radices lanatae tuberosae, superne inter se adnatae.

Squamae firmae brunneae collum longum formantes, interiores pallidae membranaceae.

Folia filiformia terna surculo, juvenia spiralia. Racemus divaricatus patens globosus

multiflorus ramis multis secundariis ferens; pedicelli patentes ca. 1 cm longi. Flores

maculati; filamenta biformia; ovula in loculis 10. Capsula globosa. Semina

verrucosa.

Plants up to 40 cm high. Roots forming one tuber in young plants increasing

to many separate tubers with age, often narrowed above where attached to rhizome.

749

with numerous straggling, lanate rootlets. Rhizome small. Older plants producing

4-7 shoots. Squamae, the outer firm, brown, forming a long, narrow neck around

shoot; inner thin, pale membranous. Leaves about 3 per shoot, filiform, up to 35 cm

long, -j-1 mm broad, erect, straight or wavy (especially so when young when the upper

half forms a corkscrew curl), sparsely pubescent or glabrous. Inflorescence a patent,

divaricate, many flowered raceme with accessory branches in many of the axils; at

first overtopping the young leaves, later, when the latter are full grown, more or less

equal in length; scape up to 24 cm long, minutely pubescent below; bracts small,

ovate, apiculate, membranous, white; pedicels patent or patent-ascending up to 1 cm

long. Flowers with brown keeled, maculate perianth segments up to 1 cm long;

filaments white, scabrous, the 3 inner bearded at the base, yellow in lower half {vide

Marloth); ovary with 10 ovules per cell. Capsule globose, 3 mm. Seeds verrucose.

Flowering Period: August-November. Sweet scented (“like a Reve d’or

rose,” Marloth).

Distribution : Western Cape, apparently abundant in Namaqualand, also collected

at Piketberg, possibly also Sutherland and Swellendam districts.

Cape. — Piketberg: Boschkloof, Leighton 114 (BOL); Plateau on Kapiteins Kloof

Mountain, Pillans 8016 (BOL), Stokoe 8512 (BOL); stony slopes on Twentyfour

River Mts. above Porterville, Esterhuysen 16196 (BOL). Namaqualand: Scully 104

(PRE, holo.!, BOL, iso!); between Khamieskroon and Springbok, Pillans 6351, 6355

(BOL); Mesklip, Esterhuysen 5831, 5832, 5847 {BOL), Leighton 1367 (BOL); Goechas

near Steinkopf, Schlechter 11373 (BOL); Kookfontein, Bolus 6605 (BOL, NBG).

Recognized by its rounded divaricate, many flowered raceme with many accessory

branches and the pedicels which are patent and arranged fairly close together at short,

regular intervals.

A specimen. Bolus in herb. BOL, 19605, from Nieuwoudtville, Calvinia district,

has a nearly simple raceme. This may have been the result of an injury which prevented

the side branches from developing normally. As a result the apical growth is stimulated

to form more flowers than usual. An analogous development was noted in T. affinis

Kunth.

The following specimens collected more to the East, Sutherland : Klein Roggeveld,

Marloth 10399 (PRE); foot of Komsberg Pass, Leistner 268 (PRE); Swellendam,

Marloth 8608 (PRE), seem to be near T. patens but the racemes are simple or with

1-2 short, basal branches. The root tubers formed purple stains on the sheets around

the area where they were mounted. They were poisoned with mercury-chloride and the

combination of this and a chemical in the tubers must have caused the colour patches.

35. T. oligotricha {Bak.) Oberm., comb. nov.

o/Zgotnc/zM/M Bak. in Bull. Herb. Boiss. Ser. 2, 4: 996 (1904). Type: Cape,

Piketberg, Schlechter 4857 (Z, holo., GRA, PRE, iso.!). The number was erroneously

published as 4887. A. flliforme 8 Thunberg, nom. tant. ex Juel, PI. Thunb. 120 (1918).

Cape Thunberg (UPS, holo. PRE, photo.). A. elongatum sensu Duthie in Ann. Stell.

Univ. 4, A: 11, t.III (1926); non Willd.

Plants up to 30 cm long. Roots swollen, pale, short, covered with old skin flakes

with many straggling rootlets, spreading horizontally. Rhizome short, vertical.

Squamae surrounding shoot yellow or white. Leaves usually about 7, erect, filiform,

ca. 15 cm long, 1 mm in diam., soft, usually with some refiexed bristlelike hairs

occasionally glabrous, rarely densely hairy. Inflorescence usually overtopping the leaves,

up to 30 cm high, simple or with a few, short, ascending branches, few flowered ; scape

slender, glabrous or minutely scabrid; bracts small, ovate, auriculate, ending in a

long, soft awn, membranous; pedicels short, ascending, up to 6 mm in fruit, buds

sessile. Elowers with a spreading perianth, white or pale pink, brown keeled, maculate

with yellow dots near the base; segments about 10 mm long, filaments subequal.

750

muricated, flattened below and there with dorsal and lateral fringes; ovary with 10

ovules per cell. Capsule oblong, apiculate, 3-6 mm long, touching rhachis. Seeds

tetrahedral, dark brown, verrucose.

Flowering Period: July-November, “Odour heavy, musklike” (Duthie).

Distribution: Cape, confined to a few south western districts, Piketberg, Ceres,

Stellenbosch, Cape. “ In clay soil, rare in sand ” {Duthie).

Duthie notes that it is the most abundant species around Stellenbosch. In

favourable conditions the plants are closely crowded, the horizontal roots interlacing

in a strking manner. Vegetative reproduction by means of root-buds is worth noting.

The pubescence is variable. Duthie found juvenile plants to be glabrous. One plant

of Schlechter 10712 (L) was found to be densely hispid.

Cape.— Cape Town: Phesantekraal, van Niekerk 160 (PRE). Stellenbosch: Duthie

522 (J. 15617, J, STE); Faure, Strey 523 (PRE). Piketberg: Kapitein’s Kloof

Mountain, Pillans 8016 (BOL); Boschkloof, Leighton 114 (BOL); Piketberg Road,

Schlechter 10712 (L.) (PRE). Ceres: Ceres Road, Schlechter 8982 (PRE, L.).

36. T. fiexifolia (L.f) Kunth, Enum. 4: 579 (1843).

Anthericum flexifolium L.f., Suppl. 201 (1781). Thunb. Prod. 62 (1794) et in FI. Cap.

edit. Schult. 318 (1823). Roem. & Schult. in Syst. Veg. 7: 461 (1829). Bak. in J.

Linn. Soc. 15: 307 1876). Type: Cape, Thunberg {lA^lA, holo., UPS, iso. PRE, photo.).

A. serpentinum Bak. in Flor. Cap. 6: 387 (1897). Adamson & Salter, Flora of the Cape

Peninsula 1950; 180. Type: Cape, Riet Valley near Cape Town, Ecklon & Zeyher

(K, syn.! PRE photo.). Malmesbury, Bachmann 870 (K, syn.). A. cirrifolius Schinz, in

Bull. Herb. Boiss. Ser. 2, 2: 937 (1902). Type: Cape: Piketberg, Piquenierskloof,

Schlechter 4921 (Z, holo.! PRE, GRA, BOL, NBG, iso.!). A. pappei Bak. in Flor.

Cap. 6: 388 (1897); Duthie in Ann. Stell. Univ. 4, A: 13 t.I, fig. 6, 8 (1926). Type:.

Cape, Tulbagh, Pappe (K, holo,! PRE, photo.).

Plants up to 25 cm high, rarely up to 60 cm. Roots tuberous short, about 4-8,

not fused, spreading, with some straggling rootlets, sometimes covered with old skin

flakes. Squamae surrounding the shoot short, membranous, dirty white. Leaves

variable, 5-12 per shoot, straight or wavy, filiform, 7-30 cm long, about 1 mm in diam.

(seldom flat and up to 5 mm wide), wiry, bright green, glossy, striate, with a minute

rough pubescence or glabrous, sometimes with some white bristles near the base on

the lower surface. Inflorescence an ascending panicle with 2 to many branches, as long

as the leaves or overtopping them; flowers fairly close together, more lax in the taller

forms; peduncles slender, straight, minutely scabrid or glabrous; bracts small, deltoid,

cuspidate or aristate; pedicels short, erect, up to 8 mm in fruit, buds sessile. Flowers

witli a white, dark-keeled perianth, maculate with yellow spots, segments 1 cm long;

filaments dimorphous, outer stamens scabrid, inner forming a tube, conspicuously

muricated above, flattened and fringed below; ovary with 6-10 ovules per cell. Capsule

globose, 5 mm in diam., glabrous, touching rhachis or ascending. Seeds black.

Flowering Period : July-October.

Distribution: The typical, small form is found on or near the Cape Peninsula,

the larger form {Anthericum pappei Bak.) is found around Tulbagh and further north

to Namaqualand. The A. cirrifolius Schlechter form with wider leaves, is from the

Piketberg and Clanwilliam districts.

Cape. — Cape Town; lower slopes of Devil’s Peak near Blockhouse, Leighton 2062

(BOL); Lion’s Head near Camps Bay, Wolley Dod 2331 (BOL); near Melkbosch

Strand, Esterhuysen 2841 (BOL). Malmesbury: Riebeeck Kasteel Mt. near Botmans-

kloof, Esterhuysen 6028 (BOL, PRE). Tulbagh: Tulbagh, Esterhuysen 17405 (BOL),

Duthie (BOL. 25716). Ceres: slopes below old Witzenberg Pass, Esterhuysen 20663

(BOL). Clanwilliam: Olifants River Barrage, Esterhuysen 5375 (BOL, PRE), 5802

(BOL); The Rest near Grey’s Pass, Gillett 3714 (BOL); N. of Pakhuis, Esterhuysen

751

25715 (BOL). Vanrhynsdorp: Gi^^bQvg, Esterhuysen 2\91() {BOIS) -, Klaver, Andreae

443 (PRE); Attys, 8327 (BOL, L). Namaqualand : near Springbok,

19331 (PRE).

Duthie who studied the species at Stellenbosch and at Tulbagh, noted their

variability in size and amount of branching in the inflorescence and suspected that

she was dealing with one variable species (cf. Ann. Stell. Univ. 4, A: 14, 1926). Dr.

C. G. Aim of Uppsala kindly compared a specimen, Leighton 2062, from Devils Peak,

Cape Town, with the type of Anthericum flexifolium L.f. in the Thunberg Herbarium

and found them to be identical. This specimen also agreed closely with the type of

A. serpentinum Bak. preserved at Kew. This typical form, found around Cape Town

and Stellenbosch, is a small form; apparently the climate does not suit it. The more

luxuriant forms are found around Tulbagh; they are taller with the much branched

inflorescence overtopping the leaves which moreover are often nearly straight. They

agree with A. pappei Bak. which must also be regarded as a synonym. The type of

A. cirrifolius Schinz from Piquenierskloof, has flat wavy leaves, up to 8 mm in width

but one plant from the type collection, on the BOL sheet shows the typical, narrow

T. flexifolia leaf. It can also be considered a synonym of T. flexifolia.

37. T. dissecta Oberm., sp. nov., distinctissima, nullis e specibus notibus propinqua.

Plantae parvae glabrae ad 30 cm altae. Radices inter se adnatae tuberosae durae.

Squamae collum durum brunneum formantes, interiores membranaceae dissectae.

Folia 1-4 filiformia. Racemus floribus paucis ramosus ; pedicelli breves tenues in fructu

ad 5 mm; ovula in loculis 10. Capsula erecta rhachidi attingens.

Plants small, up to 30 cm high, glabrous. Roots short, hard, few, tuberous, rough

in old plants where the outer skins flake off. Rhizome small. Squamae up to 3 cm long,

forming a neck, outer hard, brown, inner membranous, split into shreds. Leaves

1-4, filiform, 6-16 cm long, wiry, erect or curved outwards, bright green, glossy.

Inflorescence a divaricate, few flowered panicle with about 3 branches; scape thin,

terete, about as long as the leaves or slightly shorter; bracts minute, black or colourless,

apiculate; pedicels short, thin, up to 5 mm in fruit, erect. Flowers with a white, brown

keeled, maculate perianth, segments 8 mm long, spreading or somewhat recurved;

filaments subequal, inner surrounding the ovary at the base; ovary with 10 ovules per

cell. Capsule oblong, 5 mm long, touching rhachis. Seeds (unripe) small, with hyaline

ridges.

Flowering Period: August-October.

Distribution: Cape, Ceres, Calvinia, Vanrhynsdorp and Namaqualand districts;

apparently found in stony, karroid surroundings.

Cape. — Ceres : Bokkeveld Karroo, Paardekop near Spes Bona, Dwyka Rocks, Marloth

12531 (PRE, holo.); Ceres Karroo, Zwartkoppies near Spes Bona, Marloth 10471

(PRE). Calvinia: 53 m. S. of Calvinia, stony succulent karroo (locally fairly frequent),

Acocks 18989 (PRE). Carnarvon: Karee desert, Klipkolk, Stephens 6518 (BOL).

Vanrhynsdorp: Salt River, Stokoe 8509 (BOL).

Section 3. GLANDULIFERA Oberm., nov. sect.

Radices tenues longae, acumen versus anguste fusiformia. Folia uniformia; basi

saepe reliquiis fibrosis. Racemi ramosi rarius ad racemum simplicem redacti, leviter

vel dense glandulosi. Capsula tuberculis glanduliferis. Semina verrucosa.

38. T. scabra {L.f.) Kunth, Enum. 4: 585 (1823).

Anthericum scabrum L.f. Suppl. 202 (1781). Thunb. Prod. 63 (1794) et in FI. Cap.

ed. Schult. 320 (1823); Bak. in J. Bot. Lond. 1872, 137, in J. Linn. Soc. 15: 310 (1876),

and in FI. Cap. 6: 392 (1897). Type: Cape, sandy plains between Cape Town and

Hottentotsholland, Thunberg (UPS, sheet 2, holo., PRE, photo.).

Bulbine scabra (L.f.) Roem. & Schult. Syst. Veg. 7: 451 (1829). Fig. 8.

k

752

Fki. 8. Trachyaiidra scahni (L.f.) Kunth. a, habit, X b, flower, X 7. c, capsule, x 4„

d, stamen, x 10. e, tubercle tipped with a stipitate gland, enlarged.

753

Slender plants up to 40 cm high. Roots thin, wiry. Stem woody, branched near

the base into about 5 very short, woody contracted shoots, which are densely covered

with the congested tubular, membranous leaf-bases. Leaves many per shoot, lamina

contracted above the base, linear, up to 40 cm long, 2 mm broad, triquetrous, somewhat

leathery, muriculate or glabrous. Inflorescence an elongated, simple or few-branched,

glandular-scabrid raceme; flowers laxly arranged; scape muricate, the scabrid points

gland-tipped when young; bracts small, deltoid, apiculate; pedicels up to 2 cm long,

patent with the apex recurved in fruit. Flowers with a maculate perianth, segments

7 mm long; filaments with long retrorse bristles; ovary glandular, with 2 ovules per

cell. Capsule obtriangular, trisulcate, stipitate, 5 mm long, the apical tubercles larger;

persistent perianth-cup forming a frill at the base. Seeds usually 2 per loculus, ovoid,

5 mm in diam., grey, flattened, coarsely verrucose and with hard ridges.

Flowering Period : September-December.

Distribution: South Western Cape, on sand dunes near the coast. Apparently

absent on the Cape Peninsula, where the closely related T. sabulosa takes its place.

Cape. — Wynberg: Cape Flats near Stickland Siding, Acocks (BOL 24270). Malmes-

bury: near Darling, Esterhuysen 3S11 (PRE), Bolus 12869 (BOL). Hopefield: near

Hopefield, Leighton 2446 (BOL); near Vredenburg in sand between Hoetjies Bay and

Saldanha Bay, Bolus (BOL 12868). Clanwilliam: Olifants River Valley, Keerom,

sandy hills, Esterhuysen 17857 (BOL); Pakhuis, Esterhuysen 3195 (BOL); Lange

Vallei, Leighton (BOL 21618); Nortier Reserve, Lamberts Bay, Acocks 15191 (PRE).

Vanrhynsdorp : Klaver, Bolus (BOL 20652).

39. T. sabulosa (Adamson) Oherm., comb. nov.

Anthericum sabulosum Adamson in Journ. S.A. Bot. 1941, 99. Type: Cape, Cape

Town, Retreat Flats, Adamson 2631 (BOL, holo., PRE, iso!).

Plants up to 40 cm high. Roots thin, reddish. Stem forming a short, vertical,

woody base giving rise to several short, woody branches which end in congested, leafy

shoots, densely covered by membranous tubular leaf-bases. Leaves many per shoot,

lamina linear, up to 30 cm long, ca. 5 mm broad, keeled, glabrous or slightly muricate,

soft in texture, tapering to an elongated point. Inflorescence about as long as the leaves

or overtopping them, simple or with a few short, basal branches, glandular-hispid,

laxly flowered; scape arcuate below terete, fairly stout; bracts small, deltoid, cuspidate;

pedicels up to 2 cm long at first erect and short, later curved downwards, with the fruit

facing the scape. Flowers with a white, dark keeled, maculate perianth, glandular

hispid on the outside, segments 1 cm long; filaments muricate; ovary with 2 ovules

per cell. Capsule globose, 8 mm in diam., densely covered with dendroid, gland-tipped

excrescences, with a short stipe, hidden by covering. Seeds grey, 3 mm long, ovoid,

flattened, verrucose with hard, rough, irregular ligulae and wings around the sides.

Flowering Period : September-October.

Distribution: Cape, on sand dunes on the Peninsula, to Bredasdorp in the east,

Hopefield in the west.

Cape. — Bredasdorp: Cape Agulhas, 4386 (BOL). Caledon: Onrust River,

near coast, Esterhuysen 4247 (BOL). Peninsula: Kommetjie, Wolley Dod 3604 (BOL);

Retreat near Flora Road, Salter 8484 (BOL); Adamson 2628 (BOL). Hopefield:

near Hopefield, Wilman 2446 (BOL), Leighton 456 (BOL).

In appearance this species resembles T. scabra very much but the round, densely

covered capsule, which is also larger, separates it from that species. A small stipe is

754

present at the base of the capsule but it is not visible, being hidden by the dense covering.

The plants also seem somewhat taller and the pedicel and the outside of the perianth

are more densely glandular-hispid than in T. scabra.

40. T. glandulosa {Dint.) Oberm., comb. nov.

Anthericum glandulosum Dint, in Fedde, Rep. 29: 266 (1931); Poelln. in Fedde, Rep.

52: 236 (1943); includ. var. montis-ruschii Poelln. in Fedde, Rep. 52: 236 (1943).

Type: South West Africa, Windhoek: Ruschberg, Rusch in herb. Dinter 4381 (B,

holo.! PRE, photo.). Type of var.: from the same locality, Rusch Jnr. in herb..

Dinter 7921, (B! PRE, photo.).

Plants glandular, up to 30 cm high. Roots many, hard, stout but not swollen,

about 2 mm in diam. Stem short, lateral, woody, covered with old leaf fibres, forming

short branches which end in leafy shoots. Leaves linear up to 20 cm long, 4 mm broad,

keeled, hard, glaucous, covered with gland-tipped, scabrid points, the glands disap-

pearing with age, raphides many, distinct; tubular leaf-base membranous and with

hard dorsal nerve-bundles disintegrating into fibres. Inflorescence a divaricate raceme,

glandular hispid, many flowered; scape up to 7 cm long, much shorter than raceme,

the latter overtopping the leaves; bracts ovate, subulate, up to 6 mm long; pedicels

up to 12 mm long, patent, erect. Flowers with a white, brown keeled maculate perianth

with a few glands on the outside, segments 9 mm long; filaments retrorsely scabrid;

ovary few ovuled. Capsule small, 5 mm long, sparsely and shortly glandular-hispid,

contracted at the base into a short stipe. Seeds verrucose, 1-2 per loculus.

Flowering Period : January.

Distribution : Only known from the type locality, Ruschberg, district Windhoek

in South West Africa.

41. T. asperata Kunth, Enum. 4: 574 (1843); aggregate species.

Plants up to 1 m high, variable in size and amount of glandular pubescence. Roots

many, firm, sometimes slightly fusiform near the tips. Rhizome compact, woody,

sometimes forming a very short, woody stem, covered with fibres from old leaf-bases.

Leaves many, filiform, linear and keeled or triquetrous 20-40 cm long, 1-5 mm broad,

glabrous or minutely glandular pubescent and with some sparse setae in some; with

minutely papillate ridges, soft, glabrous, long tapered to the apex, base membranous,

tubular, the dorsal nerve bundle persisting as fibres. Inflorescence a divaricately branched

raceme, often with unequal accessory branches or vestiges of these in the lower axils,

seldom simple through reduction, overtopping the leaves: scape terete, glabrous or

sparsely glandular scabrid or with some sparse setae; bracts small, deltoid, subulate;

pedicels up to 12 mm long, erect, patent, recurved or contorted in fruit, usually glabrous.

Flowers with a spreading perianth, white or pink, dark keeled, maculate with green

spots, these occasionally faint, sometimes sparsely glandular outside, segments 1 cm

long; filaments scabrid the same colour as the perianth; ovary verrucose, each wart

capped with a rod-like, unicellular gland; with 4-6 anatropous ovules (embedded in

a fold of the placenta) in each loculus. Capsule 5 mm globose or obtriangular, with

some transverse ridges, constricted below, sparsely to densely covered with simple,

elongated, gland-tipped tubercles. Seeds verrucose, grey, ridges winged.

Flowering Period : September-December.

Distribution: Eastern Cape, Natal, Orange Free State, Swaziland, Transvaal.

Found in marshy ground or on stony hills.

755

Key to Varieties

Capsule densely and evenly covered with short, gland-tipped tubercles, or occasionally nearly

glabrous, globose; leaves flat and keeled or triquetrous, glabrous or with a pubescence of

short or long hairs; usually found in marshy places:

Pedicels scabrid, twisted in fruit ; inflorescence divaricately branched or simple through reduction ;

plants up to 25 cm high (eastern Cape, Orange Free State, Swaziland):

Inflorescence laxly flowered; leaves strai^t or twisted (eastern Cape, Orange Free State)

(a) var. asperata

Inflorescence densely flowered, densely glandular-hairy; leaves straight (Swaziland, eastern

Transvaal) {b) var. swaziensis

Inflorescence simple or branched, nearly glabrous, few to many flowered; leaves straight,

filiform, minutely pubescent (glabrous to the naked eye); small plants (Orange Free

State, eastern Transvaal) (c) var. carolinensis

Pedicels glabrescent, erecto-patent up to 15 mm long in fruit; inflorescence with ascending

branches; plants up to 50 cm high (Natal, Swaziland, eastern Transvaal) (d)vaT. nataglencoensis

Capsule sparsely covered with short, gland-tipped tubercles, obtriangular in outline, with some

transverse ridges; leaves terete, plants grasslike, nearly glabrous; usually found on mountain

grasslands :

Pedicels erect up to 15 mm long:

Inflorescence divaricately branched (eastern Cape, Orange Free State, Transvaal) (e) var. macowanii

Inflorescence simple or few-branched, flowers many, at first congested, pedicels thin and long

(Natal, East Griqualand) (/) var. stenophylla

Pedicels twisted, up to 7 mm long (Basutoland, Orange Free State, eastern Cape, Transvaal

highveld) (g) var. basutoensis

{a) var. asperata.

r. Kunth, Enum. 4: 574 (1843). Type: Cape, Albert: New Hantam Moun-

tains, in stony places, 4500-5000 ft. Feb. Drege 8735 (G, iso. ! PRE, photo.).

T. humilis Kunth, Enum. 4: 574 (1843). Type: Cape, Albany, Drege 8734 (P, iso,

PRE, photo.).

Anthericum kunthii Bak. in FI. Cap. 6: 392 (1897). As there is an Abyssinian species,

Anthericum humile Hochst., Baker changed the epithet to A. kunthii. He could not

trace the locahty but the Paris sheet bears an original label with “Albany ” on it which

must be correct.*

Cape. — Mount Currie: Kokstad, Tyson 1983 (PRE).

