Bothalia

N TYDSKRIF VIR PLANTKUNDIGE NAVORSING

A JOURNAL OF BOTANICAL RESEARCH

Vol. 16,1 May/Mei 1986

PUBLICATIONS OF THE BOTANICAL RESEARCH INSTITUTE

PUBLIKASIES VAN DIE NAVORSINGSINSTITUUT VIR PLANTKUNDE

Obtainable from the Division of Agricultural Information, De-

partment of Agriculture and Water Supply, Private Bag X144,

Pretoria 0001, Republic of South Africa. A current price list of

all available publications will be issued on request.

Verkrygbaar van die Afdeling Landbou-inligting, Departement

van Landbou en Watervoorsiening, Privaatsak X144, Pretoria

0001, Republiek van Suid-Afrika. ’n Geldige lys van alle beskik-

bare publikasies kan aangevra word.

BOTH ALIA

Bothalia is named in honour of General Louis Botha, first Premier

and Minister of Agriculture of the Union of South Africa. This

house journal of the Botanical Research Institute is devoted to

the furtherance of botanical science. The main fields covered

are taxonomy, ecology, anatomy and cytology. One or two parts

of the journal are published annually.

Bothalia isvernoem ter ere van Generaal Louis Botha, eerste Eerste

Minister en Minister van Landbou van die Unie van Suid-Afrika.

Hierdie lyfblad van die Navorsingsinstituut vir Plantkunde is gewy

aan die bevordering van die wetenskap van plantkunde. Die hoof-

gebiede wat gedek word, is taksonomie, ekologie, anatomie en

sitologie. Eep of twee dele van die tydskrif verskyn jaarliks.

MEMOIRS OF THE BOTANICAL SURVEY OF SOUTH AFRICA

MEMOIRS VAN DIE BOTANIESE OPNAME VAN SUID-AFRIKA

The memoirs are individual treatises usually of an ecological ’n Reeks van losstaande omvattende verhandelings oorvernaam-

nature, but sometimes dealing with taxonomy or economic lik ekologiese, maar soms ook taksonomiese of plantekonomiese

botany. onderwerpe.

THE FLOWERING PLANTS OF AFRICA / DIE BLOMPLANTE VAN AFRIKA

This serial presents colour plates of African plants with ac-

companying text. The plates are prepared mainly by the artists

at the Botanical Research Institute. Many well-known botanical

artists have contributed to the series, such as Cythna Letty (over

700 plates), Kathleen Lansdell, Stella Gower, Betty Connell,

Peter Bally and Fay Anderson. The Editor is pleased to receive

living plants of general interest or of economic value for illustra-

tion.

Two parts of ten plates each are published annually. A volume

consists of four parts. The publication is available in English and

Afrikaans.

Hierdie reeks bied kleurplate van Afrikaanse plante met bygaan-

de teks. Die skilderye word meestal deur die kunstenaars van die

Navorsingsinstituut vir Plantkunde voorberei. Tallebekende bota-

niese kunstenaars het tot die reeks bygedra, soos Cythna Letty

(meer as 700 plate), Kathleen Lansdell, Stella Gower, Betty Con-

nell, Peter Bally en Fay Anderson. Die Redakteur verwelkom

lewende plante van algemene belang of ekonomiese waarde vir

afbeelding.

Twee dele, elk met 10 plate, word jaarliks aangebied. ’n Volume

bestaan uit vier dele. Die publikasie is beskikbaar in Afrikaans en

Engels.

FLORA OF SOUTHERN AFRICA / FLORA VAN SUIDELIKE AFRIKA

A taxonomic treatise on the flora of the Republic of South Africa,

Ciskei, Transkei, Lesotho, Swaziland, Bophuthatswana, South

West Africa/Namibia, Botswana and Venda. Die FSA contains

descriptions of families, genera, species, infraspecific taxa, keys

to genera and species, synonymy, literature and limited specimen

citations, as well as taxonomic and ecological notes. Also available

in the FSA series are the following:

’n Taksonomiese verhandeling oor die flora van die Republiek

van Suid-Afrika, Ciskei, Transkei, Lesotho, Swaziland, Bophu-

thatswana, SWA/Namibia, Botswana en Venda. Die FSA bevat

beskry wings van families, genusse, spesies, infraspesifieke taksons,

sleutels tot genusse en spesies, sinonimie, literatuur, verwysings

na enkele eksemplare, asook beknopte taksonomiese en ekologiese

aantekeninge. Ook beskikbaar in die FSA- reeks is die volgende:

The genera of Southern African flowering plants by/deur R.A. Dyer, Vol. 1 Dicotyledons (1975); Vol. 2

Monocotyledons (1976).

Keys to families and index to the genera of Southern African flowering plants by/deur R.A Dyer (1977).

Plant exploration of Southern Africa by Mary Gunn & L.E. Codd. Obtainable from/Beskikbaar van:

A. A. Balkema, Box/Bus 3 117, Cape Town/Kaapstad 8000, RSA.

PALAEOFLORA OF SOUTHERN AFRICA / PALAEOFLORA VAN SUIDELIKE AFRIKA

A palaeoflora on a pattern comparable to that of the Flora of

Southern Africa. Much of the information is presented in the

form of tables and photographic plates depicting fossil popula-

tions. Now available:

’n Palaeoflora met ’n uitleg vergelykbaar met die van die Flora

van Suidelike Afrika. Baie van die inligting word aangebied in

die vorm van tabelle en fotografiese plate waarop fossiele popula-

sies afgebeeld word. Reeds beskikbaar:

Molteno Formation (Triassic) Vol. 1 Introduction. Dicroidium by/deur J.M. & H.M. Anderson.

Prodromus of South African Megafloras. Devonian to Lower Cretaceous by/deur J.M. & H.M. Anderson.

Obtainable from/Beskikbaar van: A.A. Balkema, Box/Bus 3117, Cape Town/Kaapstad 8000, RSA.

Republic of

South Africa

Republiek van

Suid-Afrika

BOTHALIA

’N TYDSKRIF VIR PLANTKUNDIGE NAVORSING

A JOURNAL OF BOTANICAL RESEARCH

Volume 16,1

Editor/Redakteur: O. A. Leistner

Editorial Board/Redaksieraad

D. F. Cutler

B. de Winter

D. J. B. Killick

O. A. Leistner

P. H. Raven

J. P. Rourke

M. J. Werger

Royal Botanic Gardens, Kew, UK

Botanical Research Institute, Pretoria, RSA

Botanical Research Institute, Pretoria, RSA

Botanical Research Institute, Pretoria, RSA

Missouri Botanical Garden, St Louis, USA

National Botanic Gardens, Kirstenbosch, RSA

University of Utrecht, Utrecht, Netherlands

ISSN 0006 8241

Published by the Botanical Research Institute, Department of Agriculture and Water Supply, Private Bag X101,

Pretoria 0001, South Africa

Uitgegee deur die Navorsingsinstituut vir Plantkunde, Departement van Landbou en Watervoorsiening,

Privaatsak X101, Pretoria 0001, Suid-Afrika

1986

Digitized by the Internet Archive

in 2016

https://archive.org/details/bothaliavolume1616unse

CONTENTS — INHOUD

Volume 16,1

1. Revision of the genus Dombeya (Sterculiaceae) in southern Africa. I. C. VERDOORN and

P. P. J. HERMAN 1

2. The Eriosema squarrosum complex (Papilionoideae, Fabaceae) in southern Africa.

C. H. STIRTON 11

3. Studies in the genus Riccia (Marchantiales) from southern Africa. 2. A new species of the

section Pilifer. R. sarcosa. O. H. VOLK and S. M. PEROLD 23

4. Studies in the genus Riccia (Marchantiales) from southern Africa. 3. R. schelpei, a new

species, in the new subgenus Chartacea. O. H. VOLK and S. M. PEROLD 29

5. The identity of Erica vinacea and notes on hybridization in Erica. E. G. H. OLIVER 35

6. Notes on African plants:

Acanthaceae. Notes on southern African species of Justicia. K. L. IMMELMAN 39

Adiantaceae. Cheilanthes deltoidea Kunze in the Waterberg, Transvaal. W. B. G. JACOBSEN

and N. H. G. JACOBSEN 41

Bryaceae (Musci). A new species of Anomobryum. J. VAN ROOY 42

Combretaceae. A new species of Combretum from the Transvaal. E. RETIEF 44

Cyperaceae. Cyperaceae new to the flora of Natal. C. REID 45

Ericaceae. A new species of Ericinella from the southern Drakensberg. E. G. H. OLIVER 46

Fabaceae. A fourth natural Erythrina hybrid from South Africa. E. F. HENNESSY 48

Iridaceae. The correct citation of Montbretia crocosmiiflora. D. O. WIJNANDS 51

Lamiaceae. A new species of Stachys. L. E. CODD 51

Lamiaceae. A new species of Thorncroftia. L. E. CODD 52

Liliaceae. A new method for the non-destructive examination of leaves of Aloe species by

SEM. H. F. GLEN and D. S. HARDY 53

Mesembryanthemaceae. A new combination of Lampranthus. H. F. GLEN 55

Orchidaceae. Notes on the Disinae for the Flora of southern Africa. H. P. LINDER 56

Pertusariaceae (Lichenes). A new species of Pertusaria. F. BRUSSE 57

Poaceae. Diverse notes on southern African Pooids. H. P. LINDER 59

Porinaceae (Lichenes). A new species of Porina on limestone. F. BRUSSE 62

Ricciaceae. Pteroriccia Schust., should it be upheld? S. M. PEROLD 64

7. The shape and ultrastructure of the caryopsis of Pentameris and Pseudopentameris species

(Arundinoideae, Poaceae). N. P. BARKER 65

8. William Keit and the Durban Botanic Garden. D. P. McCRACKEN 71

9. The plant ecology of the farm Groothoek, Thabazimbi District. III. An annotated checklist.

R. H. WESTFALL, N. VAN ROOYEN and G. K. THERON 77

10. A checklist of Pteridophyta of the north-eastern Orange Free State. J. P. ROUX 83

11. Miscellaneous notes:

Chromosome studies on African plants. 1. J. J. SPIES and H. DU PLESSIS 87

A brief note on TCD and DBN and the herbaria of Sonder, Hooker and Harvey.

H. P. LINDER 88

Baseline data for the vegetation of two protected plots at the Matimba Station, Ellisras,

NW Transvaal. M. D. PANAGOS, R. H. WESTFALL and J. C. SCHEEPERS 89

12. Obituary: Cythna Lindenberg Letty (1895-1985). L. E. CODD 93

13. E.A.C.L.E. (Ted) Schelpe (1924-1985) — a biography. E. G. H. OLIVER 97

14. New,taxa, new records and name changes for southern African plants. STAFF OF THE

NATIONAL HERBARIUM 109

15. Review of the work of the Botanical Research Institute 1984/1985 119

16. Book review 135

17. A new guide for authors to Bothalia. O. A. LEISTNER 137

■

.

Bothalia 16,1: 1-9 (1986)

Revision of the genus Dombeya (Sterculiaceae) in southern Africa

I. C. VERDOORN* and P. P. J. HERMAN*

Keywords: descriptions, Dombeya , key, revision, southern Africa

ABSTRACT

The species of Dombeya Cav. occurring in southern Africa are revised and a key is provided to the eight species

present in the region. A new record for southern Africa, D. quinqueseta (Del.) Exell, is included.

INTRODUCTION

This revision is a contribution towards a treatment

of the family Sterculiaceae for the Flora of southern

Africa. The format followed is that previously used

in the FSA, which differs in several respects from the

current one and from that generally employed in Bo-

thalia. Instead of recasting the work and so delaying

publication of a much-needed revision of a well

known genus of trees, it was decided to present the

account in its original form.

Dombeya Cav., Diss. Bot. 2, App. 2 (1786); op.

cit. 3: 121 (1787), nom. cons.; Harv. in FC 1: 220

(1860); Benth. & Hook, f., Gen. PI. 1: 221 (1862);

Arenes in FI. Madag. 131, Sterculiaceae: 189 (1959);

Wild in FZ 1: 518 (1961); M. Friedrich-Holzhammer

et al. in FSWA 84: 2 (1969); Dyer in Gen. 1: 364

(1975). Lectotype species: D. palmata Cav.

Assonia Cav., Diss. Bot. 2, App. 2 (1786).

Xeropetalum Delile, Cent. PI. Call. 4: 377 (1826).

Trees or, more often, lax to bushy shrubs, some-

times straggling; bark usually stringy; indumentum

mainly of stellate or tufted, uni- or multicellular

hairs, often lepidote and with simple, thick-walled or

glandular hairs as well. Leaves simple, petioled,

stipulate, usually palmately nerved (in African

species). Bracts 3, caducous, pubescent on both sur-

faces, close to the calyx or scattered, usually free.

Calyx 5-lobed, pubescent on outside, tube short,

usually with 5 patches of papillae just below the base

of the lobes within; lobes reflexed in mature flowers.

Petals 5, in African species usually obliquely obo-

vate, cuneate (butterfly-wing-shaped), white or in

varying shades of rose, sometimes flecked or veined

with purplish red or deep rose at the base, persis-

tent, turning cinnamon-rufous with age. Stamens

10-40, usually 15, united at the base into a short

tube; filaments of different lengths in groups of 3-8

(occasionally 2 or 4), alternating with 5 narrowly lin-

ear-spathulate staminodes; anthers oblong, the cells

parallel, opening by slits. Ovary 3-5-celled, sutures

often partly bristly pubescent inside; style 3-5-

branched; ovules 2-8 in a cell. Capsule subglobose,

loculicidally dehiscent. Seeds globose to obovate-ob-

long, triquetrous; testa hard, rough with minute

ridges, dots or pits.

Native of Africa, Madagascar and the Mascarene

Islands. In South Africa eight species and one var-

iety are recognized.

For the conservation of the name and for the type

species see the International Code of Botanical No-

menclature 1983, p.380. The genus was named in

honour of J. Dombey, a French botanist who travel-

led in Chile and Peru in the late eighteenth century.

KEY TO SPECIES

la Petals over 1 cm long: ovary usually 5-celled; style branches usually 5:

2a Pubescence on branchlets, petioles and peduncles usually distinctly stellate-pubescent, sometimes

hairs spreading but then mostly under 1 mm long, not glandular; leaves 3-10 cm long, 2-3 cm broad

1. D. tiliacea

2b Pubescence on branchlets, petioles and peduncles of long, spreading hairs, glandular and/or non-

glandular and acute, mostly over 1 mm long; leaves large, 6-23 cm long, 5-19 cm broad:

3a Leaves distinctly discolourous, lower surface metallic-grey with a short dense indumentum of fine

hairs giving it a felted appearance; pubescence often of predominantly glandular, light brown,

spreading hairs 2. D. pulchra

3b Leaves not obviously discolourous or if somewhat so the indumentum on the lower surface sparsely

stellate to densely stellate-tomentose; pubescence usually mainly of non-glandular, acute,

spreading hairs 3. D. burgessiae

lb Petals under or up to 1 cm long; ovary usually 3-celled; style branches usually 3:

4a Leaves ovate to broadly ovate, gradually narrowing in the upper third or obscurely 3-lobed, the

central lobe the largest; lobes obtuse or the central one acuminate:

5a Leaves rather thin in texture, acuminate at the apex, reticulate veins on the lower surface not very

conspicuous and prominently raised:

* Botanical Research Institute, Department of Agriculture and

Water Supply, Private Bag X101, Pretoria 0001, RSA.

2

Bothalia 16,1 (1986)

6a Leaves mostly under 7x6 cm; pubescence on the peduncles stellate with short (± 1 mm)

hairs 4. D. cymosa

6b Leaves variable in size up to 11 x 10,5 cm; pubescence on pedicels and peduncles mostly stel-

late-tomentose intermixed with long (up to 1,5 mm), thin, acute spreading hairs 5. D. kirkii

5b Leaves thick in texture, the upper third not acuminate into a narrow apical portion, leaves some-

times appearing shallowly 3-lobed, lobes rounded, reticulate veins on lower surface conspicuous

and prominently raised 6. D. quinqueseta

4b Leaves suborbicular to broadly ovate, broadly rounded at the apex, sometimes acute but never acumi-

nate in the upper third:

7a Trees or shrubs with very rough bark; in shrubby form rough bark only on the lower portion of the

plant; usually flowering in early spring on more or less leafless branches except on the central

plateau of South West Africa/Namibia where they flower at any time from spring to autumn

depending on rains:

8a Ovary tomentellous with stellately pubescent scales as well as setose with tufted bristle-

like suberect hairs 7a. D. rotundifolia var. rotundifolia

8b Ovary tomentose with dense, stellately pubescent scales, the hairs short, not setose; tall shrubs

with a few long virgate stems; leaves velvety stellate-tomentose; restricted to the Naukluft area

of South West Africa/Namibia, near permanent water 7b. D. rotundifolia var. velutina

7b Shrubs with several to many slender stems, rarely a small tree; bark never rough; usually flowering

in autumn; ovary stellate-tomentose with short hairs (not setose as well); restricted to mesophytic

areas in the mountainous north-eastern Transvaal 8. D. autumnalis

1. Dombeya tiliacea (Endl.) Planch, in FI. des

Serres, ser. 1, 6: 225 (1850-51). Type: Cape, Drege

s.n. (P!).

Xeropetalum tiliaceum Endl. & Fenzl, Nov. Stirp. Decad. 43

(1839). Leeuwenhoekia tiliacea E. Mey. ms. Dombeya dregeana

Sond. in Linnaea 23: 18 (1850); Harv., Thes. Cap. 1: 56, t. 89

(1859); FC 1: 221 (1860); K. Schum. in Engl., Monogr. Afr. Pfl.

5: 30 (1900); Sim, For. FI. Cape Col. 45 (1907).

D. natalensis Sond. in Linnaea 23: 17 (1850); Harv. in FC 1: 221

(1860); Wood, Natal PI. 73, pi. 90 (1899); K. Schum. in Engl.,

Monogr. Afr. Pfl. 5: 29 (1900); Sim, For. FI. Cape Col. 146

(1907). Type: Port Natal, Gueinzius 105 (SAM, iso.).

D. gracilis K. Schum. in Engl., Monogr. Afr. Pfl. 5: 30 (1900).

Type: Natal, Umvoti, MacOwan & Bolus in Herb. Norm. 562

(BOL).

D. elegans K. Schum. in Engl., Monogr. Afr. Pfl. 5: 31 (1900),

nom. illegit. , pro parte as to Flanagan 1409 (STE), non Cordem.

(1895).

Shrub, sometimes straggling, 2-3 m high or tree

up to 10 m tall. Branchlets slender, longitudinally

ridged, with prominent leaf-scars and with small,

pale, raised lenticels, glabrescent but new growth

laxly to densely stellate-pubescent with patent or ap-

pressed hairs rarely up to 1 mm long. Stipules early

caducous, linear-subulate from a rather broad base,

(rarely deltoid to narrowly deltoid-long-acuminate),

sparsely to densely stellate-pubescent, glabrescent,

sometimes subpetaloid. Leaves often thin in texture,

ovate, gradually acuminate, sometimes 3-lobed at

the apex, the 2 lateral lobes shallow, 3-10 cm long,

2-8 cm broad, shallowly to deeply cordate, rarely

rounded or subtruncate at the base, margins crenate

to crenate-dentate, the lateral veins running to the

point of the tooth but not clearly excurrent, nerves

from the base 3-7 but usually 5, upper and lower sur-

faces thinly to densely stellate-pubescent, often gla-

brescent, hairs sometimes longer and some simple

on the lower surface; petiole 2-6,5 cm long, usually

subspreading, stellate-pubescent and with simple

spreading hairs, usually under 1 mm long, glabres-

cent. Inflorescence cymose to subumbellate, usually

appearing in late summer or autumn in the axils of

the upper leaves, 2-9-flowered; peduncle usually

straight, 1,5-8 cm long, stellate-pubescent and

sometimes with simple hairs as well, hairs usually

under 1 mm long; pedicels 1,5-3 cm long, pubescent

like the peduncle. Buds ovate or spindle-shaped.

Bracts 3, free, early caducous, variable in size and

shape from broadly ovate to narrowly linear, 5,5-10

mm long, 1-4 mm broad, gradually and shortly acu-

minate at the apex or abruptly acuminate into a cau-

date apical third, margins sometimes with an odd

lobe or two, shortly stellate-pubescent on both sur-

faces, the broad bracts usually close to the rounded

base of the calyx and the narrower ones at the base

of the stipe. Calyx lobed almost to the base, united

portion 2,5 mm long, rounded at the base or with a

short or long stipe, up to 4 mm long; lobes usually

reflexed, narrowly oblong-acuminate, about 9-15

mm long, 2,5-4 mm broad, pubescent dorsally. Pe-

tals persistent, white (rarely pink) turning rust-

coloured with age, oblique, cuneate, about 12-17

mm long, 11-16 mm broad at the apex. Stamens

united at the base into a tube 2-4,5 mm long; fila-

ments of different lengths, the longest about 4 mm

long; anthers 2-3,5 mm long; staminodes about 11

mm long. Ovary tomentose and with a few project-

ing hairs in the apical portion, but these hairs usually

under 1 mm long, 5-celled but sometimes imper-

fectly so; ovules 2-4 per cell; style 8-10 mm long,

3-5-branched at the apex, glabrous or stellate-pu-

bescent in lower half. Capsule about 7 mm long, stel-

late-tomentose, hairy on the sutures within; seed 3-

sided and broadly rounded at the top, about 5x7

mm, rough, with minute ridges and dots.

Found in bush, mixed scrub, coastal short forest

or subtropical scrub forest. Recorded from the east-

ern Cape and Natal.

CAPE. — Albany: ‘Kleinemund’, MacOwan 568 (BOL). Bat-

hurst: Port Alfred, Galpin 5; Britten 1427. Butterworth: Acocks

12539. East London: Comins 1512', Rogers 17017; Smith 3745.

Kentani: Pegler359. Komga: Flanagan 56. Peddi e: Acocks 12779.

NATAL. — Without precise locality, Cooper 1105. Durban:

Medley Wood 10973; ‘Port Natal’, Ecklon & Zeyher s.n. (SAM);

Gueinzius s.n. (SAM). Estcourt: Entumeni, Medley Wood 3745

(NH); Hlatikulu Forest, Killick 1956. Hlabisa: St Lucia, Ward

3127. Ixopo: Taylor 2098. Kranskop: Jameson’s Drift, Dyer 4370.

Lions River: Howick, Schlechter 6793. Lower Tugela: near Um-

voti River, Adlam in Herb. Norm. 562. Nkandhla: Nsuze Valley,

Codd 9681. Pietermaritzburg: Table Mountain, Killick 106. Port

Shepstone: McClean 474. Umzinto: Rudatis 1370.

Bothalia 16,1 (1986)

3

In the Flora Capensis the specimens here included

in one species were separated into two species on the

width of the epicalyx bracts and the shape of the

buds. Those with broad bracts were named D. dre-

geana and were supposed to occur in the Cape Pro-

vince, whereas those with narrow bracts and ‘more

fusiform buds’ were named D. natalensis. However,

the bracts vary in width, and several intermediates

between the two extremes can be found. Although

the epicalyx bracts are early deciduous, a study of

many herbarium specimens has shown that the

broader bracts are usually borne just below the

rounded base of the calyx whereas the narrower

bracts are borne slightly lower. Bract-scars can give

a clue as to whether the bracts were broad or not. It

appeared that some specimens with broad, and some

with narrow bracts, occurred both in the Cape and

Natal. The impression was gained that the broad

bracts usually occur on specimens growing in bush in

dry country and the narrow ones on trees in forests

and that these characters are therefore not specific

but rather indicate ecotypes.

Originally Drege gave the manuscript name Leeu-

wenhoekia tiliacea to both his Cape and Natal speci-

mens, showing that he considered them to be one

species. In the Flora Capensis, Drege’s Cape ma-

terial is put under D. dregeana and the Natal speci-

mens under D. natalensis.

2. Dombeya pulchra N.E. Br. in Kew Bull. 97:

142 (1895); Burtt Davy, FI. Transv. 1: 259 (1926).

Type: Barberton, Galpin 804 (K, holo.; PRE, iso.;

SAM; BOL).

Shrub, 1-4 m high, sometimes a fairly large tree

up to 7 m tall. Branches longitudinally ridged, not

very stout, fairly smooth; bark stringy; pith soft;

branches laxly to very densely pubescent with light

brown, more or less patently spreading, glandular

hairs over 1 mm long. Very occasionally a specimen

has a few acute, non-glandular hairs intermixed, or

rarely, pubescence entirely of non-glandular, acute,

patent hairs. Stipules subovate-oblong or linear-acu-

minate, about 10-16 mm long, 2-6 mm broad, felted

on both sides with a short velvety pubescence some-

times with a few long or glandular hairs along the

margins. Leaves with blade broadly ovate, about 6,5

x 6 to 25 x 9,5 cm, deeply cordate at the base,

broadly acuminate, sometimes 3-, rarely 5-lobed in

the upper half, obscurely to shallowly crenate-den-

tate on the margins, distinctly discolourous, upper

surface with a lax to dense, short, stellate pubes-

cence, lower surface always metallic-grey, felted

with a short velvety pubescence, 5-7-palmately

nerved, veins slender, prominent on lower surface;

petiole 3-16 cm long, sparsely to densely pubescent

with light brown, patent, glandular hairs, rarely with

pointed non-glandular hairs mixed, or dominant

(specimen of hybrid origin?), usually thicker than

the peduncle, longer or shorter than the peduncle.

Inflorescence cymose, axillary in upper leaves, 2-17-

flowered; peduncle 4—14 cm long, sparsely to

densely glandular hairy; pedicels up to 2,5 cm long,

similarly hairy. Bracts inserted at or very near the

calyx base, ovate, narrowly ovate-oblique, oblong or

narrowly oblong, 5-16 mm long, 2-8 mm broad,

shortly acuminate to a subacute or caudate apex,

cordate, rounded or somewhat cuneate at the base,

tomentose with stellate or tufted hairs, often shortly

and densely so, giving a felted appearance, some-

times hairs longer and shaggy. Calyx rounded at the

base or with a short stipe up to 2 mm long, rarely

longer; lobes 8-18 mm long, gradually acuminate

from the 1,5-5 mm broad base, stellate-tomentose

dorsally, sometimes hairs short and dense giving a

felted appearance. Petals mostly white, sometimes

creamy or pink and some flowers with purplish-rose

markings at the base of the petals, persistent, turn-

ing light cinnamon-rufous with age, about 12-20 mm

long, 11-20 mm broad near the apex. Stamens united

at the base for about 3 mm, filaments of different

lengths, up to about 4 mm; anthers about 3,5 mm

long; staminodes about 10 mm long. Ovary about 3

mm long, 2,5 mm broad; velvety tomentose with a

few slightly longer tufted hairs showing; style about

8 mm long, pubescent at the base, branches 5,

strongly recurved, about 3,5 mm long; cells 5, walls

thin, with up to 6 ovules in each cell. Capsule 7-10

mm long, about 7 mm broad, with a short dense stel-

late tomentum intermixed with longer tufted hairs

(but these hairs under 1 mm long), pubescent on the

sutures within; seeds brown, pitted.

Found in the cooler regions with a fairly high rain-

fall, on rocky hills, in tall grass with scattered shrubs,

on the edge of kloofbush or in damp places on banks

of streams. Recorded from Swaziland and the east-

ern Transvaal mountainous country.

SWAZILAND. — Mbabane: near Dalriack, Bolus 11716

(BOL); Valley, Compton 25024. Pigg’s Peak: Komati Bridge.

Compton 30050.

TRANSVAAL. — Barberton: Rimer’s Creek, Galpin 804 ;

Louws Spruit, Bolus 7681 (BOL); Highlands Creek, Clarke 55.

Belfast: Waterval Boven, Rogers sub TRV 14772. Carolina:

Waterval Onder, Rogers sub TRV 2351. Lydenburg: Wilms 78\

Uitsoek, Codd & Verdoorn 10507. Nelspruit: Numbi, Van der

Schijff 3052. Pilgrims Rest: Marieps Kop, Van der Schijff 4581.

This species closely resembles D. burgessiae and

the areas of distribution of the two species overlap in

part, but D. pulchra is restricted to the cooler re-

gions with higher rainfall in the eastern Transvaal

and Swaziland. D. burgessiae is widespread, from

the Natal coast through Swaziland and the eastern

Transvaal to the tropics. A few intermediate or hy-

brid specimens were seen from localities in which the

areas of distribution overlap.

3. Dombeya burgessiae Gerrard ex Harv. in FC

2: 590 (1862); Harv., Thes. Cap. 2: 24, t. 137 (1863);

Hook. f. in Bot. Mag. 21: t. 5487 (1865); Mast, in

FTA 1: 228 (1868); K. Schum. in Engl., Monogr.

Afr. Pfl. 5: 28 (1900); Sim, For. FI. P.E.A. 20, pi. 8

(1909); Burtt Davy, FI. Transv. 1: 260 (1926); Wild

in FZ 1: 522 (1961). Syntypes: Natal, Zululand,

Me Ken s.n. (not seen); Klip River, Gerrard s.n.

(TCD).

Assonia burgessiae (Gerrard ex Harv.) Kyntze, Rev. Gen. L. 76

(1891).

Dombeya burgessiae var. crenulata Szyszyl., Polypet. Thalam.

Rehm. 137 (1887). Syntypes: Natal, Newcastle. Rehmann 7034.

7049 (not seen); Drakensberg, Rehmann 7078 (not seen). D. ele-

gans K. Schum. in Engl., Monogr. Afr. Pfl. 5:31 (1900); Burtt

Davy, FI. Transv. 1: 260 (1926), nom. illegit. , pro parte as to Nel-

son 531 (PRE), non Cordem. (1895).

4

Bothalia 16,1 (1986)

D. rosea Bak.f. in J. Linn. Soc. (Bot.) 40: 29 (1911). Type:

Rhodesia [Zimbabwe], Swynnerton 196 (BM, holo.; K; SRGH).