Orange Free State. — Bethlehem: Phillips 3115 (PRE), Potgieter 85 (PRE). Kroon-

stad: Pont 395 (PRE).

{b) var. swaziensis Oberm., var. nov. Racemus densus, dense glanduloso pubescens,

folia stricta.

Swaziland. — Mbabane, Compton 25240 (PRE, holo.; Swaziland Herb. iso.).

Transvaal. — Ermelo, Gunn (PRE, 28726).

(c) var. carolinensis Oberm., var. nov. Racemus glabrescens, folia stricta minute

pubescentia, planta minora.

Transvaal. — Carolina: Galpin 12212 (PRE, holo.). Ermelo: Ermelo, Convent pupils

85 (PRE), Louw 23 (PRE); Mavieriestad, Pott 5177 (PRE). Middelburg: Klein

Olifants River near Middelburg, Young A119, A8 (PRE).

(d) var. nataglencoensis {O. Kuntze) Oberm., stat. et comb. nov.

Anthericum nataglencoense O. Kuntze, Rev. Gen. 3, 2: 315 (1898). Type: Natal,

Glencoe, Kuntze (K, holo.). A. aggericolum Poelln. in Bol. Soc. Brot. 16, 2: 50 (1942).

Type: Natal, New Hanover: Dalton, Rudatis 7 (B, holo.! PRE, photo.). A. monti-

colum Poelln. in Bol. Soc. Brot. 16, 2: 56 (1942). Type: Natal, Estcourt: mountains

near Estcourt, Schlechter 3349 (B, holo.! PRE, photo.). A. natalense Poelln. in Bol.

Soc. Brot. 16, 2: 58 (1942). Type: Natal, Inanda: Verulam, Phoenix, Schlechter

2920 (B, holo.! PRE, photo.).

* Add to synonymy: Anthericum asperatwn (Kunth.) Bak. in J. Bot. Lond. 1872, 138; J. Linn.

Soc. 15: 310 (1876); in FI. Cap. 6 : 392 (1897). .4. Bak. in FI. Cap. 6 :390(1897). Type:

Cape, Griqualand East, MacOwan & Bolus (Bo], 1206, holo.).

756

Natal. — Weenen: Estcourt Pasture Research Station, Acocks 10815 (NH, PRE).

Inanda: Wood 457 (NH). Pietermaritzburg: Scotsville, Allsop 140 (NH). Zululand;

Mtunzini, Gingindhlovu, Lawn 1195 (NH).

Orange Free State. — Kroonstad: Pont 486 (PRE, SRGH). Harrismith: vlei near

Harrismith, Schweikerdt 867, 868 (PRE).

Transvaal. — Pretoria: Fairy Glen near Pretoria, Leendertz 1636 (PRE). Johannes-

burg: marsh. Moss 6041, 17768 (PRE). Ermelo: Ermelo, Convent Sisters 1 (PRE);

Nooitgedacht, Henrici 1668 (PRE) Gunn (PRE). Witbank: Steenkoolspruit, Kies

389 (PRE), Rogers 24304 (NH).

{e) var. macowanii (Bak.) Oberm., stat. et. comb. nov.

Anthericum macowanii Bak. in J. Linn. Soc. 15: 309 (1876) et in El. Cap. 6: 390 (1897).

Type: Cape: Albany, Grahamstown, MacOwan 64 (K, holo.).

Cape. — Albany: Grahamstown, Glass 564 (GRA). Uitenhage: Baakens Valley,

Cruden 353 (GRA, SRGH). Queenstown: mountain sides, near Queenstown, Galpin

1569 (PRE). Aliwal North: Elandshoek, F. Bolus (PRE, 176). Bathurst: Hopewell,

Acocks 11052 (PRE). Middelburg: Gryskop, Archibald 3146 (GRA). Between

Umtata and Maclear, Bolus 8710 (BOL).

Orange Free State. — Philippolis: Smith 4589 (PRE).

Transvaal. — Vereeniging: Burtt Davy (PRE, 5569). Benoni: Verdoorn, 826 (PRE).

Germiston: Elandsfontein, Prosser 1150 (PRE).

(/) var. stenophylla (Bak.) Oberm., stat. et comb. nov.

Anthericum stenophyllum Bak. in Bull. Herb. Boiss. Ser. 2: 781 (1901). Type: Natal

Lion’s River: Nottingham Road, Wood 5235 (Z, holo.! PRE, photo.).

Cape. — Mount Currie: Kokstad, Tyson 1820 (BOL); Mt. Currie, Hutchinson 1824

(PRE).

Natal. — Alfred: near Weza, Killick & Marais 2015 (PRE). Umzinto: Alexandra,

Dumisa, Rudatis 446 (L.). Lion’s River: Lidgetton, Mogg 834 (PRE).

(g) var. basutoensis (Poelln.) Oberm. stat. et comb. nov.

Anthericum hasutoense Poelln. in Bol. Soc. Brot. 16, 2: 51 (1942). Type: Basutoland,

Dieterlen 669 (B, holo.! NH, iso.! PRE, photo.).

Cape. — Aliwal North: Doctor’s Drift, Gerstner 249 (PRE). Barkly East: Gerstner

111 (PRE).

Orange Free State. — Fauresmith: Pont in herb. Henrici 2880 (PRE), Verdoorn 1041

(PRE), Smith 5609, 5576 (PRE). Ficksburg: Galpin 13919 (PRE). Bloemfontein;

Potts 1989 (BOL), Bolus (BOL 11179), Di.x (BOL 17476).

Basutoland. — Mokhotlong, Liehenherg 5678 (PRE); Mafeteng, Likhoele, Dieterlen

1221, 1220 (PRE); Mohales Hoek, Marais 1076 (PRE).

Transvaal. — Lichtenburg: Hakboslaagte, Kinges 1982 (PRE). Pretoria: Irene,

Smith 3520 (PRE).

The species is best treated as a very variable one. Although certain forms are very

uniform in certain areas, intermediates were found which upset the idea of separate

species. Moreover, it was observed that a certain form would crop up in a different

area, e.g. var. macowanii from Grahamstown could not be separated from plants

collected near Benoni and other localities. This happened repeatedly in all the varieties.

Neither did a study of the habitat assist in separating them. It was seen that similar

plants were reported to be growing on grassy mountain slopes or in marshy ground.

The varieties with the twisted pedicels showed a parallel development to those that have

erect pedicels; in both the degree of pubescence varied considerably.

42. T. capillata (Poelln.) Oberm., comb. nov.

Anthericum pilosum Bak. in El. Cap. 6: 395 (1897); non Jacq. A. capillatum Poelln.

nom. nov. for A. pUosum Bak. in Fedde, Rep. 50: 232 (1941). Type: Cape, Mount

Currie: Clydesdale, Tyson 2114 (K, holo., B. PRE, GRA, iso.!).

757

Plants up to 60 cm. Roots uniform, fibrous, many. Rhizome small covered with

leaf fibres. Leaves linear, up to 45 cm, triquetrous, each face 6 mm wide, herbaceous,

softly and patently pilose with long hairs, margins, prominent. Inflorescence branched

or simple, flowers somewhat eongested near the apex, many flowered, with a few

scattered stipitate glands; scape terete, pilose, bracts small, subulate; pedicels patent,

up to 17 mm long. Flowers with a spreading white, dark keeled, maculate perianth,

segments 9 mm long, filaments swollen and retrorsely seabrid above, glabrous below,

the 3 inner situated in the furrows of the triquetrous ovary; overy with 4 ovules per

cell, globose, with minute papillate glands. Capsule unknown.

Flowering Period: February.

D str'but on: Only known from the type locality, Griqualand East and from

the Alfred District of Southern Natal, at altitudes of ±3,000 ft.

Cape. — Mount Currie: Clydesdale, Tyson 2114 (K, holo., B, PRE, GRA, iso!).

Natal. — Alfred: near Harding “ Nongoni Veld” Acocks 13439 (PRE); Harding,

Oliver 71 (NH).

Possibly a form of T. gerrardi but this cannot be decided until the capsules are

known.

43. T. erythrorrhiza (Conrath) Oberm., comb. nov.

Anthericum erythrorrhizwn Conrath in Kew Bull. 1914, 135. Type: Transvaal, Ger-

miston, Modderfontein Conrath 111 (K, GZU, holo.! PRE, photo.).

Plants solitary or gregarious, up to 80 cm high. Roots spreading, uniform, hard,

stout, up to 4 mm in diam., white at first becoming orange-red. Rhizome creeping,

woody, covered with fibres. Leaves somewhat distichous, young tubular leaf-bases

pink; lamina flat, erect often with a twist, firm, somewhat fleshy, linear-lanceolate,

up to 40 cm long, 6 mm broad, shortly and sparsely pubescent, striate, midrib prominent

below and excentric. Inflorescence erect, simple or with 1-2 ascending basal branches,

up to 80 cm high; scape minutely and patently pubescent below, glandular pubescent

above when young; bracts small, membranous, apiculate; pedicels ascending in flower,

patent-erect in fruit and then up to 18 mm long. Flowers with perianth glabrous,

white with yellow maculae, segments 10 mm long; stamens spreading shortly seabrid;

ovary glandular-pubescent with 4 ovules per cell. Capsule globose, ca. 6 mm in diam.

stipitate, covered with short, glandular tubercles. Seeds grey, verrucose, globose,

tetrahedral, 3 mm in diam.

Flowering Period: October.

Distribution: Transvaal Highveld, black turf marshes.

Transvaal. — Johannesburg: Wattles, Moss 13659 (J, PRE); Palmietfontein, Gilliland

(Moss H. 26147); near Johannesburg, Onmianeng {in herb. Galpin 7156, PRE). Pretoria:

Baviaanspoort, Smith 788 (PRE) Leendertz (TM 6439, PRE); Derdepoort, Leendertz

(TM 4032, PRE).

44. T. gerrardii (Bak.) Oberm., comb. nov.

Anthericum gerrardii Bak. in J. Bot. Lond. 1872, 137; in Gard. Chron. 1876, 6: 100

et in FI. Cap. 6: 395 (1897). Type Natal, Zululand, Gerrard 1527 (K, BOL, holo!

PRE, photo.). A. tortifolium O. Kuntze, Rev. Gen. 3, 2: 315 (1898). Type: Cape;

Uitenhage, near Zwartkops River, Ecklon & Zeyher, Asphod. 1 10 (B, holo.! NBG, iso.!

PRE, photo.). A. montium-draconis Poelln. in Bol. Soc. Brot., Ser. 2, 16: 57 (1942).

Type: Natal, Charlestown, Thode (B. holo.! PRE, photo).

Plants up to 60 cm. Roots long, spreading, thin near the rhizome but younger

parts fusiform. Rhizome compact, woody, densely covered with fibres. Leaves

triquetrous, variable in length and width, 10-40 cm long, each face 2-10 mm broad,

with patent setae, margin prominent, often minutely glandular papillate. Inflorescence

a divaricate raceme, often with accessory branches in lower axils; flower buds at first

congested at the apex, rhachis elongating during anthesis, and the capsules eventually

758

laxly arranged; scape and side branches setose or with scabrid, obtuse tubercles,

topped with stipitate, evanescent glands and often very densely glandular-pubescent;

bracts variable in size, lowest up to 3 cm including the very long soft awn, upper floral

bracts small, deltoid, subulate, keel glandular-pubescent; pedicels at first ascending,

later patent, finally with the apex recurved or sometimes making a complete loop,

up to 15 mm long, densely to sparsely scabrid -glandular, very seldom glabrous. Flowers

with perianth glandular on the outside, white, dark-keeled, maculate with purple spots

{Munro), segments spreading 10-15 mm long; filaments scabrid, erect; ovary densely

glandular pubescent, with 4-6 ovules per cell. Capsule globose, 1 cm in diam. stipitate,

densely covered with long pectinate tubercles which are gland-tipped. Seeds round,

2 mm in diam. shortly verrucose, dark, ridges suppressed except for a small ridge near

hilum.

Flowering Period : October-February.

Distribution: Eastern Cape, Natal, Zululand, Swaziland, Eastern Transvaal.

On grassy mountain slopes or in marshes.

Cape.^ — Alexandria: Bushmans River Poort, Archibald 5325 (PRE); Ohio Bridge,

Archibald 5292 (PRE). Port Elizabeth: Port Ehzabeth, St. George’s Park, Urton 333

(GRA). Albany: Grahamstown, MacOwan 1454 (GRA); Trapp’s Valley, Daly 618

(PRE). Komgha: near Komgha, F/onagfl/? 576 (PRE). Kentani: Pcg/er 1379 (GRA).

East London: Rattray 580 (GRA). Molteno: Boesmanshoek Pass, Acocks 18691

(PRE). Mount Currie: near Clydesdale, Tyson 2123 (BOL).

Natal. — Weenen: South Downs, Evans 364 (NH). Estcourt: foot of Griffin’s HiU>

Acocks 11215 (PRE); Dalton Bridge, Munro (PRE). Inanda: Wood 432 (NH)-

Zululand: Lower Umfolosi : near Ntambanana Co<7<7 1891 (PRE). Mtunzini: Inyesani,

Gerstner (NH 22651); Amatikulu, Mogg 5794 (PRE).

Swaziland. — Mbabane, Compton 25212 (PRE).

Transvaal. — Wakkerstroom: van Dam (TM 24348, PRE).

The following specimens have very narrow leaves: —

Transvaal. — Barberton: Pott 5559 (PRE); Abbott’s Hill, Galpin 1036 (PRE). Nel-

spruit: Codd 9393 (PRE). Ermelo: Pott (TM 15119, PRE); Henrici 1092 (PRE).

Natal. — Giant’s Castle: Symons (TM 25267, PRE).

The large round, rough capsule is typical for this species. Without it some specimens

are not easily distinguished from T. asperata but T. gerrardii is usually a much coarser

and more scabrid plant.

45. T. giffenii {Leighton) Oberm., comb. nov.

Anthericum giffenii Leighton in J. S.A. Bot. 10: 59 fig. 2 (1944). Type: Cape, Victoria

East, 23 miles from Alice, farm Naude’s Hoek, shady places in kloof and on dry

hillside. Giffen 783 (BOL, holo.), 783b (PRE, NH, iso.!).

Plants succulent up to 50 cm high, in clumps. Roots somewhat fleshy. Rhizome

creeping, stout, without fibres. Leaves somewhat distichous, 10-12 per shoot, linear-

lanceolate, triquetrous, up to 50 cm long, each face ca. 12 mm broad, tapering

gradually to the apex, herbaceous, with a short, soft pubescence. Inflorescence a few-

branched raceme or simple, varying in length, up to 150 cm high, glandular pubescent;

scape pubescent; bracts small, ovate, acuminate, membranous; pedicels patent, up

to 2 cm long, apex recurved in fruit. Flowers with perianth spreading, white, green-

keeled, maculate, glandular pubescent on the outside, segments up to 15 mm long;

stamens declinate, scabrid; ovary glandular, placed on a disk; ovules 2 in each cell.

Capsule rough, glandular, producing lateral horns ca 5 mm long. Seed brownish

with a well marked hilum.

Flowering Period: September-March.

Distribution: Eastern Cape.

759

Cape. — Victoria East: near Alice, Giffen 783 (BOL, hole.), 783b (PRE, NH, iso.!).

Somerset East: Middleton, False Karroid Broken Veld, on shale, occasional on S.

aspect, Acocks 21687 (PRE).

The horned capsules are unusual.

The following Species found Outside Southern Africa are here put under

Trachyandra*

T. pyrenicarpa {Welw. ex Bak.) Oberm., comb. nov.

Anthericum pyrenicarpum Welw. ex Bak. in Trans. Linn. Soc. Ser. 2, Bot. 1 : 259 (1878)

et in F.T.A. 7: 491 (1898). Type: Angola, Huilla, in damp pastures between Humpata

and Gambas, Welwitsch (BM, holo.). Near T. saJtii but capsules with one large seed

in each locule.

T. malosana (Bak.) Oberm., comb. nov.

Anthericum malosanum Bak. in Flor. Trop. Africa 7: 492 (1898J. Type: Nyasaland,

Mount Malosa, Whyte (K, holo.! PRE, photo.). Southern Rhodesia: Inyanga,

Mtenderere Source, Wild 1465 (SRGH). Very near T. asperata and may be conspecific.

Uncertain Species

Anthericum ensifolium Solch in Mitt. Bot. Staatssamm. Miinchen II: 175, 1956. Type:

South West Africa, Maltahdhe: Farm Friedland, Walter 2113 (M. holo.! PRE, photo.).

The type consists of a rosette of leaves which emerge from a “ neck ” consisting of

scaly leaf-bases. There is a detached part of a branched raceme which resembles that

of Trachyandra laxa. I doubt whether the leaves belong to the inflorescence and am

inchned to consider it a nomen confusum.

A. omissum Poelln. in Bol. Soc. Brot., Ser. 2, 16: 59 (1942). Type: Cape, locality and

collector unknown (B, holo.). This is possibly Trachyandra saltii (Bak.) Oberm.

Nomen Confusum

A. praetermissum Poelln. in Bol. Soc. Brot. 16, 2: 62 (1942). In his description

Poellnitz notes that the collector and locality were unknown. The type is preserved

in the Berlin Herbarium and is an Ecklon & Zeyher specimen, Asphod. 106, from

mountains near Simonstown, Cape. Mounted on this sheet is a sterile plant of T.

hirsuta together with a loose inflorescence of an Urginea sp. As Poellnitz described

the leaves of T. hirsuta and the flowers of an Urginea, his species is a nomen confusum.

At Kew this Ecklon & Zeyher number is a syntype of T. brachypoda (Bak.) Oberm.

Types of Anthericum at Berlin

Dr. Buchheim, in a letter dated 29-11-58, writes that during the war the types

were stored somewhere else and survived. In the Liliaceae however, the types and

other material were sent to von Poellnitz in Thuringia. In this way much material was

saved that would have perished otherwise in the fire of March 1943. On the other hand

not all the material loaned to von Poellnitz came back to Berlin for his house was

bombed and the material there, but for a few sheets, was destroyed. It is possible

that amongst these were types of Anthericum and Chlorophytum.

Cuf., Miss. Biol. Race. Bot. 4: 308, fig. 100 (1939). Type: Kenya, Moyale,

Miss. Biol. 689 (RO, holo.). It was put in the section Trachyandra by Cufodontis. As the filaments

are glabrous, the anthers basifixed and the perianth segments persistent, etc., it does not belong to

the genus Trachyandra.

760

Acknowledgments

For their kind co-operation, loan of specimens, gifts of photos and valuable advice,

I wish to thank the Directors and Staff of the following institutions: B, BM, BOL,

G, GRA, GZU, J, K, KMG, L, M, NBG, NH, NYS, P, PRE, SRGH, UPS, Z.

I am also much indebted to Mr. Wessel Marais, South African botanist at Kew,

and to Mr. John Lewis of the British Museum for their kind assistance in comparing

types and undertaking research into literature not available here. I am also most

grateful to M. E. Connell (Mrs. Stutterheim) for the illustrations.

INDEX

Albuca sp

Anthericum Linn

acutum Wright

adscendens Poelln

affine Bak

aggericoluin Poelln

amboense Poelln

anceps Bak

angulicaule Bak

apicicolum Krause

aristatum Poelln

arvense Schinz

var. rigidum Suess

asperatum Bak

hachmannii O. Kuntze

hakerianuni Poelln

hasiitoense Poelln

betschuanicum Poelln

bipeduncidatwn Jacq

blepharophoron Roem. & Schult

bolusii Bak

brachyphylhmi Suess.

brachypodum Bak

var. caespitosian Adams . . . .

bracteatwn Thode ex Poelln

brehmeanwn Bak

hrevicaide Bak

brevifolium Thunb

brevitepalum Poelln

bninneoviride Dint

buchubergense Poelln

bidbinifolium Dint

hurkei Bak

caespitosum Dint

canaliculatum Ait

var. nifiini Lindley

canaliculatum sensu Bak

calyptrocarpum Bak

capiUatum Poelln

capitatuni Bak

cepaefoUuni Dint, apud Poelln..

chlamydophyllum Bak

ciliatuni L.f.

cirrifolius Schinz

comosiim Thunb

congestum Adams

conrathii Bak

cooperi Bak

crassinerve Bak

crassiuscidum Dint

crispuni Thunb

curvifoUum Krause

cyperaceum Kies

Page

690

670

673,686

680

728

755

724

672, 673,679

. . . .671,673,675, 676

742

724

740

755

742

677

756

725

706

734

708

684

729

689

707

689

724

740

739

743

721

685

733,734

734

733

670, 671,672, 673,685

756

672, 680

724

731

733

750

699

732

681,682

671,672, 680, 708

709

724

695

685

672, 673,675

761

delagoense Poelln

dielsii Poelln

din ter i Poelln

dipliylluin Dint

divaricatiim sensu Bak

divaricatwn Jacq

drepanophyllwn (Bak.) Schltr. ex Index Kew. 7...

duhium Roem. & Schiilt

diiriiin Suess

e latum Ail

elongatuni Willd

var. flexifoliiim Bak

var. holostachyimi Bak

clongatum sensu Bak

elongatuni sensu Dulhie

elongatuni sensu Poelln

ensi folium Soelch

erraticum Oberm

erythrorrhizum Conr

faicatum L.f

fa lea t inn sensu Bak

fasciciilatum Bak

fHifonne /i Thunb

fiUforme y Thunb

fHiforme 8 Thunb

var. reflexipHosiim O. Kuntze...

fimbriatiim Thunb

flavoviride Bak

flexifoliiim L.f

flexi folium sensu Jacq

friesii Weim

galpinii Bak

var. galpinii

var. matabelense (Bak.) Oberm., comb. nov. .

var. norlindii (Weimarck) Oberm., comb. nov.

gerrardii Bak

giflenii Leighton

gih’inn Krause

var. brunneoliim Poelln

glabrum Adams

glandidosiim Dint

var. montis-ruschii Poelln

gliitinosiim Dint. 1920, 1931

gracilitepaliim Poelln

graminifoiiiim Willd

hamatiim Poelln

haygarthii (IVood & Evans) kies

liereroen.se Schinz

var. longibracteatum Poelln

hirsutifloriim Adams

hirsutiim Thunb

hispidum L

humile Hoclist

indiitum Poelln

involucratum Bak

jacqiiinianum Roem. & Schult

var. aflinis Bak. pro parte

jacquinianum sensu Bak

jiinodii Bak

kdssneri Poelln....

kovismontanum Dint

krauseanum Dint

kinitliii Bak

Page

677

708

739

742

677

677, 737, 738

735

706

684

694

716, 737

745

724

738

749

740

759

740

757

735

734

671, 672, 673, 680,681

728

737

749

726

742

740

750

738,745

683

670, 671,672,676, 679

676

677

677,711

757

758

740

730

754

739

724

708

734

672,673,686, 689, 711

681

733

730

732

755

688

721

745

728

737

677

724, 725

706

673,674, 684

755

151752.5-4

762

Page

kyllingloides Krause 684, 685

lanatuni Dint 744

lanzae Cuf. 724, 725

la.xiim N. E. Br 739

/iliago L 669,671,709

/imosum Bak 679

loiigepediinculatum Steud. ex Roem. & Schult 736

longihracteatum Dint 706

var. brevihracteatiim Dint 706

forma siibmembranaceitin Poelln 706

longiciliatum Poelln 728

longifolium Jacq 734

var. biirchellii Bak 728

longifotiiini sensu Bak 742

longifolium sensu Duthie 742

longifolium sensu Spreng 742

longipedicellatum Poelln 690

longistylum Bak 672, 673, 686, 688, 689

lowryense Bak 729

lunatum Poelln 703

lydenburgense Poelln 681, 682

macowanii Bak 756

macranthum Bak 690

moculatum Poelln 734

magnificum Poelln 696

malosanum Bak 759

matabelense Bak 677

micranthum Bak 724

monticolum Poelln 755

montium-draconis Poelln 757

multiceps Poelln 690

multisetosum Bak 670, 671, 675

muricatum L.f 742

naraglencoense O. Kuntze 755

natalense Poelln 755

nigrobracteatum Dint 739

nonscriptum Poelln 690

nudicaule Bak 672,680

oatesii Bak 725

obtusifolium Poelln 729

odorarissimum Dint 743, 744

var. brevibracteatum Dint 743

oligolricitum Bak 749

omissum Poelln 759

oocarpum Schltr. ex Poelln 745

otavense Engl. & Krause 679

pachyphyllum Bak 680, 681

pachyrrhizum Dint 739

pallidiflavum Engl. & Gilg 740

pahistre Adams 730

pappei Bak 750

parado.Mun Roem. & Schult 732

pascuorum Poelln 680

patulum Bak 677

pauciflorum Thunb. a 706

pauciflonun Thunb. ft 708

var. minor Bak 708

pauper Poelln 690

peculiare Dint 733

pilosiflorum Poelln 734

var. subpapillosum Poelln 734

pilossisimum Poelln 733

pilosum Bak 724,756

plan i folium Thunb 699 (a, c)

pleiophyllum Schltr. ex Poelln 708

763

polyphyllwn Bak

pmetermissum Poelln

pretorieme Bak

pseiidofalcatum Poelln

pitbescem Bak

pudicum Bak

pulchellwn Bak

pimgens Poelln

pyremcarpum Welw. ex Bak

quadrifidwn Poelln

radula Bak

ramosum L

range! EngL & Krause

rautanenii Schinz

recwvatum Dint

reciirvifolium Bak

rehmaimii Bak

revolution L

revolution sensu Bak

revolution sensu Ker

revolution sensu Thunb. a.

revolution sensu Thunb. /i

rigens Poelln

rigidifoliion Poelln

rigidiim (Kunth) Bak

rigidion Krause

rigidum de Wild

robust ion Bak

rosenbrockii Poelln

roiiwenortii de Gorter

rubrovittation Poelln

rudatisii Poelln

var. angustuni Poelln

rustii Poelln

sabulosum Adams

salteri Leighton

saltii Bak

saundersiae Bak

scabromarginatum Schltr. ex Poelln,

scabriim L.f.

scabrum sensu Duthie.

scariosum Duthie

schlechteri Poelln

schultesli sensu Bak..

schultesil sensu Duthie.

secundum Krause & Dinter

seineri (Engl. & Krause) Poelln....

serpentimini Bak

spongiosum Poelln

sqiiameum L.f.

stenophyUum Adams

stenophyUum Bak

sternbergianum Roem. & Schult....

subcontortum Bak

sublanatum Dint

submaculatum Poelln

subpilosum Poelln

subulatum Bak

suessenguthii Soelch

tabulare Bak

tenellum Welw. ex Bak

tenuifolium Adams

thyrsoideum Bak

tortifolium O. Kuntze

tortile Bak

transvaalense Bak

trichophlebium Bak

1517525-5

Page

670, 686, 688

729, 759

682

728

679

684

759

690

672, 673,688

669, 671,690

.672, 673,683,684

679

734

686

677

.670,715,737, 738

737

736

737

684

683

705

684

670, 671,675

695

694

671,675

679

695

753

745

722, 724

672, 673,679

742, 743

751

737

672, 683

743

705

725

690

750

734

732

729

756

699

755

744

729

732

681,682

684

730

685

729

744

757

745

672, 673, 688, 689

672, 682

764

Page

triflorum Ait 670, 706

var. minor Bak 680

tubifenun Dint 683

tumidwn Poelln 708

undulation Jacq 708

imdulatum Thunb 732

iinilaterale Thunb 690

vaginatum Bak 682

validion Poelln 729

variegation Hort 700

vespertinion Jacq 734

vestition Bak ex Schinz 688

viscosion Bak 706

volkii Solch 677

weissianion Dint 735

whytei Bak 672, 683

wil/nsii Diels apud Poelln 681,682

wilnisii Diels ex Burtt Davy & Pott-Leendertz 681

zavatterii Cuf. 759

zebrinion Schltr. apud Poelln 746

Arthropodiion hispidion (L. f.) Spreng 732

Arthropodium miiricatum (L.f.) Spreng 742

Asphodelopsis Steud. ex Bak 690

Asphodehis capensis L 694

Bulbine Willd 669

canalicidata (Ait.) Spreng 734

caulescens (L.f.) Willd 669,712

ciliata (L.f.) Link 734

crispa (Thunb.) Roem. «fe Schult 695

falcata (L.f.) Roem. & Schult 735

frutescens (L.) Willd 690

scabra (L.f.) Roem. & Schult 751

seineri Engl. & Krause 690

sqiiamea (L.f.) Kunth 732

Bulbinella Kunth 669

biirkei (Bak.) Benth 722

fiHforme (Ait.) Kunth 737

sqiiainea (L.f.) Kunth 732

Caesia R. Br 669

brevicaulis (Bak.) Dur. & Schinz 689

coniosa (Thunb.) Spreng 699

contorta (L. f.) Dor. & Schinz 689

Chlorophytum Ker Gawler 669,690

andongense Bak 698

aridum Oberni 691 , 696

blepharophyllum Scliweinf. ex Bak 711

bowkeri Bak 691,700

brachystachyum Bak 701,703

brehmeanion Roem. & Schult 707

bitrchelUi Bak 699

capense (£.) Vos.^ 692, 694, 696, 698

comosum (Thunb.) Jacques 691,693,695,698,699

crassinerve (Bak.) Oberni 691,709

crispum (Thunb.) Bak 692,695

delagoense Bak 699

doloiniticuni Dint 703

dregei Poelln 711

drepanophylluin Bak 711, 735

dubiion Roem. & Schult 705,707

diirbanense O. Kuntze 698

elation (Ait.) R. Br. ex Spreng 694

var. burchellii Bak 699

gazense Rendle 69 1 , 698

graininifoliion (Willd.) Kunth 708

haygarthii Wood & Evans 689,711

inornation sensu Bak 707, 71 1

inornatum Gawl 690,693

765

krookianum Zahlbr

laxum R. Br

leipoldtii Poelln. (in error “ leipoldii ”)

lewisae Oberm

longipechmculatwn Forbes

longituberosum Poelln

macrosporum Bak

magnificum Weini

marlothii Poelln

modestum Bak

monophyllum Oberm

namaquense Poelln

nigricans Bak

norlindii Weim

papillosum Rendle

pauciphyllum Oberm

perlongibracteatum Poel I n

petiolatum Bak

piquetbergense Poelln

pulchellum Bak

rhodesiammi Phill

rigidiim Bak

schlechterianum Poelln

stamineum Zahlbr

strict urn Bak

trachycarpum Oberm

triflorum (Ait.) Kimth

tsiimebensis Dint

tuberculatum Duthie

undulatum (Jacq.) Oberm

vaginatum Bak

vaginatum Hua

vallis-trappii Poelln

variegatum Hort

viscosum Kunth

wilmsii Engl. & Krause

Dasystachys Bak

papillosa (Rendle) Bak

Dilanthes Salisb

revolutum (L.) Salisb

Eriospermum cernuum Bak

Hartwegia Nees

comosa (Thunb.) Nees

Liriothamnus Schltr

adamsonii Compton

involucratiis (Bak.) Schltr

Ornithogalum

leptophyllum Bak

longiscapum Bak

seineri (Engl. & Krause) Oberm

zeyheri Bak

Narthecium

Phalangium Kunth :

aethiopicum raniosum, floribus alhis, petalis reflexis Commelin,

bipedunculatum (Jacq.) Poir

canaliculatum (Ait.) Poir

comosuni (Thunb.) Poir

elatuni (Ait.) Redoute

fasciculatum” Bak

fastigiatum Poir

longifolium (Jacq.) Poir

pedunculatwn Bak

revolutum (L.) Poir

squameum (L.f.) Poir

triflorum (Ait.) Pers..

undulatum (Jacq.) Poir

vespertinum (Jacq.) Poir

viviparum Hort

Page

691, 693,696

691,693,698

711

710

696

699

692, 697

701

707

691,698

692, 693, 705

707,711

700

677,711

691,692, 703

710

700

698

708

698

697

691,692,693,703, 705,707

711

700

691,692,693,701

691,705

703,706

708

691,708,710

711

699

700

691, 693,706

700

691

703

669, 670,711

737

690

699

711,712,717, 720

720

721

690,711

690

711

690

669

669, 670

737

706

734

699

694

734

706

737

732

706

708

734

699

766

Trachyandra Kunth 711

§ Glandulifera 713, 717, 720, 751

§ Liriothamnus 713, 717, 720

§ T rachyandra 71 3 , 7 1 7, 718

acocksii Obenu 713, 721

adamsonii (Compton) Oberm 713,720

affinis Kunth 728

arvensis (Schinz) Oberm 715,716,740

asperata Kunth 713,715,716,754

var. asperata 755

var. basLitoensis (Poelln.) Oberm 756

var. carolinensis Oberm 755

var. macowanii (Bak.) Oberm 756

var. nataglencoensis (O. Kuntze) Oberm 755

var. stenophylla (Bak.) Oberm 713,756

var. swaziensis Oberm 755

blepharophora (Roem. & Schult.) Kunth 734

brachypoda (Bak.) Oberm 714,729

bracteosa Kunth 734

brehmeana (Roem. & Schult.) Kunth 707

bulbinifolia (Dint.) Oberm 714,743

burkei (Bak.) Oberm 713,721

canaliculata (Ait.) Kunth 734

capillata (Poelln.) Oberm 756

chlamydophylla (Bak.) Oberm 714,731

ciliata (Z,./.) Kunth 714,715,733

corymbosa Kunth 730

dissecta Oberm 714, 751

divaricata (Jacq.) Kunth 716,737

elongata (Willd.) Kunth 737

esterhuysenae Oberm 714, 726

erythrorrhiza (Com.) Oberm 716,757

falcata (L.f.) Kunth 711,714,716,735

fimbriata (Thunb.) Kunth 742

flexifolia (L.f.) Kunth 750

gerrardii (Bak.) Oberm 713,715,757

gifFenii (Leight.) Oberm 758

glandulosa (Dint.) Oberm 754

gracilenta Oberm 726

hirsuta ( Thunb.) Kunth 695, 716, 730

hirsutiflora (Adamson) Oberm 715,716,733

hispida (L.) Kunth 708,712,732

humilis Kunth 755

involucrata (Bak.) Oberm 712,713,721

jacquiniana (Roem. <& Schult.) Oberm 713,715,745

jacquinii Kunth 745

karrooica Oberm 748

lanata (Dint.) Oberm 744

laxa (N.E. Br.) Oberm 739,740

var. erratica (Oberm.) Oberm 714,715,740

var. laxa 739

longepedunculata (Steud. ex Roem. & Schult.) Kunth 714,736

longifolia (Jacq.) Kunth 734

malosana (Bak.) Oberm 716,759

margaretae Oberm 716,727

muricata (L.f.) Kunth 716,738,742

oligotricha (Bak.) Oberm 749

paniculata Oberm 746

paradoxa Kunth 732

patens Oberm 748

pauciffora (Thunb.) Kunth 707

peculiaris (Dint.) Oberm 733

pyrenicarpa ((Velw. ex Bak.) Oberm 716,759

reflexipilosa (O. Kuntze) Oberm 714,716,726

revoluta (L.) Kunth 737,738,740

sabulosa (Adamson) Oberm 713,753

767

saltii (Bak.) Obenn 714,715,716,722

var. oatesii {Bak.) Obenn 725

var. saltii 724

var. secunda {Krause & Dinter) Obenn 716, 725

scabra {L.f.) Kunth 713,751

schultesn Kunth 705,707

tabularis {Bak.) Obenn 714,730

thyrsoidea {Bak.) Obenn 715,744

tortilis {Bak.) Obenn 715,745

imdiilata (Thunb.) Kunth 732

vespertina (Jacq.) Kunth 734

zebrina {Schllr. ex Poelln.) Obenn 716,746

Urginea 759

769

The Identity of the Fungus Causing Anthracnose of

Olives in South Africa.

By

G. J. M. A. Gorter.

In South Africa all the diseases that affect olive trees are of fungal origin (Gorter,

1959). Of these, anthracnose is perhaps the most destructive. The fruit-decay for

which it is responsible was known in the late twenties to occur in the coastal regions

of the western Cape Province (Verwoerd, 1928). It was not before 1935, however, that

the cause of the disease was identified as an anthracnose fungus of the Gloeosporium

type.*

When the writer investigated this disease in 1938, he was struck by the fact that

the causal fungus produced a bright red pigment in its hyphae on various culture media.

This was unlike the cultural characteristics that Biraghi (1934) had described for

Gloeosporium olivarum Aim., a fungus which has, until now, been considered the only

known agent of anthracnose in olives. In addition it was found that the disease could

attack the flowers (Gorter, 1960), a symptom which had not been described before.

Thus the question arose whether the local anthracnose fungus should be considered

a variant of G. olivarum Aim. or whether it was a closely related species. It was essential

to compare it with olive anthracnose fungi from other parts of the world as well as

with fungi responsible for similar fruit tree diseases. Investigations concerning these

comparisons, a preliminary account of which has appeared elsewhere (Gorter, 1956)

are reported in detail below.

Materials and Methods

As these investigations were started immediately prior to World War II, isolates

of olive anthracnose fungi could be obtained only from a limited number of countries.

In the first place, isolates were obtained from Portugal, the country from which the

disease was originally described (Almeida, 1899). Other isolates came from Italy via

the “Centraal Bureau voor Schimmelcultures” at Baarn, Holland and from California

in the United States of America. Shortly after the war, cultures were obtained direct

from Italy.!

Two aspects of the fungi were studied in detail viz. the cultural characteristics of

the mycelium on various culture media and the morphology of spores produced on

these media. The agar media used were prepared according to the specifications of

Rawlins (1933). Colours of mycelium and acervuli were recorded by comparing with

the standard colours described by Ridgway (1912). Spore measurements were made

by suspending the spores in cooled melted water agar and examining a drop of the

suspension between a slide and a cover slip under the microscope. The congealed

agar prevented movement of the spores during measurement with an eyepiece micro-

meter.

* Specimen No. 33360, Stellenbosch-Elsenburg Mycological Herbarium, collected by Dr. B. J.

Dippenaar.

t The author is much indebted to Drs. R. V. de G. Cabral, Joh. Westerdijk, H. N. Hansen and

A. Ciccarone for providing him with the respective cultures.

770

Valid conclusions about the dimensions of spores can be drawn only if the number

of spores measured is enough to allow for a statistical analysis. According to Blumer

(1926) at least 100 spores should be measured to obtain regular variation curves for

length and width which are essential requirements for such an analysis. Thus measuring

100 spores from each isolate on a given substrate was adopted as a standard procedure.

The spore measurements were analysed in two different ways. To determine the

modus, i.e. the most common spore size, the measurements were arranged in correlation

tables similar to those used by Levine (1928). Differences in shape between the spores

of the various fungi were determined by applying the discriminant function to spore

measurements as proposed by Baten (1944).

Spore shapes of all the olive anthracnose isolates were determined. In addition

the spore shapes of the anthracnose fungi from a few comparable diseases were studied,

i.e. strains of Glomerella cingulata (Stoneman) Sp. & v. Schr., Colletorichum gloeos-

porioides Penz. and Gloeosporium limetticolum Clausen. They are respectively the

cause of anthracnose in apples, oranges and limes. Cultures of these fungi were obtained

from the “ Centraal Bureau voor Schimmelcultures ” with exception of a chromogenic

strain of Glomerella cingulata which was kindly provided by Dr. M. C. Goldsworthy

of Beltsville, Maryland in the United States of America.

Stock cultures of the fungi were maintained on oatmeal agar because it provided

the best culture medium for the production of spores. On most other agar media,

including potato-dextrose agar sporulation decreased with successive transfers and soon

stopped altogether.

Cultural Character stics

In Table 1, the cultural characteristics of olive anthracnose isolates from four

different countries situated in three continents have been compared on three agar

culture media. The most striking difference between the South African isolate and the

others is the entirely different colour of the substrate mycelium. In the three overseas

isolates it is a shade of olive green but in the local fungus it is bright red.

Differences in pigmentation were also noted in the acervuli of the two types of

fungi. This was shown by the following experiment. Sevillano olives were divided in

two groups which were respectively inoculated with the W.P. (South Africa) and U.S.A.

(California) isolates of the olive anthracnose fungi. After four days incubation under

moist conditions at 27°C, an abundance of acervuli was produced on affected fruit.

The acervuli of the W.P. isolate, had an “ orange-rufous ” colour while those of the

U.S.A. isolate were “ cinnamon rufous ”. On potato-dextrose agar their colours

were “ salmon-orange ” and “ apricot buff ” respectively.

Similar differences in colour of substrate mycelium and acervuli were encountered

if the W.P. isolate was compared with the anthracnose fungi from similar diseases

except in the case of the chromogenic strain of Glomerella cingulata. The existence

of a chromogenic strain characterized by the production of a red pigment on potato-

dextrose agar was originally discovered by Shear & Wood (1913). In 1943

a similar strain was isolated by M. C. Goldsworthy from an unknown variety of apple

near Washington, D.C. (Andes & Keitt, 1950). It was this strain that was compared

with the W.P. isolate of the olive anthracnose fungus (see Table 2). The comparison

shows that as far as cultural characteristics are concerned the differences between the

two fungi are so small that they could be identical.

771

Table 1. — Cultural characteristics of four olive anthracnose isolates from four different

countries on three agar media.

*A11 descriptions were made of 7-days-old cultures, grown in the dark at 25°C.

772

Table 2. — Cultural characteristics of Gloeosporium spec. (W.P.) from olive and Glome-

rella cingulata (chromogenic strain) from apple, on three agar media.

* See footnote Table 1.

Spore Measurements

The spore dimensions of Gloeosporium oUvarum were originally described by

Almeida (1899) as 15-27 X 4-6 p. Cabral (1941), who reinvestigated the disease in

Portugal mentions the following dimensions: 9 -4-22 -5 X 3 -7-5 -6 /<. In the case of

the local olive anthracnose isolate the writer determined the dimensions from its natural

habitat as being 12-9-19-7 X 3-7-6-4 /t (Mean: 16-48 X 4-88 p). This shows good

agreement in spore size between the fungi from the two countries and if only this

morphological character were taken into account there would be no reason to consider

the local anthracnose fungus different from Gloeosporium olivarum Aim. However,

it was observed that the olive anthracnose isolates from Italy and California, although

culturally similar to the Portuguese isolates, had slightly thicker spores and a closer

study of spore shapes was therefore considered necessary.

At first only culturally similarly looking fungi were compared including those

from related anthracnose diseases. They were grown on oatmeal agar for about four

weeks at a temperature of 25°C. Of each culture 100 acervuli spores were examined

under the microscope at a magnification of 900 X . In Table 3 the spore measurements

are analysed according to mean, variation and modus. Results of comparisons between

spore shapes according to their discriminant functions are given in Table 4. Modus

773

values as well as spore shapes show that the spores of Gloeosporium olivamm (4 Port.)

and G. limetticolum are identical. The same applies to the spores of Gloeosporium

olivarum (U.S.A.) and Glomerella cingulata. The spores of Gloeosporium olivarum

(C.B.S.) and CoUetotrichum gloeosporioides may also be considered identical in spite

of the fact that the F-value for the difference in spore shape is significant (Table 6).

Saunders (1939) has pointed out that the significance of a difference must be accepted

with reservations if the number of degrees for freedom in a statistical analysis is very

low unless the significance, i.e. the F-value, is very high. As this is not so in our case and

as the modus of the spore measurements for both fungi is the same, the shape of their

spores can be considered identical.

Table 3. — ^Spore sizes of six different anthracnose fungi from acervuli which had

developed after four weeks’ growth on oatmeal agar at 25 °C.

Table 4. — Results of comparisons of the discriminant functions of spore shapes with

reference to different combinations of anthracnose fungi.

* Determined from the Tables of Fisher and Yates (1938)

Thus, the comparison showed that the culturally similar olive anthracnose isolates

from different countries have, according to spore size, a closer resemblance to related

anthracnose fungi than to each other. From this it might well be concluded that

olives are subject to attack by a number of closely related anthracnose fungi. If this

is indeed the case then a comparison of the spores of the local olive anthracnose fungus

with those of the chromogenic strain of Glomerella cingulata as well as those of the

Portuguese anthracnose fungus might be expected to throw more light on the identity

774

of the local fungus. The spores of the said fungi were therefore compared not only

with each other but also with the spores of a number of other olive anthracnose isolates

as well as those of Gloeosporium limetticolum and of CoUetotrichum gloeosporioides.

The fungi were again grown on oatmeal agar for four weeks at 25C°, but this time the

acervuli spores were examined under the microscope at a magnification of 1100 x.

The spore sizes and differences in shape are shown in Tables 5 and 6 respectively.

Table 5. — Spore sizes of eight different anthracnose fungi from acervuli which had

developed after four weeks growth on oatmeal agar at 25 °C.

Fungus Species

Spore sizes in /(

Mean Variation I Modus

Gloeosporium species (W.P.)

Glomerella cingulata (chrom. strain)..

Gloeosporium olivarum (4 Port.). . . .

Gloeosporium limetticolum

Gloeosporium olivarum (C.B.S.)

CoUetotrichum gloeosporioides

Gloeosporium olivarum (315, Italy).. .

Gloeosporium olivarium (319, Italy)..

Table 6. — Results of comparisons between the discriminant functions of spore shapes

with reference to different combinations of anthracnose fungi

See footnote Table 4.

Table 5 shows that the modus values for the spores of both olive anthracnose

fungi from Portugal and South Africa were identical while those for the chromogenic

strain of Glomerella cingulata were but slightly different. Differences in the latter were

in fact so small that for practical purposes their modus values could be considered

identical with those of the two olive anthracnose fungi. This was confirmed by a

comparison of their spore shapes (Table 6) which showed that differences in the shape

between the spores of the local olive anthracnose fungus and those of the chromogenic

strain of Glomerella cingulata were even smaller than between the spores of the two

olive anthracnose fungi just mentioned. The identity of the spore sizes and shapes

of the local anthracnose isolate from olives and Gloeosporium limetticolum from limes

is obvious. The above findings and the fact that the spore size of G. limetticolum was

775

found identical with G. olivarum (4 Port.) all point to identical spore shapes in the

anthracnose fungi from olives in Portugal and South Africa, the anthracnose fungus

from limes and the chromogenic strain of the anthracnose fungus from apples.

On the other hand there was a distinct difference in shape between spores of the

local olive anthracnose fungus and olive anthracnose isolates from Italy. This is found

clearly expressed in the different modus values. These values greatly resemble the

values for Colletotrichum gloeosporioides, in fact they are identical for the C.B.S. isolate

and C. gloeosporioides. A study of their spore shapes revealed the interesting fact

that their differences were of the same order as between different isolates of G. olivarum

in Italy. As the variability of Colletotrichum gloeosporioides is well known (Burger,

1921) those facts strengthen the possibility that anthracnose of olives in Italy is caused

by strains of Colletotrichum gloeosporioides.

Although it has now been ascertained that the local anthracnose fungus of olives

was not only culturally identical with the chromogenic strain of Glomerella cingulata

but also identical as far as spore shape is concerned, the question remained how to

name the fungus. The answer depended on whether this chromogenic strain should

indeed be considered a form of G. cingulata. We have already seen that it differed

from this fungus in having narrower, more pointed, spores. Moreover, Andes & Keitt

(1950) stated that the chromogenic strain, and in fact all strains with narrow pcinted

atypical spores, have never produced perithecia, not even in cross breeding experiments.

However, still another difference was found, viz. the way in which spores are produced

on potato-dextrose agar. Shear & Wood (1913) already pointed out that the chromo-

genic strain produced an abundance of spores in the aerial mycelium of the fungus

when grown on potato-dextrose agar. It was also noted that these spores were slightly

smaller than those produced in the acervuli. The identical phenomenon was observed

for the South African olive anthracnose fungus. It now remained to ascertain whether

this phenomenon did also occur in cultures of Glomerella cingulata and Colletotrichum

gloeosporioides.