Shrub, 1-4 m high with several to many stems,

sometimes a slender tree. Branches stout, longitudi-

nally ridged with fairly prominent leaf scars and

many lenticels, more or less pubescent; ultimate

branchlets comparatively slender, patently pubes-

cent with a mixture of long, pointed, multicellular,

thin-walled hairs and obtuse, glandular, multicellu-

lar hairs, hairs mostly over 1 mm long, single or two

to several from the same base. Stipules variable,

broadly ovate-acuminate from a broad subcordate

base, narrowly oblong-acuminate, or narrowly lan-

ceolate-acuminate, 1-1,7 cm long, 3-8 mm broad,

densely to sparsely pubescent, glabrescent, from

rather thick-textured and dark brown to fairly thin-

textured and light brown. Leaves broadly ovate,

6-20 cm long, 5-19 cm broad, deeply cordate at the

base, usually 3-5-lobed, sparsely to very densely pu-

bescent with stellate and tufted hairs, lower surface

usually more densely so than the upper, 5-7-pal-

mately nerved, margins irregularly and unequally

crenate to crenate-dentate; petiole 2-15 cm long,

mostly thicker than the peduncles, ± 1,5-2 mm long,

usually sparsely to densely patently pilose at least in

part, with multicellular, non-glandular, pointed

hairs mixed to a greater or lesser degree with glandu-

lar hairs. Inflorescence cymose, axillary, 2- to sev-

eral-flowered; peduncle longer or shorter than the

subtending petiole and usually slightly more slender,

2-14 cm long, similarly pubescent, branched in the

upper half; pedicels up to 3 cm long. Bracts 3, free to

the base, early caducous, variable in size, shape and

pubescence, narrowly or broadly oblong to ovate-

subcordate, shortly acuminate to an acute or shortly

caudate apex, 9—15 mm long, 3-8 mm broad, pubes-

cent on both surfaces, shortly tomentose or less

densely covered with stellate and long tufted hairs,

inserted close to the calyx base or up to 1,5 mm dis-

tant. Calyx lobed almost to the base, the united por-

tion (about 1,5 mm long) consolidated with the o-

vary base which is rounded or with a short stipe;

lobes about 14 mm long, acuminate from a 3-4 mm

wide base, dorsally pubescent. Petals white or pale

pink, sometimes with a rose-pink centre, persistent,

turning light brown with age, up to 20 mm long, 18

mm broad at the apex. Stamens united at the base

for 2-4 mm, filaments of various lengths, up to 4,5

mm long; anthers 3-4 mm long; staminodes 10-13

cm long, 2 mm broad in upper half. Ovary about

6x6 mm, stellate-tomentose to hirsute with tufted

hairs, normally 5-celled; style 8-14 mm long, gla-

brous or partly to wholly stellate-tomentose, 3-5-

branched; ovules 3-4 in a cell. Capsule about 10 mm

long, 8 mm broad, stellate-tomentose to hirsute with

tufted hairs (hairs about 1 mm or more long), pubes-

cent on the sutures within.

Found at forest margins, edge of bush, on wooded

stream banks, in kloofs, on grassy slopes with tall

grass and in marshy ground, or on rocky dolerite

hillsides under trees. Recorded from Natal, Swazi-

land and the eastern and northern Transvaal; also

found in tropical Africa from Mozambique and Zim-

babwe to Kenya.

NATAL. — Durban: Berea, Medley Wood 11094. Estcourt: Ta-

bamhlope, West 1086. Hlabisa: Codd 9612\ Hluhluwe Game Re-

serve, Ward 2202. Ingwavuma: Codd & Dyer 2838. Klipriver:

Gerrard 465 (NH); Grobbelaars Kloof, Acocks 1002. Mtonjaneni:

Melmoth, Lawn 1949 (NH). Newcastle: Rehmann 7034 (BOL).

Nongoma: Gerstner 3910. Richmond: Waterfall, Thode A. 1203.

SWAZILAND. — Hlatikulu, Stewart sub TRV 10089.

TRANSVAAL. — Barberton: near Kobinja, Codd 7806. Le-

taba: ‘Houtbosch’, Rehmann 6321 (BOL); Duiwelskloof, Scheep-

ers 262; Galpin 11385; Modjadjies, Krige 105; Dublin Mine,

Miller 4262. Lydenburg: Erasmus Pass, Strey 3721. Messina: Pole

Evans 1447. Petersburg: Blaauwberg, Codd 7956. Piet Retief:

Mooihoek, Devenish 1022. Sibasa: Tshakhoma, Van-Warmelo

5157/19; Punda Milia, Van der Schijff 973. Soutpansberg: Wylies-

poort, Hutchinson 2059; near Louis Trichardt, Meeuse 10209.

For the relationship with D. pulchra see the notes

under that species.

In the Natal Herbarium there is a specimen, West

1976 from Mapumulo, Tugela, which seems to be a

hybrid between this species and D. tiliacea-, in culti-

vation one finds plants which resemble D. burgessiae

but do not agree exactly with it or with any other

South African species. They are probably cultivars

of D. burgessiae. Nurserymen sometimes use the

names D. rosea and D. calantha for pink-flowered

cultivated Dombeyas. Specimens seen to date in gar-

dens are not at all like those species. D. rosea

Bak.f., from Zimbabwe, is a synonym of D. burges-

siae, whereas D. calantha K. Schum., from Malawi,

is described as having very large, coarsely crenate

leaves and the flowers borne on long brown-tomen-

tose pedicels.

4. Dombeya cymosa Harv. in FC 2: 589 (1862);

K. Schum. in Engl., Monogr. Afr. Pfl. 5:33 (1900);

Sim, For. FI. Cape Col. 145, pi. 18, Fig. 4 (1907);

Wild in FZ 1: 527 (1961). Type: Eastern Cape, Kaf-

fraria, Bowker 216 (TCD, holo.; PRE, fragment).

Small tree or shrub, about 3 m tall, sometimes up

to 8 m tall, occasionally a straggling bush about 1,3

m tall. Branchlets slender, woody, glabrous, or new

growth with short, massed, stellate hairs, faintly lon-

gitudinally ridged, leaf-scars fairly prominent, lenti-

cels small, raised, more or less circular. Stipules

early deciduous, linear-subulate, 2-7 mm long,

densely stellate-pubescent to glabrescent, thick to

subpetaloid in texture. Leaves rather thin in texture,

ovate, gradually narrowing in the upper third and

then abruptly acuminate towards the apex, rounded

or shallowly cordate at the base, not very large, usu-

ally under 70 mm long and 60 mm broad, crenate on

the margin, sparsely and coarsely stellate-pubescent

on both surfaces, glabrescent; hairs short; petiples

on flowering branches up to 35 mm, sparsely to

densely stellate-pubescent. Inflorescence cymose-co-

rymbose to subumbellate, 3-12-flowered, axillary in

the upper leaves of the main branches and the many

short lateral branchlets; peduncles slender, 1-3 cm

long, coarsely stellate-tomentose; pedicels varying

from 2 to 7 mm long, coarsely stellate-tomentose

with hairs short and appressed. Bracts 3, scattered

on the pedicels, often distant from the calyx and

from each other or 2 about midway, subopposite,

early caducous, linear-subulate, densely stellate-pu-

bescent, 1-2 mm long. Calyx lobed almost to the

subrounded base, united for about 1 mm, stellate-

pubescent without, sometimes glabrescent; lobes

4-5 mm long, 1,5-2 mm broad, usually reflexed in

mature flowers. Petals white, turning rusty brown

Bothalia 16,1 (1986)

5

with age, persistent, 6-8 mm long, 4-5 mm broad

near the apex. Stamens united at the base for less

than 1 mm, free portion varying slightly in length,

the longest about 4 mm long; anthers 0,75 mm long;

staminodes up to 5 mm long. Ovary appressedly stel-

late-tomentose, suboblate, about 1,5 mm long, 2

mm diam., 3- or rarely 4-celled; style about 1,5 mm

long, branches 3, sometimes 4, about 3 mm long,

rather thick and revolute; ovules usually 2 in a cell,

smooth, hard, yellow. Capsule 4 mm long, 3,5 mm in

diameter, stellate-tomentose with short hairs (under

1 mm long); seeds usually only one developing and

filling the capsule, rough with raised lines.

Found on river and stream banks, forest margins,

short closed woodland, stony slopes in gorges or dry

valley bushveld scrub. Recorded from the eastern

Cape, Natal, Swaziland and the eastern Transvaal;

also found in Mozambique and Malawi.

CAPE.— Butterworth: Pegler 770. East London: Pearson 7363;

Rattray 274 (GRA). Elliotdale: ‘Krelis Country’, Bowker 216.

Komga: Flanagan 58; Kei Cutting, Barker 9253 (NBG). Queens-

town: Junction Farm, Galpin 8077. Stutterheim: Kei Valley,

Acocks 9684. Tsolo: Tsitsa Waterfall, Galpin 6587 (GRA; BOL).

NATAL. — Dundee: Codd 2414. Durban: Medley Wood 6438.

Hlabisa: False Bay, Ward 3670; Hluhluwe Game Reserve, Ward

2291. Ixopo: Huntley 356. Kranskop: Mambula, Dyer 4348. Port

Shepstone: Oribi Flats, McClean 398. Ubombo: Strey 5286. Um-

voti: Edwards 2757. Weenen: Acocks 10144.

SWAZILAND. — Stegi, Ben Dlamini in Herb. Compton s.n.

TRANSVAAL. — Nelspruit: Malelane, Codd 5258. Pilgrims

Rest: Mariepskop, Van der Schijff 5502.

In some keys to this genus the distinguishing

character between this species and D. kirkii is given

as ‘stamens 2 per fascicle’ as opposed to ‘stamens 3

per fascicle’. In the South African specimens most of

the flowers dissected had 2, 3 and sometimes 4 sta-

mens in a fascicle.

It is reported that honey produced by bees visiting

these plants is excellent.

5. Dombeya kirkii Mast, in FTA 1: 227 (1868);

K. Schum. in Engl., Monogr. Afr. Pfl. 5: 39 (1900);

Sim, For. FI. P.E.A. 20 (1909); Wild in FZ 1: 527

(1961). Syntypes: ‘Nyassaland’, Meller s.n. (K); Mo-

zambique, Lupata Gorge, Kirk s.n. (K).

D. gilgiana K. Schum. in Engl., Pflanzenw. Ost.-Afr. C.270, t.

30 (1895). Syntypes: ‘Tanganyika’, Mschusas Dorf, Holst 8993;

9171a (K).

Shrub 1-5 m tall (in the tropics sometimes a tree

up to 10 m tall). Branchlets woody, faintly longitudi-

nally ribbed with scattered lenticels and prominent

leaf-scars; young branches densely stellate-pubes-

cent. Stipules 4-8 mm long, narrowly linear, thick or

thin in texture, sometimes with incurved margins,

stellate-pubescent with interspersed glandular hairs.

Leaves ovate, gradually narrowing in the upper

third, the apex abruptly acuminate, 4-12 cm long,

3-11 cm broad, crenate on the margins, cordate at

the base, 5-9-nerved from the base (nerves tomen-

tose) stellate-pubescent on both surfaces, hairs short

and appressed, wearing off with age; petiole 1-6 cm

long, stellate-tomentose. Inflorescence a cymose

panicle, short or elongated, axillary and clustered at

the apices of the branchlets, usually appearing from

February to May, pubescent with the characteristic

patent hairs; peduncle 1,5-8 cm long, stellate-to-

mentose interspersed with long, patent pointed

hairs; pedicels usually 0,5-1 ,5 cm long, sparsely to

densely patently pubescent as well as tomentose.

Bracts 3 at the base of the calyx, linear to linear-dl-

liptic (that is narrowing slightly to base and apex),

sometimes conduplicate, shortly tomentose on both

surfaces, the outer also bristling with long patent

hairs, 5-7 mm long, 1-1,5 mm broad at the middle.

Calyx rounded at the base, united for about 1,5 mm;

lobes 5-7 mm long, 1,5-2 mm broad at the base, dor-

sally tomentose and usually patently hairy as well.

Petals white turning light brown with age, persistent,

8-10 mm long, 4,5-5 mm broad just above the mid-

dle. Stamens united for 1-1,5 mm, 3 rarely 2 be-

tween the staminodes; filaments of different lengths,

very slender, up to 4 mm long; anthers 1 mm long;

staminodes very slender, 7 mm long.

Ovary 3-celled, 2,5 mm long, 2,5 mm broad, tomen-

tose and with long erect hairs at the apex, hairs up to

1 mm long; style about 2-3,3 mm long, sparsely

hairy near base or sometimes altogether pubescent,

branches suberect or slightly recurved, about 3 mm

long; ovules usually 2 in a cell. Capsule 4x3 mm,

with tufted setose hairs at the apex, hairs about 1

mm long.

Found on dry river banks in the north-eastern cor-

ner of the Transvaal; also recorded northwards as far

as Kenya.

TRANSVAAL. — Sibasa: Pafuri, Van der Schijff 3045; 3810;

Van der Schijff & Marais 3715.

The stems are said to be used by the Masai for

making bows. Of the tropical African species that

are considered to be synonymous by certain authors,

only D. gilgiana is included here. This is because no

material of the other synonyms has been seen,

whereas the plate accompanying the original de-

scription of D. gilgiana clearly represents D. kirkii.

6. Dombeya quinqueseta (Del.) Exell in J. Bot.,

Lond. 73: 263 (1935). Type: Cailliaud in Herb. Del.

(MPU; not seen).

Xeropetalum quinquesetum Del.. Cent. PL Afr. 84 (1826).

Dombeya reticulata Mast, in FTA 1: 228 (1868). Assonia reticu-

lata (Mast.) Kuntze, Rev. Gen. PI. 1: 85 (1896). Type: Nile Land,

Spike & Grant s.n. (K).

Assonia cuanzensis Uiern, Cat. Afr. PI. Welw. 1: 86 (1896).

Dombeya cuanzensis (Hiern) K. Schum. in Engl., Monogr. Afr.

Pfl. 5: 40 (1900). Syntypes: Angola, Pungo Andongo, Welwitsch

4735; 4736 (BM!).

Assonia huillensis Hiern, Cat. Afr. PI. Welw. 1: 85 (1896).

Dombeya huillensis (Hiern) K. Schum. in Engl., Monogr. Afr.

Pfl. 5: 35 (1900). Syntypes: Angola, Huilla, near Quipungo, Wel-

witsch 4726 (BM); near Lopollo, Welwitsch 4727 (BM); 4728

(BM).

D. myriantha K. Schum. in Engl., Monogr. Afr. Pfl. 5: 33,

t.2,A (1900). Type; Angola. Buchner 527 (Bt).

D. cuanzensis sensu K. Schum. in Warb., Kunene-Samb.

Exped. Baum 301 (1903), non (Hiern) K. Schum.

D. melanostigma K. Schum. ex Engl, in Engl., Pflanzenw. Afr.

3,2: 428 (1921), nom. nud.

D. rotundifolia sensu Exell in J. Bot. 65, Suppl. Polypet. 42

(1927), pro parte quoad spec. Gossweiler 1747, non Harv.

Shrub or small tree up to about 5 m tall. Branch-

lets woody, brown to reddish brown, minutely stel-

late-pubescent when young, glabrescent, with a few

scattered lenticels and prominent leaf-scars. Stipules

6

Bothalia 16,1 (1986)

apparently early deciduous. Leaves ovate, some-

times shallowly 3-lobed, cordate to deeply cordate at

base, up to 14 cm long and 20 cm broad, margin

faintly dentate, 5-7-nerved from the base, upper sur-

face stellate-pubescent, densely so on main and sec-

ondary veins, lower surface densely stellate-tomen-

tose when young, pubescence less dense with age,

reticulate veins on lower surface conspicuous and

prominently raised; petiole up to 7,5 cm long, stel-

late-tomentose. Inflorescence of dense axillary

cymes on old branches; peduncles up to 6 cm,

densely stellate-tomentose; pedicels up to 10 mm

long, densely stellate-tomentose. Bracts 3, at the

base of the calyx, whorled or 1 at base of calyx and 2

lower down, subopposite, linear, 3 mm long, not

more than 0,5 mm broad, densely stellate-tomentose

on outside, glabrous inside. Calyx rounded at the

base, united for 1-2 mm; lobes 5-7 mm long, 1-2

mm broad near base, dorsally densely stellate-to-

mentose. Petals white, turning yellow to light brown

with age, shaped like the wings of a butterfly, up to

10 mm long and 6 mm broad. Stamens 15, united for

0,5 mm at the base; staminodes 5, linear, up to 8 mm

long; fertile stamens in 5 groups of 3 each alternating

with the staminodes; filaments of different lengths,

up to 4 mm long; anthers 1 mm long. Ovary de-

pressed globose, 3 mm in diameter, 2,5 mm long,

tomentose with long erect hairs at the apex, 3-celled

with 2 ovules in each cell; style 2-2,5 mm long;

branches 2-2,5 mm long, recurved. Fruit not seen.

Found in the Okavango area of South West Afri-

ca/Namibia and northwards through Angola to Ke-

nya, Uganda and Ethiopia.

S.W.A. — Grootfontein: Barnard 194.

7. Dombeya rotundifolia (Hochst.) Planch, in

FI. des Serres, ser. 1,6: 225 (1850-51). Type: Natal,

Pietermaritzburg, Krauss 252 (BM, iso.).

Tree or shrub, 2-10 m tall; stems with rough bark.

Stipules caducous, narrowing from a triangular base

into a linear-subulate upper portion, stellate-pubes-

cent. Leaves more or less orbicular, rarely very

broadly ovate, 2-8 cm long, 1,8-9 cm broad (larger

on sterile branches and in tropical areas up to 13 x 14

cm), sparsely to densely stellate-pubescent on both

surfaces, the stellate hairs from a scaly base, crenate-

dentate, shallowly to clearly cordate at the base; pe-

tiole 1-3 cm long, stellate-tomentose. Cymes sev-

eral- to many-flowered, crowded at the apices of the

branches and branchlets; peduncles 1-4 cm long,

stellate-tomentose, sometimes with few to many

longer, tufted hairs intermixed, pedicels 0,5-1, 5 cm

long, pubescence as on the peduncles. Bracts 3, 1 or

2 at or near the calyx-base and 1 lower down, or all

three scattered on the pedicel, linear to linear-spa-

thulate or navicular, 1,5-5 mm long, sparsely to

densely stellate-pubescent on both surfaces. Calyx

rounded at the base, united for 0,5-1, 5 mm; lobes

5-7 mm long, 1,5-2 mm broad near base, stellate-

pubescent dorsally. Petals usually white (rarely rose-

pink) turning a light cinnamon-brown, 7-10 mm

long, 4,5-8 mm broad. Stamens united at the base

for up to 0,5 mm; filaments up to 3 mm long; anthers

1 mm long or slightly longer; staminodes about 6 mm

long. Ovary subglobose, about 2,5 mm in diameter,

tomentellous with stellate hairs mixed in the upper

half with tufted bristle-like hairs, bristles 0,5-1 mm

long; style with a few short patent hairs or stellate-

pubescent, 2, 5-3, 5 mm long; branches 1, 5-2,5 mm

long; cells usually 3 with 2 ovules in each cell. Cap-

sule subglobose, about 5 mm in diameter, tomentel-

lous with stellate and tufted hairs; the large bristle-

like hairs up to 1 mm long.

Occurs plentifully in Natal, Swaziland, the Trans-

vaal and South West Africa/Namibia; also found in

Botswana and northwards to east tropical Africa.

Two varieties are recognized, one widespread (var.

rotundifolia) and the other from a restricted area in

South West Africa/Namibia.

For key to varieties, see key to species.

(a) var. rotundifolia.

Verdoorn in Bothalia 9: 144 (1966).

Dombeya rotundifolia (Hochst.) Planch, in FI. des Serres, ser.

1,6: 225 (1850-51); Harv. in FC 1: 221 (1860); K. Schum. in

Engl., Monogr. Afr. Pfl. 5: 35 (1900); Medley Wood, Natal PI. 3,

t. 229 (1902); Sim, For. FI. Cape Col. 145 (1907); Burtt Davy, FI.

Transv. 1: 259 (1926); Wild in FZ 1: 525 (1960); M. Friedrich-

Holzhammer et al. in FSWA 84: 2 (1969). Xeropetalum rotundifo-

lium Hochst. in Flora 27: 295 (1844). Type: Natal, Pietermaritz-

burg, Krauss 252 (BM, iso.!).

D. densiflora Planch, ex Harv. in FC 2: 589 (1862). Type:

Transvaal, Magaliesberg, Burke & Zeyher s.n. (K).

D. damarana K. Schum. in Engl., Monogr. Afr. Pfl. 5:36

(1900). Syntypes: Hereroland, Marloth 1346 ; 1371 (PRE). Lind-

ley s. n. ; Giirich 40.

D. dinteri Schinz in Bull. Herb. Boiss., ser. 2, 2: 1005 (1902).

Syntypes: Hereroland, Waterberg, Dinter 392 (Z); Ondjombe-

ranga-Kette Dinter 392a (Z).

Characterized by very rough bark on stems; rather

coarse suborbicular leaves which are densely or

sparsely stellate-pubescent on both surfaces and

have prominent reticulate veins beneath; the ovary

is densely stellate-tomentose and setose, that is with

tufts of bristle-like hairs which are up to 1 mm long,

in upper portion.

This typical variety occurs in two growth forms.

The commoner form is a tree 5 to 10 m tall with

dark, rough bark, and bears its flowers profusely in

spring on almost leafless branches. The other form,

growing under less favourable conditions, is shrubby

forming a low bushy growth and only occasionally

developing into a tree. This ecotype flowers at any

time from spring to autumn and, if late, then on

leafy branches.

Found in fairly dry to dry areas; the spring-flow-

ering tree form occurs in Natal, Swaziland and the

Transvaal (also north-eastwards to the Ethiopian

border); the shrubby late-flowering form is found on

the central plateau of South West Africa/Namibia

and also in parts of Botswana.

NATAL.— Durban: ‘Port Natal’, Gueinzius s.n. Camper-

down: Inchanga, Marloth 4084. Hlabisa: Hluhluwe Game Re-

serve, Ward 1408. Inanda: Umhlanga, Medley Wood 7942 (NH).

Ubombo: Mkuze Game Reserve, Oatley M6 (NH).

Bothalia 16,1 (1986)

7

SWAZILAND. — Hlatikulu: Ebataan River, Compton 27955.

Mbabane: Komati Bridge, Compton 26980. Stegi: Verdoorn

1667a.

TRANSVAAL. — Barberton: Galpin 407. Lydenburg: Sekuku-

niland, Barnard 35. Nelspruit: Pabinspruit, Van der Schijff 47.

Piet Retief: Mooihoek, Devenish 26. Pilgrims Rest: near Sku-

kuza, Codd 4391. Potchefstroom: Boskop, Louw 357. Pretoria:

The Willows, Codd 7582\ Meintjies Kop, Burtt Davy 2188\ Foun-

tains, Verdoorn 465. Rustenburg: Swartruggens, Sutton 783.

Soutpansberg: Louis Trichardt, Koker 16. Waterberg: Mosdene,

Galpin M.33.

S.W. A. /NAMIBIA. — Grootfontein: Kinges 2966. Maltahohe:

Tsarris, Marloth 5072. Okahandja: Kaiser Wilhelms Berg, Mar-

loth 1346. Okavango: Runtu, Banks 91, Otjiwarongo: Tue Kop-

pie, Bradfield 13. Swakopmund: Boss sub TRV 36121. Windhoek:

De Winter 2595.

The commoner form of this typical variety, in the

adult stage, is conspicuous during early spring be-

cause of its profuse white (rarely pink) flowers borne

on almost leafless, rough-barked trees. The same

variety on the central plateau of South West Afri-

ca/Namibia and in places in Botswana, is hardly re-

cognizable because it occurs as a bushy shrub, only

here and there reaching tree form, and flowers at

any time from spring to autumn, the flowers and

leaves appearing together.

flceiDR v. v. MERWE.

FIG. 1. — Dombeya autumnalis Verdoorn. Flowering branch taken from the type specimen when

collected at Penge in April 1965: 1, bud with reflexed calyx, x 4; 2, petal, x 4; 3, portion of

the united stamens showing one staminode and 3 stamens, x 4; 4, ovary and style, x 4; 5,

leaf, x 1.

Bothalia 16,1 (1986)

Botanical collectors look upon the latter form as

an ecotype, the differences probably being caused by

the struggle for existence under adverse conditions.

Specimens from all parts agree in the suborbicular

leaves and the bristly hairs on the ovary.

(b) var. velutina Verdoorn in Bothalia 9: 144

(1966); M. Friedrich et al. in FSWA 84: 3 (1969).

Type: Rehoboth, Naukluft, Aub Schlucht, Strey

2010 (PRE, holo.; BOL; NBG).

Tall shrubs with several virgate stems up to 5 m

tall; bark rough; new growth softly and shortly to-

mentose. Leaves suborbicular or broadly oblong-or-

bicular, broadest in the upper half, 3-9 cm long,

2, 5-9, 5 cm broad, velvety-tomentose on both sur-

faces (the tomentum made up of minute stellate-pu-

bescent scales, the hairs short and silky), crenate-

dentate, cordate at the base, palmately 4-7-nerved;

petiole 1-2 cm long, softly tomentose. Cymes

crowded on lateral and terminal branchlets; pedun-

cles 10-15 mm long; pedicels 7-10 mm long. Calyx

about 6 mm long, densely and shortly tomentose

without. Petals about 7 mm long. Stamens united at

the base for about 0,5 mm; filaments unequal, about

2,5 mm long; staminodes about 5 mm long. Ovary

densely and shortly stellate-tomentose (not setose);

style 2-3 mm long, stellate-pubescent.

To date recorded only from the banks of the per-

manent stream at Aub Schlucht in the Naukluft

Mountains, South West Africa/Namibia.

S.W. A. /NAMIBIA.- — Rehoboth: Aub Schlucht, Naukluft

Mountains, Strey 2328; Tolken & Hardy 666; Hopefield, Giess

10961.

This variety agrees with the typical one in the sub-

orbicular leaves and the rough bark on mature

stems. This latter feature can be determined (as

pointed out in notes under D. autumnalis ) by the

examination of a cross-section of a branchlet; the

outer layer is seen to be rather thick and porous. It

differs from the typical variety principally in the ha-

bit and the pubescence. The plants are tall, virgate

shrubs with several comparatively slender stems and

do not develop into trees, whereas typical D. rotun-

difolia, as found in South West Africa/Namibia, oc-

curs as a low bushy shrub which, under certain con-

ditions, grows into the characteristic rough-barked

tree. The pubescence, which is a stellate tomentum

as in the typical variety, differs in that the hairs are

much shorter and rather silky, forming a dense, vel-

vety covering. This short tomentum is of particular

significance on the ovary because in typical D. rotun-

difolia, throughout the length and breadth of its dis-

tribution, the ovary is setose as well as stellate-to-

mentellous.

8. Dombeya autumnalis Verdoorn in Bothalia

9: 143 (1966). Type: Lydenburg, Penge Mine, Ver-

doorn 2470 (PRE, holo.).

Shrub or small tree, 1,6-5 m tall. Branchlets slen-

der, leafy, new growth shortly stellate-pubescent,

the pubescence formed of short spreading hairs from

a scaly base. Stipules caducous, deltoid or linear-sub-

ulate from a deltoid base, densely pubescent. Leaves

more or less orbicular, 1,5-5 cm long, 1-5 cm broad

(on flowering branches), densely to sparsely stellate-

pubescent on both surfaces, finely crenate-dentate

on the margins, reticulate veins rather obscure be-

neath; petiole slender, 2-12 cm long, stellate-pubes-

cent. Cymes in the axils of the upper leaves often

overtopping them, sometimes shorter; peduncles

slender, about 2,5 cm long, pubescent; pedicels very

slender, 1-1,5 cm long, pubescent. Bracts 3, 1 or 2 at

the calyx-base, or all scattered on the pedicel, linear-

navicular, about 2,5 mm long, pubescent on both

surfaces. Calyx rounded at the base, united for 0,5

mm; lobes reflexed, about 4 mm long, 2 mm broad,

dorsally stellate-pubescent. Petals persistent, white

turning cinnamon-rufous with age, about 7 mm long,

5 mm broad. Stamens united at the base for more or

less 1 mm, filaments with the longest about 2,5 mm

long; anthers up to 1 mm long; staminodes about 5

mm long. Ovary globose, about 3 mm in diameter,

shortly stellate-tomentose (not setose); style gla-

brous or minutely stellate-pubescent, about 2 mm

long, branches about 2 mm long; ovules 2 in a cell.

Capsule about 5 mm in diameter, stellate-tomen-

tose. Fig. 1.

Recorded from the eastern Transvaal in moun-

tainous country on mesophytic, well wooded slopes

among rocks and in riverine bush.

TRANSVAAL. — Lydenburg: Abel Erasmus Pass, Schlieben &

Strey 8387; Codd 10027; near Penge mine, Codd & Dyer 7737;

Codd & Verdoorn 10488; Repton 5936 (partly); Verdoorn 2470;

near Weltevreden Asbestos Mine, Verdoorn 2471. Letaba: Dub-

lin Mine, Miller 4271.

D. autumnalis differs from the form of typical D.

rotundi folia, which occurs in the eastern and central

areas of southern Africa, in that it flowers in late

summer and autumn together with its leaves, and

not in early spring only, on more or less leafless

branches. From all forms of D. rotundi folia, includ-

ing those in South West Africa/Namibia, D. autum-

nalis differs mainly in that its slender stems do not

develop a rough bark. This feature can be deter-

mined by the examination of a cross-section of a

branchlet. The outer layer of D. autumnalis is thin

and solid, whereas that of D. rotundi folia is thick

and porous. A further diagnostic character is that

the pubescence on the ovary of D. rotundifolia is se-

tose as well as stellate-tomentellous, whereas in D.

autumnalis it is stellate-tomentose without setae (see

Fig. 1). In the slender peduncles and pedicels, the

comparatively thin leaves, and the habit, our species

resembles D. cymosa, from which it can be readily

distinguished by the leaf shape. In D. cymosa the

leaves are acuminate in the upper third, not suborbi-

cular, and more or less rounded at the apex.

UITTREKSEL

Die spesies van Dombeya Cav. wat in suidelike

Afrika voorkom, is hersien en 'n sleutel van die agt

spesies in die gebied word verskaf D. quinqueseta

(Del.) Exell, wat vir die eerste keer vir suidelike

Afrika aangeteken word, is ingesluit.

Bothalia 16,1 (1986)

9

INDEX TO DOMBEYA

Assonia Cav

burgessiae (Gerrard ex Harv.) Kuntze

cuanzensis Hiern

huillensis Hiern

reticulata (Mast.) Kuntze

Dombeya Cav

autumnalis Verdoo n

burgessiae Gerrard ex Harv

burgessiae var. crenulata Szyszyl

calantha K. Schum

cuanzensis (Hiern) K. Schum

cuanzensis sensu K. Schum

cymosa Harv

damarana K. Schum

densiflora Planch, ex Harv

dinteri Schinz

dregeana Sond

elegans K. Schum

gilgiana K. Schum

gracilis K. Schum 2

huillensis (Hiern) K. Schum

kirkii Mast

melanostigma K. Schum. ex Engl

my riant ha K. Schum

natalensis Sond 2

palmata Cav

pulchra N.E. Br

quinqueseta (Del.) Exell

reticulata Mast

rosea Bak. f 4

rotundifolia sensu Exell, non Harv

rotundifolia (Hochst.) Planch 6

var. rotundifolia Verdoorn 6

var. velutina Verdoorn 8

tiliacea (Endl.) Planch 2

Leeuwenhoekia tiliacea E. Mey 2

Xeropetalum Delile 1

quinquesetum Del

rotundifolium Hochst 6

tiliaceum Endl. & Fenzl 2

1

3

5

5

5

1

8

3

3

4

5

5

4

6

6

6

2

2

5

Bothalia 16,1: 11-22 (1986)

The Eriosema squarrosum complex (Papilionoideae, Fabaceae) in

southern Africa

C. H. STIRTON*

Keywords: Eriosema squarrosum complex, Fabaceae, new combinations, new taxa

ABSTRACT

Eriosema squarrosum (Thunb.) Walp. has traditionally been the dumping ground for all densely pubescent

Eriosema species in southern Africa. This study clarifies the identity of E. squarrosum ; recognizes three new taxa:

E. luteopetalum C. H. Stirton, E. rossii C. H. Stirton and E, umtamvunense C. H. Stirton; effects the combina-

tions E. latifolium (Benth. ex Harv.) C. H. Stirton and E. acuminatum (Eckl. & Zeyh.) C. H. Stirton; and re-

instates E. dregei E. Mey. The species E. preptum C. H. Stirton, described earlier, also belongs to this complex.