Therefore, a chromogenic and a non-chromogenic strain of Glomerella cingulata

were grown on potato-dextrose agar for four weeks at 25 °C. For purposes of com-

parison, cultures of Colletotrichum gloeosporioides. Gloeosporium limetticolum and

olive anthracnose isolates from Portugal, Italy and South Africa were also included

in this test. Measurements were made of spores borne on the hyphae as well as in the

acervuli. The measurements are recorded in Table 7. These show that all species

with a spore width <5 /< have a spore size which depends on whether the spores were

formed on the hyphae or in the acervuli. Fungi with a spore width >5/x like Colleto-

trichum gloeosporioides and the non-chromogenic form of Glomerella cingulata are

shown to possess negligible differences in the dimensions of the two types of spores.

While the mycelium spores were produced in abundance in the first group of

fungi — spores narrower than 5 p — they were sparsely formed in the second group

giving the impression of being widely dispersed growing acervuli spores. The difference

in the two groups of fungi is further illustrated in the following comparison of spore

size variation in mycelium and acervuli spores of Gloeosporium olivarum (Aa Port.)

and Colletotrichum gloeosporioides. For C. gloeosporioides the variation in length

and width of both types of spores were practically the same: —

variation mycelium spores: 9- 1-17-1 X 4 -0-6 -3 p,

variation acervuli spores: 12 -5-18 -3 X 4 -5-6 -3 /(.

In the case of Gloeosporium olivarum (Aa Port.) there is no such overlapping of

the variations: —

variation mycelium spores: 4-5-10-3 X 2-3-4 /«,

variation acervuli spores: 9-1-14-9 X 3 -4^ -6 p.

The figures show clearly that in this case two kinds of spores are involved.

776

Table 7. — Mean spore dimensions in n of different anthracnose fungi after four weeks

growth on potato-dextrose agar at 25°C

* Mycelium spores entirely free from acervuli spores could not be obtained.

Discussion and Conclusions

The foregoing investigations have established firstly that the local anthracnose

fungus of olives is morphologically and culturally indistinguishable from the chromo-

genic strain of Glomerella cingiilata that Dr. M. C. Goldsworthy isolated from apples.

Secondly it was shown that isolates of Gloeosporium olivarum and related anthracnose

fungi fall into two groups which are defined as follows: —

(a) the Glomerella-Colletotricinmi group which is characterized by conidio-spores

with a width modus equal or larger than 5 ix and which although very variable

in size, have similar dimensions on potato-dextrose agar whether borne on

individual hyphae or in acervuli.

{b) the Gloeosporium group characterized by conidiospores with a width modus

smaller than 5 p and which produces two types of spores — micro and macro-

conidia — on potato-dextrose agar, each with comparatively little variation

in size. In this type of anthracnose fungus production of perithecia has

never been observed.

It could be asked, however, whether distinctions based on differences in spore

size are valid reasons for distinguishing anthracnose fungi. Some investigators (Kruger,

1913, Burger 1921) maintain that spore size is not a good criterion. However, they

base their opinions largely if not exclusively on the large variability in the length of

spores and fail to appreciate the importance of spore width as a shape-giving factor.

The foregoing experiments have clearly shown that under varying conditions of growth

the width of spores varies much less than the length. In addition the statistical analyses

carried out showed that with regard to spore shape width was a more important factor

than length. Therefore, in the opinion of the author, spore dimensions, or rather spore

widths, may undoubtedly be used as a differential character for the identification of

anthracnose fungi.

From the description of the two groups it is obvious that the chromogenic strain

of Glomerella cingulata falls into the second group and its designation as Glomerella

cmgulata is therefore incorrect. In this connection it should be remembered that

Edgerton (1915) has already distinguished between perithecia-forming and non-peri-

thecia-forming anthracnose fungi from apples, identifying the latter by the name

Gloeosporium fructigenum Berk, and that the author of the name (Berkeley,

1856) has noted that the spores of this fungus were not as variable as those from

Ill

a similar perithecia-forming isolate from grapes. The chromogenic strain and

the identical South African anthracnose fungus from olives should therefore be

considered a form of Gloeosporium fructigenum Berk. They could be conveniently

called Gloeosporium fructigenum Berk. f. chromogenum Gorter form. nov. [=M. C.

Goldsworthy’s chromogenic strain of Glonierella cingulata (Stoneman) Spauld. &

V. Schrenk loc. cit.]. This fungus causes apparently not only an anthracnose disease

of apples and olives but also of peaches (Ramsay et. al. 1951).

The identity in spore size and the similarity of most cultural characteristics in

G. fructigenum Berk. f. chromogenum, G. olivarum Aim. from Portugal and G. limet-

ticolum Clausen show that these three fungi are very closely related and it is felt that this

should be expressed in the name. Hence, Gloeosporium olivarum Aim. should be

renamed Gloeosporium fructigenum Berk. f. olivarum (Aim.) Gorter comb. nov.

[= Gloeosporium olivarum Aim. he. cit.] while Gloeosporium limetticolum Clausen

should be called Gloeosporium fructigenum Berk. f. limetticolum (Clausen) Gorter

comb. nov.

The investigations have also shown that Gloeosporium fructigenum f. olivarum

and Gloeosporium fructigenum f. chromogenum are apparently not the only fungi capable

of causing anthracnose disease in olives. There is evidence that olives are also subject

to attack by CoUetotrichum gloeosporioides Penz. from citrus and Glomerella cingulata

(Stoneman) Spauld. & v. Schrenk from apples. The similarity of Italian isolates of

olive anthracnose fungi and CoUetotrichum gloeosporioides Penz. is strengthened by the

fact that Ciccarone (1947) discovered the presence of long flexible hyphae at the

periphery of the acervuli in Italian isolates which remind one somewhat of setae. Von

Arx (1957 a), who most probably studied the Italian isolate present in the “ Centraal

Bureau voor Schimmelcultures ” even goes so far as to consider Gloeosporium olivarum

Aim. a synonym for CoUetotrichum gloeosporioides Penz. As at present the latter

fungus is commonly indicated by its perithecial stage it should be called Glomerella

cingulata (Ston.) Spauld. & v. Schrenk var. crassispora Wr. in accordance with the

terminology used by Wollenweber & Hochapfel (1949) so as to distinguish it from

the apple anthracnose fungus which the same authors have called Glomerella cingulata

(Ston.) Spauld. & v. Schrenk var. brevispora Wr.

After completion of the present comparative study of olive anthracnose fungi,

von Arx (1957 b) has published an extensive revision of the genus Gloeosporium. If

his opinion — that Gloeosporium should be dropped in favour of CoUetotrichum as a

generic name — finds international recognition, then the above described forms of

Gloeosporium fructigenum should be referred to as forms of CoUetotrichum fructigenum

(Berk.) Vassil. (Vassiljewski & Karakulin, 1950). Von Arx’s opinion that this fungus

is a synonym of CoUetotrichum gloeosporioides is of course not sustained.

Summary

The fungus which causes anthracnose of olives in South Africa was found to be

a form of Gloeosporium fructigenum Berk. It has been given the name of Gloeosporium

fructigenum Berk. f. chromogenum Gorter form. nov. [=M. C. Goldsworthy’s chromo-

genic strain of Glomerella cingulata (Stoneman) Spauld. & v. Schrenk he. cit.].

Gloeosporium olivarum Aim., which was originally described as the cause of olive

anthracnose in Portugal has been renamed Gloeosporium fructigenum Berk. f. olivarum

(Aim.) Gorter comb. nov.

Evidence has been presented that olives are not only subject to anthracnose diseases

caused by the above mentioned fungi but also by two varieties of Glomerella cingulata

viz.: G. cingulata (Ston.) Spauld. & v. Schrenk var. brevispora Wr. from apples and

G. cingulata (Stonem.) Spauld. & v. Schrenk var. crassispora Wr. (--CoUetotrichum

gloeosporioides Penz.) from citrus.

Literature Cited

Almeida, J. Verissimo de (1899)

Andes, J. O. and Keitt, G. W. (1950)

Baten, W. D. (1944)

Berkeley, M. J. (1856)

Biraghi, a. (1934)

Blumer, S. (1926)

Burger, O. F. (1921)

Cabral, R. V. de G. (1941)

Ciccarone a. (1950)

Edgerton, C. W. (1915)

Fisher, R. A. and Yates, F. (1938). .

Gorter, G. j. M. a. (1956)

Gorter, G. j. M. a. (1959)

Gorter, G. J. M. A. (1960)

Kruger, F. (1913)

Levine, M. N. (1928)

Ramsay, G. B„ Smith, M. A. and

Heiberg, B. C. (1951)

Rawlings, T. E. (1933)

Ridgway, R. (1912)

Saunders, A. R. (1939)

Shear, C. L. and Wood, A. K. (1913)

Vassiljewsky, N. I. AND Karakulin,

B.P. (1950)

Verwoerd, L. (1929)

VON Arx, j. a. (1957a)

VON Arx, J. A. (1957b)

WOLLENWEBER, H. W. AND HOCHAPFEL,

H. (1947)

La Gaffa des olives en Portugal. Bull. Soc. Mycol. France 15:

90-94.

Variability of Glomerella cingulata (Stonem.) Sp. et v. Schr.

from apples. Phytopath. 40: 915-925.

The discriminant function applied to spore measurements.

Pap. Mich. Acad. Sci., Arts and Lett. 29 (1943) Part 1 : 3-7.

Gloeosporium fmctigenum. Gardner's Chron. 245.

Variazione in due ceppi di Gloeosporium olivamm Aim. di

provenienze diverse. Boll. R. Staz. Pat. veg. N.S. 14:

223-253.

Variationsstatistische Untersuchungen an Erisiphaceen. Ann.

Myc. 24: 179-193.

Variations in Colletotriclium gloeosporioides. J. Agric. Res.

20: 723-736.

Notas Sobre o Gloeosporium olivarum Aim. Agron. Lusit.

3 (1): 49-58.

Considerazioni biologiche e sistematiche sull'agente della

“ lebbra ” delle olive, recentemente osservati nel Lecesse.

Boll. Staz. Pat. veg. Roma, Ser. Ill, 5: 143-165.

Effect of temperature on Glomerella. Phytopath. 5 : 247-259.

Statistical tables for biological, agricultural and medical

research. London, Edinburgh: Oliver & Boyd.

Anthracnose fungi of olives. Nature 178: 1129-1130.

Onderzoekingen over Gloeosporium fructigenum Berk. f.

chromogemim Gorter form, nov., de oorzaak van de

anthracnoseziekte van olijven in Zuid Afrika. Unpublished

D. Sc. (agric.) thesis. University of Pretoria.

Control of anthracnose in olives. Fmg. S. Afr. 36 (3):

18-19, 22, 24.

Beitrage zur Kenntnls einiger Gloeosporien. Arb. Kais. Biol.

Anst. Land und Forstw. 9: 312-322.

Biometrical studies on the variation of physiologic forms of

Puccinia graminis tritici and the effects of ecological factors

on the susceptibility of wheat varieties. Phytopath. 18:

7-123.

Anthracnose of peaches. Phytopath. 41 : 447^55.

Phytopathological and botanical research methods. New

York: John Wiley & Sons, Inc., London: Chapman &

Hall, Ltd.

Color standards and color nomenclature. Washington D.C. :

Publ. Author.

Statistical methods with special reference to field experiments.

Sci. Bull. Dep. Agric. S. Afr. No. 200, Sec. Ed.

Studies of fungous parasites belonging to the genus

Glomerella. Bull. U.S. Dep. Agric. Bur. Plant. Ind. No. 252.

Fungi imperfecti parasitica, 2: 296.

A preliminary check list of diseases of cultivated plants in

the winter rainfall area of the Cape Province. Sci. Bull.

Dep. Agric. S. Afr. No. 88.

Die Arten der Gattung Colletotrichum. Phytopath. Z. 29:

413-^68.

Revision der zu Gloeosporium gestellten Pilze. Verb. Kon.

Nederl. Akad. Wet., Afd. Natuurk., 2 Reeks, 51 (3):

1-153.

Beitrage zur Kenntnis parasitarer und saprophytischer

Pilze. VI. Vermicularia, Colletotrichum, Gloeosporium,

Glomerella und ihre Beziehung zur Fruchtfaule. Z. Para-

sitenk. 14: 181-268.

3

INDEX TO VOLUME VII

Page

Abutilon flanaganii A. Meeiise 484

fruticosum G. & P 484

galpinii A. Meeiise 484

grandiflorum Don 485

grantii A. Meeiise 484, 487

indicum sensu Harv 485

mauritianum iJacq.) Medic 485

piloso-cinereum A. Meeiise. . . .485, 486

pseudocleistogamum Hochr 483

sonneratianum Sweet 484

Acacia lebbeck (L.) Willd 81

lophantha Willd 81

marlothii Engl 77

mollissima Willd. 87

sevicocephala Fenzl 70

ACANTHACEAE 443

Achnanthes Bory 605

exigua Gnin 605

lanceolata (Breb.) Gnm.

var rostrata Must 605

linearis W.Sm 605

microcephala Kg 605

minutissima Kg 605

Achras L 366

Acokanthera G. Don 447

deflersii 450

friesiorum 450

lamarckii G. Don 448

longiflora 450

oblongifolia (Hoclist.) L. E.

Codd 449

oppositifolia {Lam.) L. E.

Codd 448, 449

ouabaio 450

rhodesica 450

schimperi {A. DC.) Schweinf.

var. rotundata L. E. Codd 449

speclabilis (Sond.) Hook. f. . 449

venenata G. Don 448, 449

venenata sensu Stapf 448

var. spectabilis (Sond.) Sim 449

venenatiim (Burm. f.) G. Don

ex C. A. Sm. nom. illegit. 448

var. (Sond.) Markgf. 448

venatorium E. Mey 448

Actinostronia infundibiilifornie Klotzsch .... 120

Adenenia liyssopifoliiini (Willd.) G. Don.. 462

Adeniiim arabiciim Balf. f 452

coetaniim Stapf 452

honghel A. DC 452

micrantlumi Stapf 452

midtiflonini Klotzsch 452

obesum (Eorsk.) R. & S.

var. multiflorum {Klotzsch)

L. E. Codd 452

speciosum¥tnz\ 452

Aeluropus 2

Afrachneria 3

African Flowering Plants — New and

Interesting Records of 21

African Plants — Notes and New Records of 443

Afroclirysopliylliim Engl, p.p 340

Albizia Diirazz 67

adianthifolia {Schiini.) W. F. Wight 79

amara (Roxb.) Boiv.

subsp. sericocephala {Benth.)

Brenan 70

anthelmintica {A. Rich.) A. Brongn 77

var. australis Bak. f. 77

var. piibescens Burtt Davy 77

antunesiana //arms' 74

brevifolia Schinz 69

coriaria Welw 74

distachya (Vent.) MacBride 81

evansii Burtt Davy 79

fastigiata (E. Mey.) Oliv 80

forbesii Be/a/z 72

gummifera sensu C. A. Sm 80

harveyi Foiirn 71

hypoleuca Oliv 71

julibrissin Diirazz 67

lebbeck (/,.) Benth 81

var. australis Burtt Davy 75

lophantha ( Willd.) Benth 81

niossambicensis Sim 73

pallida Harv 71

parvifolia Burtt Davy 69

rhodesica Burtt Davy 75

ra^ezszY Burtt Davy 69

sericocephala Benth 70

struthiophylla Milne-Redhead 70

suluensis Gerstner 76

tanganyicensis Bak. f. 75

unihalusiana Sim 77

versicDlor Welw. ex Oliv 73, 74

var. niossambicensis Schinz 73

Albuca sp 690

Aleurodiscus Rabenhorst ex Schroeter. ... 169

acerinus

var. alliaceus 170

botryosus 169

disciformis 170

mirabilis 170

polygonioides 170

roseus 170

Allium ainpeloprasum var. B. Gawl 481

dregeanum Kunth 481

rotundum L 481

Aloe arachnoides (“ arachnoidea ”). . . .427, 428

Alsike Clover Mosaic Virus 1 498

Amauroderma rude {Berk.) G. El. Gunn 86, 104, 105

4

Page

Amphora Ehrenb 608

ovalis Kg 608

var. libyca (E.) Cl 608

submontana Hust 608

Anabaena Bory

affinis Lemm 564

catenula (Kiitz.) Born & Elah.... 564

galpinii Claassen 564

mesiana Claassen 564

pseudocatenula 564

Andropogon ravus J. G. Anderson 417

sylvaticus Hubbard 418

Aningeria adolfi-frederici (Engl.) Robyns

& Gilbert 341

Anomochloa 5

ANOMOCHLOEAE 7

ANOMOCHLOIDEAE 5

Anomoeoneis Pfitzer 606

exilis {Kg.) Cl 606

sphaerophora {Kg.) Pfitzer 606

Anthericum L//;« 670

acutum Wright 673, 686

adscendens Poelln 680

affine Bak 728

aggericoluni Poelln 755

amboense Poelln 724

anceps 672, 673,679

angulicaule Bak.. 61 \, 673, 675, 676

apicicoliim Krause 742

aristatiini Poelln 724

arvense Schinz 740

var. rigidum Suess 740

asperatum Bak 755

bachtnannii O. ¥Mr\{ZQ 742

bakerianum Poelln 677

basiitoense Poelln 756

betschuanicinn Poelln 725

bipeduncidatum Jacq 706

blepharophoron Roem & Schult 734

bolusii Bak 708

brachyphylliiin Suess 684

brachypodum Bak 729

var. caespitosum Adams 729

bract eat urn Thode ex Poelln. . . 689

brehmianum Bak 707

brevicaule Bak 689

brevifoliiini Thunb 689

brevitepahmi Poelln 724

brunneoviride Oint 740

buchubergense Poelln 739

bulbinifoliiun Dint 743

burkei Bak 721

caespitosum Dint 685

canaliculatuni Ait 733, 734

var. rufum Lindley 734

canaliculatuin sensu Bak 733

calyptrocarpum Bak.. 610, 671, 672.

673, 685

capillatum Poelln 756

capita! um Bak 672, 680

cepaefoliuni Dint, apud Poelln. 724

chlamydophyllum Bak 731

ciUatum L.f 733

cirrifoUusSckinz 750

comosum Thunb 699

congest u m A&am& 732

conrathii Bak 681 ,682

cooperiB k 671,672,680,708

Page

Anthericum crassinerve Bak 709

crassiusculum Dint 724

crispum Thunb 695

curvifoHum Krause 685

cyperaceum Kies 672, 673, 675

deiagoense Poelln 677

dielsii PocWn 708

dinteri Poelln 739

diphylluni Dint 742

divarication sensu Bak 677

divarication Jacq 677, 737, 738

drepanophyl/uni (Bak.) Schltr.

ex Index Kew. 7 735

dubiion Roem. & Schult .... 706

diirion Suess 684

elation Ait 694

elongation''H\\\A. 716, 737

var. flexifoliuin Bak 745

var. liolostachyion Bak 724

elongation sensu Bak 738

elongation sensu Duthie 749

elongation sensu Poelln 740

ensifolium Soelch 759

erraticiim Oberm 740

erythrorrliiziim Conr 757

falcation L.f 735

falcation sensu Bak 734

fasciculatum Bak. 671, 672, 673, 680,

681

filiforine fi Thunb 728

filiforine j Thunb 737

filiforine S Thunb 749

var. reflexipilosion O. Kuntze 726

finibriation Thunb 742

flavoviride Bak 740

flexifoliuin L.f 750

flexifoliuin sensu Jacq 738, 745

friesii Weim 683

galpinii Bak. .670, 671, 672, 676, 679

var. galpinii 676

var. matabelense {Bak.)

Oberm. comb, nov 677

var. norlindii (Weimarck)

Oberm., comb. nov.... 677, 711

gerrardii Bak 757

giffenii Leighton 758

gilvum Krause 740

var. briinneoliim Poelln 740

glabriim Adam% 730

glandiilosum Dint 754

var. inontis-riiscliii Poelln.. 754

gliitinosiim Dint. 1920, 1931.. 739

gracilitepaliim Poelln 724

gramini folium Willd 708

liamaliim Poelln 734

haygarthii {Wood & Evans)

Kies 672, 673, 686, 689, 711

liereroense Schinz 681

var. longibracteatiim Poelln. 681

hirsutiflorion Adams 733

hirsiitum Thunb 730

hispidum L 732

humile Hochst 755

indiitiim Poelln 688

involiicratum Bak 721

jacquinianiim Roem. & Schult. 745

var. affinis Bak. pro parte . . 728

jacquiniainon sensu Bak 737

5

Page

Anthericum jimoiliiWcxk 677

kassneri Poelln 724, 725

kovisnionicmum Dint 706

krauseanum Dint 673, 674, 684

knnthii Bak 755

kyUingioides Krause 684, 685

Icinatiim Dint 744

kmzae Cuf 724, 725

iaxiim N. E. Br 739

liliago L 669, 671, 709

limosum Bak 679

longepedimculatimi Steud. ex

Roem. & Schult 736

longihracteatuni Dint 706

var. brevihracteatuni Dint . . 706

forma suhmembranaceum

Poelln 706

longiciHatum Poelln 728

longifoUinn Jacq 734

var. hiirchellii Bak 728

longifoliiini sensu Bak 742

longifoliimi sensii Duthie 742

longifoliiim sensu Spreng 742

longipedicel/atiini Poelln 690

longistylum Bak. 672, 673, 686, 688,

689

lowryense Bak 729

liinafum Poelln 703

lydenbiirgense Poelln 681, 682

macowanii BzE 756

macranthum Bak 690

inaciilafum Poelln 734

niagnificwn Poelln 696

mahsanum Bak 759

matabelense Bak 677

tnicrantinun Bak 724

monticolnin Poelln 755

inontiiim-dmconis Poelln 757

nndticeps BotWn 690

multisetosiun Bak 670, 671, 675

nniricatuni L.f 742

nataglencoense O. Kuntze .... 755

nataknse Poelln 755

nigrobracteatiini Dint 739

nonscriptum Poelln 690

nudicaide Bak 672, 680

oatesii Bak 725

obtnsifoliuni Poelln 729

odoratissinmm Dint 743, 744

var. brevibracteati/m Dint. 743

oligotrichum Bak 749

omissiun Poelln 759

oocarpiim Schltr. ex Poelln. . . . 745

oravense Engl. & Krause 679

pachyphyUinn Bak 680, 681

pachyrrhizuni Dint 739

pallidiflavuni Engl. & Gilg 740

palustre Adams 730

pappei Bak 750

paradoxinn Roem. & Schult... 732

pascuoruni Poelln 680

patidum Bak 677

paitciflonnn Thunb. a 706

pauciflonini Thunb. /i 708

var. minor Bak 708

pauper Poelln 690

peculiare Dint 733

Page

Anthericum pHosiflorum BotWn 734

var. subpapillosum Poelln . . . 734

pilossisimuni Poelln 733

pilosuin Bak 724, 756

planifolium Thunb. a, c 699

pleiophylhini Schltr. ex Poelln. 708

polyphylluin Bak 670, 686. 688

praeterniissum Poelln 729, 759

pretoriense Bak 682

pseudo falcatum Poelln 728

pubescens Bak 728

pudiciun Bak 728

pulchellum Bak 679

pungens Poelln 684

pyrenicarpum Welw. ex Bak.. 759

quadrifiditm Poelln 690

radula Bak 672, 673, 688

ramosum L 669, 671, 690

I'angei Engl. & Krau.se . .672, 673, 683,

684

rautanenii Schinz 679

recurvatuni Dint 734

recurvifolium Bak 686

rehmannii Bak 677

revolutuniL. . . .610, 715, 737, 738

revolutum sensu Bak 737

revolutum sensu Ker 737

revolutum sensu Thunb. a.... 736

revolutum sensu Thunb. fi.... 737

rzge/r5 Poelln 684

rigidifolium Poelln 683

rigiduni (Kunth) Bak 705

rigidum Krause 684

/7g/<7zz/); de Wild 684

robustum Bak 670, 671, 675

rosenhrockii BotWn 695

rouwenort’i de Gortei 694

rubrovittatum Poelln 671, 675

rudatisii BotWn 679

var. angustum Poelln 679

rustii Poelln 695

sabulosum Adams 753

salteri Leighton 745'^

saltii Bak 722, 724

saundersiae Bak 672, 673. 679

scabromarginatum Schltr. ex

Poelln 742, 743

scahrnmE.^ 751

.vcaftrzz/z? sensu Duthie 737

scariosum Duthie 672. 683

schlechteri Poelln 743

schultesii sensu Bak 705

schultesii sensu Duthie 705

.secundum Krause & Dinter. . . . 725

seineri (Engl. & Krause) Poelln. 690

serpent inum Bak 750

spongiosum Poe 1 1 n 734

squameum L.f 732

stenophyllum Adams 729

stenophyllumBak 756

sternhergianum Roem. &Schult. 699

subcontortum Bak 755

suhlauatum Dint 744

submaciilatum Poelln 729

subpilosum Poelln 732

subulatum Bak 681, 682

stiessengut/iii Soe\ch 684

6

Page

Anthericum tabulare Bak 730

tenellum Welw. ex Bak 685

ienuifoUum Adams 729

thyrsoideum Bak 744

torti folium O. Kuntze 757

tortile Bak 745

transvaalense Bak. 672, 673, 688, 689

trichophlebium Bak 672, 682

triflorum Ait 670, 706

var. minor Bak 680

tubiferum Dint 683

tumidum Poelln 708

imdidatiim Jacq 708

imdulat um Thunb 732

imUaterale Thunh 690

vaginatum Bak 682

validum Poelln 729

variegation Hort 700

vespertinum Jacq 734

vestitum Bak. ex Schinz 688

viscosum Bak 706

vo/kn Sdlch 677

weissianum Dint 735

whytei Bak 672, 683

wilmsii Diels apud Poelln.. 681, 682

wilmsii Diels ex Burtt Davy &

Pott-Leendertz 681

zavatterii Cuf. 759

zebrinum Schltr. apud Poelln . . 746

Anthracnose of Olives in South Africa —

The Identity of the Fungus Causing — by

G. J. M.A.Gorter 769-778

APHYLLOPHORALES 161

APOCYNACEAE 447

ARACEAE 455-6

Arachis hypogaea L.. .218, 234, 284, 501, 524, 528.

533, 538, 542

Arduina bispinosa L 450

Argania spinosa (L.) 323

Aristida

aequiglumis Hack 5

barbicollis Trin. & Rupert 5

bipartita {Nees) Trin 5

gracilior PUg 5

obtusa Del 5

sericans Hack 5

Aristothamnion purpuriferum (Kiitz.) J.

,4,g.-Notes on 201

Arthropodium hispidum (L.f.) Spreng 732

Arthropodium muricatum (L.f.) Spreng 742

Arundinaria 5

ARUNDINEAE 7

Arundinella 3

ARUNDINELLEAE 3

Aseroe rubra La Bill, ex Fries 109

Asphodelopsis Steud. ex Bak 690

Asphodelus capensis L 694

Asterionella Hassall 604

africana Chy 604

Asterostroma MaMce 170

cervieolor (B. & C.)Massee . ... 170

AsterostromeUa Hdhn. & Litsch..l60. 161, 169

duriuscula (B. & Br.) Talbot 160, 169

rumpiana Talbot 160, 169

Astrebla 2

Atherstonea decus.sata Pappe 11

Auricularia tabacina Sowerby 157

Page

Austromimusops {Engl.) A. Meeuse 547

c\inQdi\.2i{Engl.) A. Meeuse 355

dispar (A. E. Br.) A.

Meeuse 352

marginata {N. E. Br.) A.

Meeuse 347, 348, 352

sylvestris {S. Moore) A.

Meeuse 352, 354

AVENEAE 3

Baillonella Pierre 348, 349-350

Bakeriella Dub. pro parte 341

Bakeriella 334

brevipes (Bak.) Dub 334

cinerea (Engl.) Dub 334

Bakerisideroxylon Eng! 341

sapinii (“ Sapini ”)

De Wild 342

stolzii Mildbr 342

BAMBUSEAE 5, 7

BAMBUSOIDEAE 4, 7

Barleria ameliae A. Meeuse 443

argillicola Oberm 444

bolusii Oberm 445

delagoensis Oberm 443, 444

elegans S. Moore 445

mosdenensis Oberm 445

saxatilis Oberm 445

Batterea stevenii (Liboscbitz) Fries 110, 115

Bean, Chlorotic Ringspot Viruses 527

,. , Local Chlorosis Virus — Strain B... 265

,, , Local Chlorosis Viruses 532

,, , Mosaic Virus Stewart and Reddick 521, 524

„ , Necrosis Virus 538

„ , Virus Complex 541

„ , Yellow Mosaic Virus Pierce 218

„ , Yellow Mosaic Virus, Necrotic

Strain 267, 547

Beard, J. S. The Protea Species of the

Summer Rainfall Area of South Africa 41-63

Belmontia and Exochaenium, synonymous

with Sebaea 464

Belmontia E. Mey 464

Belonites E. Mey 453

bispinosa (Thunb.) E. Mey 453

succidenta (Thunb.) E. Mey. . . . 453

Besenna anthelmintica A. Rich 77

Boletus lacrymans Wulfen 152

Bothriochloa 314

Botryobasidium Donk 169

Brachiaria 314

BRACK YPODIEAE 311-313

Brachy podium 313

Breviea Aubr. et Pellegr 332

Briza 2,313

Broad Bean Mosaie Virus A 284

Broad Bean Mosaic Virus B 288

Broad Beans — Virus Diseases of, by Patri-

cia J. Klesser 283-294

BROMINEAE 2

Bromus 313

Buddleia — Sinking the Genus Chiliantlius

under 13

Bulbine Willd 669

canaliculata (Ait.) Spreng 734

(L.f.) Willd 669, 712

ciliata (L.f.) Link 734

crispa (Thunb.) Roem. & Schult. . 695

falcata (L.f.) Roem. & Schult. . . . 735

7

Page

Bulbine frutescens (L.) WiUd 690

scabra (L.f.) Roem. & Schult. ... 751

seineri Engl. & Krause 690

squamea (L.f.) Kunth 732

Bulbinella Ki/nr/i 669

biirkei (Bak.) Benth 722

filiforme (Ait.) Kunth 737

squamea (L.f.) Kunth 732

Bumelia 321, 347

Caesia R. Br 669

brevicaiilis (Bak.) Dur. & Schinz. . 689

comosa (Thunb.) Spreng 699

contorta {L.f.) Dur. & Schinz 689

Calamagrostis 313

CALLITRICHACEAE 22

Callitriche 22

bolusiana Schonl. & Pax nom.

nud 22

bolusii Schonl. et Pax ex Marl. 22

Caloneis Cleve 605

aequatorialis Must.

var. tugelae Chy 605

bacillum {Grun.) Cl 605

chasei Chy 605

incognita Hast 605

silicula (E.) Cl 605

var. peisonis Hast 605

var. truncatula Grun 605

Calvaria inermis (L.) Dubard 324

var. zanzibarensis Pierre

ex Dub 324

diospyroides (Bak.) Dub. . . 324

Cantharellus corniicopioides Linn, ex Fr.. . 117

Carissa L 449

acokanthera Pichon 449

acuminata E. Mey 450

arduina Lam 450

arduina Stapf 451

bispinosa (£.) Desf. ex Brenan. . . . 450

var. acuminata (E. Mev.) L. E.

Codd 451, 452

var. bispinosa 450

cordata Dinter 450

cordata (Mill.) Fourc 450

dinteri Markgf 450

erythrocarpa (Eckl.) A. DC. ...450, 451

ferox (E. Mey.) A. DC 450

haematocarpa (Eckl.) A. DC.... 450

megaphyl !a Gdgr 451

myrtoides Desf 450

oblongifolia Hochst 449, 450

oppositifolium (Lam.) Pichon.... 449

spectabilis (Sond.) Pichon 449

Catabrosa micrantha Hochst. ex A. Rich. . 388

Catha fasciculata Tul 383

CELASTRACEAE 381

Celastrus L.f 383

acuminatus L.f 383

albatus N. E. Br 383

angularis Sond 385

buxifolius L 385

concinnus N. E. Br 385

“ cordata ” Sim 384

cordatus E. Mey. ex Sond 383

crispus Thunb 403

cymosus Soland 385

elliptic us Thunb 385

heterophyllus E.&Z 385

Page

Celastrus ilicinus Burch 383

lanceolatus E. Mey. ex Sond 384

laiirinus Thunb 383

linearis Linn, f 384

lucidus Linn 383

mossambicensis Klotzsch 385

nemorosus E. & Z 385

oleoides Lam 383

peduncularis Sond 384

polyacanthus Sond 384

ruber Harv 385

senegalensis Lam 384

tenuLspinus Sond 384

undatus Thunb 383

Cenchrus 314

Centotheca 3

CENTOTHECEAE 7

CENTOTHECINAE 3

Ceriomyces venulosus (B. & C.) Torrend. . 140

Ceropegia 21

caffrorum Schltr 21

occulta R. A. Dyer 21

Cestrum 449

citrifolium Retz 449

oppositifolium Lam 448

venenatum Thunb 427, 448, 449

Chaetobromus 313

Chilianthus Burch 13

corrugatus 13

Chironia 458

arenaria 459

bachmannii GWg 462

emarginata Jarosz 460

gracilis Salisb. ex Prain 460

var. macrocalyx Prain 460

humilis Gilg 462

var. wilmsii (Gilg) Prain 462

var. zuluensis Prain 462

jasminoides 462

linoides L 459

subsp. emarginata (Jarosz) Ver-

doorn 460

,, linoides 459

,, macrocalyx {Prain) Ver-

doorn 460

,, nana Verdoorn 460

maxima Schoch 460

palustris Burch 460

subsp. palustris 460

,, rosacea {Gilg) Verdoorn 460

„ transvaalensis {Gilg) Ver-

doorn 460

var. foliata (Griseb.) Prain. . . . 460

var. radicata (E. Mey.) Schoch 460

purpurascens {E. Mey.) Benth. &

Hook.f. 461

subsp. humilis {Gilg) Verdoorn 462

,, purpurascens 461

var. impedita Prain 462

var. tysonii (Gilg) Prain.... 461

rosacea Gilg 460

stokoei Verdoorn 458

transvaalensis Gilg 460

tysonii Gilg 462

wilmsii Gilg 462

Chlamydopiis meyenianus {Klotzsch) Llovd 110,

114

CHLORIDEAE 314

Page

Chloris 314

Chlorophytum A'fE Gawler 669, 690

andongense Bak 698

aridum Obenn 691, 696

blepharophyllum Schwemf.

ex Bak 711

bowkeri Bak 691, 700

brachystachyum Bak . . . ,10\, 703

brehmeaminiRocm. &Schult. 707

bwchellii Bak 699

capense (L.) Voss 692, 694, 696, 698

comosum (Thimb.) Jacques 691,

693, 695, 698, 699

crassinerve (Bak.) Obenn. 691, 709

crispum ( Thiinb.) Bak .... 692, 695

delagoense Bak 699

dolomiticum Dint 703

dregei BotWn 711

drepanophyllum Bak 711, 735

diibiwn Roem. & Schult . . 705, 707

durbanense O. Kuntze 698

eto;/w(Ait.)R.Br. exSpreng 694

var. bwchellii Bak 699

gazense Rendle 691, 698

graminifoliuni (WJiUd.) Kunth 708

hay gait hi i Wood & Evans. 689, 71 1

inomatum sensu Bak 707, 711

inornatum Gawl 690, 693

krookianum Zahlbr . .691, 693, 696

laxum R. Br 691, 693, 698

leipoldtii Poelln. (in error

“ leipoldii ”) 711

lewisae Obenn 710

longipedunculatum Forbes . . 696

longitiiberosum Poelln 699

macrosporiim Bak 692, 697

magnificum IVeini 701

marlothii Poelln 707

modestum BaA' 691, 698

monophyllum Obenn. 692, 693, 705

namaquense Poelln 707, 711

nigricans Bak 700

norlindii Weim Gil, 1 1 1

papillosum Rendle.. 69\, 692, 703

pauciphyllum Obenn 71q

perlongibracteatiim Poelln. . 700

petio latum BaA: 698

piqiietbergense Poelln 708

pulchellum Bak 698

rhodesianum Phill 697

rigidum Bak.. 691, 692, 693, 703,

705, 707

schlechterianum Poelln. ... 711

St amine umZcMbr 700

strictinnBaV 700

trachycarpum Obenn . .69\, 692,

693, 701

triflorum (Ait.) Kunth. .. .69), 705

tsuniebensis Dint 703, 706

tuberculatum Duthie 708

undulatLim (Jacq.) Obenn. 691, 708,

710

vaginal uni Bak 711

vaginatum 7/aa 711

vallis-trappii Poelln 699

variegatum Hort 700

viscosum Kunth. . . .691, 693, 706

wilmsii Engl. & Krause. . . . 700

Page

Chrysophy lum L 326

subgenus Donella (Pierre

ex Baill. ) A . Meeuse 327

subgenus Gambeya (Pierre)

A. Meeuse 327

antunesii Engl 335, 339

argyrophyllum Hiern. . . .335, 338

bangweolense R. E. Fries 329

cainito L 327

carvalhoi Engl 335, 337

cerasiferum (Welw.) Hiern 341

disaco Hiern 341

fulvum S. Moore 330

gorungosanum Eng! 329

gossweHeri de Wild 335

holtzii Engl, et Krause .... 340

lanceolatum (Bl.) DC... 327

magalismontanum Sond. 330, 335,

338

natalense Sond 330, 339

pentagonocarpum Engl. &

Krause 329

pruniforme Engl 329

roxburghii Don 327

subnudum Bak 328

tessmannii Engl. & Krause 340

viridifolium Wood &

Franks 328, 330

welwitschii Engl 329

wilmsii Engl... 330, 335, 337

Chrysopogon 314

Cicer arietinum L 209

Cineraria mitellaefolia VHerit 25

Claassen. M. I. A Contribution to Our

Knowledge of the Ereshwater Algae

of the Transvaal Province. Enumera-

tion of the Taxa 559-666

Vegetative Reproduction in Desmids. . . 613

Cladoderris Pers. ex Berk 168

australica Berk, ex Saccardo 120, 121

australis Kalchbr. forma

minima Bres 120

dendritica Bcrs 169

elegans (Jungh.) Fries 120

funalis P. Henn 122, 169

infundibuliformis (Kl.) Fries. .120, 121

spongiosa Fr 120, 121

var. subsessilis Fr 120, 121

thwaitesii Berk. & Broome. . . . 123

thwaitesii sensu Doidge 123

Cladophora — Notes on some Species from

South Africa 189

capensis (Ag.) Kiitz 191, 192, 193.

198

ecklonii (Rud.) Kiitz 191

flagelliformis (Suhr) Kiitz 190, 191

humilis 191

isaacii Simons 193, 194, 668

neesiorum 191

ramosissima 190

virgata (Ag.) Kiitz 190, 194

Closterium Nitzsch 569

abruptum West 569

var. brevius W. & G. S. West 569

var. westiorum Claassen .... 569

acerosum (Schrank) Ehrenb.

var. elongatum Breb 570

var. waterbergense Claassen 570

9

Page

Closterium acutum (Lyngh.) Breb.

var. linea iPertv) f'V. &

G. S. Wesi 570

forma minor Claassen 570

atomicum Claassen 570

barbaricLim Claassen 570

boyanum Claassen 570

caffrorum Claassen 571

calosporum Wittr.

var. minor Claassen 571

ceratium Perty

var. angustum Claassen 571

cornu Elirenh.

var. angustum Claassen 571

var. minor Claassen 571

cymbellaeformis 571

Cynthia De Not 570

var. waterbergense Claassen 571

decorum, Breb 572, 577

var. minor Claassen 572

ehrenbergii 572

var. minutissimum Claassen 572

gracile Breb.

var. brevius Claassen 572

incurvum Breb.

var. elaboratum Claassen.. 572

insolitum C/aas'S’c/; 572

intermedium Ralfs

var. hibernicum ^Vest 573

var. mesianum Claassen 573

jenneri Ralfs 573

var. dubitabilis Claassen. . . . 573

var. robustum G. S. West.. 571

kranskopense Claassen 573

kiitzingii Breb.

var. transvaalense Claassen 573

lagoense Norclst 577

leibleinii Kiitz 574

lunula {Mill.) Nitzsch 575

malinvernianum De Not 573

moniliferum (Bory) Ehrenb... 574

var. epithemioides Claassen 574

parvLilum Niig 570, 574

var. minor Claassen 574

peracerosum Gay

var. elegans G. S. West. . . . 574

praelongum Breb.

forma brevior West 574

pritchardianum Arch 574, 576

pseudoleibleinii Claassen. . . . 574

pseudolunula Claassen 575

ralfsii Breb.

var. glentigianum Claassen 575

var. hybridum Rabenh 575

var. minor C/aa.wc/7 575

var. nodosum Claassen. . . . 575

var. SLibralfsii Claassen.... 575

rostratum Ehrenb 576

schweickerdtii Claassen 576

siliqua W. & G. S. West 577

spetsbergense Borge 576

var. africanum Eritzsch &

Rich 576

var. subafricanum Claassen 576

spinosporum Hodgetts 578

stellenboschense Hodgetts.... 576

Page

Closterium striolatum Ehrenb 570

var. nylstromicum Claassen 576

var. subnylstromicum Claas- 576

sen 576

subdecorum C/aa.ss'e/7 577

sublagoense C/arme/7 577

subsiliqua Claassen 577

truncatum Claassen 577

tumid um Johnson 577

var. angustum Claassen.... 578

Venus Kiitz.

var. inflatum C/aa.wc77 578

warmbadianum Claassen

var. porulosum Claa.ssen.. 578

Codd, L. E. Albizia Species of South

Africa — The 67-80

Codd, L. E. South African Labiatae. . . .429, 434

Codd, L. E. Review of Flora Zambesiaca. . 667

COLEANTHEAE 4

Coleanthus 4, 6, 7

Coleus caninus (Roth) Vatke 433

carnosus Dinter 432

comosus Hochst. ex Giierke 434

neochilus (Schltr.) L. E. Codd.... 432

omahekensis Dinter 434

pent her i Guerke 432

.schinzii Guerke 432

.spicatus Benth 434

Colletotrichum gloeosporioides Penz..llQ. 773,

775, 111, 778

COMPOSITAE 22, 414

Conferva capensis Ag 191

ecklonii Rud 191

virgata Ag 190, 193, 194, 197

Coniodictyum Har. & Pat. emend. G.

Melanfon 137

chevalieri Har. & Pat 137

evansii (P. Magn.) P. Magn. 137

Coniophora DC. ex Pers 165

arida (Fr.) Karst . .81, 104, 138, 165

atrocinera (Karst.) Karst. ... 138

hetulae (Schum.) Karst 138

cerebella Pers 139

fodinarum 165

incrustata Talbot 4, 139

olivacea 139

papillosa 165

pulveridenta (Lev.) Massie. . 139

puteana (Schum. ex Fr.) Karst. 87,

140, 165

puteana sensu Doidge 140

Coniophorella Karst 165

CONVOLVULACEAE 26, 415

Corchorus angolensis Exell & Mendonfa 422, 423

asplenifolius Burch 423, 424

confusus Wild 422

pinnatipartitus Wild 423

serraefolius

var. laucifolius Szyszyl 422

trilocularis L 422, 423

Corticium Pers. ex Fries 143

abeuns Burt 143

aridum Fries 138

armeniacum 172

bombycinum (Sommer f) Bres. . 143

caeruleum 171

cinereum Pers. ex Fries 147

10

Page

Corticium confiuens 172

conigerum Shear & Davidson. . 140

coronilla Hohn. apud Hohn. &

Litsch 159

dregeanum Berk 155

gloeosporum Talbot 147, 171

lac team Fr 143

laetiim (Karst.) Pers 144

litschaueri Burt 144

macrosporum Ell. & Everh. ... 132

moniliforme Talbot 144

nudum Fries 149

pallidum 171

pelliculare Karst 145

porosum Berk. & Curt, apud

Berk. & Br 145, 171

portentosum B. & C 160

praetermissum (Karst.) Bresa-

dola 151

pruinosum Pat 150

pulverulentum (Lev.) Cooke.. 139

punctulatum Cooke. .146, 152, 155, 171

radiosum Fr 144

rimicolum Karst 151

salmonicolor 172

scutellare 172

septentrionale Burt 144

solan! (Prill. & Delacr.) Bourd &

Galz 136, 146

tuberculosum Pat. apud Patouil-

lard & Lagerheim 140

tumulosum 172

vagum B. & C 135, 146

vagum sensu Burt 136

var. solan! Burt ex Rolfs.. 136, 146

Cosmarium Corda 588

barbaricum Claassen 588

binum Claassen 588

botesii Claassen 588

caffrorum Claassen 588

connatum Breb 589

var. subellipticum Claassen 589

var. truncatum IVest 589, 595

contractum Klrchn 589

var. gartanense W. & G. S.

M^est 589

forma jacobsenii (Rov.) W.

& G. S. West ■ 589

var. ellipsoidum (Eh.) West 589

var. pseudogartanense Claa.s-

sen 589

cucumis (Corda) Ralfs 594

debaryi Arch 589

var. minor Claassen 589

decoratum W. & G. S. West. . 590

var. galpinii Claassen 590

var. waterbergense Claassen 590

entochondrum W. & G. S.

West 588

galpinii Claassen 590

granatum Breb 590

var. africanum Frltsch 590

hammeri Relnsch 591

var. minor Claassen 591

holmiense Lund 597

impressulum Elfv 591

istmochondrum Nordst 590

Cosmarium laeve Rabenh 591

var. majus Claassen 591

maximum (Borg.) W. & G. S.

West 591

var. minor West 591

meneghinii Breb 591

nagelianum Breb 592

nanum Claassen 591

norimbergense Relnsch 592

nylstromicum Claassen 592

obliquum Nordst 592

forma minima West 592

obsoletum (Hantzsch) Relnsch 592

var. transvaalense Claassen 592

orthostichum Lund 592

var. compactum W. & G. S.

West 592

var. pseudopumilum Claas-

sen 592

var. pumilum Lund 592

var. transvaalense Claassen 592

pachydermum Lund 593

var. aethiopicum W. & G.

S. West 593

var. waterbergense Claassen 593

praemorsum Breb 588

protractum (Nag.) De Bary. . 593

pseudamoenum Wide 593

pseudoarctiim Nordst 591

pseudoprotractum Claassen . . 593

pseudopyramidatum Lund. . . . 593

subsp. maximum Borgesen. . 594

forma minor Eritsch & Rich 594

pseudotaxichondrum Nordst. . 593

var. atomicum Claassen. . . . 593

var. siamense W. & G. S.

West 593

pyramidatum Breb 594

var. elongatum Claassen. . . . 594

var. majus Claassen 594

quadratum Ralfs 594

var. africanum Eritsch 594

reniforme (Ralfs) Arch 594

schweickerdtii Claassen 594

sexangulare Lund. 595

sexnotatum Gutw 595

var. simplex Claassen 595

var. tristriatum (Liitkem)

Schmidle 595

speciosum Lund 595

var. simplex Nordst 595

forma africanum Claassen.. . 595

subauriculatum W. & G. S.

West 595

subconnatum Claassen 595

subcostatum Nordst 595

var. warmbadianum Claas-

sen 595

subtumidum Nordst 596

var. klebsii (Gutw.) W. & G.

S. West 596

var. theronii Claassen 596

tetragonum (Nag.) Arch 596

var. elegans (Roy. & Biss.)