Rhynchosia barbertonensis C. H. Stirton is given as a new name for E. rogersii Schinz.

INTRODUCTION

The Eriosema squarrosum (Thunb.) Walp. com-

plex remains the only unresolved complex among

the Eriosema species of southern Africa. As in the

E. cordatum E. Mey. complex (Stirton 1978, 1981a)

it is partly a nomenclatural muddle and partly a taxo-

nomic problem. Once again hybridization has played

a prominent role in the development of the complex

(Stirton 1981b).

The complex comprises the majority of the

densely pubescent Eriosema species in southern

Africa. Most of these plants have in the past been

referred to either E. zeyheri E. Mey. or E. squarro-

sum.

Central to the complex is E. squarrosum. It was

originally described by Thunberg as Hedysarum

squarrosum (Prodr. 132, 1800), and later transferred

to Desmodium by De Candolle (Prodr. 2: 333,

1825). Ecklon & Zeyher (Enum.: 251, 1836) ac-

cepted Desmodium squarrosum (Thunb.) DC. but

divided it into 3 varieties: squarrosum, acutifolium

and acuminatum. Ernst Meyer, whose Commenta-

riorum is predated by Ecklon & Zeyher’s Enumera-

tio by a few months, published the binomial Erio-

sema zeyheri E. Mey. for the same taxon (Comm.

129, 1836). At the same time he described E. dregei

E. Mey., a species completely new to science. Here

the matter rested until Walpers (Linnaea 13: 536,

1839) realized that Thunberg’s Hedysarum squarro-

sum was not a Desmodium, as De Candolle and Eck-

lon & Zeyher had thought, but was as Meyer had

noted, really an Eriosema. He accordingly effected

the new combination, E. squarrosum (Thunb.)

Walp.

Twenty three years later Harvey (FI. Cap. 2: 260,

1862) re-investigated the genus. He accepted Walp-

er’s combination, effected the combination for var.

acuminatum (Eckl. & Zeyh.) Harv., reduced E. dre-

gei to varietal rank and described the new variety

latifolium Benth. ex Harv. Thirty three years were to

pass until Baker (/. Bot., Lond. 33: 146, 1895) ap-

plied the now disallowed Kew Rule and thereby

caused considerable confusion with the attendant

combinations. Problems have also arisen from the

additional collections that have accumulated since

Harvey’s and Baker’s treatments. Several new taxa

have been discovered this century and with the

known cases of hybridization in Natal the complex

had become quite a muddle by the time this study

was initiated in 1974.

This study recognizes eight species in the complex.

Firstly E. squarrosum (Thunb.) Walp. is retained as

a variable species. E. dregei E. Mey. is reinstated,

whereas Meyer’s E. zeyheri is placed in synonomy

with E. squarrosum. Harvey’s E. squarrosum var.

latifolium Benth. ex Harv. is raised to specific rank;

E. latifolium (Benth. ex Harv.) C. H. Stirton. Three

new species are described: E. luteopetalum C. H.

Stirton, E. umtamvunense C. H. Stirton and E. rossii

C. H. Stirton. E. preptum C. H. Stirton was de-

scribed earlier (Stirton 1981c). Ecklon & Zeyher’s

var. acuminatum is raised to specific rank: E. acumi-

natum (Eckl. & Zeyh.) C. H. Stirton.

KEY TO SPECIES

la Flowers yellow or greenish yellow (drying yellow):

2a Stems and leaves silvery; upper surface of leaflets sparsely appressed pubescent; flower bracts

caducous 5. E. dregei

2b Stems and leaves tawny, especially veins of leaves; upper surface of leaflets densely appressed pubes-

cent; flower bracts persistent:

3a Flowers 9-10 mm long, yellow; wing petals much longer than the keel; flower bract shorter than

the flower; pistil 7 mm long 2. E. latifolium

* The Herbarium, Royal Botanic Gardens, Kew, Richmond,

Surrey, TW9 3AE, U.K.

12

Bothalia 16,1 (1986)

3b Flowers 12-15 mm long, greenish yellow; wing petals equal in length to the keel; flower bract equal

in length to flower; pistil 12 mm long 3. E. luteopetalum

lb Flowers pinkish orange, pale red, or orange with red venation (drying dark);

4a Flower bracts persistent, equal to or longer than the flower:

5a Lower surface of leaflets pale green, finely pubescent with margins and veins appressed hairy; inflo-

rescence hidden in leaves at anthesis, 30-60-flowered; flowers 12 mm long; fruits

10 mm wide 6. E. rossii

5b Lower surface of leaflets whitish, densely woolly with yellow appressed hairs along the veins; inflo-

rescence exserted from leaves at anthesis, 20-30-flowered; flowers 8-10 mm long; fruits

7-8 mm wide 2. E. acuminatum

4b Flower bracts caducous, shorter than the flower:

6a Flowers 13-14 mm long; calyx 8 mm long; appendages of standard fused but free from auricles;

pistil 11 mm long 4. E. umtamvunense

6b Flowers 6-9 mm long; calyx 4-6 mm long; appendages of the standard fused and extending to the

auricles; pistil 6-8 mm long:

7a Undersurface of leaflets finely and shortly pubescent; leaflets 15-20 mm wide; racemes 10-20-

flowered; flowers 6-7 mm long; seeds black or dark brown; eastern Cape 1. E. squarrosum

7b Undersurface of leaflets densely woolly with veins distinctly appressed with longer hairs; leaflets

20-30 mm wide; racemes 25-35-flowered; flowers 7-9 mm long; seeds grey or light brown with

darker speckles and blotches; Natal 8. E. preptum

' QUICK-SORT CHARACTERS

Flowers yellow: luteopetalum, latifolium, dregei

Inflorescence with less than 20 flowers: squarrosum

Flower bracts caducous: dregei, squarrosum, umtamvunense,

preptum

Flower bracts equal to or longer than flowers: acuminatum, rossii

Calyx teeth shorter than the calyx tube: dregei, latifolium, luteo-

petalum

Appendages on front of standard free from the auricles: acumina-

tum, umtamvunense

Wing petals equal in length to the keel petals: luteopetalum

Seeds black: acuminatum, squarrosum

Seeds pale chestnut brown: luteopetalum

Seeds grey or brown with speckles and blotches: preptum

1. Eriosema squarrosum (Thunb.) Walp. in Lin-

naea 13: 536 (1839); Harv. in FI. Cap. 2: 260 (1862).

Type: Cape, ‘crescit in campis graminosis cis et trans

Camtoos-river, prope Galgebosch et alibi’, Thun-

berg s.n. (UPS, Herb. 17271, microfiche).

Hedysarum squarrosum Thunb., Prodr. 132 (1800); FI. Cap.

595 (1823). Desmodium squarrosum (Thunb.) DC., Prodr. 2: 233

(1825); Eckl. & Zeyh., Enum. 251 (1836).

Crotalaria lineata Thunb., Prodr. 125 (1800); FI. Cap. 573

(1823); non Jacq. (1786). Type: ‘e Cap. Bon. Spei’, Thunbergs.n.

(UPS, Herb. 16559, microfiche).

Desmodium squarrosum (Thunb.) DC. var. acutifolium Eckl.

& Zeyh., Enum. 251 (1836). Type: Cape, ‘in collibus gramineis

terrae, Adow’, Ecklon s.n. (S, holo.; FI; K; P; W, iso.).

Eriosema zeyheri E. Mey., Comm. 129 (1836); Bak. in J. Bot.,

Lond. 33: 146 (1895). Rhynchosia zeyheri (E. Mey.) Steud.,

Nom. 2: 54 (1841). Lectotype: Cape, ‘Zwartkopsrivier, ad ipsus

ripas, iv.C.c.20’, Drige s.n. (BM; K; P, isolecto.)

Eriosema reticulatum E. Mey. var. canescens Meisn. in J. Bot.,

Lond. 2: 80 (1843). Lectotype: ‘in solo argillaceo in Zitsikamma’,

Krauss 926 (NY). This is marked in some herbaria as E. ambi-

guum Krauss (nom. nud.)

Perennial herb up to 300 mm tall. Stems ascend-

ing, strongly branched from the base, closely clothed

with deflexed hairs. Leaves trifoliolate, 25-55 x

15-20 mm, elliptic, becoming narrower and longer,

grading into lanceolate near the ends of branches,

lower leaves often obovate; apex acute, sometimes

obtuse, base cuneate, dark green above, whitish be-

neath, upper surface glabrous to strigillose, lower

surface finely and shortly pubescent, glandular; lat-

eral leaflets smaller, asymmetrical. Stipules 7-10 mm

long, narrowly lanceolate, free, appressed, persist-

ent, softly pubescent and sparsely covered in glands.

Petioles 2-3 mm long. Racemes axillary, up to 100

mm long, exceeding the leaves; peduncles 26-55 mm

long, densely racemose beyond the middle with

10-20 reflexed imbricating flowers. Flowers 7-9 mm

>o , UlUf.,

t urCitajkc/y' .

■X S- " /?, -<'

FIG. 1. — Representative specimen of Eriosema squarrosum

(PRE 56220).

Bothalia 16,1 (1986)

13

long, pinkish orange; bracts caducous, about half the

length of the flower. Calyx 5-6 mm long, lobes

equal; teeth lanceolate, equal to or slightly longer

than the tube, softly white pubescent, glandular.

Standard 8-9 mm long, narrowly obovate, clawed,

auriculate; gland-covered and pubescent on back;

appendages present, well developed, situated above

the auricles and tapering towards them, fused. Wing

petals 9x2 mm, oblong, auriculate, longer than the

keel. Keel petals 7 mm long, 3 mm wide at maxi-

mum, encrusted with yellow glands, distinctly pock-

eted. Staminal sheath 10 mm long; free stamen geni-

culate. Pistil 8 mm long; ovary 4 mm long, subses-

sile, densely hairy almost up to point of flexure;

height of curvature 3 mm; stigma capitate. Nectary

present, margin undulate. Fruit 11-15 x 8-10 mm,

glandular, covered in fine hairs overlain by long stiff

yellow hairs. Seeds 5x3 mm, black or dark brown.

Fig. 1.

This species is almost entirely confined to the east-

ern Cape (Fig. 2) where it is confined to grassveld

and sandy flats. It is sympatric with E. salignum E.

Mey. and in the eastern part of its range with E. cor-

datum E. Mey., E. dregei E. Mey. and E. latifolium

(Harv.) C. H. Stirton. It has been extremely difficult

to assign rank to any of the several distinctive local-

ised variants as these overlap and blur into each

other. There is however a cline of increasing pubes-

cence, especially on the upper surface of the leaflets

as well as a general increase in size as one moves

northwards and eastwards.

TRANSKEI. — 3129 (Port St Johns): Coffee Bay (-CC), Tyson

22 (PRE). 3227 (Stutterheim): Kabaku Hills (-CB), Acocks 9344

(PRE); near Komgha (-DB), Flanagan 704 (PRE). 3228 (Butter-

worth): Idutywa (-AB), Schlechter 1377, 6271 (PRE); River

Mouth (-BC), Hilner 485 (PRE); Kentani (-CB), Pegler 123

(PRE).

CAPE. — 3225 (Somerset East): Selborne (-DA), Smith 3711

(PRE); Boschberg (-DC), Macowan 475 (P); Stockenstrom

(-DD), Scully 155 (PRE). 3226 (Fort Beaufort): Katberg (-BC),

Moss 15400 (BM). 3227 (Stutterheim): Cathcart (-AC), Kemp

s.n. (NGB); Hang River (-BA), Spearman 25 (NBG); west of

East London (-BB), Maguire 605 (NBG); Dohne (-CB), Acocks

9381 (K; PRE); Pierie (-CC), Sim 4021 (PRE); King William’s

Town (-CD), Tyson 2942 (NBG; PRE); 1 km from Amabele

(-DA), Marais 237 (PRE); Port Alfred (-DB), South s.n. (PRE

56218, 56219). 3228 (Butterworth): 12 km E of East London

(-CC), Comins 1256 (PRE). 3322 (Oudtshoorn): George (-CD),

Guthrie 4293 (NBG). 3323 (Willowmore): Wynandskraal (-CD),

Burchell 5263 (K). 3324 (Steytlerville): Zwartkopsrivier (-DB),

Drege s.n. (BM; GBH; K; P; W); Tsitsikamma (-DC), Krauss

926 (NY). 3325 (Port Elizabeth): Zuurbergen (-AD), Drege s.n.

(P); Van Stadens Flower Reserve (-CC), Dahlstrand 2533 (PRE;

STE); Van Stadens Gorge (-CC), Long 267 (PRE); Van Stadens-

berg (-CC), MacOwan 475 (BM); Krakakamma (-CD), Burchell

4573 (K); Uitenhage (-CC), Penther 2559 (W); Addo (-DA),

Drege s.n. (GBH; K; P; Z); Addo (-DA), Zeyher s.n. (K); Eck-

lon s.n. (K; P; W); flats near Port Elizabeth (-DC), West 461

(PRE). 3326 (Grahamstown): Rautenbach's Drift (-AC), Bur-

chell 4191 (K); mountains near Grahamstown (-AD), Britten s.n.

(PRE 56210); Gane s.n. (PRE); near Alexandria Lombards

(-DA), Burchell 4155 (K); near Port Alfred, between Rietfontein

and Kowie River (-DB), Burchell 4002 (GBH; K); Port Alfred

(-DB), Sonta s.n. (PRE 56218); Kowie (-DB), Tyson s.n. (PRE);

5 km from Port Alfred on road to Kenton-on-Sea (-DB), Germis-

huizen 1531 (PRE); between Bathurst and Port Alfred (-DB),

Stirton 764a; Rietfontein (-DB), Burchell 4042 (K); 3 km NNW of

Southwell (-DA), Acocks 12053 (PRE). 3423 (Knysna): Knysna

(-AA), Herb. STE 13509 (STE); Breyer s.n. (PRE 23904); Plet-

tenberg Bay (-AB), Zeyher s.n. (NBG). 3424 (Humansdorp):

Slang River (-BA), Spearman 25 (PRE); Fourcade 1860 (BOL);

Humansdorp (-BB), Fourcade 1727 (BOL). Without precise

locality: Bouvin s.n. (P); Bowie s.n. (BM; K); Bunbury s.n.

(BM); Cooper s.n. (NY); Duthie 516 (STE); Drdge s.n. (NY);

Ecklon s.n. (TCD, W); Fourcade 1939 (BOL); Germishuizen

1745 (PRE); Macowan 475 (P); Masson s.n. (BM); Verreaux s.n.

(G; TCD).

This species has been and is easily confused with

E. acuminatum (below) and E. preptum (no. 8). For

differences see under the latter species.

Eriosema squarrosum is the smallest of all the

Cape, Transkeian and Natal species. It flowers from

September through to March.

2. Eriosema acuminatum (Eckl. & Zeyh.) C. H.

Stirton, comb, et stat. nov. Type: ‘In collibus montis

Wintersberg prope Phillipstown’, Ecklon s.n. (S,

holo.!; FI; K; P, iso.!).

Desmodium squarrosum (Thunb.) DC. var. acuminatum Eckl.

& Zeyh., Enum. 251 (1836). Eriosema squarrosum (Thunb.)

Walp. var. acuminatum (Eckl. & Zeyh.) Harv., FI. Cap. 2: 260

(1862). E. zeyheri E. Mey. var. acuminatum (Eckl. & Zeyh.)

Bak., J. Bot., Lond. 33: 147 (1895).

Perennial herb up to 250 mm tall. Stems ascending

to erect, branching near the base, densely clothed in

golden-brown hairs. Leaves trifoliolate, 40-50 x

25-30 (-35) mm, broadly elliptic but becoming nar-

rower and longer towards the ends of branches,

lower leaves often obovate; apex acute, base cu-

neate, appressed pilose above, densely woolly below

with yellow appressed hairs massed along the veins,

glandular; lateral leaflets smaller, gibbous. Stipules

11-16 (-20) mm long, falcate-lanceolate, free, clasp-

ing the stem, sparingly appressed pubescent with

long hairs interspersed from the centre thickening

towards the apex. Petioles shorter than 5 mm. Ra-

cemes axillary, (40-) 60-90 (-140) mm long, exceed-

ing the leaves; peduncles 40-55 mm long, densely

racemose beyond the middle and bearing 20-30

flowers. Flowers 8-10 mm long, pale red or orange,

rarely yellowish; bracts persistent, equal to or ex-

ceeding the flower. Calyx 6 mm long, lobes equal;

teeth narrowly lanceolate, equal to the tube, keel

tooth almost acicular, slightly longer than the vexil-

lar and lateral teeth; covered in 3 mm long yellowish

brown hairs and a few scattered glands. Standard 10

x 5-7 mm, narrowly to broadly obovate, subtended

14

Bothalia 16,1 (1986)

by a 3 mm long claw, auriculate; appendages pres-

ent, situated 4 mm from the base of the claw but

above the auricles and free of them. Wing petals

9-10 mm long, 2, 5-3, 0 mm wide, cultrate, strongly

auriculate, upcurving, longer than the keel. Keel pe-

tals 8 mm long, up to 3,5—4, 0 mm wide, covered in

yellow glands, distinctly pocketed. Staminal sheath

7-9 mm long, free stamen geniculate. Pistil 8 mm

long; ovary 3 mm long, subsessile, densely hairy at

least until halfway to the point of flexure of the style;

height of curvature 2,5 mm; stigma capitate. Nectary

present, margin erose. Fruit 11-12 x 7-8 mm, glan-

dular, covered in reddish shaggy hairs. Seeds 5x3

mm, black, oblong. Fig. 3.

Eriosema acuminatum occurs mainly in grasslands

in the Transkei (Fig. 4). It has not often been col-

lected and is probably more common within its

known overall distribution area than its representa-

tion in herbaria suggests. According to herbarium

labels this species flowers between October and

November, but also occasionally in December and

January. No ecological data were found on herbar-

ium labels.

NATAL. — 2929 (Underberg): Injassuti Heights (-AB), Thode

8225 (STE). 3030 (Port Shepstone): Umtamvuna Nature Reserve

(-AA), Abbott 2195 (NH); Shelley Bay (-CD), Mogg 11920

(PRE). 3130 (Port Edward): Port Edward (-AA), Stirton 5643,

5672 (PRE); S of Port Edward (-AA), Germishuizen 1745

(PRE); Ngwenya 215 (NH).

FIG. 3. — Holotype of Eriosema acuminatum (Ecklon s.n.)

FIG. 4. — Known distribution of Eriosema acuminatum.

TRANSKEI. — 3029 (Kokstad): Cabane River (-AB), Tyson

2653 (NBG). 3128 (Umtata); Umtata (-DB), Sole s.n. (NBG);

Umtata District (-DB), Penther 2610 (W). 3129 (Port St Johns):

Ntsubane Forest Station (-BC), Galpin 10994 (PRE); 73,2 km

from Umtata to Port St Johns (-CB), Grobbelaar 2311 (PRU);

Coffee Bay (-CC), Tyson 22 (PRE; NY); Port St Johns area

(-DA), Swinney & Baker 14146 (PRE). 3228 (Butterworth): Idu-

tywa (-AB), Schlechter 1377, 6271 (NBG; P; STE); 1 km from

Butterworth to Komgha (-AC), Grobbelaar 2306 (PRU);

Komgha (-CB), Compton 17657 (NBG); Kabonqaba (-CB), Tay-

lor 3706 (NBG). 3226 (Fort Beaufort): Klipplaats River (-BB),

iv.a.2., Drege s.n. (K; G; P; W). 3227 (Stutterheim): Fort Cu-

nynghame (-AD), Taylor 4239 (NBG); Donga Range (-CB),

Acocks 9344 (PRE). 3327 (Peddie): Igoda Holiday Camp (-BB),

Steyl 4 (STE). 3228 (Butterworth): Qora Mouth (-BC), Hilner

485 (PRE); Kentani (-CB), Pegler 123 (PRE). Without precise

locality: Barber s.n. (TCD).

E. acuminatum can be separated from E. squarro-

sum (above) by its much longer, persistent flower

bracts that are equal to or longer than the flowers,

by the appendages on the standard being above and

well free of the auricles and by the presence of

golden or russet pubescence especially on the calyx

and the undersurface of the leaflets. In E. squarro-

sum the flower bracts are caducous, shorter than the

flower, the pubescence is white and the appendages

are better developed extending into the auricles of

the standard. From E. rossii C. H. Stirton it differs

in its much narrower stipules, larger fruits and pu-

bescence of the leaflets.

3. Eriosema luteopetalum C. H. Stirton, sp.

nov., E. lati folio (Harv.) C. H. Stirton affinis, sed

floribus maioribus, bracteis longioribus differt.

Suffrutex usque ad 600 mm altus, vere florens. Fo-

lia trifoliolata, 45-60 x 35-45 mm, lateralia minora,

asymmetrica, obovata vel anguste obovata. Stipulae

12-15 mm longae, liberae. Racemi axillares, 24-45-

flori, folia subtendentia superantes. Flores 12-15

mm longi, lutei, bracteae 10—15 mm longae, persist-

entes. Calyx lobis aequalibus. Vex ilium 10-15 x 6-7

mm, obovatum, unguiculatum, reflexum, callis bene

evolutis connatis, sursum crispis, supra unguem in

auriculas extensis. Alae carinam subaequilongae.

Vagina staminalis 10 mm longa. Gynoecium 10 mm

longum; ovarium 5 mm longum, dense pubescens.

Fructus 14-16 x 8-9 mm, sericeus, tenuiter pubes-

Bothalia 16,1 (1986)

15

hairs and short white pubescence, glandular; upper

margin double convex, lower margin convex-con-

cave, beaked. Seeds 6 mm long, 4 mm wide, oval,

pale chestnut brown; cotyledons well developed,

radicle short, barely protruding from apex, plumule

exserted. Fig. 5.

This species is endemic to southern Natal (Fig. 6)

but may yet be discovered in the Transkei. It is com-

mon along the coast and seems to grow best on

sandy soils in previously burnt grassland. (Fig. 7).

The range of this species appears to be extending as

a result of roadbuilding activities.

FIG. 5. — Holotype of Eriosema luteopetalum (Stirton 5652).

cens. Semina 6 mm longa, 4 mm lata, pallide casta-

nea.

TYPE. — Natal, 3030 (Port Edward): Roselands

(-CD), Stirton 5652 (PRE, holo.; K, iso.).

Suffrutex up to 600 mm tall. Stems erect, branch-

ing from the base, densely covered with golden de-

flexed appressed hairs. Leaves trifoliolate, 45-60 x

35-45 mm, obovate, inland populations with leaflets

becoming narrower, more acute and ovate; strigose

above, densely white pubescent below with veins

prominently yellowish strigose; lateral leaflets

smaller. Stipules 12-15 mm long, broadly lanceolate,

free, persistent. Racemes axillary, 24-45-flowered,

greatly exceeding the leaves. Flowers greenish yel-

low, 12-15 mm long with persistent 10-15 mm long,

3 mm wide boat-shaped bracts. Calyx lobes equal.

Standard 10-15 x 6-7 mm, obovate, prominently

clawed and auriculate, appendages well developed,

fused, upcurled, extending from above the claw into

the auricles. Wing petals equal in length to the keel,

pouched. Keel petals pocketed, encrusted with small

yellow glands. Staminal sheath 10 mm long; free sta-

men geniculate. Pistil 10 mm long; ovary 5 mm long,

subsessile, densely hairy, extending halfway along

style to point of flexure, height of curvature 4 mm;

stigma capitate, exserted beyond stamens. Nectary

present, margin undulate. Fruit 14-16 x 8-9mm,

thickly covered with a mixture of long yellowish

FIG. 7. — Habit and habitat of Eriosema luteopetalum.

16

Bothalia 16,1 (1986)

NATAL. — 3030 (Port Shepstone): The Valley, Port Shepstone

(-CB), Martin s.n. (PRE); Margate (-CD), Stirton 5660 (K;

PRE), 10355 (NU); Shelley Beach (-CD), Stirton 5664 (K; PRE);

Roselands (-CD), Stirton 5652 (K; PRE); Uvongo (-CD), Grob-

belaar 1009 (PRE); near Izotsha turn-off on Ramsgate-Port Shep-

stone Road (-CD), Stirton 1407 (PRE). Without precise locality:

Drege s.n. (L: P; W); Wood 3139 (K).

Eriosema luteopetalum is a very showy shrub

worthy of consideration as a garden plant. It pro-

duces masses of inflorescences in spring providing a

flash of yellow colour, soon to be followed by

colourful brown fruits that persist on the plant long

after the fruits have explosively scattered their

seeds.

The specific epithet luteopetalum, was chosen to

draw attention to the massed yellow flowers. It

seems remarkable that this distinctive and locally

abundant species has, until recently, been so rarely

collected. This species appears to have been missed

by most of the early collectors. Its nearest allies are

E. latifolium (no. 7) and E. dregei (no. 5) from

which.it differs in its very much larger flowers and

distinctive pubescence.

4. Eriosema umtamvunense C. H. Stirton, sp.

nov., E. squarroso (Thunb.) Walp. affinis, sed

planta maiora, robustiora, floribus maioribus dif-

fert.

Herba perenna usque ad 50 mm alta, vere florens.

Folia trifoliolata, 57-70 x 28-40 mm, lateralia min-

ora asymmetrica, elliptica. Stipulae 13 mm longae,

FIG. 8. — Eriosema umtamvunense (Stirton 5624):

libri. Racemi axillares, 20-25-florati, folia subten-

dentia superantes. Flores 13-14 mm longi, rosei fla-

vique; bracteae 8 mm longae, caducae. Calyx 6 mm

longa, lobi tubam subaequantes. Vexillum 13 x 9

mm, unguiculatum, reflexum, calli bene evoluti,

conferruminati cucullati, ab auriculis liberi. Petala

carinae breviora quam alae. Vagina staminalis 10-12

mm longa. Gynoecium 11 mm longum; ovarium 5

mm longum, dense pubescens. Fructus 15-16 mm

longus, 10-11 mm latus, molliter flavo-pubescens.

TYPE. — Transkei, 3130 (Port Edward): near Ku-

Mankenbeya, Imizizi location (-AA), Stirton 5624

(PRE, holo.). Fig. 8a.

Erect perennial shrub up to 500 mm high. Root-

stock horizontal, branched. Stems up to 20, branch-

ing from lower nodes, densely recurved, appressed

fulvous above but less dense towards the base.

Leaves trifoliolate, 57-70 x 28-40 mm, length-

breadth ratio 1, 6-1,9, scalloped, symmetrical, ellip-

tic; laterals smaller, 50-65 x (19-) 24-32 mm, gib-

bous, length-breadth ratio 1,3-1, 9, asymmetrical; fi-

nely appressed hirsute above but dull green; tertiary

venation visible in fresh leaves if held against the

light; lower surface finely woolly grey to white with

longer fulvous hairs on the primary veins, small yel-

low glands visible; both terminal and lateral leaflets

have a hairy midrib above; lowest leaves of the plant

are obovate, apiculate. Stipules 13 x 6 mm, widest at

middle, free, rapidly senescent, semipatent, tip re-

a, holotype; b, inflorescence; c, habit and habitat.

Bothalia 16,1 (1986)

17

curving, glabrous inside, pubescent outside, hairier

along the margin. Petioles 3 mm long. Racemes up to

125 mm long, floriferous section 55 mm long, elon-

gating with anthesis, 20-25-flowered. Flowers 13-14

mm long; red and yellow (Fig. 8b); bracts 8 mm

long, caducous. Calyx 8 mm long, lobes equal; teeth

more or less equal to the tube. Standard 13 x 9 mm;

claw 3 mm long; emarginate sides recurved; bright

brick red on the back, venation black, base above

the claw yellow; glands present, yellow; appendages

present, hooded, free from auricles. Wing petals 13

x 4 mm, cultrate, longer than the keel; orange, suf-

fused with pink. Keel petals 12 mm long, 5 mm wide

at broadest point, sparsely covered in yellow glands.

Staminal sheath 10-12 mm long, free stamen genicu-

late. Pistil 11 mm long; ovary 5 mm long; height of

curvature 4 mm high; stigma small, exserted. Nec-

tary present. Fruits 15-16 x 10-11 mm, with 3-4 mm

long beak; staminal sheath shrivelled but persistent

during fruiting, densely covered in long, golden, ap-

pressed hairs. Seeds unknown.

Eriosema umtamvunense is endemic to the rolling

grasslands decking the plateaux on either side of the

Umtamvuna Gorge (Fig. 8c). So far it has been re-

corded only above 300 m. It is restricted, with E.

latifolium, to Acocks’s Pondoland Coastal Plateau

Sourveld (Fig. 9). Flowering takes place in Novem-

ber and December.

FIG. 9. — Known distribution of Eriosema umtamvunense.

TRANSKEI. — 3130 (Port Edward): near Ku-Mankenbeya in

the Umizizi area (-AA), Stirton 5624 (K; PRE).

NATAL— 3030 (Port Shepstone): Blencathra Farm (-AA),

Stirton 8063 (PRE); Izingolweni Hill (-AA), Hilliard 1709 (NU);

7 km from Port Edward to Izingolweni (-AA), Stirton 8099

(PRE); 10 km from Izingolweni to Port Edward (-CC), Stirton

1389 (K; PRE); Beacon Hill East (-CC), Strey 7242 (NU); Sky-

line Farm (-CC), Schrire 320 (NU; NH); Skyline Farm (-CC),

Germishuizen 1713 (PRE); 15 km from Izingolweni to Port Ed-

ward (-CC), Stirton 1388, 1391 (PRE); Umtamvuna River (-CC),

Nicholson 1306 (PRE); Umtamvuna Nature Reserve (-CC), Van

Wyk 5149 (PRU).

E. umtamvunense is a very distinctive, locally

abundant Eriosema, yet like E. latifolium it has been

collected rather infrequently. Strey 7242, collected

as recently as 1967, is the first record of the species.

It seems to have been missed by all the early collec-

tors. This is not surprising as it is distributed on top

of the escarpment and generally grows in grassland

that would have been largely inaccessible to early ex-

plorers. One wonders what other treasures are still

to be discovered in the Umtamvuna Gorge and its

escarpment.

This species has been consistently lumped with E.

squarrosum (no. 1), E. dregei (below), E. latifolium

(no. 7) and E. luteopetalum (no. 3). It differs from

all of these species in its red and yellow flowers and

golden-haired calyces; occasional yellow morphs can

occur. The latter species all have yellow flowers and

silver-haired calyces. From E. preptum (no. 8) it can

be separated by its much larger flowers and fruits

and by the wing petals exceeding the keel petals.

5. Eriosema dregei E. Mey., Comm. 129 (1836).

Rhynchosia dregei (E. Mey.) Steud., Nom. 2: 454

(1841). Eriosema squarrosum (Thunb.) Walp. var.

dregei Benth. ex Harv., FI. Cap. 1: 260 (1862). E.

zeyheri E. Mey. var. dregei (E. Mey.) Bak. f. in J.

Bot., Lond. 33: 147 (1895). Lectotype: Natal, Um-

zimkulu River, Drege s.n. V.c. 18 (P; K, isolecto.).

Suffrutex up to 400 mm high. Stems 4-10,

branched from the base, finely appressed white pu-

bescent. Leaves trifoliolate; upper leaflet 60-70 x

30-33 mm, ovate to narrowly ovate; laterals some-

what gibbous, 45-50 x 21-25 mm; finely grey woolly

beneath, finely sericeous, greyish green above; rha-

chis channelled. Stipules 12 x 4 mm, free, senescing

before leaves expand. Petioles 3-4 mm long. Race-

mes axillary, up to 67-flowered, exceeding leaves,

55-60 mm long. Flowers 14 mm long, yellow; bracts

5x2 mm, boat-shaped. Calyx 7-8 mm long, lobes

equal; teeth 3-4 mm long, shorter than the tube,

finely covered in grey hairs and minute yellow

glands. Standard 14 x 9 mm, obovate, emarginate;

claw 3 mm long; auricles present 4 mm apart; back

of standard finely pubescent and densely covered in

minute yellow glands; appendages present, fused,

hooded, extending to auricles. Wing petals 13 mm

long, up to 3 mm wide, slightly longer than the keel

blades, basal part held horizontally, but other edges

drooping. Keel petals 12,5-13,0 mm long, up to 7

mm wide, densely covered in yellow glands. Stami-

nal sheath 11 mm long, tenth stamen free. Pistil 11

mm long; ovary 4 mm long; height of curvature 4

mm, style thickened at point of flexure. Nectary

present, margin erase. Fruit 15 x 11 mm, beak 2 mm

wide; chestnut-brown covered in soft 2 mm long,

red-brown hairs. Seeds unknown. Fig. 10.

Eriosema dregei is endemic to the low-lying

coastal dune and riverine grasslands, below 200 m

altitude, and extending from Port Edward in Natal

to the Mkambati River Mouth in the Transkei (Fig.

11). The area between these localities and Port St

Johns is little explored and this species can be ex-

pected to occur there. Flowering takes place be-

tween August and October.

NATAL. — 3130 (Port Edward): Port Edward (-AA), Stirton

5671 (PRE), 8068 (K; PRE; NU); Germishuizen 1532, 1740

(PRE); Ngwenya 214 (NH).

TRANSKEI.— 3129 (Port St Johns): 4 km inland from Port

Grosvenor (-BD), Strey 8905 (K; PRE); Mkambati (-BD), Van

18

Bothalia 16,1 (1986)

»■

FIG. 10. — Representative specimen of Eriosema dregei (Stirton

5671).

Wyk 1551 (PRE); Msikaba River Mouth, Venter & Vorster 204

(PRE). 3130 (Port Edward): Mzamba River Mouth (-AA), Stir-

ton 5604 (K; PRE). Without precise locality: Umzimkulu River,

Drege V.c. 18 (K; P).

Eriosema dregei is most commonly confused with

E. luteopetalum (no. 3) from which it differs in its

silvery stems and immature leaves, narrower wing

and keel petals, smaller caducous flower bracts, and

the silvery appressed pubescence of the upper sur-

face of the leaflets. In E. luteopetalum the stems and

leaves are russet- or golden-brown, the flower bracts

are nearly twice as long and wider, and the pubes-

cence on the upper surface of the leaflets is short,

appressed and yellowish.

Like E. luteopetalum this species is also rather at-

tractive, especially when in full flower. These small

silvery plants stand out quite strikingly in the coastal

grasslands where they occur. E. dregei is sympatric

with E. acuminatum (no. 2) but is allopatric with E.

luteopetalum (no. 3) E. umtamvunense (no. 4) and E.

lati folium (no. 7).

Strey 8905, collected near Port St Johns, has a very

characteristic facies and may turn out to be a new

species. This distinctive plant should be searched for

and compared with E. dregei.

6. Eriosema rossii C. H. Stirton, sp. nov., affini-

tate incerta.

Herba perennis ad 350 mm alta, vere florens.

Caules erecti, basi ramificantes, subtiliter pubes-

centes pilis reflexis. Folia trifoliolata, 45-65 x 25-30

mm, elliptica; lateralia minora asymmetrica. Stipu-

lae 15-18 mm longae, liberae. Racemi axillares,

30-60-florati ut pseudospicati congesti, foliis aequi-

longi. Flores 12 mm longi, aurantiaci venis rubribus

flavique; bracteo ad 12 mm longo. Calyx 7 mm

longo. Vexillum 11 mm longum, 6-7 mm latum,

anguste obovatum, auriculatum, reflexum. Petala

carinae breviora quam alae. Vagina staminalis 1 mm

longa. Gynoecium 7-8 mm longum; ovarium seri-

ceum. Fructus 15 mm longus, 10 mm latus, pilis

patentibus, 2-3 mm longis vestitus.

TYPE. — 3030 (Port Shepstone): 1 km from Hluta-

kungo on road to Highflats (-AD), Stirton 1205

(PRE, holo.; K, iso.).

Erect herb up to 350 mm high, arising from a short

vertical rootstock with constricted outline; lateral

branches very constricted, horizontal. Stems up to

10, covered in semi-patent, downward pointing hairs

and short appressed hairs. Leaves trifoliolate; ter-

minal leaflet 45-65 x 25-30 mm, elliptic; laterals

38-55 x 16-20(-27) mm, gibbous, length-breadth

ratio 1, 4-2,0; densely covered in fine erect silky

hairs; lower surface covered with numerous small

yellow glands, margins and veins appressed hairy,

finely pubescent between; upper surface dark green,

lower pale green. Stipules 15—18 mm long, up to 6-8

mm wide, free, green, erect, clasping the stem.

Racemes axillary, 30-60-flowered, hidden by the

leaves during anthesis but elongating thereafter.

Flowers 12 mm long, pale orange, equal in length to

the subtending bracts. Calyx 7 mm long, lobes equal,

tube 2 mm long, teeth triangular, covered in golden

patent hairs. Standard 11 mm long, up to 6-7 mm

wide, narrowly obovate, auriculate, clawed, apex

truncate; back red with darker venation and packed

with yellow glands, inside pale orange with red vena-

tion and yellow nectar guide. Wing petals 10 x 2,3

mm, claw 3 mm long; auriculate, triangular; exceed-

ing the keel blades; pale pink with red venation and

Bothalia 16,1 (1986)

19

with a few small yellow glands and hairs along the

main vein. Keel petals lined with red along the lower

margins, densely covered with yellow glands. Stami-

nal sheath 7 mm long, tenth stamen free. Pistil 7-8

mm long; ovary silky; height of curvature 2 mm;

stigma large, capitate. Fruits 15 x 10 mm, very

shortly beaked, bracts still persistent during fruiting;

constricted, clothed in 2-3 mm long reddish hairs.

Seeds unknown. Fig. 12.

-noou C H <Ust^

c.u. mo

SS3& HERBARIUM PRETORIA

FIG. 12. — Isotype of Eriosema rossii (Stirton 1205).

Eriosema rossii is restricted to the higher-lying

Ngongoni Veld (Acocks’s Veldtype 5) of southern

Natal and the eastern Transkei (Fig. 13). The Um-

komaas River Valley bisects Acocks’s Veldtype 5

and it may be significant that many legumes found in

the southern portion of this veld type have not been

recorded from the area north of the Umkomaas

River. Another example in Eriosema is E. populifo-

lium Harv.

NATAL. — 3029 (Kokstad): Ingeli Forest area (-DA), Stirton

8113 (PRE). 3030 (Port Shepstone): 10 km from Highflats to Um-

zinto (-AB), Stirton 8202 (PRE); 1 km from Hlutakungo to High-

flats (-AD), Stirton 1205 (K; PRE); 4 km from Umsawoti to

Highflats (-AD), Stirton 750, 751 (K; PRE); Hlutakungo (-AD),

Stirton 5563 (K; PRE); Umtwalumi Falls (-AD), Stirton 743

(PRE); Vernon Crookes Nature Reserve (-BC), Balkwill & Man-

ning 980, 988 (NU). Umgaye (-BC), Rudatis 559 (BM), 717 (BM;

STE). Without precise locality Krauss 475 (K; BM; US).

TRANSKEI. — 3029 (Kokstad): Malowe Mountain (-BD), Ty-

son 2698 (NBG; PRE), 5846 (PRE). 3129 (Port St Johns): 13 km

NE of Ludongo Store (-AD), Acocks 13425 (PRE); Ntsubane

Forestry Station (-BC), Galpin 10994 (PRE).

Eriosema rossii is named in honour of Dr James

Ross in recognition of his important contribution to

our knowledge of southern African legumes, par-

ticularly the subfamilies Caesalpinioideae and Mim-

osoideae.

Although this species is fairly common within its

distribution range, it was until recently poorly repre-

sented in herbaria. With its large stipules and small

compact inflorescences it is generally easily separ-

ated from the other species in the E. squarrosum

complex.

There is, however, a group of plants which may

form part of this species. I have previously anno-

tated this group in various herbaria as E. superposi-

tum mss. It is allopatric with E. rossii and is separ-

ated from it by its long-peduncled, few-flowered

box-like inflorescences held high above the sparsely

pubescent subtending leaves. But since the interven-

ing area between the two ranges has not been col-

lected I do not wish to recognize it formally until

more is known about the variation present in both

groups.

The following brief description may be useful to

collectors who might be fortunate to find plants of

this unnamed group which grows in small scattered

colonies.

Erect perennial arising from a daucate rootstock

with side branches arising from just below the stylo-

podium; younger rootstocks constricted. Stems 1-5,

branching once or twice near base, densely clothed

in 1-1,5 mm long, fulvous, semi-patent, downward

pointing hairs with shorter hairs interspersed.

Leaves trifoliolate, first leaves regularly narrowly

ovate and unifoliolate; terminal leaflet 45-70 x

12-20 mm, sparsely covered in short, stiff, semi-

erect hyaline hairs, under surface sparingly pubes-

cent with numerous yellow glands present. Stipules

9-15 x 3-4 mm. Racemes axillary, with up to 12

20

Bothalia 16,1 (1986)

flowers, box-like, congested at the apex of a long pe-

duncle, greatly overtopping the leaves, 13 x 20 mm;

peduncle 40-100 mm long. Flowers 7-9 mm long,

longer than the subtending flower bract.

NATAL. — 2930 (Pietermaritzburg): Inchanga (-DA), Stirton

387b (K; PRE); Key Ridge (-DC), Stirton 555, 1126, 1365, 5077

(K; PRE); 3 km from Key Ridge to Durban (-DC), Stirton 5546

(K; PRE); Botha’s Hill (-DC), Stirton 542 (K; PRE); near Dur-

ban (-DD), Gerrard 423 (TCD); Wentworth (-DD) Ward 6112

(PRE; NU), 5207 (NU) 6474 (NU); Treasure Beach (-DD),

Bluff, Ellery 39 (NU). 3030 (Port Shepstone): Umkomaas (-BB),

Stirton 8044 (PRE). Without precise locality: Sanderson 278, 378.

(TCD); Hutton s.n. (TCD).

This taxon has disappeared rapidly from the area

between Durban and Key Ridge and is now found

only in a few isolated patches of the natural grass-

land that has not yet been built on or been destroyed

through overgrazing.

7. Eriosema latifolium (Benth. ex Harv.) C. H.

Stirton, comb, et stat. nov.

Eriosema squarrosum (Thunb.) Walp. var. latifolium Benth. ex

Harv., FI. Cap. 2: 260 (1862). Eriosema zeyheri E. Mey. var. lati-

folium Benth. ex Bak. f. in J. Bot., Lond. 33: 147 (1895). Lecto-

type: Natal, ‘in graminosis circa stationem St. Andrews dictam’,

Tyson 2834 (SAM; K, isolecto.).

Erect suffrutex up to 1 m tall. Stems up to 10,

branching from the lower nodes, velvety. Leaves tri-

foliolate, basal leaves unifoliolate, 50-75 x 35—45

mm, symmetrical, obovate but also elliptic; laterals

smaller, 45-60 x 20-30 mm, asymmetrical, gibbous,

length-breadth ratio 1,3-1 ,5; densely pubescent

above, dull greenish, densely woolly grey-white be-

neath with fulvous veins; densely glandular below

but hidden beneath the hairs. Stipules 8— 9(— 11) x

5-6 mm, broadly ovate, tip somewhat falcate, free,

clasping the stem. Petiole 3-5 mm long. Racemes ax-

illary, 30-45-flowered, 11-22 mm long, held well

above foliage; peduncles 30-75 mm long. Flowers

(9)10-11 mm long, yellow; bracts 6-7 mm long, 2

mm wide, boat-shaped. Calyx 5, 0-5 ,5 mm long,

lobes equal; teeth 2 mm long, equal, shorter than the

tube, long tawny-haired becoming appressed on the

tube, glandular. Standard 11 mm long, narrowly

obovate, clawed, weakly auriculate, glandular and

pubescent on the back; appendages weakly devel-

oped, fused, thinly ridged, extending on each side

downwards to the auricles but ending 1,5 mm away

from them. Wing petals 11,0-11,5 x 2, 5-3,0 mm,

narrowly cultrate, slightly hairy along the base,

somewhat pouched near the poorly developed auri-

cle, sparsely glandular, longer than the keel blades.

Keel blades 8, 5-9,0 mm long, up to 3, 5-4,5 mm

wide; densely glandular, hairy along the base. Stami-

nal sheath 8 mm long; free stamen geniculate; pollen

variable in size. Pistil 7 mm long; ovary 2,8-3 ,0 mm

long, densely pubescent; height of curvature 3-4

mm; stigma capitate. Fruits unknown. Fig. 14.

This species is endemic to southern Natal and the

north-eastern Transkei (Fig. 15). It occurs in open

grassland, both near riverine and mountain forests.

Flowering takes place in October and November.

NATAL.— 3030 (Port Shepstone): 18 km from Izingolweni to

Port Edward, (-CC), Stirton 1385 (K; PRE).

TRANSKEI. — 3029 (Kokstad): 86,5 km from Lusikisiki to Port

Edward (-DD), Grobbelaar 2321 (PRU); Bizana (-DD), Stirton

5599 (PRE; K). 3129 (Port St Johns): St Andrews Station (-BC),

FIG. 14. — Representative specimen of Eriosema latifolium

(Strey 10132).

Tyson 2834 (K; NBG); Goss Point f-BD), Strey 10132 (PRE; K).

Without precise locality: near Umkwani River, Tyson 2633

(NBG); Anonymous 559 (W).

Although described by Harvey as long ago as 1862

this species has been collected only rarely. It occu-

pies Acocks’s Veld Type 3, his Pondoland Coastal

Plateau Sourveld. This veld type has until recently

been little explored and I am certain that once a full

Enumeration has been made of its constituents it will

receive the recognition it deserves as an area of

endemism.

Eriosema latifolium is closely allied to E. dregei

(no. 5) and E. luteopetalum (no. 3), two other yel-

low-flowered suffrutices from the same general re-

gion. It differs from these two species in its much

smaller flowers, the appendages on the standard be-

ing well-free of the auricles, its shorter stipules and

narrowly oblong racemes.

The presence of variably sized pollen grains

strongly suggests that this species may be of hybrid

origin. It is perhaps significant that this species oc-

curs as scattered individuals or small colonies. The

few fruits that have been found contained shrivelled

seeds only.

8. Eriosema preptum C. H. Stirton in Bothalia

13:323 (1981). Type: Natal, 2930 (Pietermaritz-

burg): Scottsville, Pietermaritzburg (-CB), Stirton

1242 (PRE, holo.; K, iso.).

Bothalia 16,1 (1986)

21

Perennial herb or suffrutex 200-600 mm tall.

Stems 1-15, clothed in short white hairs with longer

hairs interspersed. Rootstock with long stylopodium,

thin and beaded when young, becoming wavy or

constricted but finally daucate when mature. Leaves

trifoliolate, basal leaves usually unifoliolate, 45-60

x 20-30 mm; laterals smaller, less symmetrical, el-

liptic to narrowly obovate, if unifoliolate then obo-

FIG. 16. — Representative specimen of Eriosema preptum (Stir-

ton 1244).

vate, apex subacute, base cuneate; sparsely pubes-

cent above, densely woolly below with veins promi-

nent due to dense covering of longer appressed

hairs, glandular. Stipules 8-14 mm long, free. Race-

mes axillary, (8-)25-35-flowered, overtopping the

subtending leaves. Flowers 6-7 mm long, up to 3 mm

wide, orange with red veins or yellow-orange; bracts

4-6 mm long, rapidly caducous. Calyx 4 mm long,

tube 2 mm long; keel lobe up to 3,5 mm long, teeth

triangular. Standard 6-10 x 6 mm, emarginate with

well developed downward curving auricles, clawed;

back hairy and glandular; appendages present,

fused, extending from auricle to auricle just above

the apex of the claw. Wing petals 8-9 x 2,0-2,75

mm, auriculate, upcurving, longer than keel. Keel

petals 6-7 mm long, up to 3-4 mm wide, pouched,

gland-dotted. Staminal sheath 6 mm long; tenth sta-

men free. Pistil 6-7 mm long; ovary 2,5 mm long,

densely hairy; style thickest at point of flexure, hairy

for | its length, height of curvature 2 mm. Nectary

present, 0,2-0, 3 mm high, margin wavy. Fruits 10-13

x 8-10 mm, sericeous. Seeds grey or light brown,

with speckles or blotches. Fig. 16.

Eriosema preptum is endemic to Natal and ex-

tends some 100 km inland from the coastal belt (Fig.

17). It occurs in Acocks’s Coastal Forest and Thorn-

veld (VT1), Ngongoni Veld (VT5) and his Zululand

Thornveld (VT6). It favours sandy, well drained

sites along roadsides and ditches but is also com-

monly found in regularly burned grassland. Flow-

ering extends from September to February but oc-

curs mainly in October.

NATAL. — 2830 (Dundee): Scottspoort (-CC), Thode 4418

(STE). 2831 (Nkandla): 6 km S of Hlabisa (-BB), Codd 2003 (K;

PRE); 10 km from Eshowe to Gingindlovu (-CD), Stirton 5349

(PRE). 2930 (Pietermaritzburg): Scottsville (-CB), Stirton 1242,

1410, 5516 (K; PRE); Goossens 126 (G); behind Oribi Aero-

drome (-CB), Stirton 1244 (K; PRE); Camperdown (-DA), Stir-

ton 5542 (PRE); 5 km from Table Mountain to Pietermaritzburg

(-DA), Stirton 1032 (K; PRE). 2931 (Stanger): 43 km from

Stanger to Mtunzini (-AB), Stirton 407, 1001 (K; PRE); Tugela

Monument (-AB), Grobbelaar 1810 (PRU); Umhlali (-AD),

Meebold 13364 (NY); 10 km from Durban to Stanger (-AD), Stir-

ton 1254 (K; PRE); Gingindlovu (-BA), Stirton 1256 (K; PRE);

near Compensation (-BA), Stirton 1160 (K; PRE); Kanyile

Monument (-CD), Grobbelaar 2325 (PRU). 3030 (Port Shep-

stone): Pumula (-BB), Stirton 10343 (NU); Clydesdale (-BD),

Tyson s.n. (NBG); 8 km from Eston to Winklespruit (-BB), Stir-

ton 1122 (K; PRE); 19 km from turn-off to Oribi Gorge Hotel on

road to Paddock (-CA), Germishuizen 1690 (PRE); Southbroom

(-CB), Schrire 318 (NU); 3 km from Port Shepstone to Margate

(-CD), Stirton 8050 (PRE).

Eriosema preptum hybridizes with E. cordatum E.

Mey. and E. salignum E. Mey. (Stirton 1981b). The

hybrid progeny are rather robust and are well repre-

sented in herbaria. Their presence in herbaria has

however obscured the boundaries of what are three

quite distinct species.

This species is related to E. rossii (no. 6) and its

variants but is separated by its fewer-flowered well

exserted racemes, much smaller, rapidly caducous

flower bracts, smaller flowers and more woolly leaf-

lets.

SPECIES EXCLUDED FROM THE COMPLEX

During the course of this study I have come across

several specific epithets which have been attributed

22

Bothalia 16,1 (1986)

to the E. squarrosum complex. I have not been able

to verify the identity of all of these names. The fol-

lowing notes should however clarify the position of

some of them.

1. Eriosema dregei Meissn., in Krauss ( Flora 27:

357, 1844), nomen. Krauss never published this

name but used it to identify material he saw of Erio-

sema parviflorum E. Mey.

2. Eriosema puberulum Eckl. & Zeyh. ( Enum .

256, 1836) = Rhynchosia puberula (Eckl. & Zeyh.)

Steud.

3. Eriosema reticulatum E. Mey. {Comm. 129,

1836) = Rhynchosia sp. There is a specimen of E.

salignum E. Mey. in P, collected by Drege, which is

annotated by E. Meyer as E. reticulatum E. Mey.

This is clearly a misidentification as the specimen

does not match the protologue of E. reticulatum.

4. Eriosema rogersii Schinz = Rhynchosia bar-

bertonensis C. H. Stirton, nom. nov. Eriosema ro-

gersii Schinz in Vjschr. naturf. Ges. Zurich 71: 138

(1926); non R. rogersii Schinz in Vjschr. naturf. Ges.

Zurich 71: 137 (1926). Type: Transvaal, Barberton,

Thorncroft leg. Rogers 19157 (Z, holo.; BM, frag-

ment).

5. Eriosema sericeum Eckl. & Zeyh. {Enum. 256,

1836). = Rhynchosia sp.

6. Eriosema transvaalense Moss ex P. Glover (5.

Afr. J. Sci. 34: 247, 1937), nomen.

7. Eriosema trinerve E. Mey. {Comm. 130, 1836).

I have seen only one specimen annotated as such by

E. Meyer. This specimen is not an Eriosema. Dr. R.

M. Harley (Kew) has kindly named it as Micromeria

sp. (Lamiaceae). There is a note written by the late

Dr J. Raynal in the Paris Herbarium (27-6-1963) to

suggest that ‘this is probably an incorrectly labelled

plant’. The protologue of this species is too vague

and as it could be applied to any of a number of

species it cannot be applied until a specimen so

named is found.

8. Eriosema villosum (Meissn.) C. A. Sm. ex

Burtt Davy {FI. Transv. 2: 413, 1932). = Rhynchosia

villosa (Meissn.) Druce.

ACKNOWLEDGEMENTS

I thank the Directors and staff of all the cited her-

baria who supplied material and other information;

Messrs G. Germishuizen, G. B. Harding and B. Sch-

rire who accompanied me on field trips to study ma-

terial in situ; Mrs A. Romanowski for the photo-

graphs of type specimens, and Mr M. Svanderlik

(Kew) for photographs of the distribution maps.

UITTREKSEL

Alle digbehaarde Suider-Afrikaanse spesies van

Eriosema is oor die jare gewoonweg onder E. squar-

rosum (Thunb.) Walp. geplaas. Hierdie studie klaar

die identiteit van E. squarrosum op; drie nuwe tak-

sons word erken: E. luteopetalum C. H. Stirton, E.

rossii C. H. Stirton en E. umtamvunense C. H. Stir-

ton; twee nuwe kombinasies word gemaak: E. latifo-

lium (Benth. ex Harv.) C. H. Stirton en E. acumina-

tum (Eckl. & Zeyh.) C. H. Stirton; en E. dregei E.

Mey. word herstel. Die spesie E. preptum C. H. Stir-

ton, wat reeds beskryf is, behoort ook tot hierdie

kompleks. Rhynchosia barbertonensis C. El. Stirton

word as nuwe naam vir E. rogersii Schinz gegee.

REFERENCES

STIRTON, C. H. 1978. The Eriosema cordatum complex. I. The

E. populifolium group. Bothalia 12: 395 — 404.

STIRTON, C. H. 1981a. The Eriosema cordatum complex. II.

The Eriosema cordatum and E. nutans groups. Bothalia 13:

281-306.

STIRTON, C. H. 1981b. Natural hybridization in the genus Erio-

sema (Leguminosae) in South Africa. Bothalia 13:

307-315.

STIRTON, C. H. 1981c. Studies in the Leguminosae — Papilio-

noideae of Southern Africa. Bothalia 13: 317-325.

Bothalia 16,1: 23-27 (1986)

Studies in the genus Riccia (Marchantiales) from southern Africa. 2.

A new species of the section Pilifer : R. sarcosa

O. H. VOLK* and S. M. PEROLD**

Keywords: dorsal epithelium, Marchantiales, Riccia sarcosa sp. nov., section Pilifer

ABSTRACT

Riccia sarcosa Volk & Perold, a new species endemic to southern Africa is described. This species belongs to the

section Pilifer Volk (1983), which now comprises 8 species characterized by the dorsal epithelium consisting of

loose cell pillars. R. sarcosa is recognized by the distinct white margins of the thallus, by inconspicuous hyaline

scales that do not extend above the thallus margins and by the spore ornamentation consisting of round, deep-set

areolae or foveae.

Riccia sarcosa Volk & Perold, sp. nov. sectionis

Pilifer, R. duthiae similis sed thallo marginibus albis

sporisque foveolis bene impressis differt.

Thallus dioecius mediocris, lobis ad 10 mm longis,

1,5-3 mm latis, 1-1,5 mm crassis, obovatis, apice

sulcatis; superficies dorsalis velutina, virella, leviter

concava, marginibus albis. Frons in sectione trans-

versali: stratum piliferum (epithelium) pilis liberis

gradatim contractis 3-4 cellulis seriatis; chloren-

chyma columnis 8-10-cellularibus canalibusque aeri-

feribus tenuibus. Squamae imbricatae, hyalinae,

margines thalli aegre superantes. Sporangia dorsa-

liter protuberantia. Sporae 90-130 pm diametro, tri-

angulo-globosae polares, ochraceae, ala angusta,

6-12 foveolis in diametro. Chromosomatum nume-

rus n = 8 (Bornefeld 1985).

TYPE. — Cape, 3224 (Graaff-Reinet): Aberdeen,

next to road R57, 2 km north-east of junction with

R61, at shallow edges of vleis temporarily damp or

occasionally inundated (-AC), 1981.04.11, Volk

81-274b (M; PRE), associated with R. duthieae,

Marsilia burchellii, Crassula spp., Ruschia spp.,

Chloris virgata and thick layers of Cyanophyceae.

On clayey soil, pH 6. 5-6. 9.

Thallus dioecious, perennial, gregarious or in in-

complete rosettes up to 20 mm across, medium-

sized, lobes up to 10 mm long, 1,5-3 mm broad,

1-1,5 mm thick (Fig. 1.1, 1.2; Table 1), obovate,

narrow at base, widening distally, occasionally sin-

gle, usually 2 to 3 times furcate, some segments

branching again close to apex; branches variously di-

vergent: main branches often parallel at initial di-

chotomy, subsequent branches usually more widely

divergent (Fig. 1.1); in dense populations segments

mostly elongated and small; dorsal surface when dry

whitish green, somewhat felt-like, apex and sides

with hyaline scales inflexed; when turgid, bright

green, glistening, velvety, older parts and along mar-

gins white, sulcate at apex and shortly emarginate,

slightly concave to flat, scales only prominent at

* Botanische Anstalten d. Univ. Wurzburg D 8700, Germany,

B.R.D.

** Botanical Research Institute, Department of Argiculture &

Water Supply, Private Bag X101, Pretoria 0001, RSA.

apex. Thallus branches in transverse section 1,5 to

2,5 times as broad as thick; dorsal surface with shal-

low depression in centre, margins acute; flanks

steeply ascending, sloping slightly outwards near

apex, pale green, sometimes flecked with reddish

purple; ventral surface slightly convex, greenish;

dorsal covering of epithelial cells (Figs 1.2, 1.3;

2. 2-2. 4) about the thickness of transverse sec-

tion, consisting of loose cell pillars varying from

130-7 S0-22O pm in length (Tables 1 & 2), each pillar

with 3-4 empty, delicate and inflated, hyaline cells,

basal cells up to 80 pm wide and occasionally giving

rise to unicellular globular outgrowths in spaces be-

tween pillars, thus reducing sizes of air-spaces above

intercellular channels (Fig. 2. 2-2. 4), pillars tapering

to smaller terminal apical cells about 33 pm wide,

their shapes conical, mammillate or globular, giving

dorsal surface of thallus a somewhat papillose ap-

pearance when viewed from above (Fig. 2.1); assimi-

lation tissue (chlorenchyma) the thickness of

section, 320-570-640 pm thick (Fig. 1.2; Table 1),

consisting of columns or plates one cell thick, 8-10

cuboidal cells high, and enclosing 4-6-8-sided air-

channels (Fig. 1.4); storage tissue ^ the thickness

of section, 320—700-470 pm thick (Fig. 1.2; Table 1),

with closely packed, rounded or hexagonal cells, av-

erage diameter 50 pm; rhizoids arising from flat epi-

dermal cells of ventral surface or from bases of

scales, about 30 pm wide, mostly smooth. Scales

closely imbricate, semicircular, 1 000-1 500 pm long

and 600 pm broad, hyaline or white, basal cells often

reddish purple; almost as broad as thickness of thal-

lus, scarcely projecting above thallus margin; cells

oblong-hexagonal, 110 pm long x 35 pm wide,

smaller at margin, 35 pm long x 30-40 pm wide, cell

walls straight (Fig. 1.2, 1.5). Antheridia not seen.

Archegonia near middle of thallus, their necks pur-

ple-brown. Sporangia bulging dorsally, 0,6-0, 8 mm

across, containing 250-450 spores. Spores 90-130 pm

in diameter, triangular-globular, polar; ochre-

coloured to dark brown, semi-transparent becoming

opaque with age; wing narrow 2,5 pm wide, project-

ing slightly more at marginal angles, incised or with

pore in spore wall, margin smooth to finely crenu-

late; ornamentation on distal face incompletely re-

ticulate: in centre of spore 5-7 smaller areolae about

10 pm wide, surrounded by larger 25 pm wide areo-

24

Bothalia 16,1 (1986)

FIG. 1 .—Riccia sarcosa (Volk 81-274b and Volk 81-292b , M, PRE). Structure of thallus, spores and chromosomes. 1, different

habits; 2, transverse section of the thallus; 3, epithelial cells; 4, horizontal section through chlorenchyma; 5, part of ventral

scale, 6, ornamentation of distal spore face; 7, chromosomes. (1-6, by O. H. Volk; 7, by T. Bornefeld). Scale bar on 1 - 4

mm; 2 = 500 pm; 3-5 = 100 pm; 6 = 50 pm; 7 = 1 pm.

Bothalia 16,1 (1986)

25

FIG. 2. — Riccia sarcosa (Volk 81-274b and Volk 81-292b, M, PRE). Thallus, epithelium and spores. 1, surface view of distal part of

thallus; 2 & 3, epithelium viewed from above, showing interspatial outgrowths; 4, oblique view of same; 5-7, distal face of

spores. [1-4, SEM micrographs by O. H. Volk; 5, 6, LM (light microscope) photographs by O. H. Volk; 7, LM photograph

by S. M. Perold.] Scale bars on 1-4 = 50 pm; diameter of spores on 5-7 ca 100 pm.

lae, their borders thick, ring-like, often not reaching

to wing (Figs 1.6; 2. 5-2. 7; 3.5, 3.6), some spores

with 6-12 small pits or foveae across diameter; prox-

imal face with triradiate mark distinct, areolae of

facets (Fig. 3. 1-3.4) about 7,5 pm wide, often with

raised papillae at nodes; an area 10 pm wide be-

tween areolae and margin of spore, as well as narrow

strips flanking parts of triradiate mark, without or-

namentation (Fig. 3. 1-3.4). Chromosome number n

= 8 (Bornefeld 1984); the letters A-E (Fig. 1.7)

identify the chromosomes according to Bornefeld

(1984).