W. & G. S. West 596

var. transvaalense Claassen 596

II

Page

var. natalensis Fritsch &

Rich

var. pseudonatalensis Claas-

sen

forma irregulare Claassen. .

var. spinosum West

var. subspinosum Claassen . .

transvaalense Claassen

trilobulatum Reinsch

tumidum Lund

waterbergense Claassen

Cotilydia diaphana ISchw.) Lentz

Cowpea, Mosaic Virus A

„ „ Virus B

„ Virus Diseases of, by Patricia

J. Klesser 23

Craterellus cornucopioides (Linn, ex Fr.)

Pers

Craterellus, Cymatoderma (Cladoderris)

508, 524, 528

spectabilis Roth.. 209, 234, 284, 498,

501, 508, 524, 528, 533, 538, 542

Ctenium 314

Cucumber Mosaic Virus Strain 241

Curtisia dentata {Bunn. /.) C. A. Smith. . . . 326

faginea Ait 326

Cutandia 2

Cyclotella Kute 603

ktitzingiana Tim 603

Cylindrocystis Menegh 566

barbarica Claassen 566

brebissonii Menegh 566

caffra Claassen 566

crassa De Bary 566

Cymatoderma 118, 120, 168

elegans Junghun 120, 169

Cymatopleura W. Smith 610

solea (Breb) W. Sm 610

Cymbella Agardh 608

amphicephala Nag 608

var. hercynica (N.S.) Cl.

608

bengalensis Grun 608

608

cistula (Hempr.) Grim.

var. africana Chy.

claasseniae Chy

^acilis Cl..

javanica Hust

kappii Chy 608

kolbei Hust.

microcephala Grun

rniilleri (O.M.) Hust...

naviculiformis Auersw.

nylstroomensis Chy. . .

pretoriensis Chy

pseudoincerta Chy. . . .

raytonensis Chy

var. debegenica Chy.

608

Cynosurus.

spicula Hust 608

spuria Cl

turgida (Greg.) Cl

forma minor Chy

var. pseudogracilis Chy. . .

ventricosa Kg

608

609

1

Danthonia — Leaf Anatomy and Morpho-

logy in South Africa by J. M. J. de

Wet 308-310

Danthonia DC 308, 311, 313, 315

andongensis Rendle 303

arundinacea {Berg.) Schweick . .304,

306, 307, 309

brachyphylla Nees 304, 308

calycina Roem. & Schult . . . .307, 309

chilensis Desv 307

cincta Nees 304, 306, 308

curva Nees 304, 305, 306, 307

disticha Nees. .304, 305, 306, 307, 308

drakensbergensis Schweick. .304, 306,

419, 420

dura Stapf. 304, 306, 308

forskalii {Vahl) R. Br. .303, 304, 307,

308

glauca Nees.... 303, 304, 306, 309

lanata Schrad 304, 306, 308

lechleri Steud 307

lupulina (Thunb.) Roem. &

Schult 304, 306, 308

macowanii Stapf 306, 307, 308

macrantha Schrad 304, 308

malacantha (Steud.) Pilger. . . . 308

montana Doell 308

montevidensisT/ack. & ,4 rec/!. . 308

mossamedensis Rendle. .303, 304, 306,

308, 309

oresigena Phil 307

papposa Nees 304, 305, 308

pumila Nees 304, 306, 309

purpurea Beaiiv 304, 306, 307

range! Pilger. .304, 305, 306, 307, 308

secundifiora Presl 307, 308

stereophylla J. G. Anderson. . 419

stricta (Nees) Schrad . . 304, 306, 308

sufTrutescens Stapf.. 303, 304, 306, 308,

309

tandilensis Kotze 308

tenella Nees 306, 307, 309

zeyheriana Steud 304, 306, 308

var. trichostachya 304, 306

DANTHONIEAE 311, 313

Danthoniopsis 3, 314

12

Page

Dasystachys Bak 691

papillosa (Rendle) Bak 703

Dendrothele Hdhnel & Litsch 170

duthieae 170

Desmids, Vegetative Reproduction in —

Claassen 613

De Wet, J. M. J. Culm Anatomy in Relation

to Taxonomy 31 1-316

De Wet, J. M. J. Leaf Anatomy and Mor-

phology in South African Species of

Danthonia 303

De Wet, J. M. J. Leaf Anatomy of Six South

African Grass Genera 299-301

De Wet, J. M. J. Leaf Anatomy in the Grass

Series Phragmitiformes (Herz.) Avdulov 1-10

De Wet, J. M. J. Chromosome Morphology

in Kniphofia 295-297

De Winter, B. New Genus of Gramineae. 387

De Winter, B. Notes on the Genus Royena

in South Africa 17-19

Notes on Some South African

Species of the Genus Euclea 403

Diandrochloa de Winter 387-390

namaquensis (Nees) de Win-

ter 388

pusilla (Hack.) de Winter. .388-389

Diectomis 314

Digitaria 314

Digitariella de Winter 467

remotigluma de Winter 467

Dilanthes SaUsb 669, 670, 711

revolutnm (h.) SaMsb 737

Dinebra 314

Diospyros L.f 457

acocksii (r/e Winter) de Winter.. 457

angustifolia Aiidib. ex Spach.. 457

austro-africana de Winter 457

var austro-africana 457

var. microphylla (Burch.) de

Winter 457

var. rubriflora de Winter. . . . 457

var. rugosa (E. Mey. ex A.

DC.) de Winter 457

cordata (Hiern) Bakhnizen . . . . 458

galpinii (///e/v;) f/e Winter 457

gXabra. (L.) de Winter 457

glandulifera de Winter 457

glandulosa Lace 457

hirsuta Desf. 457

hirsuta L. f. 457

lycioides Desf. 457

subsp. guerkei (O. Ktze.) de

Winter 458

subsp. lycioides 457

subsp. nitens (Harv. ex Hiern)

de Winter 458

subsp. sericea (Bernh.) de

Winter 457

microphylla Bedd 457

opaca C.B. Cl 458

pallens (Thunh.) E. White 458

ramulosa (E. Mey. ex A. DC.)

de Winter 458

rugosa Sap 457

scabrida (Harv. ex Hiern) de

Winter 458

var. cordata (E. Mey. ex A.

DC.) de Winter 458

Page

Diospyros var. scabrida 458

simii (O. Ktze.) de Winter 458

v\\\o%a(L.) de Winter 458

var. parvifolia (de Winter) de

Winter 458

whyteana (Hiern) E. White 458

Dipholis 347

Diplachne 314

Diploneis Ehrenb 606

ovalis (Hilse) Cl 606

subovalis Cl 606

Dischisma

squarrosum Schltr 33, 35

struthioloides Killick 33-35

Distichlis 2

Dolichos lablab..224, 234, 498, 501, 508 524, 528,

533,538

Donella (Pierre ex Baill.) A. Meeuse 327

Dumoria Chev 348

Duportella Patouillard 166

tristicula (B. & Br.) Reinking. . 157-166

DUTHIEINAE 3

Dyer, R. A. The Opening of the Pretoria

National Botanic Garden 391^01

EBENACEAE 17-19, 403, 457-8

Echinaria 2

Echinochloa 314

Echites L.

bispinosa L.f 452, 453

snccnlenta L..f 452, 453

Echium hispidum 427

spicatum 427, 428

EHRHARTEAE 311. 313

Eichleria Hartog 376

discolor (Sond.) Elartog 377

Eichleriella Bresadola 132

macrospora (Ell. & Everh.)

Martin 132

Elymus 313

Elyonurus 314

Enicostema 462

hyssopifolium ( Willd.) Ver-

doorn 462

//Ytora/e Blume 462

verticillare (Retz) Baill 462

Enneapogon 4, 313

Entoplocamia Stapf 2, 299, 300, 314

Epithemia Breb 609

zebra (£.) Kg 609

ERAGROSTEAE 314, 387

ERAGROSTIFORMES 5

Eragostis //oi't 314

aristata de Winter 468

bergiana 469

diarrhena Stead 388

diplachnoides Stead 388

ciliaris 387

confertiflora Black 387

crassinervis Hack 468

echinochloidea Stapf 477

glandulosipedata de Winter 469

glomerata (Watt) Dewey 388

homomalla Nees 479

horizontalis Peter 473

hygrophila Hubb. & Schweick.. . 479

japonica (Thanb.) Trin 388

kingesii de Winter 470

lamprospicula r/e Winter 471

13

Page

Eragostis membranacea Hack 47

micrantha 479

namaquensis Nees 388

var. rohusta Stapf 388

var. iininoclis Hack 390

obtLisa 477

omahekensis cle WhUev 469, 473

pilosa 479

porosa Nees 469

procumbens Nees 470, 471

pseudo-obtiisa cle Winter 474-477

piisilla H&ck 388, 389

racemosa 473

remotiflora c/e Winter 477

tenella 387

truncata 469

Erica hispida 427, 428

Eriochloa 314

Eriospermum cernuum Bak 690

Euastrum Ehren/} 581

ansatum Rolfs 581

brasiliense Barge

var, theronii Claassen 581

cuneatum Jenner

var. minor Claassen 581

denticulatum (Kirchn.) Gay 581

divaricatLim Land

var. transvaalense Claassen.. 582

divergens Joshua

var. galpinii Claassen 582

var. subbifidum Claassen 582

dubitabilis Claassen 582

elegans {Breh.) Kiitz 582

var. transvaalense Claassen... 582

galpini Claassen 583

inerme ( Ralfs) Lund 582

insulare ( lk77/r.) /^o.v 583

obesum Josh 583

pseudovalidum Claassen 583

rostratum Ralfs 583

subsp. umbonatum W. & C.

S. West 583

var. kranskopense Claassen. . 583

schweickerdtii Claassen 584

spinulosiim Delponte 583

subsp. africanum Nordst . . . . 583

var. transvaalense Claassen.. 584

Euclea coriacea 403, 404

crispa (Thunh.) Sand, ex Guerke . . 403

var. crispa 403

var. ovata (Burch.) de Winter . . . . 403

daphnoides Hiern 404, 405

kellaua Hochst 404

lanceolata E. Mey. ex A. DC 403

inacrophylla E. Mey. ex A. DC. . . . 404

natalensis 403

ovaro Burch 403

racemosa Murr.

var. burchelUi Hiern 404

schimperi (A. DC.) Dandy 404

var. daphnoides (Hiern) de Win-

ter 405

var. schimperi 405

schimperi (A. DC.) Dandy 405

Euglena Ehrenb.

spirogyra Ehrenb 565

var. major Claassen 565

var. suprema Skuja 565

Page

Eimotia Ehrenb 604

alpina (Nag.) Hast 604

exigua (Breb.) Grun 604

garusica Chy 604

lunaris (E.) Grun 604

var. subarcuata (Nag.) Grun.. 604

mogolensis Chy 604

montana Hust 604

pectinalis (Kg.) Rabh 604

var. minor (Kg.) Rabh 604

porcellus Chy 604

praerupta E 604

var. infiata Grun 604

pseudotlexuosa Hu.st 604

SLibaequalis Hust 604

tenella (Grun.) Hust 604

tschirchiana O.M 604

zygodon E 604

var. elongata Hust 604

EUPHORBIACEAE 28

Euphorbia

connnalis R. A. Dyer 28

cussonioides Bally 29, 30, 39

obovalifolia .4. Rich 29

rowlandii R. A. Dyer 28, 38

waterbergensis R. A. Dyer 28

Eustachvs 314

EUTRITICINAE 3

E.xacum hyssopi folium Willd 462, 463

verticiflatuni (L.) Willd .- . 463

Exochaenium Griseb 464

Faurea galpinii 42

Festuca 2, 313

FESTUCEAE 3, 311, 313

FESTUCl FORMES 5, 6

FESTUCINAE 1

FESTUCOIDEAE 1-10

Fingerhuthia Nees 2, 7, 299, 300, 314

Flora Zambesiaca Vol. 1 Part ' 667

Fragilaria Lyngbye

capucina Des/n 604

’■ar. acuta Grun 604

fonticcla Hust 604

ungerian,! Grun 604

Freshwater Algae of the Transvaal Province

— A Contribution to Our Knowledge of

the 559-666

Frustulia Agardh 605

rhomboides (E.) de Toni 605

var. saxonica (Rabh.) de Toni . . 605

vulgaris (Thw.) de Toni 605

var. angusta Chy 605

Fungi — Studies of Woodrotting 83

Fungus causing Anthracnose of Olives in

South Africa 769-778

Gambeya Pierre 327-340

Ganoderma colossum (Er.) Bres....Sl, 101, 104

GENTIANACEAE 458, 464

Gentiana verticillata L.f 462

Geranium ovale 427

GESNER 1 ACE AE 457^65

Gilgiochloa 3

Gladiolus junceus 427

Gloeosporium 769

fructigenum 776-778

forma chromogenum .... 777-778

forma limetticolum 778

forma olivarum 777

14

Page

Glomerella cingulata (Stoneman) Sp. & U.

Schr 770, 773-776

var. brevispora

var. crassispora

Glyceria 26

GLYCERIINAE 2

Glycine javanica Z,. ..224, 234, 498, 501, 508

max (L.) Merr. .209, 234, 284, 498, 501,

508, 528, 533, 538, 542

Gnaphalium spathulatum (“ spatulatum ”) 427

Gomphonema Agardh 609

acuminatum E.

var. turris (£.) Cl 609

augur E 609

clevei Ericke 609

constrictum E 609

gracile E 609

var. lanceolatum (Kg.) Cl 609

lanceolatum E 609

parvulum (Kg.) Gnin 609

var. lagenula (Grim.) Must 609

subtile E 609

Gonatozygon monotaenium De Bary 567

Gorter, G. J. M. A. The Identity of the

Fungus causing Anthracnose of Olives in

GRAMINEAE 467, 480, 417, 422

Grandinia Fries 158

muscicola (Pers.) Bres. apud

Bourd & Galz 158

rosea P. Hemiings 158

Grass Genera — Leaf Anatomy of Six South

African— J. M. J. de Wet 299-301

Gymnosporia acanthophora Loes 386

acuminata Szyszyl 383

albata Sim 383

angiilaris Sim 385

var. grancliflora Davison. . 385

var. orbiciilata Davison . . 385

angustifolia Loes 383

baclunannii Loes 384

botsabelensis Loes 384

biixifolia Szyszyl 385

conciensata Sprague 385

crataegiflora Davison 385

creniilata Engl 386

defiexa Sprague 383

dinteri Loes 384

elliptica SchonX 385

erenioeciisa Loes 385

fascicidata Loes 383

filiformis Oa.v\son 384

heterophylla Loes 385

ilicina Loes 383

integrifolia (L.f.) Glover 386

lanceolata Loes 384

laitrina Szyszyl 383

linearis Loes 384

liicida Loes 383

monococca Loes 383

mossambicensis Loes 385

nemorosa Szyszyl 385

peduncularis Loes 384

peglerae Davison 383

polyacantha Szyszyl 384

procumbens Loes 383

rehmannii Szyszyl 383

rubra Loes 385

Gymnosporia saxatilis (Burch.) Davison . . 386

schlechteri Loes 386

senegalensis Loes 384

tenuispina Szyszyl 384

undata Szyszyl 383

itniflora Davison 385

vaccinifolia Conrath 384

woodii Szyszyl 385

zeyheri Szyszyl 383

Gyrosigma Hassal 605

kiitzingii (Grim.) Cl 605

spenceri ( W. 5'w'.) Cl 605

var. nodifera Grim 605

Hantzschia Grimow 609

amphioxys (E.) Grim 609

var. africana Must 609

forma minuta Chy 609

var. intermedia Grim 609

var. vivax (Hantzscli) Grim. . 609

Harpechloa 314

Hartwegia Nees 690

comosa (Thunb.) Nees 699

Heeria argentea (Thunb.) Meisn 326

Helichrysum 22

albirosulatum Zf////ck 22

album N.E. Br 414, 415

drakensbergense Killick 23

flanaganii Bolus 22

marginatum DC 415

milfordiae Killick 414

Helicosporium aureum (Corda) Linder. ... 110

Helictotrichon 313

Hemarthria 314

Henrardia 3

HENRARDIINAE 3

Heterochaete Patouillard. 132

byliana Talbot 132

grandispora Talbot 133

Heteropogon 314

Hippion hyssopifoliimi (Willd.) Spreng. . . . 462

Hippiomim verticillatiim O. Ktze 463

Holcus 313

HOMOBASIDIOMYCETAE 161

HORDEAE 311, 313

Hyalodema evansii P. Magnus 137

Hydniim bicolor Alb. & Schw. ex Fr 157

muscicola Pers 158

pinastri Fries 154

Hymenochaete Levc///c 147, 155, 166

(Berk.) Massee 155

fasciculata 166

fiilva Burt sensu Doidge 155

fiisco-violascens (Mont.) van der

Byl 155

luteobadia (Fr.) Hdhnel & Litsch 139,

155, 156, 166

nigricans (Lev.) Bresadola ... .156, 166

riibiginosa (Dicks, ex Fr.) Lev.

sensu Doidge 156

semistupposa 166

spadica BqxV. 156

tabacina (Sow. ex Fr.) Lev. sensu

Doidge 157

tenuissima Berk, sensu Doidge. . 157

tristicida (B. & Br.) Massee 157

15

Page

Hyparrhenia 314

Hypochniis Fries 165

cucumeris Frank 134

eylesii van der Byl 146, 155

filamentosus Pat 136

puniceus (A. & S. ex Fr.) Saccar-

do 154

solani Prill. & Delacr 134

HYPOXIDACEAE 480

Imbricaria obovala N. ab E. ex Sond 358

Imperata 314

Inga sericocephala A. Rich 70

Inhambanella (Engl.) Dub 343, 347, 348

hemiquesii (Engl, et Warb.)

Dub 344

natalensis (Schinz) Dub. . . . 351

Ipomoea 27

aquatica Forsk 27

biloba Forsk 27

bisavium A. Meeiise 26

heterosepala Bak 27

Irpex dregeanus (jSerA.) Talbot 155, 159

vellereus B. & Br 159

Ischaemum 314

Ischnurus 2

Jasminum breviflorum Harv. ex Wright-

lectotype of 15

gerrardii 15

Jasminoneriinn bispinoswn (L.) O. Ktze. . . . 450

Jouvea 7

Justicia lorteae RemHe 407

minima A. Meeiise 407

montis-salinarum A. Meeuse 407

odora Vahl 407

orchioides L.f. 407

phillipseae RencUe 407

spergulifoHa sensu C. B. Clarke. . 407

Kaokochloa de Winter 479

nigrirostis de Winter 480

Kel/ana schimperi A. DC 405

Killick, D. J. B. South African Species of

Satureia 435-437

Klesser, Patricia J. The Virus Diseases of

Beans 521-558

Klesser, Patricia J. Virus Diseases of Broad

Beans 283-294

Klesser, Patricia J. Virus Diseases of

Cowpeas 23 3-257

Klesser, P. J. The Virus Diseases of Crota-

laria. Glycine and Medicago Species. .497-519

Klesser, Patricia J. Virus Diseases in Peas

and Sweet Peas 253-282

Kniphofia — Chromosome Morphology in,

by J. M. J. de Wet 295-297

bachmanii Bak 296

baurii Bak 296

caulescens Bak 296

conrathii Bak 296

ensifolia Bak 296

var. albiflora E. A. Bruce. . 296

ichopensis Schinz 296

macowanii Bak 296

multiflora Bak 296

obtusiloba Diels ex Berger. . . 296

porphyrantha Bak 296

praecox Bak 296

rhodesiana Rendle 296

rigidifolia E. A. Bruce 296

Page

Kniphofia rivularis Berger 296

rooperi (Moore) Lam 296

rufa Bak 296

splendida E. A. Bruce 296

tuckii Bak 296

tysonii Bak 296

uvaria (L.) Hook 296

Koeleria 313

LABIATAE — South African, by L. E. Codd 429

Labourdonnaisia Boj. ex parte 376

discolor Sond 376, 377

Lachnopylis HochsX 13, 14

glomerulata C. A. Sm 14

suave ole ns C.A.Sm 14

Laeticorticium Donk 160

polygonioides (Kfl/'j/.) DowA 160

roseum (Pers. ex Er.) Donk 160

Lamarckia 2

Lasiochloa KuntliA, 299, 300, 311, 314, 315

Lathyrus odoratus L. ...209, 234 284, 498, 501

508, 524, 528, 533, 538, 542^

Laxitextum 160

bicolor (Pers. ex Fr.) Lentz. . 160

crassum (Lev.) Lentz 160

Lecomtedoxa (Engl.) Dub 318, 321, 343

henriquesii (Engl, et Warb.)

A. Meeuse 344, 365

klainiana (Pierre ex Engl.)

Dub 343, 365

Leersia 4

LEGUMINOSAE 67

Lentinus sajor-caju Fr....8S, 89, 101, 104

Lenzites palisoti F>- 89, 104

Lepinema verticilata Raf 462

Leptaspis 4

Leptocarydion 314

Lepturus 2

LILIACEAE 295, 481

Limeum aethiopicum 427, 428

Limonium Mill.

amoenum (C. H. Wright) R. A.

Dyer 489

anthericoides (Schltr.) R. A.

Dyer 490

avenaceum (C. H. Wright) R. A.

Dyer 491

corymbulosum (Boiss.) O.K 491

decumbens (Boiss.) O.K. 489

depauperatum R. A. Dyer. . . . 490

dregeanum (Presl.) O.K 488

equisetum (Boiss.) R. A. Dyer. . 489

fergusonae L. Bol 490

foliosum R. A. Dyer 490

linifolium (L.f.) O.K 490

var. linifolium 490

var. maritinum (E. & Z. ex

Boiss.) R. A. Dyer 490

longifolium (Thunb.) R. A. Dyer 490

kraussianum (Bitch, ex Krauss)

O.K 489

membranaceum R. A. Dyer. . . . 490

pedicellatum (Wallr. ex Boiss.)

O.K 489

penicillatum Adamson 491

perigrinum (Bergius) R. A. Dyer 491

var. avenaceum (C. H. Wright)

R. A. Dyer

491

16

Pace

Page

Limonium scabrum (T/iiinb.) O.K.

var. corymbulosum (Boiss.)

R. A. Dyer 491

var. scabrum 491

Lintonia 1

Liriothanmus Schltr 711, 712, 717 720

adamsonii Compton 720

involucratus (Bak.) Schltr... . 721

Lithachne 4

Lobelia volubilis 427, 428

Loewia thomasii (Urb.) Lewis 493, 496

LOCiANlACEAE (in South Africa) 11-14

Lolium 2

LOLllNAE 2

Lophacme 314

Eopharia Kalchbr. & MacOwan 160

Lophatherum 3

Loudetia 3, 314

Eucerne (Alfalfa) Mosaic Viruses 508

Lucerne Mosaic Virus 508

Glycine Strain 508

Necrotic Strain 508

Necrotic Ring Strain 262

Lucinna Molina 332

Lupin Virus A 209

Virus B 212, 547

Virus C 216

Lupins — Virus Diseases of, by Patricia J.

Klesser 207, 232

Lupinus affinis L 208

albus L. .209, 234, 284, 498, 533, 538, 542

var. Bitter 209, 234, 524

var. Sweet 209, 234, 524

angustifolius L 209

mutabilis L.’ . 209,' 234, 284, 498, 501, 508,

524, 528, 533, 538, 542

virus 1 207

Lyciiini co) datum Mill 450

Lygaeum sparteum 4

LYGEAE 6, 7

Lyngbya Ag 565

uliginosa Claassen 565

Mahea Pierre 376

nataleusis (Pierre) Engl 318, 377

MAEVACEAE 483

Manilkara /li/a/w emend. C/Z/y. . 3 18, 321. 349,

350, 365

concolor (Harv. ex Wright)

Gerstner 367, 370

concolor [sphalm. (“ E. Mey.”)

Gerstner] 367, 370, 373

cuneifolia (Bak.) Diibard 365

demiflora Dale ex Bren. &

Green way 369

" densi flora Engl.” 369

densiflora (Er\g\.) H. i Lam.... 369

fischeri (Engl.) Lam 369

kauki (L.) Dub 366

macaulayae (Hutch. & Corb.)