R. sarcosa is distinguished from the other seven

species in the section Pilifer by the distinct white

margin of the thallus, by its mostly inconspicuous

scales that do not project above the thallus margin,

and by the dorsal epithelial cell pillars which often

TABLE 1. — R. sarcosa, measurements on transverse section (cul-

tivated plants of Volk 81-274b and Volk 81-292b)

Breadth of thallus : 2, 1-2, 3-2, 7 mm

Thickness of thallus : 1,0-1, 3-1, 5 mm

Thickness of epithelium : 130-180-220 pm, ca thickness of

thallus

Thickness of chlorenchyma : 320-510-640 pm, ca ^thickness

of thallus

Thickness of storage tissue : 320-400-470 pm, ca^- thickness of

thallus

26

Bothalia 16,1 (1986)

have inflated basal and smaller terminal cells. The

ornamentation of the spore wall is also distinctly dif-

ferent with deep-set, ringed areolae or foveae.

The type specimen, Volk 81-274b , was part of a

mixed collection of R. duthieae (Volk 82-274) and

was only recognized as a new species when plants

from this collection were cultivated. A further speci-

men of R. sarcosa was isolated from a gathering of

R. albomarginata, ( Volk 81-292) (see Volk 1983)

collected in the Willem Pretorius Wildtuin about 25

km E of the Park office on shallow soil over flat rock

plates, temporarily wet and growing together with

Crassula spp., Ruschia indurata (L. Bol.) Schwant.,

Anacampseros spp., Oropetium spp., Riccia volkii

etc. pH of soil 6,2. A third specimen of this rare

species was recently collected by J. M. Perold 10 km

S of Ladybrand on shallow soil overlying a flat rocky

outcrop. Fig. 4.

FIG. 3.— Riccia sarcosa (Volk 81-292b, M, PRE). Spores. 1 & 2, proximal face; 3, areolae on one of proximal facets;

4, viewed from side; 5, distal face; 6, areolae on distal face. (SEM micrographs by S. M. Perold). Scale bars =

50 (rm.

Bothalia 16,1 (1986)

27

TABLE 2. — R. sarcosa, size (in |xm) of the 3-4 cells of the epithelial pillars and of cells of the chlorenchyma in transverse section

(cultivated plants of Volk 81-274b and Volk 81-292b)

OFS. — 2827 (Senekal): Willem Pretorius Game Reserve

(-AC), Volk 81-292b (M; PRE). 2927 (Maseru): Ladybrand, 10

km S (-AB), J. M. Perold 35 (PRE).

FIG. 4. — Map showing distribution of R. sarcosa.

ACKNOWLEDGEMENTS

The authors wish to thank Dr. habil. T. Borne-

feld, Am Reele 1, D-8706 Hochberg, Wurzburg,

Germany, for the chromosome counts and figures.

UITTREKSEL

Riccia sarcosa Volk & Perold, ’n nuwe spesie ende-

mies in suidelike Afrika, word beskryf. Hierdie spesie

behoort tot die seksie Pilifer Volk (1983), wat nou 8

spesies behels en wat gekenmerk word deur ’n dorsale

epiteel wat uit los selpilare bestaan. R. sarcosa word

erken aan die wit rande van die tallus, aan die onop-

vallende deurskynende skubbe wat nie by die tallus-

rand verbysteek nie en aan die ornamentasie van die

spore wat uit ronde diep ingesonke areole of putjies

bestaan.

REFERENCES

BORNEFELD, T. 1984. Chromosomenanalyse der Gattung Ric-

cia von Slid- und SW-Afrika und allgemeine Bemerkungen

zur Zytogenetik der Lebermoose. Nova Hedwigia 40:

313-328.

VOLK, O. H. 1981. Beitrage zur Kenntnis der Lebermoose (He-

paticae) aus Siidwest-Afrika (Namibia), II. Riccia alboves-

tita. Mitt. bot. StSamml., Munch. 17: 245-252.

VOLK, O. H. 1983. Vorschlag fur eine Neugliederung der Gat-

tung Riccia L. Mitt. bot. StSamml., MiXnch. 19: 453-465.

VOLK, O. H. 1984. Beitrage zur Kenntnis der Marchantiales aus

Siidwest-Afrika, Namibia IV. Zur Biologie einiger Hepati-

cae mit besonderer Beriicksichtigung der Gattung Riccia.

Nova Hedwigia 39: 117-143.

VOLK, O. H. & PEROLD, S. M. 1984. Studies in the liverwort

genus Riccia (Marchantiales) from the south-west Cape.

Bothalia 15: 117-124.

VOLK, O. H. & PEROLD, S. M. 1985. Studies in the genus Ric-

cia (Marchantiales) from southern Africa. 1. Two new

species of the section Pilifer: R. duthieae and R. alatos-

pora. Bothalia 15: 531-539.

Bothalia 16,1: 29-33 (1986)

Studies in the genus Riccia (Marchantiales) from southern Africa. 3.

R. schelpei, a new species, in the new subgenus Chartacea

O. H. VOLK* and S. M. PEROLD**

Keywords: air-pores, dorsal epidermis, Riccia, subgenus Chartacea

ABSTRACT

Riccia schelpei Volk & Perold, sp. nov., endemic to the western Cape, is described. It is characterized by the

parchment-like epidermis of the thallus, thick-walled hyaline epidermal cells and by dorsal air-pores encircled by a

raised ring of smaller thin-walled cells. This species is the type of the new monotypic subgenus Chartacea Perold.

Chartacea Perold, subgen. nov. Ricciae L.

Textura thalli chartacea, inde nomen; epidermis

indurata cellulis hyalinis parietibus incrassatis; pori

aerii (stomata) annulo cellularum superpositarum

parietibus tenuibus circumcincti.

TYPE. — Riccia schelpei Volk & Perold

Thallus dorsally with papery texture; epidermis

parchment-like with thick-walled hyaline cells; air-

pores surrounded by a ring of thin-walled superim-

posed cells.

Riccia schelpei Volk & Perold, sp. nov. Thallus

monoicus (?), mediocris ad magnus; lobi ad 12 mm

longi, 1,5-2 mm crassi, oblongo-ligulati; superficies

dorsalis in sicco pallido-flavescens, chartacea, pro-

funde sulcata, apice emarginato, marginibus late ala-

tis. Frons in sectione transversali: chlorenchyma cav-

ernulis aeriis latis polyedricis. Epidermis unistratosa,

cellulis hyalinis parietibus incrassatis, poris aeriis an-

nulo cellularum parvarum superpositarum parietibus

tenuibus circumcinctis. Squamae inconspicuae, mar-

ginem frondis non superantes. Sporangia in sulco ag-

glomerata. Sporae 95-105 p diametro, triangulo-

globulares, polares, rubello-bruneae, alatae, reti-

culo-foveolatae, 10-12 foveolis in diametro, plerum-

que granulosae. Chromosomatum numerus n = 8

(Bornefeld, 1984).

TYPE. — Cape, 2917 (Springbok): Hester Malan

Res. Carolusberg (W), seepage area (-DB),

1977.09.14 Schelpe 7775 (BOL; PRE) associated

with Bryum radiculosum Brid., B. argenteum

Hedw., Chamaebryum pottioides Ther. & Dix., Ric-

cia parvo-areolata Volk & Perold, Crassula spp. and

Cyanophyceae. On decomposed granitic soil, pH 7.

Thallus monoecious (?), perennial, in gregarious

patches or single and scattered, medium-sized to

large, lobes up to 12 mm long, simple or irregularly

furcate, branches widely divergent, oblong-ligulate,

winged, 1,5-2 mm thick, 2-3 times broader than

thick, when dry 3^4 mm broad, dorsal surface yellow

and parchment-like, only apical sides inflexed (Fig.

1.1a & lb), opposing each other and sometimes

clasped together, otherwise wings of thalli expanded

* Botanische Anstalten d. Univ. Wurzburg D 8700, Germany,

B.R.D.

** Botanical Research Institute, Department of Agriculture &

Water Supply, Private Bag X101, Pretoria 0001, RSA.

and irregularly undulate; when turgid, up to 6 mm

broad, dorsal surface green, somewhat greasy, re-

ticulate with many scattered areolae, formed by

faintly visible outlines of air-chambers, each with a

single air-pore (Fig. 2.2); apex rounded, emarginate,

sulcus deep towards apex, sides convex and sloping

steeply; proximally groove shallow and wide (Figs

1.2; 2.1). Thallus branches in transverse section with

deeply grooved surface on apical sections (Fig.

1.3a), on more proximal sections with wide shallow

channel (Figs 1.3b; 2.5); margins of wings acute

(Figs 1.3a & 3b; 1.7), attenuate and undulating,

flanks sloping steeply up and outwards, occasionally

flecked with purple near base; ventrally thickened,

slightly rounded, greenish, with numerous smooth

and some tuberculate rhizoids, laterally abruptly

forming a wing which is without rhizoids (Fig. 1.3a &

3b); assimilation tissue (chlorenchyma) the

thickness of section, about 600-850 (-1 000) pm

thick, with polyhedral air-chambers up to 150 pm

wide (Figs 1.4; 2.6), in larger plants 25-30 chambers

across width of thallus, sloping very obliquely at

bases, gradually curving upwards and becoming al-

most vertical towards surface, where each one opens

via a small pore; each air-chamber enclosed by 6-8

chlorophyllose lamellae or plate^, 1 cell thick, cells

isodiametric, about 55 pm wide; storage tissue^-,

the thickness of section 300-500 pm thick, cells

roundish, up to 70 pm wide, without starch, but rich

in oil.

Epidermis unistratose, thick-walled, except for

ventral rhizoid-bearing part; dorsal cells variously

shaped, polygonal, with rounded corners (Fig. 1.5),

35-70 pm long, 30-50 pm wide and 10-40 pm high;

air-chambers roofed over by 5-8 thick-walled epi-

dermal cells centred around air-pores, each pore en-

circled by a slightly raised ring of 5-7 fragile, smaller

cells 15-20 pm wide (Figs 1.5; 2.3), overlying and

overlapping thick-walled epidermal cells, thus re-

ducing diameters of pores to 5-20 pm; the pores re-

semble those of Oxymitra; distances between pores

80-140 pm; some of the epidermis cells bear unicel-

lular club-shaped ‘slime’ papillae (Fig. 2.4), 40 x 20

pm, very thin-walled and easily destroyed, especially

numerous around archegonia (Fig. 1.11 & 1.12); on

transverse section of wing (Fig. 1.7), both dorsal and

ventral cells thick-walled, but epidermal cells on

dorsal surface of wings with irregularly thickened

walls (Fig. 1.6), and on ventral surface of wings cells

30

Bothalia 16,1 (1986)

FIG. 1. — Riccia schelpei (S. M. Perold 535, PRE). Structure of the thallus, scales and chromosomes. 1, dry thalli: a, with several

ripe sporangia; b, shrunken, sterile thallus; 2, fresh thallus; 3, branch of thallus in transverse section: a, during transition to

resting phase, emptied cells in wing shaded, near apex deeply grooved; b, showing thick-walled epidermis, assimilation and

storage tissue, proximally with shallow groove; 4, transverse section enlarged, showing epidermal cells, pores, air-chambers

and chlorophyllose lamellae; 5, dorsal epidermis from above: pores (black) encircled by a ring of fragile cells overlying thick-

walled epidermis cells; 6, epidermis on margin of wing, from above; 7, transverse section of wing: walls of dorsal and ventral

epidermis thick-walled, but different in shape; 8, ventral epidermis at wings, cells thick-walled; 9, scale; 10, transverse sec-

tion of bistratose base of scale; 11, longitudinal section of thallus through gametangia, ‘slime' papillae and wing; 12, mouths

of archegonia in deep depressions, from above, surrounded by ‘slime’ papillae; 13, chromosomes. (1-12, by O. H. Volk; 13,

by T. Bornefeld). Scale bar 1, 2, 3 = 2 mm; 4, 5, 8, 10, 12 = 50 pm; 6, 7 = 100 pm; 9 = 500 pm; 11 = 1 mm; 13 = 1 pm.

Bothalia 16,1 (1986)

31

uniformly thick-walled, rectangular (Fig. 1.8), up to

60 pm long, without air-pores or rhizoids. Scales im-

bricate, 300-500 pm broad and up to 1 500 pm long

(Fig. 1.9), flush with and not projecting above thal-

lus margin (Fig. 2.1), edge nearly smooth, hyaline,

some scattered cells near base purple, cells oblong,

5-6-sided, 110 pm long and 50 pm wide, cell walls

straight, at margins cells smaller, about 40 x 30 pm;

in cross section cells bulging and base bistratose

(Fig. 1.10). Gametangia in transient but well-defined

groups along dorsal groove (Fig. 1.11); epidermis

less complex here, but with large numbers of the

fragile, blunt, ‘slime’ papillae; antheridia with short,

hyaline ostioles; archegonia opening in deep cup-

like depressions, surrounded by papillate cells (Fig.

1.12); subsequently, as the spores develop, the necks

protrude up to about 200 pm, with bases purple-

brown and tips hyaline, thin, almost thread-like.

Sporangia crowded together or scattered along

groove, bulging dorsally, containing 600-800 spores

(over 1 000 in capsules of large plants) enclosed in

red-brown sac, which later disintegrates. Spores

(90— )95— 105(— 1 15) pm in diameter, triangular-globu-

lar, polar, reddish or yellowish brown when young,

darkening to mahogany brown, opaque; wing 7,5 pm

wide, margin crenulate and somewhat eroded, pore

FIG. 2. — Riccia schelpei (E. G. H. Oliver 8041, PRE). Thallus, epidermal pores and air-chambers. 1, thallus

near apex, showing scales; 2, air-pores on part of dorsal surface with margin and scales in foreground; 3,

air-pore with 5 surrounding cells; 4, epidermis, air-pores, ‘slime' papillae; 5, transverse section of thallus

branch; 6, transverse section of air-chambers. (SEM micrographs by S. M. Perold). Scale bar on 1, 2, 5,

6 = 100 pm; on 3, 4 = 50 pm.

32

Bothalia 16,1 (1986)

FIG. 3. — Riccia schelpei ( E . A. Schelpe 7775, BOL). Spores. 1, proximal face; 2, apex; 3, viewed from side; 4,

marginal pore and margin; 5, 6, distal face. (1-5, SEM micrographs; and 6, LM (light microscope)

photographs, by S. M. Perold). Scale bar on 1-5 = 50 pm; diameter of spore on 6, ca 100 pm.

at each marginal angle 5-7,5 jam across (Fig. 3.4),

distal face convex, reticulate-foveolate, with

9-10(-12) deep cup-like areolae across diameter of

spore (Fig. 3.5, 3.6), each areola 10-12,5 pm wide,

becoming somewhat smaller near margin, high

ridges surrounding areolae and raised at nodes usu-

ally heavily encrusted with granules and papillae,

but occasionally smoother; proximal face with apex

blunt and sometimes acute, triradiate mark present,

but partly obscured by granules, each facet with

15-20 small shallow areolae, surrounding ridges

granulose (Fig. 3. 1-3.3). Chromosome number n =

8 (Bornefeld 1984); the letters A-E (Fig. 1.13) iden-

tify the chromosomes according to Bornefeld (1984) .

Under adverse conditions, the thalli become

shrunken and transformed to dormant bulbils, as the

peripheral cells lose their contents and form a pro-

tective covering (Fig. 1.1b). As with other Riccia

species, the walls of the epidermal cells, of the

empty cells and of the rhizoids, are stained a deep

blue when treated with dilute Toluidine Blue N,

whereas all other cell walls are reddish violet (Volk

1984).

R. schelpei is endemic to the western Cape, which

is a winter rainfall region. It grows at seepages or on

rocky outcrops, fully exposed to the sun, on acid to

neutral (pH 5. 0-7.0), well-drained soils, composed

Bothalia 16,1 (1986)

33

of finely or coarsely decomposed granite, rich in

dust. It may be associated with lichens, several small

Crassula species, other Riccia species and with small

mosses like Archidium and Bryum. R. schelpei has

been named in honour of the late Prof. E. A.

Schelpe, former curator of Bolus Herbarium, Uni-

versity of Cape Town, who collected the type speci-

men. Fig. 4.

FIG. 4. — Map showing distribution of R. schelpei.

The parchment-like, and somewhat greasy epider-

mis, the air-pores with a ring of fragile cells superim-

posed over thick-walled, hyaline, epidermal cells, as

well as the gametangia in well-defined stands, bear a

resemblance to other genera of the Marchantiales

e.g. Oxymitracea. These characters have necessi-

tated placing R. schelpei in the new subgenus Char-

tacea, setting it apart from other members of the re-

lated subgenus Spongodes.

In addition to the type locality, collections were

made at the following sites:

CAPE. — 3018 (Kamiesberg): Plateau N of Leliefontein to-

wards Draaiklip (-AC), Oliver 8041 (PRE); 3218 (Clanwilliam):

N of Citrusdal, above Olifants River (-BD), S.M. Perold 535

(PRE).

ACKNOWLEDGEMENTS

The authors are grateful to the late Prof. E. A.

Schelpe for the loan of his specimen from BOL and

Dr habil. T Bornefeld, Am Reele I, D-8706

Hochberg, Wurzburg, Germany, for the chromo-

some counts and figures. Sincere thanks are also due

to Mr E. G. H. Oliver, Stellenbosch, for collecting

and kindly sending us the specimen from Kamies-

berg.

UITTREKSEL

Riccia schelpei Volk & Perold, sp. nov., endemies

tot die Wes-Kaap, word beskryf. Die spesie word ge-

kenmerk deur die perkamentagtige dorsale oppervlak

van die tallus, verdikte selwande van die dorsale epi-

dermisselle, en deur dorsale poriee wat omring is deur

’n opgehewe kring van kleiner, dunwandige selle.

Hierdie spesie is die tipe van die nuwe monotipiese

subgenus Chartacea Perold.

REFERENCES

BORNEFELD, T. 1984. Chromosomenanalyse der Gattung Ric-

cia von Slid- und SW-Afrika und allgemeine Bemerkungen

zur Zytogenetik der Lebermoose. Nova Hedwigia 40:

313-328.

VOLK, O. H. 1983. Vorschlag fur eine Neugliederung der Gat-

tung Riccia L. Min. bot. StSamml., Munch. 19: 453-465.

VOLK, O. FI. 1984. Beitrage zur Kenntnis der Marchantiales aus

Siidwest-Afrika, Namibia IV. Zur Biologie einiger Hepati-

cae mit besonderer Beriicksichtigung der Gattung Riccia.

Nova Hedwigia 39: 117-143.

VOLK, O. H. & PEROLD, S. M. 1984. Studies in the liverwort

genus Riccia (Marchantiales) from the south-west Cape.

Bothalia 15: 117-124.

Bothalia 16,1: 35-38 (1986)

The identity of Erica vinacea and notes on hybridization in Erica

E. G. H. OLIVER*

Keywords: Erica fastigiata, E. fervida, E. x vinacea, hybridization in Erica

ABSTRACT

The rediscovery of Erica vinacea L. Bol. among sympatric populations of two other species led to a comparison

of their morphological characters. Results of this study indicate a putative hybrid origin: E. x vinacea L. Bol. = E.

fastigiata L. x E. fervida L. Bol.

Recently Mr A. W. Schumann of Cape Town dis-

covered E. vinacea L. Bol. north-west of Platberg in

the Kogelberg Reserve. In this locality, which falls in

the area indicated on the type specimen, E. vinacea

was found growing as scattered plants throughout a

large marsh dominated by the restiad, Chondropeta-

lum mucronatum (Nees) Pillans. In the same marsh

there occurred numerous plants of E. fastigiata L.

and groups of E. fervida L. Bol. It was immediately

suspected that the scattered plants were hybrids be-

tween the two more common species.

E. fastigiata is a common species in the area from

the Hottentot’s Holland Mountains to Hermanus

where it can be sometimes abundant on moister

slopes and in seepage zones. It is a highly variable

species and has several closely allied species such as

E. walkeria Andr., E. daphniflora, E. hendricksei

H. A. Baker and E. turrisbabylonica H. A. Baker,

none of which, however, occurs within the Kogel-

berg Reserve. E. fastigiata can easily be recognized

by its very large spreading corolla lobes which are

usually white to pale pink, the outside colour of the

flowers being deep red. In the centre of each flower

(see Fig. 1.1a) there is a darker area of red or green

which acts as a nectar guide. The long foliage-like

bract, bracteoles and sepals, together with the

crowded imbricate leaves, hide most of the flower

from view. The flowers are typically arranged in

groups of four at the ends of the branches. In the

area surrounding Platberg most of the collections of

this species exhibit corolla tubes which are very

slightly and finely hairy whereas elsewhere they are

mostly glabrous.

E. fervida is a remarkable species closely allied to

that rare and endangered species, E. pillansii H.

Bol. Both have vivid scarlet, noticeably puberulous

flowers and are endemic in the Kogelberg Reserve.

The flowers are borne at the ends of very short lat-

eral branchlets (brachyblasts) which are crowded

along the main branches. E. pillansii has, however,

much larger flowers and flowers in May/June,

whereas E. fervida flowers in October/November.

E. vinacea has dark pinkish red flowers with small

spreading lobes which have a slightly paler upper

surface. The flowers are borne at the ends of short

* Botanical Research Unit, Department of Agriculture and

Water Supply, P.O. Box 471, Stellenbosch 7600.

lateral branchlets (dolichoblasts) which are more

loosely crowded towards the ends of the branches

than in E. fervida.

A comparison of the main distinguishing charac-

ters of the three taxa is given in Table 1. From this

table it can be seen that the main differences lie in

the degree of hairiness of nearly all the organs, the

inflorescence arrangement, flower shape and details

of the anther and ovary. In most cases the characters

of E. vinacea are intermediate between those of E.

fastigiata and those of E. fervida and tend towards

either one or the other.

The occurrence of scattered plants of E. vinacea

amongst numerous plants of the other two species in

the locality north-west of Platberg indicates that the

plants are chance first generation crosses that have

not become reproductively established. The plants

of the putative hybrid appeared to be of the same

age as those of the other two species and are thought

to have grown up soon after the last fire in the area

about 40 years ago. All three species are single-

stemmed and must regenerate from seed after a fire.

It was not possible to revisit the locality to check on

seed-set and seed fertility.

Two other isolated populations of E. vinacea

growing together with E. fastigiata and E. fervida

have been reported (A. W. Schumann pers.

comm.).

The pollen of the collection Oliver 8611 is typical

of the genus Erica, namely in tetrads, and did not

show any abnormalities. The few species of Erica

which have so far been investigated all have n=12

with extremely small chromosomes. A thorough in-

vestigation of the cytology of the three taxa may re-

veal some additional evidence on their relationships.

The facts as discussed above and set out in Table 1

indicate a putative hybrid origin of E. vinacea L.

Bol. (E. x vinacea L. Bol. = E. fastigiata L. x E.

fervida L. Bol.).

HYBRIDIZATION IN ERICA

The hybrid origin of E. x vinacea poses an in-

triguing question as to how such crosses took place.

In a paper on pollination syndromes of Erica species

in the south-western Cape (Rebelo et al. 1985), E.

fervida is classified as being ornithophilous. involv-

ing species of sunbirds (Nectariniidae). E. fastigiata

on the other hand is classified as being rhinomyio-

36

Bothalia 16,1 (1986)

FIG. 1. — A, Erica fasligiata, B, Erica x vinacea and C, Erica fervida: 1, flower, X 6; la, flower from above, X 6; 2, bracteole and

bract, x 12; 3, sepal, x 12; 4, anther, front, side and back views, x 25; 5, ovary and diagrammatic longitudinal section of

upper half of ovary, x 12; 6, leaf, x 12. Drawn from Oliver 8609, 8611 & 8610 (STE) respectively.

Bothalia 16,1 (1986)

37

TABLE 1. — Comparison of morphological characters of E. fastigiata, E. X vinacea and E. fervida

philous, a term coined to describe the most discrete

insect-pollination syndrome in Erica where pollina-

tion is performed by hovering dipterans with very

long proboscises. In the latter situation the opening

to the corolla tube is extremely small and the nectar

guides are very prominent. It would seem then that

there are two very different pollinating agents for

the two parent species. This problem needs investi-

gation in the field.

This is not the first time that hybridization has

been recorded for Erica. Oliver (1977) considered

E. x flavisepala Guth. & Bol. to be a putative hy-

brid between the very unlikely parent species E.

thunbergii Montin and E. spaerocephala Wendl.

(Oliver 1977). Salter (1950) mentions the following

hybrids occurring between some 28 species of Erica

on the Cape Peninsula, one of which is described

and documented as E. x fontensis Salter (Salter

1935):

E. fontana Salter x E. capensis Salter

E. fontana Salter x E. laeta Bartl.

E. eburnea Salter x E. laeta Bartl.

E. capensis Salter X E. laeta Bartl.

E. heleogena Salter x E. laeta Bartl.

E. hirtiflora Curt, x E. heleogena Salter

E. hirtiflora Curt. X E. mauritanica L.

E. hirtiflora Curt. X E. pyxidiflora Salisb.

E. mauritanica L. x E. baccans L.

E. phylicifolia Salisb. x E. nudiflora L.

E. curvirostris Salisb. X E. nudiflora L.

E. pulchella Houtt. x E. nudiflora L.

E. lutea Berg, x E. corifolia L.

E. gnaphaloides L. X E. palliiflora Salisb.

Specimens examined:

Erica fastigiata

CAPE PROVINCE.— 3418: Kogelberg (-BB), Compton

16454 (NBG; STE); Buffelskloof (-BD), Barker 8033 (NBG;

STE); Platberg (-BD), Boucher 183 (STE); 877 (STE); 2653

(STE); road to Platbos (-BB), flaynes 696 (STE); Wynand

Louwsbos (-BD), Lamb 124 (STE); Buffelstalberg (-BD), Oliver

sub STE 30052 (STE); neck NW of Platberg towards Buffelstal-

berg (-BD), Oliver 8609 (STE); Platbos (-BD), Stehle 254 (STE);

Platberg (-BD), Thompson 495 (STE).

Erica fervida

CAPE PROVINCE. — 3418: Kogelberg Reserve, neck NW

of Platberg towards Buffelstalberg (-BD), Oliver 8610 (BM; E;

K; MO; P; PRE; S; STE).

Erica x vinacea

CAPE PROVINCE. — 3418: Kogelberg Reserve, neck NW

of Platberg towards Buffelstalberg (-BD), Oliver 8611 (BM; E;

K; MO; P; PRE; S; STE); between Platteberg and Kogelberg

(-BB/BD), Stokoe sub BOL 17599 (BOL, holo.).

38

BothaKa 16,1 (1986)

UITTREKSEL

Die herontdekking van Erica vinacea L. Bol. in

simpatriese populasies van twee ander soorte het aan-

leiding gegee tot ’n vergelyking van hul morfologiese

eienskappe. Die resultate van die ondersoek dui op 'n

vermeende hibridiese oorsprong: E. x vinacea L.

Bol. = E. fastigiata L xE, fervida L. Bol.

REFERENCES

BOLUS, H. M. L. 1928. Novitates Africanae. Ann. Bol. Herb. 4:

135-136.

OLIVER, E. G. H. 1977. The identity of Erica flavisepala. Bo-

thalia 12: 195-197.

REBELO, A. G., SIEGFRIED, W. R. & OLIVER, E. G. H.

1985. Pollination syndromes of Erica species in the south-

western Cape. S. Afr. J. Bot. 51: 270-280.

SALTER, T. M. 1935. Erica x fontensis in Plantae Novae Afri-

canae. Jl S. Afr. Bot. 1: 82-86.

SALTER, T. M. 1950. In R. S. Adamson & T. M. Salter, Flora of

the Cape Peninsula, 633-657. Cape Town: Juta.

Bothalia 16,1: 39-64 (1986)

Notes on African plants

VARIOUS AUTHORS

ACANTHACEAE

NOTES ON SOUTHERN AFRICAN SPECIES OF JUSTICIA L.

As the result of a revision of Justicia L. in south-

ern Africa, to be published in the Flora of southern

Africa (FSA), the number of species represented in

the area has been reduced from 30-40 [Dyer, R. A.,

Genera 1: 598 (1975)] to 22 with six subspecies. One

species and one subspecies are new and are here de-

scribed. A list of name changes of the species in the

area is given, with an abridged synonomy containing

only basionyms and names until recently considered

to be those of distinct species. A full synonomy will

be given in the FSA. Changes in rank are also given.

Justicia parvibracteata Immelman, sp. nov., J.

protractae (Nees) T. Anders, subsp. rhodesianae (S.

Moore) Immelman affinis, sed planta minor, brac-

teis reductis, triangularibus differt.

Suffrutex vel herba perennis, 0,12-0,5 m alta, om-

nino subtiliter et dense puberula. Folia 4-25 x 1-10

mm, anguste vel late lanceolata, apice late acuto vel

obtuso, basi cuneata, margine interdum glanduloso,

petioli graciles, 0-6 mm longi. Inflorescentia 1 (-2)

floribus in cyma. Bractea bracteolaeque reductae,

subulatae vel triangulares, c. 2 x 1 mm. Calyx lobis

5, subaequalibus, anguste lanceolatis. Corolla (tubus

et labium superum) (4-) 5-8 mm longa, alba, lineis

testaceis in palato. Capsula quadriseminalis, usque

ad 8 mm longa, delicatula, subtiliter puberula. Pol-

len bicolporatum, sexinio fasciato.

TYPE. — Cape Province, 2722 (Olifantshoek): in

Toto Mountains, kloof, in rock crevices and under

shrubs (-DD), Tolken & Schlieben 1176 (PRE,

holo.!).

Shrublet or perennial herb, 0,12-0,5 m high, all

parts minutely and densely puberulous. Leaves 4-25

x 1-10 mm, sometimes glandular on margins, nar-

rowly to broadly lanceolate, apex broadly acute to

obtuse, base cuneate; petiole slender, 0-6 mm long.

Inflorescence of 1 (-2) flowers per cyme, sessile,

scattered in leaf axils. Bract and bracteoles reduced,

subulate to triangular, c. 2 x 1 mm. Calyx of 5 sub-

equal, narrowly lanceolate lobes. Corolla (tube and

upper lip) (4) 5-8 mm long, white with dark red lines

on palate. Capsule 4-seeded, cylindrical, with a

stipe, puberulous, delicate, up to 8 mm long. Pollen

2-colporate, with sexine in areoles on either side of

the colpus.

Endemic to the Northern Cape, where J. protracta

does not occur.

Justicia orchioides L.f. subsp. glabrata Immel-

man, subsp. nov. a subspecie typica omnibus parti-

bus sine pilis longis rectis albo-opacis differt.

Fruticulus lignosus, 0,14—0,6 m altus; caules

crassi, nodosi, interdum spinescentes, partes omnes

glabratae vel pilis brevibus vel papillis crassis pyra-

midalibus, sine pilis candidis ut in subsp. orchioide;

cortex cinereus, rimosus sulcatus; rami juvenes sub-

herbacei. Folia ovata vel lanceolata, sessilia, 3,5-13

x 1,5-7 mm, apice obtusa vel acuta, basi cuneata,

coriacea; costa prominens sed nervi laterales

obscuri. Inflorescentia cymarum remotarum, una-

quaeque ad florem solitarium pedunculatum axilla-

rem reducta. Pedunculi et pedicelli validi, 1-10 mm

longi. Bractea nulla. Bracteolae duae, triangulares,

1-2 mm longae, basi pedunculum transverse junc-

tae. Corolla (tubus et labium superum) 7-10 mm

longa, lactea striis testaceis palato. Capsula unisemi-

nalis, glabrata, dura, usque ad 17 mm longa. Pollen

bicolporatum, sexinium fasciatum.

TYPE. — Cape Province, 3326 (Grahamstown):

between Piggots Bridge and Hounslow, 400 m, road-

side on dry clay soil (-AB), A. Jacot Guillarmod

6902 (PRE, holo.!; GRA!).

Woody shrublet, 0,14-0,6 m high; stems thick,

gnarled, may become spiny; all parts glabrous or

with short hairs or with stout pyramidal papillae,

without white-opaque hairs as in subsp. orchioides;

bark grey, cracked and furrowed; young branches

subherbaceous. Leaves ovate to lanceolate, sessile,

3,5-13 x 1,5-7 mm, apex obtuse to acute, base cu-

neate, leathery; midrib prominent but side veins not

visible. Inflorescence of scattered cymes, each re-

duced to a single pedunculate flower. Peduncles and

pedicels stout, 1-10 mm long. Bract absent. Brac-

teoles 2, triangular, 1-2 mm long, joined at base

across peduncle. Corolla (tube and upper lip) 7-10

mm long, cream with red striping on palate. Capsule

1- seeded, glabrous, hard, up to 17 mm long. Pollen

2- colporate, sexine banded on either side of the

colpi.

Very like the typical subspecies except for the lack

of the characteristic long stiff white-opaque hairs.

The typical subspecies is furthermore confined to the

area around Port Elizabeth, whereas subsp. glabrata

is widespread in the eastern half of the Karoo and

the southern Cape.

The type of subsp. orchioides is a Thunberg speci-

men at UPS. Although I have seen it only on a

microfiche, which does not show fine detail, and do

not know precisely where it was collected, informa-

tion on the hairs received from UPS indicates that at

least the right hand plant on the sheet belongs with

the Port Elizabeth subspecies. I have therefore pro-

vided a name and a description for the more wide-

spread subspecies for which no existing name was

found.

J. orchioides and J. cuneata have often been con-

fused, both in the literature and in herbaria, and the

40

Bothalia 16,1 (1986)

characters used to distinguish them in the Flora Ca-

pensis key are not reliable. The two species are

nevertheless quite distinct, with J. cuneata having

hooded flowers longer than 10 mm and the sexine of

the pollen areolate, while in/, orchioides the flowers

are 10 mm or shorter, not hooded, and the sexine of

the pollen forms a raised band along the smooth area

on either side of each colpus.

Justicia betonica L., Sp. PI. 15 (1753). Type:

Ceylon [Sri Lanka], Hermann vol. 3, fol. 2 (BM;

photo at PRE!).

/. trinervia Vahl, Enum. 1: 156 (1804). Type: E India, Rottler

s.n. (C).

Adhatoda variegata var. pallidior Nees in DC., Prodr. 11: 385

(1847). Justicia pallidior (Nees) C.B. Cl. in FI. Cap. 5,1: 58

(1912). Type: Transvaal, Apies River, Burke 514 (K!).

A. cheiranthifolia Nees in DC., Prodr. 11: 387 (1847). Justicia

cheiranthifolia (Nees) C.B. Cl. in FI. Cap. 5,1: 58 (1912). Type:

Transvaal, Magaliesberg, Burke s.n. (K!).

Justicia betonicoides C.B. Cl. in FI. Trop. Afr. 5: 184 (1900), in

FI. Cap. 5,1: 58 (1912). Syntypes: Sudan, Jur, Jur Ghattas,

Schweinfurth 1423; Gabon, Bongo, Schweinfurth 2543; French

Equatorial Africa, Mittu (Mittou), Schweinfurth 2793; Kenya,

along Gilgil River, north of Lake Naivashu, 6-7000 ft, Scott-Elliot

6647; Tanzania, Tanganyika Plateau, at Fort Hill, 3500-4000 ft,

Whyte s.n.

Justicia petiolaris (Nees) T. Anders, in J. Linn.

Soc., Bot. 7: 39 (1864). Adhatoda petiolaris Nees in

DC., Prodr. 11: 402 (1847). Syntypes: Natal, Um-

zimvubu River, wooded rocky shaded valley and ra-

vine by river, below 1000 ft, Drege s.n. (K!; P!); Na-

tal, Umgeni, on hills, 200 ft, Drege s.n. (P!).

(a) , subsp. petiolaris.

(b) . subsp. bowiei (C. B. Cl.) Immelman, stat.

nov.

J. bowiei C.B. Cl. in FI. Cap. 5,1: 59 (1912). Syntypes: Cape

Province, near Kei Mouth, in woods, 300 ft, Flanagan 882 (BOL!;

GRA!; PRE; SAM!); Cape Province, moist situations in George,

Uitenhage and Albany Divisions, Bowie s.n. (K!) ; no locality,

Guthrie 4711 (BOL!).

J. mutica C.B. Cl. in FI. Cap. 5,1: 61 (1912). Type: Cape Prov-

ince, wooded situations in Uitenhage and Albany Districts, Bowie

s.n. (K!).

(c) . subsp. incerta (C.B. Cl.) Immelman, stat.

nov.

J. incerta C.B. Cl. in FI. Cap. 5,1: 66 (1912). Type: Transvaal,

bushveld between Elandsrivier and Klippan, Rehmann 5058.

The distribution of the subspecies is unusual: it is

found in N Natal and in the Nylstroom-Thabazimbi-

Rustenburg area, with a single record from the cen-

tral Kruger National Park.

Justicia protracta (Nees) T. Anders, in J. Linn.

Soc., Bot. 7: 41 (1864). Syntypes: Cape Province,

between shrubs in field by Zwartkops River, Ecklon

456 (BOL!; MEL!); Cape Province, Bosmans River

Mountains, Ecklon s.n. ; near Grahamstown, Ecklon

s.n. (S!); between Great Fish River and Ceded Ter-

ritory, Ecklon s.n. (PRE!; S!).

Gendarussa protracta Nees in Linnaea 15: 371 (1841), partim

excl. syn. Thunb.

(a), subsp. protracta.

J. kraussii C.B. Cl. in FI. Cap. 5,1: 62 (1912). Syntypes: Natal,

between Mlazi River and Durban Bay, Krauss 61 (BM!; K!); Na-

tal, Inanda, Wood 423; Natal, Zulu land, Gerrard 1272 (BM! ; K!).

J. kraussii var. florida C.B. Cl. in FI. Cap. 5,1: 62 (1912). Type:

Natal, Inanda, Wood 566 (BM!).

/. pulegioides C.B. Cl. in FI. Cap. 5,1: 62 (1912) partim excl.

syn. Chaetacanthus persooni Nees. Syntypes: Cape Province, Uit-

enhage, Ecklon & Zeyher 436; Cape, Komadagga, Burchell 3300

(G-DC!); Cape Province, Komgha, Flanagan 725 (NH!; GRA!;

PRE!; SAM!); Natal, Durban Flats, Wood in Herb. Norm. Aust.-

Afr. 1019 (BOL!; PRE!; SAM!); Natal, Inanda, Wood 718

(PRE!; NH!), Wood 309; Natal, Durban Bay, Krauss 304; Natal,

Pondoland, between St Johns River and Umsikaba River, Drege

s.n. ; Transvaal, Houtbosch Rand, Schlechter 3324; without locali-

ties, Peddie s.n., Sanderson 433, Grant s.n.

J. woodii C.B. Cl. in FI. Cap. 5,1: 64 (1912). Type: Natal,

Noodsberg, Wood 112 (K!).

Probably the most common of the southern Afri-

can species, J. protracta subsp. protracta is found in

the eastern half of the country as far south as Port

Elizabeth. In the northern Transvaal as well as in

SWA/Namibia and Botswana it is largely replaced

by subsp. rhodesiana, though the transition is grad-

ual. The great range in pubescence and leaf size and

shape has led to the description of numerous species

and subspecies, but examination of a larger number

of specimens shows that these form a continuous

range rather than discrete entities.

(b). subsp. rhodesiana (S. Moore) Immelman,

stat. nov.

J. rhodesiana S. Moore in J. Bot., Lond. 51: 188 (1913). Syn-

types: Botswana, Mahalapye, Rogers 6069 (BOL!; SAM); Zim-

babwe, Bulawayo, Rogers 5740 (BOL!; SAM!).

Differs from the typical subspecies in having all

parts puberulous rather than pilose. The distribution

is also more northerly, as it occurs in SWA/Namibia,

Botswana and the northern Transvaal, while subsp.

protracta is found in the southern and eastern Trans-

vaal, Natal and the eastern Cape. The transition,

however, is not an abrupt one.

Justicia cuneata Vahl, Symb. bot. 2: 10

(1790-94), Enum. 1: 163 (1804). Type: Cap. bon.

spei (Cape of Good Hope), Sparrman s.n. in herb.

Dahl (C!).

(a) , subsp. cuneata.

(b) . subsp. latifolia (Nees) Immelman, comb, et

stat. nov.

Gendarussa orchioides var. latifolia Nees in Linnaea 15: 369

(1841). Type: Cape Province, Kanaquasberg, Ecklon s.n.

(BOL!); Cape, Clanwilliam, between Olifantsrivier and Brakfon-

tein, Ecklon & Zeyher s.n. (S ! ; MEL!).

(c) . subsp. hoerleiniana (P. G. Mey.) Immel-

man, stat. nov.

J. hoerleiniana P. G. Mey. in Mitt. bot. StSamml., Munch. 2:

300 (1957). Type: SWA/Namibia, Alicetal, Pomona (probably on

the coastal plain in the Liideritz District), Dinter 6401 (BOL!;

PRE!).

For the differences between J. cuneata and J. or-

chioides, see under J. orchioides (above).

Subsp. cuneata occurs only around Port Elizabeth,

and is glabrous, while subsp. latifolia is relatively

widespread in Namaqualand and the western half of

the Karoo, with one record from Port Elizabeth, and

is densely and minutely puberulous on leaves and ca-

lyx. Subsp. hoerleiniana is confined to a small area

on the southern coast of SWA/Namibia, and is also

Bothalia 16,1 (1986)

41

puberulous, but the hairs have very swollen, anvil-

shaped heads instead of being pointed as in subsp.

latifolia.

Justicia matammensis (Schweinf. ) Oliv. in

Trans. Linn. Soc. Lond. 29: 130 (1875). Type: E Su-

dan, Gallabat, Matamma (Metemma), Schweinfurth

130c (K!).

Adhatoda matammensis Schweinf. in Verh. zool.-bot. Ges.

Wien. 18: 674 (1868).

J. exigua S. Moore in J. Bot., Lond. 38: 204 (1900). Type: Zim-

babwe, Bulawayo, Rand 389 (BM!).

Species insufficiently known or excluded

Justicia brycei C.B. Cl. in FI. Cap. 5,1: 67

(1912). Type: Lesotho, near the summit of Macha-

cha, 10,000 ft, Bryce s.n. (K!). The specimen be-

longs to J. elegantula S. Moore, but this species does

not occur further south than Zimbabwe. Also, no

southern African species is known to grow at such

high altitudes. Probably, as Jacot Guillarmod sug-

gests in her Flora of Lesotho, the locality on the

specimen is incorrect, and should possibly be Ma-

cheke, Rhodesia (Zimbabwe).

Rhytiglossa rubicunda Hochst. in Flora (1845):

71 (1845); C.B. Cl. in FI. Cap. 5,1: 67 (1912). Type:

Cape Province, Tsitsikamma Forest, Krauss 1128

(K!). Placed tentatively in Justicia under ‘species in-

sufficiently known’, but without formal transfer, by

C.B. Clarke. Type not found, nor does the descrip-

tion fit any Justicia species known from near that

area.

Justicia pulegioides subsp. late-ovata C.B. Cl. in

FI. Cap. 5,1: 62 (1912). Type: Cape Province, on the

rocks of Zwartwater Poort, Burchell 3405 (K!), 3364

(K!). The specimens belong to Siphonoglossa tubu-

losa (Nees) Benth. ex Lindau.

ACKNOWLEDGEMENTS

I am grateful for the assistance of Dr H. P. Linder,

then Liaison Officer at Kew, for sending me many

photographs of types there, and to Mr Mikael

Hendren at UPS who answered my queries about

the hairs on the type of J. orchioides subsp. or-

chioides. Dr H. Glen assisted with the Latin transla-

tion, and Dr L. E. W. Codd with nomenclature and

typification, especially in interpreting Nees van

Esenbeck’s publication.

K. IMMELMAN

ADIANTACEAE

CHEILANTHES DELTOIDEA KUNZE IN THE WATERBERG, TRANSVAAL

In March 1980 N. H. G. Jacobsen discovered a

tiny, bright green, gregarious fern in cracks and shal-

low humus pockets on north-facing cliffs on the farm

Leeuwpoort 573 KR, about 20 km north-west of

Nylstroom, Grid no. 2428: (-CB) (Fig. 1). Samples

were at first thought to be a dwarf form of Chei-

lanthes viridis (Forssk.) Swartz var. viridis, but the

different rhizome scales and filiform stipes raised

doubts. Material secured and examined during two

visits to the site is mounted on sheet numbers N. Ja-

cobsen 5209 and W. Jacobsen 5500, at the National

Herbarium, Pretoria (PRE).

The plants grew on very roughly hollowed out

cliffs on a highly ferruginous vesicular lava of the

Waterberg System, obviously representing the iron-

rich scoria of an andesitic lava flow. Such scoriae are

usually locally limited. The total length of outcrop

was about 200 m. The vegetation both at the foot

and top of the cliffs consisted of grasses and herbs

with tufts of Cheilanthes viridis var. glauca (Sim)

Schelpe & N. C. Anthony in sheltered position and

frequent, elongate and rather contracted plants of

Cheilanthes hirta Swartz, especially on top of the

cliffs. Occasional small shrubs of Diospyros lycioides

Desf. and Ziziphus mucronata Willd. grew along the

ridge.

On the whole the characteristics of the fern from

the Waterberg agree with those of Cheilanthes del-

toidea Kunze, a species known so far only from the

arid and semi-arid regions of Namaqualand and

southern SWA/Namibia. The rhachis is somewhat

atypical as it is not always winged with green laminar

tissue above the basal pinnae (N. C. Anthony 1984:

110). Also slightly deviating are the occasional fork-

ing of the stipe about halfway up, allowing the devel-

opment of two laminae on one common stipe, and

the more widely spaced and narrower triangular pin-

nules of the lowest pinnae, the latter resembling

Sim’s var. laxa of the species (Sim 1915, PI. 105).

Spores of Jacobsen 5209, however, were found to

match those of typical C. deltoidea (N. C. Anthony

pers. comm.).

The occurrence of this species outside its pre-

viously recognized distribution range suggested vari-

42

ous possibilities. Spore dispersion from Namaqua-

land or southern SWA/Namibia could possibly ac-

count for the apparently completely isolated pre-

sence in the Transvaal (Fig. 2).

In view of the harsh climatic conditions (north

face, hot position on bare cliffs) of the Transvaal site

it appears to be more likely that this population is a

relict from a former, much drier period. Similar re-

cently discovered occurrences in the Transvaal of

some species thought so far to be confined to the

western Cape or to SWA/Namibia, such as Chei-

lanthes parviloba Swartz and C. marlothii (Hieron.)

Schelpe (W. B. G. Jacobsen 1983; N. C. Anthony

1984) or C. contracta (Kunze) Mett. ex Kuhn (N. C.

Anthony 1984) support this hypothesis.

ACKNOWLEDGEMENTS

N. H. G. Jacobsen thanks the Director of Nature

Bothalia 16,1 (1986)

Conservation, Transvaal for permission to publish

this paper.

REFERENCES

ANTHONY, N. C. 1984. A revision of the Southern African

species of Cheilanthes Swartz and Pellaea Link (Pterida-

ceae). Contr. Bolus Herb. 11: 1-293.

JACOBSEN, W. B. G. 1983. The ferns and fern allies of southern

Africa. Durban, Pretoria: Butterworths.

SIM, T. R. 1915. The ferns of South Africa , edn 2. Cambridge:

University Press.

W. B. G. JACOBSEN* and N. H. G. JACOBSEN**

* P. O. Box 1178, White River 1240, South Africa.

** Division of Nature Conservation, Private Bag X209, Pretoria

0001.

FIG. 2. — Cheilanthes deltoidea

in situ.

BRYACEAE (MUSCI)

A NEW SPECIES OF ANOMOBRYUM

Anomobryum drakensbergense Van Rooy, sp.

nov., A. sharpii A.J. Shaw similis, sed costa percur-

renti, capsula pyriformi breviori et endostomio seg-

ments late perforatis differt. Plantae caespitosae fo-

lds imbricatis, costa percurrenti; laminae cellulis su-

perioribus brevi-rhomboidalibus vel lineari-rhom-

boidalibus interdum vermicularibus, endostomium

ciliis rudimentalibus.

TYPE. — Natal, 2929 (Underberg): Organ Pipes

Pass (-AA), 7000-8000 ft., Esterhuysen 34594

(PRE, holo.; MO; BOL).

Plants small to medium-sized, caespitose, yellow-

ish green to brownish above, yellow-brown to brown

below; saxicolous to terricolous. Stems to 20 mm

tall, branching by forks or subperichaetial innova-

tions, occasionally tomentose below, yellowish

green or reddish brown to brown; in section round,

central strand of thin-walled cells present, inner cor-

tical cells in 2-4 rows, thin-walled, outer cortical

cells in 1-2 rows, thin-walled to incrassate. Leaves ±

equidistant, about equal in size or subperichaetial

leaves larger, imbricate, frequently concave, erect

when dry, erect-spreading when wet, shortly oblong-

acute or ovate to ovate-lanceolate, (0,5-) 0,6-1 ,3

mm long; apex acute; margins plane or rarely re-

curved, entire; border absent. Costa percurrent or

occasionally mucronate, generally yellow, fre-

quently reddish below; in section subround to

round, lamina inserted ventrally, ventral surface

Bothalia 16,1 (1986)

43

FIG. 3. - Anomobryum drakensbergense : 1, habit, x 2; 2, habit, x 5; 3, stem in cross section, x 175; 4 rhizoid, x 350; 5, leaves

x 35- 6 part of leaf in cross section, x 350; 7, basal leaf cells, (left side), x 175; 8, upper laminal cells, x 350, 9, leaf apex

showing upper laminal cells, x 175; 10, penchaetial leaf, x 35; 11, capsule, X 5; 12 portion of. ^

apparatus, x 350; 13, part of capsule mouth showing cells and peristome, x 175; 14, spore, x 700. (1, 4, 6, 11 14, Este h y

sen 34594 ; 2, 7, 10, Smook 1095; 3, 5, Van Rooy 21; 8, 9, Smook 1096a).

44

cells present, dorsal stereid band strong, dorsal sur-

face cells incrassate, guide cells incrassate. Upper

laminal cells short-rhomboidal to linear-rhomboidal,

occasionally vermicular, frequently incrassate, (25-)

35-63 (-85)x7-19 pm; basal laminal cells frequently

reddish, quadrate.

Dioicous. Perichaetia terminal but quickly over-

grown by innovations; leaves ovate-lanceolate to

lanceolate, 1-1,5 mm long, apex acute to acuminate,

margins frequently recurved, costa percurrent to

mucronate, upper laminal cells vermicular, basal

laminal cells quadrate to rectangular. Seta 10-16 mm

long, yellowish red or reddish brown; capsule pyri-

form, inclined to horizontal, yellowish to reddish or

brown, frequently contracted below mouth when

dry, urn 1-1,5 mm long, neck 0,8-1 ,5mm long,

wrinkled when dry; exothecial cells irregularly rec-

tangular to quadrate, incrassate, smaller at mouth;

stomata present on neck, phaneropore; annulus

present; peristome teeth narrowly oblong-acumi-

nate, frequently irregular in outline, 220-300 pm

long, bordered, trabeculate, yellowish to reddish,

frequently hyaline above, minutely papillose; endo-

stome segments broad below, tapering above,

keeled, broadly perforated, yellowish to hyaline, ci-

lia rudimentary, basal membrane high, yellow, mi-

nutely papillose; operculum conic, blunt to mucro-

nate; calyptra cucullate; spores round, 12-18 pm,

granulose. Fig. 3.

The species is known from the Drakensberg of

Natal and Lesotho. It is found on soil in rock crev-

ices in the Subalpine and Alpine Belts, from

2 100-3 050 m.

NATAL. — 2828 (Bethlehem): on rocks below cliffs at Sentinel

(-DB), Smook 1095, 1096a (PRE; MO). 2929 (Underberg):

Bothalia 16,1 (1986)

Giant’s Castle Game Reserve, on the escarpment at the Judge

Pass (-AB), Van Rooy 21 (PRE; MO; H).

LESOTHO. — 2828 (Bethlehem): Oxbow, on cliff above road 8

km west of Lodge (-DC), Magill 4604 (PRE). 3028 (Matatiele):

15 km west of Ongeluksnek, cliffs above Lake Letsie (-AC), Ma-

gill 4705 (PRE).

The slender plants, branching pattern of the stem,

imbricate, erect leaves, leaf shape and areolation in-

dicate the gametophytic relationship to other species

of Anomobryum. The upper laminal cells are vari-

able in shape, size and degree of cell wall thickening.

Stem and innovation leaves have short-rhomboidal

to rhomboidal or occasionally vermicular cells with

thickened walls. Subperichaetial and perichaetial

leaves have linear-rhomboidal to linear-vermicular

cells with incrassate walls.

Sporophytically the peristome structure of this

species falls within the infrageneric variation in per-

istome development. Both exostome and endostome

show signs of reduction. Some exostome teeth are

relatively short, blunt and irregular in outline and

the endostome cilia are rudimentary.

A. drakensbergense is related to the Mexican A.

sharpii A.J. Shaw but differs in the percurrent costa,

shorter pyriform capsule and the broadly perforated

endostome segments. A. filiforme (Dicks.) Solms,

the other species occurring in southern Africa, has

taller, julaceous and glossy stems, longer, narrower

and strongly incrassate upper laminal cells, costae

ending below the leaf apices and well developed

peristomes.

The specific epithet drakensbergense refers to the

Drakensberg mountains of Natal and eastern Leso-

tho where this species occurs.

J. VAN ROOY

COMBRETACEAE

A NEW SPECIES OF COMBRETUM FROM THE TRANSVAAL

Combretum petrophilum Retief, sp. nov., C.

apiculato Sond. subsp. apiculato affinis, sed petalis

non ciliatis et lamine anguste ovate differt.

TYPE. — Transvaal, 2430 (Pilgrim’s Rest): Stry-

dom Tunnel (-BC), Carr 203 (PRE, holo.; K).

A shrub or small slender tree, up to 4 m high; de-

ciduous; bark ± smooth, longitudinally reticulate,

grey. Stems with few or no lateral branching for c.

1-1,5 m above ground; young twigs dull reddish

brown. Leaves opposite, petiolate; lamina narrowly

ovate to elliptic, occasionally broadly ovate to ovate,

(14) 36-60(75) x (10)15-23 mm, discolorous, base

asymmetrical, rounded, apex acute, obtuse or

rounded, apiculate, often twisted, both surfaces

sparsely to densely lepidote, sometimes with tri-

chomes along margin and main vein, principal lat-

eral veins alternate or opposite, in 5-7 pairs, main

vein and reticulate tertiary veining of under surface

prominent, margin smooth, occasionally undulate;

petioles (2)5-8mm long, lepidote, trichomes some-

times present. Inflorescence an axillary spike; pe-

duncles 11-16 mm long, rachis (3)8-11 mm long,

glutinous and lepidote; bracts caducous. Flowers

4-merous. Receptacle glutinous and lepidote; lower

receptacle c. 1,5 mm long, cylindrical, upper recep-

tacle c. 1,5 mm long, campanulate. Sepals yellow, c.

1 mm long, lobes broadly triangular with a few tri-

chomes at apices of lobes. Petals obtriangular,

shortly unguiculate, margins not ciliate, c. 1x1,5

mm, dull yellow. Stamens 8, 1-seriate; filaments

5,5-6 mm long; anthers c.,1 mm long. Disk free for

c. 0,5 mm, purplish pink, tauter part pilose. Style 5

mm long. Fruit a 4— winged samara; subglobose,

16-18x10-15 mm; apical peg 0,5-1 mm long; stipe

3-4 mm long; glabrous but lepidote; light reddish

brown when mature. Cotyledons 2, epigeal. Scales

circular in outline, c. 55-75 mm in diameter deli-

mited by 8 primary radial walls and 6-8 tangential

walls. Fig. 4.

TRANSVAAL. — 2429 (Zebediela): 12,8 km from Malipsdrif

to Ganspoort (-BB), Van Wyk 5243 (PRE; PRU). 2430 (Pilgrim’s

Rest): Strydom Tunnel (-BC), Van der Schijff 7318 (PRE); Abel

Erasmus Pass (-BC), Strey 3454 (K; MO; PRE); Swadini Na-

tional hiking trail (-BD), Van Greuning 513 (PRE; PRU); Ma-

Bothalia 16,1 (1986)

45

riepskop picnic spot (-DB), Van der Schijff 6094 (PRE; PRU).

2529 (Witbank): Fonteinsonderend, Loskop Dam (-AD), Theron

2171 (PRE; PRU); Doornkop (-CB), Du Plessis 422 (PRE;

PRU).

FIG. 4. — Combretum petrophilum. 1, branch with leaves and

inflorescences, X 0,8; 2, flower, X 3 (Strey 3454).

Combretum petrophilum is endemic to the Trans-

vaal. The species usually occurs on north-western or

southern slopes in sourish mixed bushveld. Speci-

mens of the species are found growing between

rocks, in fissures or along ledges. The specific epi-

thet refers to the habitat preference of the species.

C. petrophilum is placed in the section Ciliatipe-

tala Engl. & Diels even though [as in the case of C.

psidioides Welw. subsp. glabrum Exell (1978), which

was also placed there] its petals are not ciliate at the

apex. All other characteristics of the new species are

typical of the section Ciliatipetala.

C. petrophilum is similar to C. apiculatum subsp.

apiculatum. Apart from its petals being without cilia

at the apex it differs mainly in its leaf lamina which is

ovate to elliptic rather than broadly to narrowly obo-

vate-elliptic. The main and secondary veins of the

new species are also much less prominently raised

and not markedly yellow and the flowers are ar-

ranged in laxer spikes than in C. apiculatum subsp.

apiculatum.

The first record of C. petrophilum was from the

Abel Erasmus Pass, where Mr R. G. Strey collected

some fruiting material in March 1960. In November

of the same year he went back and collected flow-

ering specimens. Exell examined the material in

1967 and tentatively identified the taxon as a form of

C. apiculatum.

REFERENCE

EXELL, A.W. 1978. Combretaceae. FI. Zambesiaca 4: 100-183.

E. RETIEF

CYPERACEAE

CYPERACEAE NEW TO THE FLORA OF NATAL

During March 1985 a collecting trip to the Muzi

Swamps in the Ingwavuma District of KwaZulu was

undertaken. The following additions to Ross’s Flora

of Natal (1972) were collected:

1. Cy perns cuspidatus Kunth: Reid 1025.

Annual, previously known from South West Afri-

ca/Namibia, Botswana and Transvaal, also Tropical

Africa. Rodin 4539 (PRE!), collected at Stegi (Sit-

eki), Swaziland, is also this species.

2. Cyperus tenuispica Steud.: Reid 1057.

Annual, previously known from South West Afri-

ca/Namibia, Botswana and Transvaal, also Old

World Tropics.

3. Eleocharis atropurpurea (Retz.) Presl : Reid

1053.

Annual, previously known from South West Afri-

ca/Namibia, Botswana and Transvaal, also world-

wide. Hitchins 827 (PRE!) from Hluhluwe Game

Reserve is also this species.

4. Hemicarpha micrantha (Vahl) Pax : Reid

1026.

Podlech in Prodr. FSWA (1967) is followed in main-

taining Hemicarpha Nees & Arn. separate from Scir-

pus L. s. str. and from Lipocarpha R. Br. Annual,

previously known from South West Africa/Namibia

and Transvaal, also Tropical Africa and throughout

America.

5. Mariscus paradoxus (Cherm.) Cherm. : Reid

1027.

Annual, previously known from South West Afri-

ca/Namibia and Transvaal, also Mozambique and

Madagascar.

6. Pycreus atribulbus (Kukenth.) Napper : Reid

1029.

Perennial, previously known from the coasts of Mo-

zambique and Tanzania. One of the syntypes,

Schlechter 12254, (PRE, iso.!), was collected at 25

Miles Station (Dondo), near Beira, Mozambique.

The species is represented in PRE by a further two

46

Bothalia 16,1 (1986)

collections from Natal : Ward 2918 and Ward 7731,

both from Hlabisa District. In the Muzi Swamps the

species is locally dominant.

7. Pycreus pumilis (L.) Nees : Reid 1034.

Hooper in FWTA edn 2 (1972) is followed in no

longer upholding the var. patens (Vahl) Kiikenth.

Annual, previously known from South West Afri-

ca/Namibia, Botswana, Transvaal and Swaziland,

also widespread in Old World Tropics and Central

America. Pentz & Acocks 10265 (PRE!) collected at

Wessels Nek (Klip River District, Natal) is also this

species.

During the course of routine herbarium work a

further new record for the Flora of Natal was noted:

8. Pycreus macrostachyos (Lam.) J. Raynal.

Previously known from South West Africa/Namibia,

Botswana, Transvaal and Swaziland, also Tropical

Africa. In PRE the following collections from Natal,

which were previously misidentified, represent this

species: Tinley 603 from Mkuze Game Reserve, and

Ward 5574 from the vicinity of Ncemane Station,

Hlabisa District.

C. REID

ERICACEAE

A NEW SPECIES OF ERICINELLA FROM THE SOUTHERN DRAKENSBERG

Ericinella hillburttii E.G.H. Oliver, sp. nov.,

Dracomontibus Capitis Ericinellae multiflorae

Klotzsch Capite orientali affinis sed lobis corollae vi-

ridi-flavae cucullatis, staminibus inclusis, antheris

terminalibus ovario insidentibus, filamentis brevibus

apice latis aristis decurrentibus, foliis adaxiale prae-

cipue basin versus pubescentibus.

Frutex erectus ad 1,5 m altus caulibus multis.