H. J. Lam 365, 370, 373

menyhartii (Engl.) H. J. Earn.. 369

mochisa (Bak.) Dub.. 369, 370, 373

multinervis (Bak.) Dub 365

spiculosa (Hutch. & Corb.) H. J.

Earn 373

umbraculigera (Hutch. & Corb.)

H. J. Lam 373

Manilkara zanzibarensis (Engl.) Dub ...365, 367

Marais, W. An Enumeration of the Mayte-

nus Species of Southern Africa 381-386

Maytenus — An Enumeration of Species of

Southern Africa 381-386

Maytenus H.B.K 381

acuminatus (L. f.) Loes 383

angustifolius Loes 383

bachmannii (Loes.) Marais 384

buxifolius Griseb 385

capitatus (E. Mey. ex Sond.)

Marais 384

cordatus Loes 384

cymosus (Soland.) Exell 385

ellipticus Krug. & Urb. ex Duss. 385

fasciciilatus Loes 383

ilicinus Loes 383

linearis (L.f.) Marais 384

lucidus (L.) Loes 383

monococcus Loes 383

mossambicensis (Klotzsch)

Blake lock

var. mossambicensis 385

var. ruber (Harv.) Blakelock . . 385

nemorosus (E. & Z.) Marais.. 385

oleoides (Lam.) Loes 383

peduncLilaris (Sond.) Loes 384

polyacanthus (Sond.) Marais.. 384

procLimbens (L.f.) Loes 383

rudatisii (Loes.) Loes 386

senegalensis (Lam.) Exell 384

tenuispinus (Sond.) Marais. . . . 384

undatus (Thunb.) Blakelock. . . . 383

zeyheri Loes 383

Medicago lupulina L..209, 234, 284, 498, 501,

508, 524, 528, 533, 538, 542

sativa L 212, 234, 501, 508, 524

Meeuse, A. D. J. Notes on the Sapotaceae

of Southern Africa 317-379

Meeuse, A. D. J. and de Wet, J. M. J. Rut-

tyruspolia, a Natural Intergeneric Hybrid

in Acanthaceae 439^41

Melia 6

Melica 2, 313

MELIACEAE 313

MELICINAE 2

Melilotus officinalis Willd..lD9, 234, 284, 498,

501, 508, 524, 528, 533, 538, 542

Melinis 314

Melosira Agardh 603

granulata (E.) Ralfs 603

nyassensis O.M. 603

roesana Rabh 603

Menodora j uncea 15

Merismopedia Meyen 563

convoluta Breb 563

glauca (Ehrenb.) Nag 563

punctata Meyen 563

Merremia ampelophylla Hallier f. 415

bipinnipartita (Engl.) Hallier f. 415

guerichiana Engl, ex Hallier f. .415-417

guerichii A. Meeuse 415

quercifolia Hallier f. 415

Merulius Haller ex Fries 152

confluens Schweinitz 152

confluens sensu Nel 152

corium (Pers. ex Fr.) Fr 152, 164

17

Page

Merulius fugax sensu Burt. ... ... 153

gelatinosus 165

himantioides 153, 165

lacrymans Wulf. ex Fries 152, 164

laeticolor Berk. & Fries 153

molluscus 153, 164

pinastri (Fr.) Burt 154

pinastri sensu Doidge 154

rufus Pers. ex Fries 154

rufus sensu Doidge 154

SQxp&ns Tode ex Fries 154

je/Tje/w sensu Doidge 154

strigoso-zonata Schw 143

tremellosus Schrad. ex Fries 154

treniellosus sensu Doidge 154

MICRAIREAE 6

Micrasterias Ag 584

americana (Elirenb.) Rolfs . . 584

var. transvaalensis Claas-

sen 584

apiculata (E/ireub.) Menegh 584

conferta Lund 586

crux-melitensis (Ehrenb.)

Hass 584, 614

var. evoluta IF. B. Turner 585

var. transvaalensis Claas-

sen 585

decemdentata (Nag.) Arch . . 585

var. galpinii Claassen.... 585

denticulata Breb 585

var. africana Claassen. . . . 585

var. angustosinuata Gay . . 585

var. subnotata (Vest 585

forma cornuta Claassen. . 585

groenewaldii Claassen 586

mahabuleShwarensis Hobson 586

var. tetracerum W. & G.

S. West 586

var. transvaalensis Claas-

sen 586

var. wallichii (Grun.) W

& G. S. West 586

nylstromica Claassen 586

pinnatifida (Kiitz.) Rolfs.... 587

var. divisa W. West 587

forma major Schmidle. . . . 587

var. transvaalensis Claas-

sen 587

sol (Elirenb.) Kiitz 587

tropica Nordst 587

var. elongata Schmidle . . 587

var. elongatissima Claas-

sen 587

\.xur\c^^^^.(Corda) Breb 587

var. africana Fritsch &

Rich 587

var. minor C/aaMew 588

Microchloa 314

Microlaena 3

Micromelia 435

biflora (Buch. Hamilt. ex D.

DonjBenth 435

grandiflora Killick 435

ovata Benth 436

pilosa Benth 436

purtschelleri Guerke 436

schimperi Vatke 436

Mimosa adianthifolia (Schum.) W. F.

Wight 79

distachya Merit 81

lebbeckL 81

lebbek Forsk 81

MIMUSOPOIDEA H. J. Lam 347

Mimusops Z, 321, 349, 350, 355

sensu Engl 318

subgenus Lecomtedoxa Pierre

ex Engl 343

„ Quaternaria sect. In-

hambella Eng 343

,, Ternaria (A. DC.)

Engl. sect. Euterna-

ria Engl 376

section Muriea Hartog 376

ex parte sensu Wright Phillips. . . 376

altissima Engl 364, 379

angolensis Engl 364

buchananii Engl 364, 379

caflfra E. Mey. ex A. DC. .317, 356, 358

co«ca/ar Harv. ex Wright 364

cuneata Engl 355, 365

ca«c//a//a Bak 365

densiflora Bak 365

discolor (Sond.) Hartog 344, 365

dispar N. E. Br 365

elengi L 355, 365

fi.scheri (Engl.) Engl 365, 369, 372

henriquesiana Sim 365

henriquesii Engl. & Warb. 344, 348, 365

“ henriquezii ” sphalm 344

kaukiE 355

kirkHEak 361

klainiana Pierre ex Engl 343, 365

“ kleiniana ” sphalm. in text. . . 343

macaulayae Hutch. & Corb. . . . 365

marginata N. E. Br 365

menyhartii Engl 365

mochisia E'ak 365

natalensis (Pierre) Engl. 318, 365, 377

natalensis Sdainz 348, 365

natalensis Schinz sensu Wright 365, 377

obovata Sond 358, 361

oleifolia N. E. Br

revoluta Hochst. ex Krauss .... 356, 358

schinzii Engl 365, 377

spiculosa (Hutch. & Corb.) H. J.

Lam 365, 373

sy/i’cs'tra S. Moore 354, 365

“ transvaalensis ” Schinz

(sphalm.) ex Radi, (recte

natalensis) 351, 365

umbraculigera Hutch. & Corb. 365, 375

wehvitschii Engl 365

woodii Engl 358, 361, 365

zanzibarensis Engl 365, 367

zeyheri Sond 361

var. laurifolia Engl 361

Miscanthidium 315

Monanthochloe 3

MONANTHOCHLOINAE 3

Monechma namaense C. B. Clarke 407

Monelytrum 313

MONERMEAE 3

Monocymbium 315

Mosdenia 313

2541564-2

18

Page

Muriea Hartog 376

discolor [Soiul.) Hartog.. 364, 365, 370,

376, 379

Myrsine qiierimbensis Klotsch 323

NARDEAE 6, 7

Narthecium 669

Navicula Bory 606

acidophila Chy 606

anglica Ralfs 606

bryophila Petersen 606

cari E 606

var. angusta Grim 606

cincta (£.) Kg 606

cinctaeformis Hast 606

confervacea Kg 606

cryptocephala Kg 606

var. intermedia Grim 606

var. veneta (Kg.) Grim 606

cuspidata Kg 606

var. ambigua (E.) Cl 606

dicephala (£.) (V. Sin 606

var. neglecta (Krasske) Must . . 606

elaborata Hast 606

exigua (Greg.) O.M 606

exiguiformis Hast 606

gastrum E 606

graciloides A. Mayer 606

gregaria Donk 606

gregarioides Chy 606

grimmei Krasske 606

hungarica Grim 606

var. capitata (£.) Cl 606

krasskei Hast 606

lanceolata (Ag.) Kg 606

longicephala Hast 606

mengeae Chy 606

menisculus Schiini 607

microcephala Grim 607

minima Grim 607

var. atomoides (Grim.) Cl 607

minuscula Grim 607

muralis Grim 607

mutica Kg 607

forma cohnii (Hilse) Hast 607

var. nivalis (E.) Must 607

var. pseudolagerheimii Chy. . . . 607

nyassensis O.M. 607

var. capitata O.M. 607

perpusilla Grim 607

pupulaA^ 607

var. capitata ///«/ 607

var. rectangularis (Greg.) Grim. 607

pygmaea Kg 607

radiosa Kg 607

var. tenella (Breh.) Grim 607

rhynchocephala Kg 607

ruttneri Hast 607

schroeteri Mels ter 607

seminuloides Hast 607

var. sumatrana Must 607

seminulum Grim 607

simplex Krasske 607

var. minor Chy 607

subtilissima C/ 607

suecorum Carlson 607

tenelloides Hast 607

terrestris Petersen 607

tridentula Krasske 607

Page

Navicula viridula Kg 607

var. linearis Must 607

var. rostellata (Kg.) Cl 607

vulpina Kg 607

zanoni Hast 607

Neidium Pfitzer 605

affine (E.) Cl 605

var. amphirhynchus (E.) Cl. . 605

var. longiceps (Greg.) Cl 605

gracile Must 605

forma aequale Must 605

iridis (E.) Cl 605

var. amphigomphus (E.) van

Heiirck 605

var. ampliata (£.) Cl 605

javanicum Must 605

productum (IV. Sni.) Cl 605

Nephelochloa 2

Neriiiin obesuni Eorsk 452

Netrium Nag 566

digitus (Ehrenb.) Itzigs. & Rothe 566

oblongum (de Bary) Liitkeni. . . . 566

var. cylindricum W. & G. S.

West 566

Nicotiana glutinosa £ 215, 509

Nitzschia Hassal 609

accomodata Hast 609

amphibia Grim 609

capitellata Grim 609

claasseniae Chy 609

clausii Haiitzsch 610

confinis Hast 610

debilis (Arnott) Grim 610

denticula Grim 610

diserta Must 610

dissipata (Kg.) Grim 610

fonticola Grim 610

frustulum (Kg.) Grim 610

var. perpusilla (Rabh.) Grim.. 610

goetzeana O.M. 610

kutzingiana Hilse 610

linearis (Ag.) W. Sni 609

palea (Kg.) W. Sni 610

var. tropica Host 610

perminuta Grim 610

pseudobacata Chy 610

purvuloides Chy 610

recta Hantzsch 610

romana Grim 610

sigma (Kg.) W. Sni 610

subadapta Chy 610

subcommunis Hast 610

subvitrea Hust 61o

tarda Hust 610

transvaalensis Chy 610

tropica Hust 610

tryblionella Hantzsch 610

var. levidensis (W. Sni.) Grim 610

von-hauseniae Chy 610

Nostoc Vaucher 563

commune Vaucher 563

gelatinosum Schousboe 564

muscorum Kiitz 563

nylstromicum Claassen 563

pseudogelatinosum Claassen 564

Notes and New Records of African Flower-

ing Plants 407, 443

Notochloa 2

19

Page

Notosceptrum natalense Bak 296

Nuxia Lam 13

Reasons for Again Reducing the

Genus Lachnophylis to — 13-14

congesta Fres 14

glomerulata (C. A. Sm.) Venlooni. . 14

verticillata 14

Obermeyer, A. A. A Revision of the

South African Species of Antheri-

cum, Chlorophytum and Trachy-

andra 669

Odontia Pers. ex S. F. Gray 157

bicolor (Alb. & Schw. ex Fr.)

Bresadola 157

hvinkmaniii Bres 159

Odyssea 314

Oedogonium Link 611

boy anum Claassen 611

crassum Wittr. sec. Him. ... 612

fragile Wittr. sec. Him 612

intermedium Wittr. sec. Him 612

mattiei Claassen 612

mesianum Claassen 612

pseudofragile Claassen 612

subintermedium Claassen. . . 612

subvaucherii Claassen 612

varians Wittr. & Land. sec.

Him 612

vaucherii A. Braun sec. Him 612

zig-zag Cleve sec. Him.... 612

Olea enervis 12

OLEACEAE 15

Olyra 4

OLYREAE 4, 5, 7

OLYROIDEAE 7

Ononis prostata 427, 428

strigosa 427, 428

Onychonema G. C. Wallich 603

laeve Nordst 603

var. micracanthum Nordst 603

sp 603

Oplismenus 314

Orcutia 2

Ornithogalum 482, 690, 711

caudatum Jacq 482

filibracteatiim Oberm 482

leptophyllum Bak 690

longibracteatum Jacq 482

longiscapum RaA' 711

pulchrum Schinz 482

seineri (Engl. & Krause) Oberm 48 1 ,

690

wilmaniae Leight 482

zeyheri Bak 690

Orobanche capensis 427

ORYZEAE 6

Oscillatoria Vaucher 564

annae van Goor 565

nylstromica Claassen 564

princeps Vaucher 565

pristleyi W. & G. S. West 565

simplicissima Gom 565

subpristleyi Claassen 565

tenuis Ag 565

waterbergensis Claassen 565

Osteospermum incanum 427

Othonna linifolia 427, 428

Olive, Anthracnose, Eungi 769-778

Pachypodium bispinosum (L.f.) A. DC 453

succulentum (L.f.) A. DC.. . . 453

Pachy. stela Pierre ex Engl 332

argyrophylla (Hiern) H. Lee. . . . 335

brevipes (Bak.) Bail!, nom. nud. 334,

335

brevipes (Bak.) Engl. ..334, 335, 341

cinerea (Engl.) Pierre ex Engl. . 334

magalismontana (Sond.) H. Lee. 335

msolo (Engl.) Engl 341

PANICIEORMES 5

PANICEAE 314

PANICOIDEAE 5

Panicum 314

hippothrix K. Schum 421

obscurans (Woodr.) Stapf. 420

volutans J. G. Anderson 420

PAPPOPHOREAE 4, 7, 313

PARIANEAE 7

Paspalum 314

Pavonia 487

clathrata Mast 487

commutata Conr 487

schumanniana Giirke 487

var. parviflora Schinz 487

var. transvaalensis Ulbr 487

transvaalensis (Ulbr.) A. Meeuse. . 487

Pea Mosaic Doolittle & Jones 254

Pea Mosaic Virus H. Zaumeyer 221

Pea Stunt Virus Zaumeyer 257

Pea Virus 2 Osborn 256

Pea Wilt-Virus-New Strain 259

Pea Wilt Virus Strain 547

Peas and Sweet Peas-Virus Diseases of, by

Patricia J. Klesser 253-282

Pediastrum Meyen 611

boryanum (Turpin) Menegh . . 611

var. forcipatum 7?aa7) 611

var. granulatum (Kiitz.) Al

Braun 611

var. longicorne Reinscli.... 611

var. minor Claassen 611

duplex Meyw 611

var. reticulatum Lagerheim 611

forma cohaerens Bohlin. ... 611

tetras (Ehrenb.) Ralfs 611

forma evoluta lEcir 611

var. excisum Rabenh 611

Pelliciilaria Cooke sensu Rogers 169

asperula Rogers 137, 169

filamentosa (Pat.) Rogers . .\?>A, 136,

146, 169

fodinarum Talbot & Green . .Vi5, 146

Va23 \ • OC L t »v J-tlHilL / I D J J

146

Peniophora Cooke 147

arenata Talbot 171

aspera (Pers.) Sacc 170

carnosa Burt 147

carnosa sensu Doidge 147

chordalis Hohn. & Litsch 150

cinerea (Pers. ex Er.) Cooke.. . 147

cremea Bres 147

cremea sensu Doidge 147

filamentosa (B. & C.) Burt 170

gigantea (Er.) Massee 171

glebulo.sa Btes 147,148

2541564-3

20

Page

Peniophora gracillima EH. & Everli. ex

Rogers 147, 148, 171

incarnata (Pers. ex Er.) Karst.. 148

incarnata sensu Doidge 148

longispora (Pat.) Hohnet. . . . 149, 171

var. brachyspora Talbot &

Green 148, fig. 26

nuda (Fries) Bresadola 149

papyri na (Mont.) Cooke 149

pelliculosa Talbot 171

pruinata (B. & C.) Burt 150

priiinata sensu Doidge 150

pruinosa (Pat.) Jackson. . 150, 170 fig.

30

rimicola (Karst.) Eldhnel. ... 151, 170

roumeguerii Bres. . . .\A1, 152, 171

soraria G. El. Gunn

subgelatinosa Litsch..\50 fig. 27

tenuis (Pat.) Massee 151, 171

velutina (DC. ex Fr.) Cooke. . 152

velutina sensu Doidge 152

Penium Breb.

barbaricum Claassen 567

cruciferum (de Bary) Wittr 567

cucurbitinum Biss 567, 569

forma botesii Ciaassen 567

forma minor 567

curtum Breb 568

var. waterbergense Ciaassen. . . . 568

jenneri Raifs 568

var. interruptum W. & G. S.

(Vest 568

libellula (Focke) Nordst 568

var. schweickerdtii Ciaassen .... 568

margaritaceum (Eiirenb.) Breb. . . . 568

var. brevior Ciaassen 568

var. incognitum Ciaassen 568

mesianum Ciaassen 569

navicula Breb 569

phymatosporum Nordst 569

pseudorufescens Ciaassen 569

subcucurbitinum Ciaassen 569

Pentameris 313

Pentaschistis 313

Perotis 313

Petalidium linifolium T. Anders 409, 410

lucens Oberin 409, 410

luteo-album A. Meeuse 409

Peziza cornucopioides Linn 117

Phacus Dujardin 566

pleuronectes (O.F.M.) Duj 566

Phaeophiebia strigoso-zonata (Schw.) W. B.

Cooke 143

Phalangium Kunth 669, 670

aethiopicu/n ramosimi, fioribus

albis, petalis re/lexis Comme-

lin 737

bipeduncuiatum (Jacq.) Poir. . 706

canaiicuiatum (Ait.) Poir 734

comosum (Thunb.) Poir 699

eiatum (Ait.) Redoute 694

“ fascicuiatum ” Bak 694

fastigiatuin Poir 694

longifolium (Jacq.) Poir 734

peduncuiatum Bak 706

revolutum (L.) Poir 737

.squameum (L.f.) Poir 732

trifiorum (Ait.) Pers 706

Page

Phaiangium undulatum (Jacq.) Poir 708

vespertinum (Jacq.) Poir 734

viviparum Hort 699

PHALARIDEAE .’3,’ '311, 313

Phalaris ’ 313

PHAREAE 5, 7

Pharus 4’ 5

Phaseolus acutifolius Gray 211

var. latifolius Freem. .209, 2il, 234

284, 498, 501, 508, 522, 524, 528, 533^

538, 542

capensis 427

lunatus L. .209, 234, 284, 498, 501, 509

522, 524, 528, 533, 538, 542

mungo L. .218, 234, 284, 498, 501,509

528, 533, 538, 542

vulgaris L..209, 234, 284, 498, 509

524, 528, 533, 538, 542

var. Black wonder 211

var. Canadian Wonder 207

var. Haricot 21I

var. Idaho Refugee 211

var. Long Tom 2II

var. S. A. Black 211

var. Tendergreen 211

var. Victor 2II

Phiebia strigoso-zonata (Schw.) Lloyd 143

Pholiurus 2

Phragmites 313

PHRAGMITIFORMES 5 7

PHYLLORACHIEAE ! ! ! 4’ 7

Pinnularia Ehrenb 607

acoricola EJust 607

acrosphaeria Breb 607

var. controversa Chy 607

amaniensis Elust 607

borealis E 607

braunii (Grun.) Ci 607

var. amphicephala (A. Mayer)

Hast 607

claasseniae Chy 607

dactylus E 607

divergens W. Sm 607

var. undulata Hmb. & Perag 607

dubitalis Must 607

gibba £■ 608

var. parva (E.) Grun 608

var. sancta Grun 608

forma subundulata Mayer. . 608

graciloides 608

interrupta IV. Sm 608

var. minutissima Nust 608

maior (Kg.) Ci 608

mesolepta (£■.) W. Sm 608

microstauron (E.) Ci 608

forma biundulata O.M. 608

var. brebissonii (Kg.) Must.. 608

ruttneri Hast 608

subcapitata Greg

var. hilseana (Janisch) O.M. . 608

viridis (A/rz^c/i)^ 608

vav.sudet\ca(Hilse) Must 608

Piptochaetium 4

Pisum sativum L..209, 234, 284, 498, 501, 509

522, 524, 528, 533, 538, 542

var. Greenfeast 207

Plagiochloa Adamson & Sprague . .299, 300, 311,

21

Page

Planchonella Pierre 333

Platygloea opalina Talbot 133

Plectrachne 5

Plectranthus caninus Roth 433

neochilus Schltr 432

succulent us Dyer & Bruce. ... 431

thorncroftii S. Moore 430

thunbergii 430

Pleonosporium purpuriferum (Ktitz.) de

Toni 201

Pleuropogon 2, 6, 7

Pleurotaenium Nag 578

basiundatum IV. & G. S.

West 578

cafFrorum C/aassen 578

ehrenbergii (Breb.) de Bary 579

var. waterbergense Claas-

sen 579

var. undulatum Schaar-

sc/im 579

maximum (Reinsch) Lund . . 579

monoliferum W. & G. S.

West 580

ovatum Nordst 579

pseudoehrenbergii Claassen 579

trabeculum (Ehrenb.) Nag 579

var. angustum Claassen 579

var. barbaricum Claassen 580

var. brevis Claassen.... 580

var. rectum (Delp.) W. &

G. S. West 580

trochiscum W. & G. S.

West 580

var. galpinii Claassen. . 580

truncatum (Breb.) Nag. . . . 580

var. granulatum West.. 580

var. mattiei Claassen. . . . 580

westiorum Claassen 580

Plocandra E. Mey 460

albens E. Mey 460

var. radicata E. Mey 460

palustris (Burch.) Griseb 460

var. foliata Griseb 460

purpurascens E. Me,)/ 462

PLUMBAGINACEAE 488-491

Poa 2, 313

Poagrostis 2, 313

Pogonarthria 314

Polygala stipulacea 427, 428

Polyporus 90, 101

arcularius Batsch ex Fries 90, 101

cinnabarinus 91

sanguineus Linn, ex Fries. .91, 102, 104,

hirsutus Wulf. ex Fries.. 9\, 102, 104,

105

pubescens 92

zonatus 92

Pommerculla 3

POMMERCULLINAE 3

POOIDEAE 5

Poria vaillantii (DC.) Fries. .92, 102, 106, 107

Porotheleum (Fr. ex Fr.) Fries 163

incanum 163

Potamophila 4, 313

POUTERIEAE H. J. Lam 347

Page

Pouteria Aubl 321, 332, 341

section Bakersideroxylon (Engl.)

Baehni 341

adolfi-frederici (Engl.) Meeu.se. . 341

argyrophylla (Hiern) H. Lee. . . . 335

brevipes (Bak.) Baehni 333, 331

cerasifera ( Welw.) A. Meeu.se . . . . 344

guyanensis Aubl 332

magalismontana (Sond.) A.