Rami breve lanati glabrescentes, saepe sterigmati-

bus. Folia 3-nata appressa imbricata anguste ovata

ad anguste elliptica ad oblonga 1-1,8 x 0,4-0, 8 mm,

abaxiale glabra, adaxiale pubescentia praecipue ba-

sin versus, juventute longe ciliata, caespite apicali

demum strigulis; petiolo abaxiale puberulo juven-

tute, saepe glabrescenti sed nonnullis pilis persisten-

tibus. Flores l-3(6)-nati extremis brachyblastorum

lateralium coarctatorum; pedicello 1,5-1 ,7 mm

longo puberulo viridi; bractea toto recaulescenti;

bracteolis deficientibus. Calyx 4-lobatus viridis; lobo

majore 0,7-0, 8 mm longo, anguste triangulari, om-

nibus glabris ciliatis apice sulcatis. Corolla 4-lobata

1,2-1 ,8 x 1-1,3 mm late obovoidea, viridi-flava; lo-

bis latis rotundatis partim cucullatis interdum emar-

ginatis. Stamina 4 libera inclusa; filamentis 0,3-0, 6

mm longis linearibus apice expansis glabris; antheris

0,5-1, 1 mm longis ellipsoideis ad obovoideis ovario

insidentibus glabris muticis ad distincte aristatis,

aristis ad 0,2 mm longis partim decurrentibus; poro

longitudine thecae partes aequanti. Ovarium 3-

cellulare, 0,5-0, 8 mm longum late ellipsoideum ad

globosum longitudine porcatum in dimidio su-

periore, supra lanatum; stylo 0, 6-0,9 mm longo gla-

bro; stigmate infundibuliformi 0,4-0, 7 mm diam.

manifesto ad paulo exserto. Fructus capsularis

1,0-1 ,2 mm longus, sparse lanatus, septis base i-i

capsulae partes aequantibus; seminibus anguste el-

lipsoideis c. 0,75 x 0,42 mm testa reticulata cellula-

rum elongatarum impressarum marginibus irregular-

iter undulatis. Fig. 5.

TYPE. — Cape, Elliott District, Baster-voetpad

between Saamwerk and Mt Enterprise, 2 130 m, 16

November 1983, Oliver 8151 (PRE, holo.; BM;

BOL; E; GRA; K; MO; NBG; NU; NY; S; STE).

FIG. 5. — Ericinella hillburttii. 1, flower; 2, bract (as large lobe of calyx); 3, lateral sepal; 4, anther, front, side and back views; 5,

ovary; 6, fruit with one valve removed; 7, leaf; all drawn x 25 from the holotype, Oliver 8151 (STE).

Bothalia 16,1 (1986)

47

Erect compact shrubs up to 1,5 m tall, many-

stemmed from a woody rootstock. Branches shortly

lanate soon becoming glabrous, often with distinct

internodal sterigmata. Leaves 3-nate, appressed im-

bricate, narrowly ovate to narrowly elliptic to ob-

long, 1,0-1 ,8 x 0,4-0, 8 mm, glabrous abaxially, pu-

bescent adaxially mainly towards the base, long ci-

liate and pubescent all over when young and with an

apical tuft which remains as strigulae; petiole pube-

rulous abaxially when young, often glabrescent but

with some long hairs remaining. Flowers l-3(6)-nate

on the ends of short lateral brachyblasts crowded at

the ends of the branches; pedicel 1,5-1 ,7 mm long,

puberulous, green; bract fully recaulescent; brac-

teoles wanting. Calyx 4-lobed, green; larger lobe

0,7-0, 8 mm long, narrowly triangular to subspathu-

late from a broadened base; remaining lobes 0,6-0, 8

mm long, more or less triangular; all lobes ciliate,

the smaller ones pubescent on the inner surface, sul-

cate at the apex, the larger ones more so. Corolla 4-

lobed, 1,2-1 ,8 x 1,0-1 ,3 mm, broadly obovoid,

greenish yellow soon turning pale brown and

papery; lobes broad, rounded, partially cucullate

sometimes emarginate. Stamens 4, free, included;

filaments 0,3-0, 6 mm long, linear, expanded at the

apex, glabrous, the length of the anther; anthers

0,5—1 ,1 mm long ellipsoid to obovoid, seated on top

of the ovary, glabrous, muticous to distinctly aris-

tate, awns up to 0,2 mm long, partially decurrent;

pore the length of the thecae. Ovary 3-celled,

0,5-0, 8 mm long, broadly ellipsoid to globose,

ridged longitudinally in the upper half, lanate above

and mainly down the ridges; style 0,6-0, 9 mm long,

glabrous; stigma infundibuliform, 0,4-0, 7 mm in

diam., manifest to slightly exserted. Fruit a dehis-

cent loculicidal capsule 1,0-1 ,2 mm long, sparsely

lanate with well developed septa at the base 3-5 the

length of the capsule; seeds narrowly ellipsoid, 0,75

x 0,42 mm with reticulate testa of elongate sunken

cells with irregularly undulate margins.

This species was first collected by Prof. Olive Hil-

liard and Mr Bill Burtt as recently as February 1983

when only fruiting material was available. From this

it was nevertheless possible to determine that the

material represented a new species. During Novem-

ber 1983, I visited the locality of their collection to

collect flowering material and to study the species in

the field. En route, material of Ericinella multiflora

Klotzsch, the only other species in the genus in the

eastern Cape, was collected on the Katberg Pass.

This made it possible to compare the two species in

the fresh state. Fig. 6.

FIG. 6. — Distribution of Ericinella hillburttii.

E. hillburttii is allied to E. multiflora, but differs in

a number of characteristics as set out in Table 1.

E. hillburttii is known to date only from the type

locality, the Baster-voetpad in the mountains north-

west of Elliott in the north-eastern Cape.

Accessibility in these mountains is very poor, and

other populations may well occur farther north to-

wards Naude’s Neck.

The plants were among the tallest in the vegeta-

tion, but because of the very dull colour and small

size of the flowers were not very striking even when

in full bloom. The above features, coupled with the

larger stigma and lack of nectaries, suggest the wind

pollination syndrome.

TABLE 1. — -Comparison of characteristics of the allied species, Ericinella hillburttii and E. multiflora

E. hillburttii

E. multiflora

Multi-stemmed from woody rootstock

Distinct internodal sterigmata often present on branches

Leaves adaxially pubescent mainly near the base

Corolla pink, lobes slightly cucullate

Filaments 0,3 -0,6 mm long, expanded at apex

Anthers included, seated on ovary

Anthers terminally attached

Awns partially decurrent

Ovary lanate

Seeds ellipsoid

Single-stemmed

No distinct sterigmata

Leaves adaxially evenly pubescent

Corolla greenish yellow, lobes erect to spreading

Filaments c. 1,4 mm long, linear

Anthers exserted, placed well above ovary

Anthers dorsally attached near base

Awns free

Ovary pubescent

Seeds broadly ellipsoid to subsphaerical

48

Bothalia 16,1 (1986)

One striking feature of E. hillburttii is the large

number of erect stems which arise from the basal

woody rootstock. This is a sure indication that the

plants regenerate quickly from the rootstock after a

fire. This characteristic contrasts strongly with the

single-stemmed habit found in its nearest relative, E.

multiflora , and has probably evolved in response to

the different habitat factors: E. hillburtii grows in

shorter scrub vegetation on more open grassy slopes

whereas E. multiflora grows on the edges of forest

patches and in more sheltered woody scrub patches,

only occasionally on open grassy slopes.

The plants of E. hillburttii occurred in a limited

area on grassy rocky south-facing slopes with a sur-

face layer of very loamy soil acting as a seep. At the

high altitude of 2 100 m the plants have to tolerate

very cold conditions during the winter months, often

with a good covering of snow.

Specimen examined:

CAPE PROVINCE.— 3127(Lady Frere): Baster-voetpad

(-BB), Hilliard & Burn 16662 { STE);ibid. Oliver 81 51 (see type).

E.G.H. OLIVER

FABACEAE

A FOURTH NATURAL ERYTHRINA HYBRID FROM SOUTH AFRICA

Krukoff and Barneby (1974) described two natu-

rally occurring Erythrina hybrids from South Africa,

E. x coddii, an intersubgeneric hybrid of E. latis-

sima E. Mey. (subgenus Chirocalyx ) and E. lysiste-

mon Hutch, (subg. Erythrina), and E. x hennessyae,

an intersectional hybrid of E. lysistemon (subg.

Erythrina section Caffrae) and £. humeana Spreng.

(subg. Erythrina sect. Humeanae).

Subsequently a third natural South African hy-

brid, E. x johnsoniae E.F. Franklin Hennessy, was

described (Hennessy 1985) the parents of which are

E. latissima subg. Chirocalyx) and E. caffra Thunb.

(subg. Erythrina ).

Neither fruit nor seeds have been obtained from

any individuals of these three hybrid taxa, which

suggests that they are sterile.

A fourth, previously unnamed hybrid taxon exists

(Hennessy 1972) which is fertile. Codd (1956) had

already mentioned two specimens with characters in-

termediate between E. lysistemon and E. caffra.

These are the two Codd collections cited below.

Erythrina x dyeri E.F. Franklin Hennessy, hy-

brid. nov. inter parentes E. caffram Thunb. et E.

lysistemon Hutch, (subgen. Erythrina sect. Caffrae )

quasi intermedia, ab ambobus vexillo breviore dif-

fert. Typus: Natal, 2931 (Stanger): Durban, Went-

worth and Brighton Beach (-CC), Hennessy 445

(holotypus Durban-Westville; isotypi NH; PRE).

Tree, c. 10 m tall, branched; bark grey-buff, thin,

smooth with shallow longitudinal fissures; branches

sparingly armed with short, conical or falcate brown

prickles 5 mm long. Leaves pinnately trifoliolate,

terminal leaflet 25-50 mm remote from laterals,

green, chartaceous, minutely pubescent on both sur-

faces when young, becoming glabrous, deciduous;

stipules ovate-lanceolate, 5-6 mm long, caducous;

TABLE 2. — Comparison of measurements of corolla parts of Erythrina caffra, E. lysistemon and E. X dyeri

Bothalia 16,1 (1986)

49

petiole glabrescent, adaxially grooved, sparingly "UP

armed with falcate prickles or unarmed, 70-130 mm

long; stipellae 4, paired, one pair at apex of petiole,

one at apex of rhachis, green, glandular; petiolules 75_

glabrescent, 6-10 mm long; terminal leaflet usually

unarmed, broadly ovate with cuneate base, apex

acute, obtuse or acuminate, 50-130 x 45-130 mm;

lateral leaflets usually unarmed, broadly ovate,

equal- or unequal-sided with cuneate base, apex

acute or acuminate, 50-120 x 45-85 mm. Inflores- x 65_

cence a subterminal pseudoraceme, precocious; pe- £

duncle pubescent, olive green, brown or purple, te- z

rete, 55-200 mm long; bracts ovate-lanceolate, pu- ^

bescent, caducous. Flowers subverticillate in groups

of 3, crowded; pedicel pubescent, 3-6 mm long; s

bracteoles linear-lanceolate, pubescent, caducous, j

distally situated, 2-4 mm long. Calyx tube ± cam- - 55 -

panulate, pubescent, splitting laterally to become bi- 2

labiate at anthesis, olive-green proximally, reddish >

brown distally becoming brown, 12-19 mm long;

lobes 5, obsolescent, thickened, abaxial lobe prog-

nathous in bud. Vexillum conduplicate-falcate, scar-

let, glabrous, spread and re flexed at maturity, 49-58 45 _

E. lysistemon

E. Xdyeri

E. caff ra

•

a • •

▲ AAA

AAA •

A A AAA A •

AAA A A •«

AAA

AAA A •

AA A

AAA •

• •

M •••

• •

• •

FIG. 7. — Superimposed polygonal graphs of eight corolla

characters of plants of Erythrina caffra, E. lysistemon

and E. X dyeri :

\

/-iIq I 26 I 3b mm

ALA LENGTH

FIG. 8. — Scatter diagram of two corolla characters of plants of

Erythrina caffra , E. lysistemon and E. x dyeri.

x 29-36 mm. Alae falcate, long-clawed, greenish

white flushed scarlet with violet or purple distal mar-

ginal zone, 16-21 x 7-9 mm. Carina of two broadly

boat-shaped, abaxially partly connate or rarely free

petals, green spotted with scarlet with purple or vi-

olet border distally, each 14-17 x 9-12 mm; alae and

carina partially exposed at anthesis. Stamens 10, di-

adelphous, vexillary stamen coherent or free, with a

distinct genuflexion proximally; filaments green

proximally, purple distally, of two alternating

lengths 34-41 and 39^16 mm, connate proximally for

24-34 mm; anthers uniform, bithecate with longi-

tudinal dehiscence, dorsifixed, ochreous, 3 mm long.

Gynoecium 47-55 mm long; gynophore green, his-

pid, c. 10 mm long; ovary linear, multiovulate, olive-

green, pubescent, c. 20 mm long; style terminal, te-

rete, hispid proximally, purple, 17-25 mm long; stig-

ma terminal, small, capitate, green. Fruit stipitate,

subligneous, falcate, moniliform, glabrescent, black-

ish, dehiscing adaxially, up to 200 mm long x 13 mm

in diameter in broadest part. Seeds scarlet, elliptic,

8-10, x 5-6 mm; hilum oval, depressed, blackish, c.

5x2 mm.

NATAL. — 2931 (Stanger): Wentworth towards Brighton

Beach (-CC), Hennessy 445 (Durban-Westville; NU; PRE). 3030

(Port Shepstone): Marburg Mission (-CB) Codd 7999 ( K; PRE).

TRANSKEI.— 3128 (Umtata): 6 miles NW of Elliotdale,

(-DC), Codd 7983 (PRE).

This hybrid of the winter-flowering species E. caf-

fra and E. lysistemon, both members of section Caf-

frae, occurs on the coastbelt of Natal and Transkei

where the parent species are either sympatric or

grow together in cultivation.

50

Bothalia 16,1 (1986)

Of the four hybrid taxa E. x dyeri is the most dif-

ficult to recognize. The difficulty is compounded by

introgression.

Living specimens of the parent species are not dif-

ficult to distinguish on inflorescence and floral

characters. Vegetative differences are not as well de-

fined although, in general E. cafjra is a bigger tree

with fewer prickles and larger leaflets than E. lysiste-

mon.

The two most conspicuous differences between

the parent species are the shape of the inflorescence,

which is attributable to the shape and attitude of the

vexilla of the open flowers, and the colour of the

vexillum. Because the vexillum of E. caffra is

strongly arcuate, spread and re flexed at anthesis

thus exposing the inner whorls, the inflorescence is

broader than that of E. lysistemon in which the mar-

gins of the slightly falcate, conduplicate vexillum re-

main contiguous at anthesis, concealing the inner

whorls. The shape of the inflorescence is fairly well

preserved in dried specimens. The colour of the vex-

illum of both parent species varies in intensity in dif-

ferent individuals. That of E. caffra ranges from

deep vermilion-red through orange to creamy-white

whereas that of E. lysistemon ranges from deep scar-

let through pillar-box red to pink to white. Albino

forms of both species are very rare. Colour is not

preserved in dried specimens.

The shape of the inflorescence of E. x dyeri re-

sembles that of E. caffra whereas the colour of the

vexillum is like that of E. lysistemon. Living speci-

mens of E. x dyeri are often misidentified as red-

flowered plants of E. caffra. No pale colour forms of

the hybrid have yet been found. Presumably both

FIG. 9. — Outlines of corolla parts of A, Erythrina caffra ; B, E.

x dyeri and C, E. lysistemon.

FIG. 10. — Inflorescence: A, E. caffra ; B, E. lysistemon; C, E. x dyeri. Young infructescence: D, E. x dyeri.

Bothalia 16,1 (1986)

51

parents would have to be pale forms for such lack of

colour to manifest itself in the hybrid.

In order to determine the floral characters which

can be reliably and easily used to identify the F, hy-

brid and to distinguish it from its parents, inflores-

cences were obtained from each of five different

trees of E. caffra and of E. lysistemon and one tree

of E. x dyeri in the Durban area. Measurements

were made of 65 flowers of each parent species and

of 55 flowers of the hybrid. Those of the corolla

parts are summarized in Table 2. The apparent

anomaly of one measurement, vexillum length,

which is not intermediate in the flower of the hybrid

vanishes when the relative proportions of the corolla

parts are considered. These are shown in Table 3

and Figs 7 & 8.

Shrinkage of flower parts occurs in drying, but the

relative proportions of these parts do not change. By

estimating the relative proportions of the corolla

parts of dry herbarium specimens, rehydrated speci-

mens or fresh material, it is possible to distinguish E.

caffra, E. lysistemon and the Fj hybrid, E. x dyeri.

Corolla parts of these three taxa are shown in Fig. 9

and photographs of living inflorescences in Fig. 10.

The name Erythrina x dyeri is proposed for this

taxon in honour of Dr R.A. Dyer.

ACKNOWLEDGEMENTS

The assistance of Mr V. K. Bansi and of Misses J.

R. Dhanjee, Y. Moodley, S. Naidoo, S. Naidu and

P. Reddy in compiling measurements is acknow-

ledged and I thank Mrs R. Bunsee for typing the

manuscript and Mrs E. L. van Hooff for the photo-

graphs.

REFERENCES

HENNESSY, E. F. 1972. South African Erythrinas. Natal

Branch, Wildlife Protection and Conservation Society of

South Africa, pp. 14-19.

HENNESSY, E. F. 1985. Erythrina x johnsoniae. Flower. PI.

Afr. 50: t. 1911.

KRUKOFF, B. A. & BARNEBY, R. C. 1974. Conspectus of

species of the genus Erythrina. Lloydia 37,3: 440-449.

E. F. HENNESSY*

* Department of Botany, University of Durban-Westville, Pri-

vate Bag X54001, Durban 4000, South Africa.

IRIDACEAE

THE CORRECT CITATION OF MONTBRET1A CROCOSMIIFLORA

The name Montbretia crocosmiiflora is generally

attributed to Lemoine ex E. Morren in La Belgique

Horticole 31: 299 & t. 14 (1881). Dr M. P. de Vos

also cited it in this form in her revision: The African

genus Crocosmia Planchon, published in Jl S. Afr.

Bot. 50,4: 497 (1984). However, Morren’s publica-

tion is antedated by an unsigned note in the Floral

Magazine of October 1881, t. 472, as noted by Dr P.

J. Kostelijk in The Plantsman 5,4: 248 (1984).

On p. 287 of La Belgique Horticole 31, mention is

made of the death of Gerard Calopin on 18 Decem-

ber 1881. The issue containing Morrens’s description

of Montbretia crocosmiaeflora must therefore have

been published in the last days of December 1881 or

in 1882, whereas the Floral Magazine is dated Oc-

tober 1881. I am not aware of any data indicating

that this dating is incorrect.

The editors of the Floral Magazine, Burbidge and

Dean, must be held responsible for the description

with Fitch’s drawing. It therefore appears that the

correct citation of the hybrid garden Crocosmia is

Crocosmia x crocosmiiflora (Lemoine ex Burbidge

& Dean) N.E. Br.

The plants shown in the drawings in La Belgique

Horticole and in the Floral Magazine are very much

alike; they could well belong to the same cultivar.

However, it does not seem possible to assign a culti-

var name to these plants. The firm of Lemoine at

Nancy introduced the plant into commerce in

1882 (Lemoine, J. Roy. Hort. Soc. 25: 128-132,

1900-1901). Named selections have been sold from

1883 onwards.

D. O. WIJNANDS*

* Department of Plant Taxonomy, Agricultural University Wag-

eningen, P.O. Box 8010, 6700 ED Wageningen, The Netherlands.

LAMIACEAE

A NEW SPECIES OF STACHYS

Stachys comosa Codd, sp. nov., a S. natalensi

Hochst. var. galpinii (Briq.) Codd inflorescentia

compacta, bracteis longioribus et angustioribus dif-

fert.

Herba, perennis; caules decumbentes, graciles,

parce ramosi, 0,25-0,4 m longi, villosi. Folia subses-

silia vel breviter petiolata; lamina ovata vel ovato-

deltoidea, 20-35 x 14-24 mm, dense villosa, subtus

glanduloso-punctata, apice obtuso, basi truncata vel

subcordata, margine crenata; petiolus usque ad 7

mm longus. Inflorescentia compacta, dense villosa,

30-40 mm longa; bracteae lineari-lanceolatae vel

lanceolatae, acuminatae, 10-14 mm longae; verticil-

lastri 2-flori; pedicelli 1 mm longi. Calyx tubuloso-

campanulatus, 8-10 mm longus, aequaliter 5-denta-

tus, villosus; tubus 4-5 mm longus; dentes lineari-

lanceolati, aristati, 4—5 mm longi. Corolla alba vel

52

Bothalia 16,1 (1986)

dilute malvina, saepe purpureo-maculata, 14-16 mm

longa; tubus 7-8 mm longus; labium posticum hori-

zontale, concavum, 5-6 mm longum; labium anti-

cum descendens, 8-9 mm longum. Stamina 4, intus

labio postico ascendentia; antherae brunneae. Stylus

filiformis, 4—5 mm exsertus ; stigma breviter bifidum.

TYPE. — Natal, Umzinto District, Vernon

Crookes Nature Reserve, Balkwill & Cadman 2082

(PRE, holo.).

Perennial herb; stems decumbent, slender, spar-

ingly branched, 0,25-0,4 m long, villous, with long

spreading hairs and shorter gland-tipped hairs.

Leaves subsessile or shortly petiolate; blade ovate to

ovate-deltoid, 20-35 x 14 — 24 mm, densely villous on

both surfaces with minute gland-dots on the under

surface, apex obtuse, base truncate to subcordate,

margin crenate; petiole up to 7 mm long. Inflores-

cence compact, densely villous, 30-40 mm long;

bracts linear-lanceolate to lanceolate, acuminate,

10-14 mm long, the lowermost pair larger and leaf-

like; verticils 2-flowered; pedicels 1 mm long. Calyx

tubular-campanulate, 8-10 mm long, equally

5-toothed, villous; tube 4-5 mm long; teeth linear-

lanceolate, aristate, 4-5 mm long. Corolla white to

pale mauve often flecked with purple, 14—16 mm

long; tube 7-8 mm long; upper lip horizontal, con-

cave, 5-6 mm long; lower lip descending, 8-9 mm

long. Stamens 4, ascending in the upper lip; anthers

brown. Style filiform, bifid, exserted in the upper lip

by 4-5 mm. Fig. 11.

Found on grassy slopes, often among rocks, in

southern Natal; flowers in spring and early summer,

especially after the grass has been burnt.

NATAL. — 3030 (Port Shepstone): Dumisa (-AD), Rudatis

2033\ Vernon Crookes Nature Reserve (-BC), Balkwill, Man-

ning, Brophy & Getliffe Norris 1013: Balkwill & Cadman 2082;

2209 ; near Harding, Bedford Farm (-BC), Balkwill & Cadman

2794.

During recent studies in Stachys (Codd in Bothalia

12: 181, 1977; FI. S. Afr. 28,4: 51, 1985), the Rudatis

specimen listed above was included, with some hesi-

tation, in S. natalensis var. galpinii because of its

somewhat similar densely villous pubescence,

though it differed in the compact inflorescence and

longer and narrower floral bracts. Since then several

matching specimens have been collected by Balkwill

and his colleagues of the University of Natal, Pieter-

maritzburg, which support the view that a suffi-

ciently clear-cut entity is involved to warrant sepa-

rate species status.

In the keys previously provided, S. comosa would

tend to run to S. flexuosa Skan because of its rela-

tively compact inflorescence. However, plants of the

latter species are smaller, with stems up to 0,25 m in

length and the bracts and calyx are hispid, not

densely villous as in S. comosa.

FIG. 11. — Holotype of Stachys comosa, Balkwill & Cadman

2082.

A NEW SPECIES OF THORNCROFTIA

Thorncroftia media Codd, sp. nov., a T. succu-

lenta (Dyer & Bruce) Codd pubescentia pilis simpli-

cibus,foliissubintegris, inflorescentialaxioradiffert.

Frutex semisucculentus, c. 0,6 m altus, basi ramo-

sus; caules erecti, parce ramosi, teretes, demum gla-

brescentes. Folia petiolata, semisucculenta; lamina

elliptica vel ovato-elliptica, 35-65 x 25-35 mm,

utrinque tomentosa et glanduloso-punctata pilis sim-

plicibus, apice rotundato, basi cuneata, margine sub-

integra; petiolus 20-30 mm longus. Inflorescentia

terminalis, paniculata, satis condensata, usque ad

250 mm longa, 100 mm lata; rhachis glanduloso-to-

mentosa; bracteae persistentes, basi racemi foli-

aceae, superne sensim reductae demum c. 4 mm

longae, floribus axillaribus solitariis, pedicellis 1-2

mm longis. Calyx campanulatus, glandulosus, ali-

quantum bilabiatus, demum 5-6 mm longus; lobus

posticus ovato-deltoideus, 2 mm longus; lobus anti-

cus subaequaliter 4— dentatus, dentibus lanceolato-

deltoideis, acuminatis, 1,5 mm longis. Corolla tubu-

losa, apice 4-lobata, lilacina, lobis purpureo-macula-

tis; tubus anguste cylindricus, 20-22 mm longus, 2

mm diam., leviter compressus, breviter glanduloso-

tomentosus; lobus posticus erectus, obcordatus 5-6

x 4 mm, lobi laterales deflexi, lineari-lanceolati,

acuminati, 5-6 x 1 mm; lobus anticus cymbiformis,

6-7 mm longus, demum reflexus. Stamina 4, fila-

mentis liberis, 3-4 mm longis, fauce corollae insertis.

Bothalia 16,1 (1986)

53

Ovarium 4-lobatum, glabrum; stylus filiformis, 7

mm exsertus; stigma breviter bifidum.

TYPE. — Transvaal, Drakensberg range, west of

Trichardtsdal (2430CC), cultivated in BRI nursery.

Hardy 3966 (PRE, holo.).

Semi-succulent shrub, c. 0,6 m tall, branching at

the base; stems erect, sparingly branched, terete,

eventually glabrescent. Leaves petiolate, semi-suc-

culent; blade elliptical to ovate-elliptical, 35-65 x

25-35 mm, tomentose and gland-dotted on both sur-

faces with simple hairs, apex rounded, base cuneate,

margin subentire; petiole 20-30 mm long. Inflores-

cence terminal, paniculate, fairly dense, up to 250

mm long, 100 mm broad; rhachis glandular-tomen-

tose; bracts persistent, those at the base of the

raceme leaf- like, becoming smaller towards the apex

and eventually c. 4 mm long, with the flowers axil-

lary, solitary, pedicels 1-2 mm long. Calyx campanu-

late, glandular, somewhat bilabiate, eventually 5-6

FIG. 12. — Holotype of Thorncroftia media, Hardy 3966.

mm long; posticous lobe ovate-deltoid, 2 mm long;

anticous lobe subequally 4-toothed, teeth lanceo-

late-deltoid, acuminate, 1,5 mm long. Corolla tubu-

lar with the apex 4-lobed, lilac with the lobes purple-

flecked; tube narrowly cylindrical, 20-22 mm long, 2

mm in diam., slightly compressed, shortly glandular-

tomentose; posticous lobe erect, obcordate, 5-6 x 4

mm; lateral lobes deflexed, linear-lanceolate, acumi-

nate, 5-6 x 1 mm; anticous lobe cymbiform, 6-7 mm

long, eventually reflexed. Stamens 4, filaments free,

3-4 mm long, attached in the throat of the corolla.

Ovary 4-lobed, glabrous; style filiform, exserted by 7

mm; stigma shortly bifid. Fig. 12.

Known only from the type gathering ( Hardy 966),

collected on rocky slopes of the Drakensberg Range

in the eastern Transvaal, north-west of the Olifants

River poort and more or less due west of

Trichardtsdal, and cultivated in the BRI nursery.

Access to this part of the Drakensberg Range is diffi-

cult, as there are no roads into the mountains be-

tween the Olifants River and the road over the

mountains at The Downs, some 45 km to the north-

west. The opportunity to make a small collection in

the area arose when Mr Hardy accompanied a heli-

copter expedition engaged on the eradication of

Cannabis plantings.

In corolla characters and growth habit T. media

resembles T. succulenta but, in the latter species, the

leaf blade tends to be smaller (16-30 x 15-20 mm)

with the margin crenate in the upper half, and the

tomentum on leaves and stems consists mainly of

dendroid hairs. In T. succulenta the inflorescence is

also smaller and denser than in T. media. T. succu-

lenta occurs on the Soutpansberg and has been re-

corded on the escarpment opposite Mariepskop and

from the mountains east of Barberton. It is possible,

therefore, that it may occur at other localities along

the eastern escarpment but, in the short time at his

disposal, Mr Hardy did not encounter it. Obviously

more information is desirable on the distribution of

the two species and whether there is any intergrada-

tion between them, but there seems to be little

chance of further study being possible in this moun-

tainous area in the near future.

Another related species, T. longiflora N.E. Br.,

which has so far been recorded only from the moun-

tains above Joe’s Luck Siding east of Barberton, has

a similar greyish tomentum of simple hairs but the

leaves are very much smaller (10-20 x 4-10) and the

corolla tube is considerably longer (30-38 mm) than

in T. media.

L. E. CODD

LILIACEAE

A METHOD FOR THE NON-DESTRUCTIVE EXAMINATION OF LEAVES

OF ALOE SP

Cutler and Brandham (1977; see also Cutler 1969,

1972 and Brandham & Cutler 1978) have shown that

the leaf surface structure of members of Aloineae is

not only under precise genetic control with little if

BY SEM

any environmental influence, but is also species (or

infra-specific taxon) specific. This means that it can

be used to identify otherwise unidentifiable speci-

mens, and to assist in the identification of others.

54

Bothalia 16,1 (1986)

Techniques are known for preparing surface-ana-

tomical material of dried herbarium specimens with-

out disturbing them and causing minimal distortion.

The making of replicas is a well known method of

examining otherwise difficult material by both scan-

ning and transmission electron microscopy. The ap-

plication of replicas to scanning electron microscopy

(SEM) is reviewed by Hearle et al. (1974). As part of

a study of the southern African species of Aloe for

the Flora of southern Africa, we experimented with

a number of replica techniques, notably those of

Chapman (1967) using polystyrene foam dissolved in

toluene, of De Winter (pers. comm.) using clear nail

varnish, and of Watkins (pers. comm.). We found

Watkins’ technique of using a small piece of cellu-

lose acetate film soaked in acetone was ideal for ap-

plication to valuable specimens, because the dry ace-

tate replica lifts off the specimen without damaging

it, and removes at most some of the surface wax and

dust. A piece of cellulose acetate film c. 10 x 5 mm

(the dimensions are not critical) is softened for a few

seconds in acetone and laid on a suitable piece of

leaf epidermis. A drop of acetone on the film will

further soften it so much that all but the smallest air

bubbles are removed, and an impression of the leaf

is formed. After ten to twenty minutes the film dries

and releases itself from the specimen. It may then be

trimmed, if necessary, and mounted and coated for

viewing by SEM in the normal manner.

As the material actually examined is a negative

replica, it must be noted that papillae, micropapillae

and ridges appear on the pictures as hollows of vari-

ous shapes, and stomatal cavities appear as humps.

Wax deposits appear ‘right way round’, but are seen

FIG. 13. — Live material of leaf of Aloe namibensis, Hardy 6330

in PNBG 28193. The scale bar is approximately 135 pm

long.

FIG. 14. — Acetate replica of a specimen of Aloe namibensis,

Giess 10459 in PRE. The scale bar is approximately 140 /am

long.

FIG. 15. — Negative of Fig. 14. The scale bar is approximately

140 pm long.

Bothalia 16,1 (1986)

55

from below rather than from above. The practical

effect of this may be seen by comparing the picture

of live material of Aloe namibensis in Fig. 13 with

the replica in Fig. 14.

An interesting optical illusion may be noted when

using these replicas (see Fig. 15). This is the negative

of Fig. 14. In negatives of SEM pictures of replicas

made as described above, hairs, papillae, micropa-

pillae, ridges and hollows are seen in an approxima-

tion to their correct perspective; this is seen when

Fig. 15 is compared with Fig. 13.