Meeuse 326, 330, 335

msolo (Engl.) A. Meeuse 341

natalensis (Sond.) A. Meeuse. .330, 335,

339

robusta (Mart. & Field.) Eyina. . 341

trident ata Baehni 342

zenkeri A. Meeuse 341

Pretoria National Botanic Garden — the

Opening of. 391-401

Prionanthium 3

Priva

auricoccea A. Meeuse 424

curtisiae Kobuski 424, 425

Protea — The Species of the Summer Rainfall

Area of South Africa 41 -63

Protea R. Br. (nom. cons.)

aftvAWHca Willd 49, 57

angolensis Welw 42

baurii Phillips 50, 51

bella Hauman 42

6o//w/7 Phillips 51, 52

caffra Meisn 49, 50, 52

chionantha Engl. & Gilg 42

comptonii Beard 42, 61

curvata N. E. Br 42, 62

doddii Phillips 55

dracomontana Beard 53 tab. 1, 65

flanaganii PhiWips 50, 51, 55

gaguedi Gniel 41, 42, 57

gaguedi X rhodantha 62

grandiflora Thunb 63

hirta Klotzsch 41, 58

subsp. glabrescens Beard 60

subsp. hirta 59

hirta sensu Phillips 60

hirta X simplex 62

lacticolor Salisb 41, 47, 48

lanuginosa K. Schum 45

madiensis Oliv 42

multibracteata Phillips . .50, 51, 52, 55

mundii Klotzsch 49

natalensis Phillips 50, 51

orientalis Sim 47, 48

parvula Beard 54

pegleri PhWWps 50, 51

rhodantha Hook. f.

var. falcata Beard 53

var. rhodantha 50, 51-53

rouppelliae Mm/; 41, 45-47

rubropilosa Beard 42-60

rudatisii Engl 45

simplex Phillips 51, 55

stipitata Phillips 51

subvestita N. E. Br 41, 48

tenax R. Br 63

transvaalensis Phillips 52, 56

transvaaliensis Gdgr. & Schinz.... 45

trigona Phillips 57

22

Page

PSILURINAE 2

Psilurus 2

Psoralea linearis 427, 428

Puccinella 2

Punctularia T’ar 140

affinis (B. & C.) Talbot 140

atropurpurascens (R. & Br.)

Petch 140

strigoso-zonata (Schw.) Talbot 143,

164

tuberculosa (Pat.) Pat 140, 141

Punctulatum 177

Raddia 4

Rendlia 314

Reseda capensis 427

Resupinate Hymenomycetae-Studies of

some South African 131-171

Reticularia

affinis B. & C 140

atro-nifa B. & C 140

vemilo.sa B. & C 140

RHAMNACEAE 31

Rhapalodia O. Muller 609

gibba (£.) O.M. 609

Rhinacanthus communis Nees 412

gracilis Klotzsch 412

nasutus (L.) Kurz 412

rotundifolius C. B. Cl 412

xerophilus A. Meeuse 410

Rhodohypoxis

baurii (Bak.) Nel 480

var. millioides 480

forma platypetala 480

palustris Killick 480

Rhopalodla O. Muller 609

gibba (£■.) O. M. 609

Rhus thunbergii Elook 326

Rhynchelytrum 314

Richardia melanoleuca Hook, f 456

var. tropicalis N. E. Br 456

Royena L 17

acocksii de Winter 18, 457

angustifolia Willd 457

brachiata E. Mey. ex DC 19

cordata E. Mey. ex DC 17

var. scabrida (Harv. ex Hiern)

de Winter 17

galpinii Hiern 457

glabra L 457

glandulosa Harv. ex Hiern 457

guerkei O.K 17, 457

liirsuta L 18, 457

var. microphylla (Burch.) de

Winter 18, 457

var. rubriflora de Winter 18, 457

lycioides (Desf.) A. DC 17, 457

subsp. sericea (Bernh. ex Krauss)

de Winter 17

subsp. guerkii (O. Ktze.) de

Winter 17, 458

microphylla Burch 18

nitens Elarv. ex Hiern 18, 458

opaca E. Mey. ex A. DC 458

pallens Thmh 19, 458

pubescens 19

ramulosa E. Mey. ex A. DC 458

rugo.sa E. Mey. ex A. DC 457

scabrida Harv. ex Hiern 17, 458

Page

Royena sericea Btrfih. ex Krauss 17, 457

simii O. Ktze 458

villosa L.

var. de Winter 18, 458

whyteana Hiern 458

Ruspolia hypocrateriformis (Vahl) Milne

Redhead

var. australis Milne-Redhead. . . . 439

Ruttya ovata //arv 439

Ruttyruspolia, A Natural Intergeneric

Hybrid in Acanthaceae by A. D. J.

Meeuse and J. M. J. de Wet 439-443

Salvadora 14

angustifolia Turn'll 14

var. australis (Schweick.) Ver-

doorn 15

australis Schweickerdt — New Sta-

tus for 14

persica L 14

SALVADORACEAE 14-15

SAPOTACEAE 317

— Notes, Southern Africa . 3 1 7-379

— Comparative analysis. .319-320

Sapota cerasifera Welw.

Satureia — South African Species of, by

D. J. B. Killick 435^37

Satureia

biflora (Bitch. Hamilt. ex D. Don)

Brig 435

var. rhodesiaca Walther 436

var. villosa Walther 436

compacta Killick 437

grandibracteata Killick 435, 437

grandiflora Scheele 435

pilosa Velen 437

reptans Killick 437

Satyrium cornutum 427-428

Scenedesmus 611

bijugatus (Turpin) Kiitz 611

quadricauda (Turpin) Breb. . 611

Schismus 3

Schizachne 2, 6

Schizophyllum commune Fries 93, 102, 109

Schizostoma laceratum Ehrenb. ex Lev. . 109, 113

Schmidtia 313

Schoenus splcatus 427

SCROPHUL ARI ACEAE 491

Sclerochiton

scissipalor C. B.Cl 446

triacanthus A. Meeuse 446

Scleropoa 2

Scolymocephalus lanuginosus O. Ktze 45

Scytinostroma Donk 160

duriuscula Donk 169

Scytonema Ag 563

myochrous (Dillw.) Ag 563

Scytophyllum

angustifolium Sond 383

laurinum E. & Z 383

Sebacina africana Burt 151, 152

macrospora (Ell. & Everh.) Burt 132

Sebaea 463

acutiloba Schinz 464

burchellii Gilg 464

capitata Cham. & Schlechtdl 463

var. sclerosepala (Schinz) Marais 463

cleistantha R. A. Dyer 464

23

Page

Sebaea compacta Hill 464

confertiflora Schinz 464

corclata

var. intermedia Cham. &

Schlechtdl 464

exacoides 464

fourcadei Marais 463

laxa N.E. Br 464

micrantha Cham. & Schlechtdl.

var. intermedia (Cham. &

Schlechtdl.) Marais 464

pentandra E. Mey.

var. burchellii (Gilg) Marais. . . . 464

schoeidandii Schinz 464

sclerosepala Gilg ex Schinz 463

sedoides Gilg

var. confertiflora (Schinz) Marais 464

var. schoenlandii (Schinz) Marais 464

thomasii 464

zeyheri Schinz

subsp. acutiloba (Schinz) Marais 464

subsp. cleistantha (R. A. Dyer)

Marais 464

Sehima 315

SELAGINACEAE 33

Senecio

brevidentatus M. D. Henderson .... 25

cordifolius L.f. 95

crypt olanatus Killick 24

praeteritus Killick 25

tugelensis Wood et Evans 24

Sersalisia R. Br 332

cerasifera (Welw.) Engl 341

disaco (Hiern) Engl 341

Sesleria 2

SESLERINAE 2

Setaria 314

SIDEROXYLOIDEAE H. J. Lam 347

SIDEROXYLOIDEAE— BUMELIEAE. . . 347

Sideroxylon L 332

adolfi-frederici Engl 341

argenteiim Thunb 326

brevipesBzk 333

cinerenm Lam 323

densiflorum Bak 341

dent at um Burm. f 326

diospyroides Hak. 323

fischeri Engl 369

inerme L....317, 323. 325, 330

var. schlechteri Engl 330

;w/i7/7 S. Moore 335, 338

spinosum L 341

toxiferum Thunb 448

Sideroxylon sensu Bak. (=Vincentella

Pierre) 341

Simons, R. H. Notes on Aristothamnion

purpuriferum (Kiitz.) J. Ag. . .201-205

Simons, R. H. Notes on some Species of

Cladophora from South Africa 189

Sistotrema Fn'es emend. 158, 161, 162

brinkmanni (Bres) J. Erikson 159, 163

muscicola (Pers.) Lnndell. . . . 158, 163

Slevogtia orientaiis Griseb 462

verticillata D. Don nom. illeg. . . . 462

Solan um capsicum 215

Page

Solenia Pers. ex Fr ... t 163

minima 164

natalensis 164

rhoina 164

Sorghum 315

Soybean, Mosaic Virus 506

Sphaerozosma Corda 603

aubertianum West 603

St at ice 488

avenacea C. El. Wright 491

corymhidosa Boiss 491

equisetina

var. depauperata Boiss 490

linifoUa

var. hrachyphylla Boiss 489, 490

var. maritima E. & Z. ex Boiss.. . . 490

longifolia T\\\xx\b 488, 490

pedicellata Boiss 489

perigrina Bergius 488, 490

pnrpurata L 488

var. longifolia Boiss 490

rosea 488

Thunb 491

Stauroneis £7/re«6 606

anceps E 606

forma gracilis (E.) Cl 606

borrichi (Peterson) Lund 606

claasseniae Chy 606

crucicula Grim 606

phcenicenteron E 606

forma gracilis (Dippe!) Hast . . 606

Stenopterobia

intermedia (Lewis) Hast. . . . 610

rautenbachiae Chy 610

Stereum Persoon ex S. F. Gray 166

acerimtm (Pers. ex Fr.) Fr 123

affine Lev 167

australe Lloyd 167, 168

cellum sensu van der Byl 168

bicolor (Pers. ex Fr.) Fr 160, 168

cinerascens (Schw.) Massee 168

co«co/or (Jungh.) Mont 160

cyphelloides B. & C 167

diaphannm (Schw.) Cooke 160, 167

durbanense 167

dtiriuscidinn B. & Br 168

elegans Mey 120

fasciatum (Schw.) Fr 160

friesii sensu van der Byl 168

fulvum (Lev.) Sacc 149, 166, 167

fuscum 123

hirsutum (Willd.) Pers. ex S. F.

Gray 94, 102, 167

involutum (Klotzsch) Fr 167

lobatum (Kunze ex Fr.) Fr 160, 167

hiteohadium Fr 155

murraii (B. & C.) Burt 168

nitidulum Berk 167

ostrta (Blume Sc Nees) Fr 160

papyrinum Mont 149

perlatum Berk 160

pruinatum Berk. Sc Curt 150

purpureum (Pers. ex Fr.) Fries. .94, 102,

107, 168

ravenelii B. Sc C 167

rimosum Berk.

var. africanum Talbot 16

24

Page

Stereum ntbiginosum (Dicks, ex Fr.) Fr. . . 156

sanguinolentum {Alb. & Schw. ex

Fr.) Fr 95

thozetii 120, 167

umbrinum B. & C..147, 149, 150, 168

villosiim Lev 156

Stigonema Ag 563

minutum (Ag.) Hass 563

pseudominutum Claassen 563

Stipa 3, 313, 315

STIPEAE 4, 313

Straurastrum Meyen 598

barbaricum Claassen 598

brevispinum Breb.

var. masoganum Claassen 598

var. obversum W. & G. S.

West 598

caffrorum Claassen 598

connatum Roy. & Biss

var. americanum W. & G.

S. West 599

var. warmbadianum Claas-

sen 599

crenatum Claassen 599

dickiei Ralfs 599

dilatatum Ehrenb 599

ellipticum West 599

excavatum W. & G. S. West 599

furcatum (Ehrenb.) Breb. . . . 599

galpinii Claassen 599

gemelliparum Nordst.

\ar. afnc&num Claassen... 600

leptocladum Nordst.

var. cornutum Wide 600

mattiei Claassen 600

mesianum Claassen 600

orbiculare Ralfs

var. hibernicum W. & G.

S. West 601

var. ralfsii W. & G. S.

West 600

var. waterbergense Claas-

sen 601

pseudogemmulatum 601

var. warmbadianum Claas-

sen 601

pygmaeum Breb.

var. apiculatum W. & G.

S. West 601

var. botesii Claassen 601

quadrangulare Breb.

var. armatum W. & G. S.

West 601

var. subarmatiim Claassen 601

rotula Nordst.

setigerum Cleve

var. pectinatum W. & G.

S. West

forma australe Claassen . . 602

subgemmulatum W. & G.

S. West

var. gracilis W. & G

West 602

var. mattiei Claassen.... 602

subtrifurcatum West

forma major W. & G. S.

West 602

Page

Straurastrum teliferum Ralfs

var. transvaalense Claas-

sen 602

trihedrale Wolle

var. australe Claassen 602

tumidum Breb 602

xiphiophorum Wolle

var. brachyacanthum W.

& G. S. West 602

var. westiorum Claassen. . 602

Streptocarpus

pumilus Burtt 465

rimicola Story 465

vandeleuri 466

STREPTOCHAETEAE 5

STREPTOGYNINAE 2

Strophanthus

gerrardii 5rap/. 454, 455

hirsutus H. Hess 455

luteolus L. E. Codd 454

petersianus Klotzsch 454

Strychnos

atherstonei Harv 11

decussata (Pappe) Gilg — Nomi-

nation of neotype for 11

dysophylla Benth 11, 12

gerrardii N. E. 5r.— Relationship

between S. innocua, S. dyso-

phylla and 11, 12

innocua Z>e/ 11, 12

subsp. burtonii (Bak.) Bruce &

Lewis 12

subsp. dysophylla (Benth.) Ver-

doorn 12

subsp. engleri (Gilg) Bruce &

Lewis 12

subsp. gerrardii (N. E. Br.) Ver-

doorn 12

subsp. innocua 12

Surirella Turpin 610

angusta Kg 610

delicatissima Lewis 610

linearis W. Sm 610

robusta E 610

var. splendida (E.) van Heurck 610

tenera Greg 610

forma minor CAy 610

Sutera cooperi Hiern 491

dentatisepala Overkott 491

pristisepala Hiern 491

Synedra Ehrenb 604

acus Kg 604

var. radians (Kg.) Hust 604

rumpens Kg 604

var. fragilarioides Grun 604

var. meneghiniana Grun 604

var. scotica Grun 604

ulna (Nitzsch) E 604

var. biceps (Kg.) Hust 604

var. danica (Kg.) Grun 604

vaucheriae Kg 604

Synsepalum A. DC 333

Talbot, P. H. B. New and Interesting Re-

cords of South African

Fungi, Part 111 109-111

25

Page

Talbot, P. H. B. The Genera Craterellus,

Cymatoderma (Cladoder-

ris) and Thelephora in

South Africa 117-129

Talbot, P. H. B. Studies of Some South

African Resupinate Hyme-

nomycetes 131-171

Tetmemorus Ralfs

laevis (Kiitz) Ralfs 581

Tetrachne 2, 299, 301, 470

Tetrapogon tenellus 470

Thanatephorus cucumeris (Frank) Dank . . 134

Thecaphora bulbinellae P. H. Talbot Ill

Thelephora Ehrhart ex Fries 123, 165

acerina (Pers.) Pers. ex Fr ... . 123

arida Fries 138

airopiirpurascens B. & Br. . . . 140

biennis Fries 123

biennis sensu Kalchbr 123

fulva Lev 124

fuscoviolascens }Aoni 124, 155

hirsiita (Willd.) Pers. ex Fr. . . . 124

intybacea Pers. ex Fr 124

laciniata Pers. ex Fr 124, 126

nuda Fries 149

pa I inata (Scop.) Ft 125

pedicellata Schw 125

penicillata Lloyd 125, 126, 165

pidvendenta Lev 126, 139

pimicea A. & S. ex Fries. ... 126, 154

sinuansFcrs 126

tabacina So'fiCThy 157

terrestris Ehrh. ex Er 126, 165

THELEPHORACEAE (Pers.) Saccardo . . 161, 162

THELEPFIORACEAE, Species Index of —

Recorded South African

Themeda

triandra

Thorncroftia N. E. Br

longiflora N. E. Br

succulenta (Dyer & Bruce) L.

E. Codd

thorncroftii (S'. Moore) L. E.

Codd

Tieghemella Pierre

Thunberg and Burman — Homonyms in the

Prodromi of, by J. E. Dandy 427, 428

Thymus biflorus Buch. Hamilt. ex D. Don . . 435

Thysanolaena 3

THYSANOLAENEAE 3

TILIACEAE 422

Tilletia ehrhartae Talbot 110

Tomato Spotted Wilt Virus. Samuel et al 270

T omcnTcWz. Patouillard 154, 165

punicea (A. & S. ex Er.) Schroet.

154, 165

Toxicophlaea 449

spectabilis Sond 449

thunbergii Harv 448

Trachyandra Kunth 711

§ Glandulifera. .713, 717, 720, 751

8 Liriothamnus 713, 717, 720

§ Trachyandra 713,717,718

acocksli Oberm 713, 721

adamsonii (Compton) Oberm.l 13 , 720

affinis Kunth 728

arvensis (Schinz) Oberm. 715, 716, 740

172

315

417

429

430

431

430

348

Trachyandra asperata Kunth.. 113, 715, 716, 754

var. asperata 755

var. basutoensis (Poelln.)

Oberm 756

var. carolinensis Oberm.. 755

var. macowanii (Bak.)

Oberm 756

var. nataglencoensis (O.

Kuntze) Oberm 755

var. stenophylla (Bak.)

Oberm 713, 756

var. swaziensis Oberm .... 755

blepharophora (Roem. &

Schult.) Kunth 734

brachypoda (Bak.) Oberm. .714, 729

bracteosa Kunth 734

brehmeana (Roem. & Schult.)

Kunth 707

bulbinifolia (Dint.) Oberm. .714, 743

burkei (Bak.) Oberm 713, 721

canaliculata (Ait.) Kunth. . . . 734

capillata (Poelln.) Oberm. . . . 756

chlamydophylla(5o^.) Oberm. 714,

731

ciliata (L.f.) Kunth. . . .714, 715, 733

corymbosa KunTh 730

dissecta Oberm 714, 751

divaricata (Jacq.) Kunth... .716, 737

elongata (Willd.) Kunth.... 737

esterhuysenae Oberm 714, 726

erythrorrhiza (Conr.) Oberm. 716, 757

falcata (L.f.) Kunth . .Ill, 714, 716,

735

fimbriata (Thunb.) Kunth.. 742

fiexifolia (L. f.) Kunth 750

gerrardii (Bak.) Oberm. 713, 715, 757

giffenii (Leight.) Oberm. . . . 758

glandulosa (Dint.) Oberm. . . . ISA

gracilenta Oberm 726

hirsuta (Thunb.) Kunth 695, 716, 730

hirsutiflora (Adamson) Oberm. 715,

716, 733

hispida (L.) Kunth.. 70S, 712, 732

humilis Kunth 755

involucrata (Bak.) Oberm. Ill, 713,

721

jacquiniana (Roem. & Schult.)

Oberm 713, 715, 745

jacquinii Kunth 745

karrooica Oberm 748

lanala (Dint.) Oberm 744

laxa (N.E. Br.) Oberm ... .139, 740

var. CTxaiica (Oberm.) Oberm. .114,

715, 740

var. laxa 739

longepedunculata (Steud. ex

Roem. & Schult.) Kunth .114, 736

longifolia (Jacq.) Kunth.... 734

malosana (Bak.) Oberm. . . .716, 759

margaretae Oberm 716, 727

muricata (L.f.) Kunth. 716, 738, 742

oligotricha (Bak.) Oberm. . . . 749

paniculata Oberm 746

paradoxa Kunth 732

patens Oberm 748

paucifiora (Thunb.) Kunth... 707

peculiaris (Dint.) Oberm 733

26

Page

Trachyandra pyrenicarpa {,We!w. ex Bak.)

Obenn 716,759

refiexipilosa (O. Kunize)

Obenn 714, 716, 726

revoluta (L.) Kunth. . . .12)1, 738, 740

sabulosa (Adamson) Obenn. 713, 753

saltii (Bak.) Obenn . .114, 715, 716,

722

var. oatesii (Bak.) Obenn. . 725

var. saltii 724

var. secunda (Krause &

Dinter) Obenn 716, 725

scabra (L.f.) Kunth 713, 751

schultesii Kunth 705, 707

tabularis (Bak.) Obenn. . . .714, 730

thyrsoidea (Bak.) Obenn.. 1\5, 744

tortilis (Bak.) Obenn 715, 745

iWwtoa (Thunb.) Kunth. .. . 732

vespertina (Jacq.) Kunth. . . . 734

zebrina (Schltr. ex Poelln.)

Oberm 716, 746

Trachypogon 315

Tragus 31 .3

Trametes im

cingulata Berk 96, 103, 107

meyenii (Klotzsch) Lloyd.. 91, 103, 165

proteus (Berk.) Fr 97, 103, 107

Trechispora brinkinaimi (Bres.) Rogers &

Jcickson 1^^

muscicola (Pers.) Rogers 158

Tricholaena 314

Trichopteryx 3, 314

Trichosporium curtisii Massee 140

Trifolium fragiferum L 221, 234, 528, 533

hybridum L. .209, 234, 284, 498, 501,

509, 524, 528, 533, 538, 542

incarnatum L. .209, 234, 284, 498, 501,^_

509, 522, 524, 528, 533, 538, 542

pratense L..209, 234, 284, 498, 501,

509, 524, 528, 533, 538, 542

repens L. .209, 234, 501, 509, 524, 528,

542, 553

Triodia 2

TRIODINAE 2

Tripogon 314

Triraphis

Tristachya ^*4

TRITICEAE 3

Turnera oculata Story 493

thomasii (Urb.) Story 493

ulmifolia L.

var. thomasii Urb 493

TU RNER ACEAE 493-496

Uniola 2

UNI OLE AE 1

Urelytrum 315

Urginea 259

dimorphantha Bak 483

langii Brem 482

Urochlaena blees 1, 299, 301, 311, 314

Urochloa 314-315

Van der Westhuizen G. C. A. Studies on

Wood-rotting Fungi 83-1 1 5

Vararia Karst 160

VERBENACEAE 424

Verdoorn I. C. Notes on the Loganiaceae,

Salvadoraceae and Oleaceae in South

Africa 11-15

Vida faba L. .209, 234, 284, 498, 501, 509, 522,

524, 528, 533, 538, 542

Vignacatjang 233

cylindrica 233

sesquipedalis (L.) Wight.. 224, 234, 284,

498, 501, 509, 524, 533, 538, 542

unguiculata (L.) Walp . .209, 234, 284, 498,

501, 509, 522, 524, 528, 533, 538, 542

Vincentella Pierre 333, 341

densiflora (Bak.) Pierre 341

longistyla (Bak.) Pierre 341

sapinii (de Wild.) Brenan .... 342

stolzii Mildbr. ex Hutch 342

Virus Diseases of Beans — The, by Patricia

J. Klesser 521-558

Virus Diseases of Crotalaria, Glycine and

Medicago Species — The, by Patricia J.

Klesser 497-519

Virus Diseases of Lupins by P. J. Klesser. 207-233

Vitis capensis 427

Voandzeia subterranea Thouars . .209, 234, 284,

498, 501, 509, 524, 528, 533, 538, 542

Vulpia 2, 313

White Clover Mosaic Virus 501

Xanthidium Ehrenb 598

cristatum Breb 598

var. delpontei Roy & Biss.

forma laevis Claassen. . . . 598

var. transvaalense Claassen 598

Xantholis 333

Zantedeschia

jucunda C. Petty 455

melanoleuca Hook. f.

var. tropicalis N. E. Br. . . . 456

pentlandii ( Whyte ex Watson)

tropicalis (N. E. Br.) C. Petty 456

Zeugitis 3

Zeyherella (Pierre ex Baill.) nom. nud 332

section Zeyherella sensu

Wright pro parte 326

Zizania 4

Ziziphus Mill 33

jujuba Mill 33

rivularis L. E. Codd 31-33

ZOISEAE 313

Zygia sensu E. Mey 68

fastigiata E. Mey 80

. '^v t- ■