An example of the use of characters made avail-

able by this method, that would otherwise have been

unavailable to the investigators, is given by Glen &

Hardy (in press).

We thank Mr R. M. Watkins (Wirsam Scientific)

for suggesting this method to us, and Mrs S. M. Per-

old for help with the minutiae of SEM specimen

preparation. Thanks are also due to Drs G. E. Gibbs

Russell and O A. Leistner for reading drafts of this

note and offering helpful comment and advice.

REFERENCES

BRANDHAM, P. E. & CUTLER, D. F. 1978. Influence of chro-

mosome variation on the organization of the leaf epidermis

in a hybrid Aloe (Liliaceae). Bot. J. Linn. Soc. 77: 1-16.

CHAPMAN, B. 1967. Polystyrene replicas for scanning reflexion

electron microscopy. Nature, Lond. 216: 1347-1348.

CUTLER, D. F. 1969. Cuticular markings and other epidermal

features in Aloe leaves. Notes Jod. Lab. 6: 21-27.

CUTLER, D. F. 1972. Leaf anatomy of certain Aloe and Gasteria

species and their hybrids. In A.K.M. Ghouse & M. Yunus,

Research trends in plant anatomy, 103-122. New Delhi:

Tata McGraw-Hill.

CUTLER, D. F. & BRANDHAM, P. E. 1977. Experimental evi-

dence for genetic control of leaf surface characters in hy-

brid Aloineae (Liliaceae). Kew Bull. 32: 23-32.

GLEN, H. F. & HARDY, D. S. in press. The identity of Aloe

spicata L.f. Bothalia.

HEARLE, J. W. S., SPARROW, J. T. & CROSS, P. M. 1974.

The use of the scanning electron microscope. Oxford: Per-

gamon Press.

H. F. GLEN and D. S. HARDY

MESEMBRYANTHEMACEAE

A NEW COMBINATION IN LAMPRANTHUS

A proposal was made (Glen 1980) and accepted

(Brummitt 1983) to conserve the generic name Lam-

pranthus N.E. Br. against the earlier name Oscularia

Schwant. if both names were considered to refer to

the same genus. Glen (1978) showed that all three

names published in the genus Oscularia belong to

the same species, that this species could not be held

separate from the genus Lampranthus, and that it

belongs to the section Lunati of that genus. How-

ever, the necessary new combination has not been

published until now.

Lampranthus deltoides (L.) Glen, comb. nov.

Mesembryanthemum deltoides L., Sp. PI. edn 1, 482 (1753);

Gouan, Hort. Monspel. 244 (1765); Mill., Gard. Diet, edn 8, n.

11 (1768); Soland. in Ait., Hort. Kew. edn 1, 2: 183 (1789);

Gmel., Syst. Nat. edn 14, 2: 844 (1791); Haw., Obs. Gen. Me-

semb. 2: 364 (1795); Willd., Sp. PI. edn 5, 2: 1052 (1799); DC.,

Hist. PI. Succ. t. 53 (1800); Haw., Misc. Nat. 74 (1803); Willd.,

Enum. PI. Hort. Berol. 539 (1809); Haw., Syn. PI. Succ. 296

(1812); Hornem., Hort. Reg. Hafniae 465 (1815); Haw., Rev. PI.

Succ. 133 (1821); DC., Prodr. 3: 433 (1828); Salm Dyck, Monogr.

Gen. Aloes Mesemb. § 30 f. 3 t. 24 (1840); D. Dietr., Syn. PI. 3:

140 (1843); Sond., F. C. 2: 421 (1862); Berger, Mesemb. Portu-

lac. 190 (1908); N.E. Br. in J1 Linn. Soc. Bot. 45: 118 (1920). M.

deltoideum L., Syst. Nat. edn 10, 1059 (1758). Oscularia deltoides

(L.) Schwantes in Mollers dt. Gartn.-Ztg 42: 187 (1927); Jacob-

sen, Handb. Succ. PI. 3: 1338 (1960); in Lex. Succ. PI. 535 (1974).

Iconotype: Dill., Hort. Eltham. t. 195 f. 246 (1732). Typotype:

Dillenius s.n., hort. (OXF!).

M. caulescens Mill., Gard. Diet, edn 8, n. 12 (1768); Haw.,

Obs. Gen. Mesemb. 2: 367 (1795); in Misc. Nat. 74 (1803); in

Syn. PI. Succ. 296 (1812); Hornem., Hort. Reg. Hafniae 465

(1815); Haw., Rev. PI. Succ. 133 (1821); DC., Prodr. 3: 433

(1828); Salm Dyck, Monogr. Gen. Aloes Mesemb. § 30 f. 3 t. 23

(1840); D. Dietr., Syn. PL 3: 140 (1843); Sond., F.C. 2: 421

(1862); Berger, Mesemb. Portulac. 188 (1908). M. deltoides L.

var. 6 L., Sp. PI. edn 2, 690 (1762); Soland. in Ait., Hort. Kew.

edn 1, 2: 183 (1789). O. caulescens (Mill.) Schwantes in Mollers

dt. Gartn.-Ztg 42: 187 (1927); Jacobsen, Hand. Succ. PI. 3: 1338

(1960); in Lex. Succ. PI. 535 (1974). Iconotype: Dill., Hort. El-

tham. t. 195 f. 243-4 (1732). Typotype: Dillenius s.n., hort.

(OXF!).

M. deltoides L. var. majus Weston, Universal Botanist 1: 169

(1770); Haw., Obs. Gen. Mesemb. 2: 366 (1795); N.E. Br. in J1

Linn. Soc. Bot. 45: 118 (1920). O. deltoides (L.) Schwantes var.

major (Weston) Schwantes ex Jacobsen in Fedde Reprium Beih.

106: 158 (1938); Jacobsen, Handb. Succ. PI. 3: 1338 (1960); in

Lex. Succ. PL 535 (1974). O. deltata (Mill.) Schwantes var. major

(Weston) Schwantes in Mollers dt. Gartn.-Ztg 42: 187 (1927).

Iconotype: Dill., Hort. Eltham. t. 196 f. 247 (1732). Typotype:

Dillenius s.n., hort. (OXF!).

M. muricatum Haw., Obs. Gen. Mesemb. 2: 364 (1795); in

Misc. Nat. 75 (1803); in Syn. PL Succ. 297 (1812); in Rev. PL

Succ. 133 (1821); DC., Prodr. 3: 433 (1828); Salm Dyck, Monogr.

Gen. Aloes Mesemb. § 30, f. 3 t. 25 (1840); D. Dietr., Syn. PL 3:

140 (1843); Sond., F.C. 2: 421 (1862). M. deltoides L. var. murica-

tum (Haw.) Berger, Mesemb. Portulac. 190 (1908). O. muricata

(Haw.) Schwantes ex Jacobsen, Sukk. Lex. 478 (1970) in syno-

nymy, comb, illegit. Iconotype: Dill., Hort. Eltham. t. 195 f. 246

(1732). Typotype: Dillenius s.n., hort. (OXF!).

M. deltoides L. var. simplex DC., Hist. PL Succ. t. 53 (1800).

Type not stated.

M. deltoides L. var. pedunculatum N.E. Br. in J1 Linn. Soc.

Bot. 45: 118 (1920). O. deltoides (L.) Schwantes var. pedunculata

(N.E. Br.) Schwantes ex Jacobsen in Fedde Reprium Beih. 106:

158 (1938). O. pedunculata (N.E. Br.) Schwantes in Natn. Cact.

Succ. J1 4: 58 (1949); Jacobsen, Handb. Succ. PL 3: 1338 (1960);

in Lex. Succ. PL 535 (1974). Syntypes: Schlechter 9045, Nuwe-

kloof (K!); Scott Elliot 228, Nuwe Kloof (K!).

Oscularia deltata (Mill.) Schwantes in Mollers dt. Gartn.-Ztg

42: 187 (1927). Type: not stated, but apparently based on Miller’s

description of M. deltoides.

The pre-Linnaean citations listed by Glen (1978)

are not repeated here, in order to save space.

REFERENCES

BRUMMITT, R. K. 1983. Report of the Committee for Sperma-

tophyta 25. Taxon 32: 279-284.

GLEN, H. F. 1978. A taxonomic monograph o/Lampranthus and

allied genera. Ph. D. thesis, University of Cape Town.

GLEN, H. F. 1980. Proposal (537) to conserve the generic name

2405 Lampranthus N.E. Br. (1930) against Oscularia

Schwantes (1927) (Mesembryanthemaceae). Taxon 29:

693-694.

H. F. GLEN

I

56 Bothalia 16,1 (1986)

ORCHIDACEAE

NOTES ON THE DISINAE FOR THE FLORA OF SOUTHERN AFRICA

While checking the account of the Disinae for the

Flora of southern Africa , the following problems re-

quiring discussion or resolution were found.

1. Disa subgen. Micranthae has not been validly

published. In a previous account (Linder 1981: 9) I

referred to a ‘subgen. Micranthe’, but did not indicate

the type or basionym. Disa Berg, subgen. Micran-

thae (Lindl.) Linder, stat. nov., Disa Berg. sect. Mi-

cranthae Lindl., Gen. Sp. Orch. 347 (1838). Lecto-

type species: Disa chrysostachya Swartz.

2. Disa subgen. Hircicornu (Kraenzl.) Linder is

based on Disa sect. Hircicornes Kraenzl. I changed

the ending of the name in 1981 in order to satisfy

Recommendation 21 B 1 of the ICBN (1978), which

states: ‘The epithet of a subgenus or a section is pref-

erably a substantive. . However, article 73.1

states that ‘The original spelling of a name or epithet

is to be retained, except for the correction of typo-

graphic or orthographic errors’. Hircicornu cannot

be regarded as having an ‘incorrect Latin termina-

tion’, which can be corrected. Recommendation 21

B 2, stating that new epithets for subdivisions of gen-

era should have the same form as already existing

names of co-ordinate rank, implies that the epithets

are not to be ‘corrected’, so the correct name of the

subgenus would be Disa Berg, subgen. Hircicornes

(Kraenzl.) Linder.

3. Disa longicornu L.f. is the correct spelling of

what all subsequent authors (except Linder 1982),

have called ‘Disa longicornis’, according to article

73.1 of the Code.

4. Disa maculomarronina is the name that Mc-

Murtry (1984) gave to a population that Linder

(1981: 146) described, but did not name, as a hybrid

between Disa versicolor and D. hircicornis.

Taxonomically, this is a difficult species. Disa

maculomarronina can readily be separated from D.

hircicornis by the petals which curve over the anther,

and which are ovate and acute, and it can be separ-

ated from the South African collections of D. versi-

color by its constant colouration and only gradually

decurved spur. However, the material of D. versico-

lor from Zimbabwe is problematic, as in spur shape

and orientation it ranges from D. maculomarronina

to D. hircicornis, it is generally robust as in D. versi-

color and according to various reports, the coloura-

tion is as in D. maculomarronina. The type of D.

versicolor, which is also the only collection from An-

gola, is the same as the South African D. versicolor

in all respects. The resolution of the problem in Zim-

TABLE 4. — Application of early names of Disa patens and

D. filicomis based on the names and types of Linnaeus

the Younger

L.f. (1784) Orchis filicomis Ophrys patens

Thunberg (1794) Limodorum longicome Serapias patens

Swartz (1800) Disa patens Disa tenui folia

Thunberg (1807) Disa filicomis Disa patens

Lindley (1838) Penthea filicomis Penthea patens

babwe will probably have to wait until the popula-

tions can be studied in the field.

5. Disa patens I filicomis. The nomenclatural his-

tory of these two quite distinct species has been

much confused. The problem dates to the early his-

tory of the usage of the names (Table 4).

The diagnoses and typification of the names of

Linnaeus the Younger, despite his incompetence at

the generic level, are clear and sound. Thunberg

(1794), in his Prodromus, transferred Ophrys patens

L.f. to Serapias, but from his diagnosis (‘Serapias fo-

lds lanceolato-setaceis, spica ovata, floribus perpen-

dicularibus’) it appears as if he is referring to O. fil-

icornis L.f. For Orchis filicomis he published a new,

superfluous name, Limodorum longicome, which

refers to a Mystacidium.

The confusion started with Olof Swartz’s (1800)

treatment of the group (Fig. 16). In his paper, he

lists the species with their synonyms indented, and

with descriptions provided as footnotes. New names

are printed in italics, while everything else is in ro-

man type. He clearly followed the circumscription,

rather than the typification method of nomencla-

ture, and he also recognized the two species. How-

ever, following the circumscription method, he

placed Serapias patens sensu Thunberg in the same

species as Orchis filicomis L.f. (from their descrip-

tion they are the same), under the name Disa patens.

He is consistent throughout his paper, preferring

Thunberg epithets (usually superfluous) to those of

Linnaeus the Younger or the Elder. Consequently

‘patens’ is printed in roman type, as Swartz regarded

it as a ‘new combination’, not as a new name. How-

ever, as Swartz explicitly excluded the type of Sera-

pias patens (which is Ophrys patens L.f., which

Swartz placed into the other species), by the type

method ‘he is considered to have published a new

name that must be ascribed solely to him’ (ICBN,

1983, article 48.1), so it should read: Disa patens

Swartz. Further support for the notion that Swartz

regarded D. patens as a ‘new combination’ is the fact

that Swartz nowhere in his paper replaced earlier

names with ‘more appropriate’ names, the way that

Thunberg did.

For Ophrys patens L.f. Swartz proposed a new

name, Disa tenuifolia Swartz. This name can be re-

garded as being superfluous, the correct name being

Disa patens, which was being blocked simultane-

ously by being erroneously applied to Orchis filicor-

nis. This appears to have been the interpretation of

all authors to date. The other possible interpretation

is that it is an avowed substitute (nomen novum), as

there is already a Disa patens. This latter interpreta-

tion seems better. Swartz, using the circumscription

method, and consistently preferring the last epithet

applied to a species, would have regarded Serapias

patens, and hence Disa patens, as the ‘correct’ name

for Orchis filicomis, thus blocking Disa patens (L.f.)

Thunb., a later homonym for Disa patens Swartz,

which Thunberg established in his Flora Capensis of

Bothalia 16,1 (1986)

57

514 i8oot £ful. Aug . Sept.

D. tmuifolia Sw. patens.

Serapias Th.

Ophrys patens Orchis filicornis

fuppl, fuppL

3. SATYRIUM. (Thunb.) Tab. III. C.

Char, ejfent. Calyx ringens : foliolo fuperiore for-

nicato, poftice bicalcarato, ceteris

labeltoque baft coalito.

Anthera Jlylo elongato adnata fub

Stigmate terminals

Char after naturalis.

Calyx ringens, 5-phyllus:

Foliola omnia bafi coalita. Tria exteriora,

qvorum unum fuperius f. pofterius maxi-

mum , fornicatum , ball Calcaria duo variae

longitudinis poftice exferens; duo anteriora,

lanceolato-linearia.

Duo interiora minora ftylo foliolisque ex-

terioribus baft accreta.

Cor.

D. tenuifolia: galea acuminata ere£lo-patens concava

ecalcarata , labello fililormi; caule fubbifloro,

foliis fetaceis.

D. patent: galea acuminata erefto-patens concava

ecalcarata; labello filiformi; fpica ovata multi-

flora; foliis lineari-lanceolatis.

FIG. 16. — The treatment of Disa tenuifolia and D. patens by

Swartz. The synonymy and names are given at the top of

the page, new names are printed in italics. The diagnoses

are given at the bottom of the page.

1807, a work in which he ignored Swartz’s earlier

work. Unfortunately, authors in the 19th and 20th

centuries managed to get the names confused. Lind-

ley (1838) transferred both Ophrys patens and Or-

chis filicornis to Penthia, a treatment followed by

Rolfe (1913). However, Schlechter (1901), Kraenz-

lin (1900), Bolus (1911) and Linder (1981, 1982) mis-

takenly upheld Disa patens Swartz, a name which is

clearly superfluous, and so illegitimate. Linder

(1985) interpreted Disa tenuifolia Swartz as super-

fluous, and proposed D. lutea Linder as an avowed

substitute, a name which would now have to be re-

garded as superfluous.

The full synonymy of the two species is now:

1. Disa filicornis (L.f.) Thunb. (1807); Orchis

filicornis L.f. (1784); Limodorum longicorne Thunb.

(1794); Disa patens Swartz (1800); Penthea filicornis

(L.f.) Lindl. (1838). Penthea reflexa Lindl. (1838);

Disa reflexa (Lindl.) Reichb.f. (1865).

2. Disa tenuifolia Swartz (1800); Ophrys patens

L.f. (1784); Serapias patens Thunb. (1794); Disa pa-

tens (L.f.) Thunb. (1807) non Swartz (1900); Pen-

thea patens (L.f.) Lindl. (1838); Disa lutea Linder

(1985).

REFERENCES

BOLUS, H. 1911. leones Orchidearum Austro- Africanarum. Vol.

2. London: Wesley.

KRAENZLIN, F. 1900. Orchidacearum , Genera et species. I.

Berlin: Mayer & Muller.

LINDER, H. P. 1981. Taxonomic studies on the Disinae. III. A

revision of Disa Berg, excluding sect. Micranthae Lindl.

Contr. Bolus Herb. 9.

LINDER, H. P. 1982. Disa. In J. Stewart et al., Wild orchids of

southern Africa. Johannesburg: MacMillan.

LINDER, H. P. 1985. A new name for Disa patens. Bothalia 15:

553.

LINDLEY, J. 1830-1840. Genera and species of orchidaceous

plants. London: Ridgways.

LINNAEUS, C. (fil.) 1781. Supplementum Plantarum. Brunsvi-

gae.

McMURTRY, D. 1984. Disa maculomarronina. S. Afr. Orchid J.

15,3: 91.

SCHLECHTER, R. 1901. Monographic der Diseae. Bot. Jb. 31:

134-313.

STAFLEU, F. A. 1983. International Code of Botanical Nomen-

clature. The Hague: Junk.

SWARTZ, O. 1800. Orchidernes slagter och arter upstallde. K.

svenska VetenskAkad. Handl. 21: 202-254.

THUNBERG, C. P. 1794. Prodromus plantarum Capensium.

Uppsala.

THUNBERG, C. P. 1807. Flora Capensis. Stuttgart.

H. P. LINDER

PERTUSARIACEAE

A NEW SPECIES OF PERTUSARIA

Pertusaria salax Brusse, sp. nov.

Thallus sulphureus, saxicola, usque ad 110 mm di-

ametro, 0, 2-1,0 mm crassus, areolatus, areolis

0, 2-2,0 mm latis, rimis saepe 0, 1-0,2 mm latis, isidiis

sorediisque destitutus. Algae 5-14 pm diametro, ad

Trebouxiam pertinentes. Ascomata hyalina, perithe-

cioidea, 0,5-0, 8 mm diametro, 2-6 in verrucis thalli-

nis apothecioideis immersa, interdum confluentia.

Verrucae 1, 5-3,0 mm diametro, basin versus con-

strictae (substipitatae). Parietes peritheciorum hya-

lini, 13-40 pm crassi, periclinate prosoplectenchy-

mati. Hymenium hyalinum, circa 500 pm altum.

Paraphyses pernumerosae, graciles, laxae, ramosae

et anastomosae, 1, 5-2,0 pm crassae. Gelatinum

copiosum, J-. Asci cylindrici, 350-500 x 80-110 pm,

parietibus aequaliter crassis, J+ valde cyaneis.

Ascosporae singulares vel binae, hyalinae, ellipsoi-

deae, pergrandes, 175-250 x 70-100 pm, halonatae,

lumenibus etruncatis; paries externus tenuis, 1, 0-3,5

pm crassus, pagina interna reticulate undulata; par-

ies internus 17-20 pm crassus, polis non incrassatis.

58

Bothalia 16,1 (1986)

Pycnidia globosa, immersa, hyalina, circa 200 pm di-

ametro. Pycnidiosporae teretes, pertenues, extremi-

tatibus truncatis, hyalinae, rectae, 18-24 x 1,2-1 ,3

pm. Thallus unum xanthonorum et acidum norsticti-

cum continens.

TYPUS. — South West Africa/Namibia, 2715 (Bo-

genfels): Diamond area no. 1, summit of Buchu-

berg, c. 90 km N of Oranjemund. On schist on steep

E slope (-DD), G. Williamson 2898, 1982.04 (PRE,

holo.; LD, iso.). Fig. 17.

Thallus yellow, saxicolous, up to 110 mm in di-

ameter, 0, 2-1,0 mm thick, areolate, areoles 0, 2-2,0

mm across, fissures often 0, 1-0,2 mm wide, isidia

and soredia absent. Algae 5-14 pm in diameter, Tre-

bouxia. Ascomata hyaline, perithecioid, 0,5-0, 8 mm

in diameter, 2-6 immersed in thalline, apothecioid

verrucae. Verrucae 1, 5-3,0 mm in diameter, con-

stricted at the bases (substipitate). Perithecial walls

hyaline, 13-40 pm thick, periclinally prosoplecten-

chymatous. Hymenium hyaline, about 500 pm high.

Paraphyses abundant, slender, flaccid, branched and

anastomosed, 1, 5-2,0 pm thick. Gel copious, J-.

Asci cylindrical, 350-500 x 80-110 pm, 1- or 2-

spored, walls evenly thick, J+ strongly blue. Asco-

spores very large, oval, hyaline, 175-250 x 70-100

pm, halonate, lumens not truncated; outer wall thin,

1-3,5 pm thick, with the inner surface (junction be-

tween the two walls) reticulately rippled; inner wall

17-20 pm thick, poles not thickened. Pycnidia glo-

bose, immersed, hyaline, about 200 pm in diameter.

Pycnidiospores hyaline, straight, very narrowly cyl-

indrical with truncate ends, 18-24 x 1,2-1, 3 pm.

Chemistry: one of the xanthones and norstictic acid

present (TLC).

This genus is fairly well known in Europe (Erich-

sen 1936; Poelt 1969) and in the United States and

Canada (Dibben 1980), but is otherwise rather

poorly understood. However, this conspicuous new

lichen has very large ascospores and a yellow thallus

and is distinct in these two characters alone. The as-

cospores are so large, that they are visible with the

naked eye, and are easily discernible with a lens.

Another yellow saxicolous species, Pertusaria di-

aziana Massal., is common in the Cape floral area

further south, but has a less coarse thallus habit,

smaller perithecioid ascomata and asci, and 4 (6, 8)

much smaller ascospores, with truncated lumens, in

the asci. The perithecioid ascomata fuse much more

rarely than in Pertusaria salax, and the ascomatifer-

ous verrucae are rounded and K-, and not apothe-

cioid and K+ red (norstictic acid), as in the latter.

Both species are KC-t- bright orange due to the pre-

sence of lichen xanthones in the upper zone.

At present this lichen is known only from the type

locality, the Buchuberg, some 90 km north of the

Orange River mouth.

REFERENCES

DIBBEN, M. J. 1980. The chemosystematics of the lichen genus

Pertusaria in North America north of Mexico. Milwaukee

Public Museum, Publications in Biology and Geology, no.

5. pp. 162.

ERICHSEN, C. F. E. 1936. Pertusariaceae. Rabenh. Krypt.-Fl.

[2] 9, 5(1): 319-728.

POELT, J. 1969. Bestimmungsschliissel europaischei Flechten pp.

71 + 757. Lehre: Cramer.

F. BRUSSE

TTFTnTrmi 1 1

FIG. 17. — Pertusaria salax

Brusse, G. Williamson

2898, holotype. Scale in

mm.

Bothalia 16,1 (1986)

POACEAE

DIVERSE NOTES ON SOUTHERN AFRICAN POOIDS

59

While preparing the account of the Poeae for the

Flora of southern Africa, I have encountered the fol-

lowing problems, which require a more detailed

treatment than is desirable in the Flora. A full taxo-

nomic treatment of the tribe will shortly appear in

the Flora, so I will not repeat it here.

1. Festuca elatior L.

This taxon has been known in the last decades

under the name F. arundinacea Schreb. When Lin-

naeus described Festuca elatior in 1753, he included

in his protologue elements of two taxa, F. pratensis

Hudson and F. arundinacea Schreb. Subsequently

the name has been used for either taxon, or for both,

or as F. elatior subsp. arundinacea (Schreb.) Wimm.

and F. elatior subsp. pratensis (Hudson) A. Gray.

Terrell (1967) dealt in detail with the typification of

F. elatior L. , and argued convincingly that this name

appliesi to F. arundinacea and not to F. pratensis

Hudson. Following European practice he then re-

jected the name as a nomen ambiguum. However,

subsequent to the Leningrad Conference, nomina

ambigua no longer exist, and the name must be

used. According to Terrell (1967: 131) the type

specimen is specimen no. 92.17 in LINN. I formally

propose it here as the lectotype:

Festuca elatior L., Sp. PI. 75 (1753). Type: Hor-

tus Upsaliensis, specimen no. 92.17 (LINN, lecto.)

(lectotypified here).

2. Festuca dracomontana Linder, sp. nov., a

Festuca camusiana Saint-Yves arista subterminali, a

Festuca simensis A. Rich, auriculis parvioribus et

aristis brevioribus recedit. Type: Lesotho, 3028 (Ma-

tatiele): Letsing La Letsie slopes (-AC), P. C. V. du

Toit 2714 (PRE, holo.!; K!; MO!).

Plants perennial, 0,5-0, 8 m tall. Culms smooth,

terete, rhizomatous at the base and occasionally

swollen. Leaves cauline; sheaths smooth; ligules

membranaceous, glabrous, less than 1 mm long and

often lacerated; auricles glabrous, 1-2 mm in di-

ameter; blades expanded, 2-8 mm wide, acute,

smooth or somewhat rough, to 200 mm long. Inflo-

rescence an open panicle, exserted, to 250 mm long;

rhachis smooth, branches somewhat compressed and

rough on the edges. Spikelets pale green, 10-12 mm

long, 3-7-flowered. Glumes unequal, acute; lower

glume 2,5^1 mm long, 1 (3)-nerved; upper glume

4,5-5 mm long, 3-nerved. Lemmas acute, dorsally

rounded, 5-nerved, 5,5-8 mm long, scabrid in the

upper half with the upper margins membranous;

awn subterminal, scabrid, 1-4 mm long. Paleas

about as long as the lemmas, keels scabrid. Stamens

3; anthers 3 mm long. Styles 2; mature ovaries not

seen. Fig. 18.

Habitat montane grassland in the Natal and

Transvaal Drakensberg; rather rare.

LESOTHO. — 3028 (Matatiele): Letsing La Letsie slopes

(-AC), P. C. V. du Toit 2714 (K; MO; PRE).

TRANSVAAL.— 2528 (Pretoria): Pretoria (-CA), Meebold

13800 (PRE). 2329 (Petersburg): Haenertsburg, grassy hillsides

(-DD), Ellis 1874 (PRE).

This curious species, with its disjunct distribution,

is quite distinct from all other southern African fes-

cues, but approaches the tropical African species F.

camusiana Saint-Yves and F. simensis A. Rich. The

former has terminal awns, and the transverse section

(ts) of the leaf is somewhat different. The leaf ts of

the latter is the same as F. dracomontana, but it has

much longer awns and larger auricles. F. dracomon-

tana is probably a southern vicariant of F. simensis,

but it is surprising that F. simensis does not reach

further south than Kenya.

3. Leaf anatomy of southern African fescues

The arrangement of the vascular bundles and the

distribution of sclerenchyma is generally held to be

almost species-specific in Festuca, especially in the

narrow-leaved fescues. Saint-Yves (1929) published

drawings of the leaf anatomy of four of the nine

southern African species.

Transverse sections of the leaves were prepared

from herbarium material, which was briefly softened

in boiling water, the silica removed by soaking in

HF1 for 12 hours, and the leaves sectioned at 20 pm

on a slide microtome. The sections were stained with

FIG. 18. — Festuca dracomontana.

60

Bothalia 16,1 (1986)

Safranin, and mounted in 50% glycerol. The results

are presented in Fig. 19.

The following groups may be recognized:

(a) Festuca caprina Nees: leaves less than 1,5 mm

wide, usually folded, with 5-7 vascular traces, and

small bundles of sclerenchyma opposite the bundles,

more or less embedded in the epidermis, and not

reaching the bundles (Fig. 19a; see also Saint-Yves

1929: 97, Fig. 15). Schelpe 1396 differs from this pat-

tern by having small sclerenchyma bundles on the

inside of the vascular traces. This collection also dif-

fers morphologically from the typical F. caprina, but

other collections that agree with Schelpe 1396 in ex-

ternal morphology have a typical leaf ts.

(b) Festuca .costata group (F. costata Nees and F.

killickii K.-O’Byrne): leaves with at least 9 traces,

often flat. Vascular traces of alternating size. Scleren-

chyma reaching from the inner to the outer epider-

mis, expanded on the inner surface in a T-shaped

girder. In F. costata the sclerenchyma is continuous

FIG. 19. — Transverse sections of leaves of southern African fescues. Shaded areas indicate sclerenchymatous tissue: a, F.

caprina , x 100, from McCallum Webster N 483a ; b, F. costata , x 40, from Hilliard & Burtt 7200 ; c, F. killickii , x 100,

from Hilliard & Burtt 1521 1; d, F. costata, x 100, from Manning, Hilliard & Burtt 16008 ; e, F. scrabra, x 100, from

Acocks 19731 ; f, F. longipes, x 40, from Rogers 12744; g, F. vulpioides, x 100, from Liebenberg 3811; h, F. dracomon-

tana, x 100, from Du Toit 2714.

Bothalia 16,1 (1986)

61

under the outer epidermis (Fig. 19b, d). In some

specimens the leaves are about 0,5 mm wide, and are

folded, (Fig. 19d) and agree exactly with Saint-

Yves’s (1929: 99, Fig. 17) illustration of F. obturbans

Saint-Yves. However, F. obturbans is a northern vi-

cariant of F. caprina, occurring from Mt Kilimanjaro

northwards to the Yemen, and is in external mor-

phology scarcely distinct from F. caprina. F. killickii

(Fig. 19c) lacks the subepidermal sclerenchyma that

characterizes F. costata (Fig. 19d).

(c) Festuca scabra group (F. scabra Vahl, F. long-

ipes Stapf, F. vulpioides Steud., F. africana (Hack.)

W. D. Clayton and F. dracomontana Linder): this

group has wide, expanded leaves with numerous vas-

cular bundles, all about the same size, seated be-

tween two sclerenchymatous bundles, which connect

the vascular bundles to the upper and the lower epi-

dermisses (Fig. 19e, f, g, h). Usually the inner epi-

dermal cells in the folds are enlarged. In F. vul-

pioides, F. africana and F. dracomontana the scle-

renchyma of only some bundles reaches the epider-

mis.

4. The typification of many of the Poeae and

Bromeae indigenous to Southern Africa is difficult.

Most of them were described by Nees ab Esenbeck

(1841).

Nees frequently recognized several varieties in his

species. When he recognized varieties, he included

all the material in these varieties, that is, he did not

recognize a ‘typical’ form. As in his study of the Res-

tionaceae (see Linder 1985), these varietal names

are not correct, as he did not consistently employ

binomials [ICBN 1983, art. 23 (c)]. For example, p.

21: ‘Panicum monodactylum var. a. inferioribus va-

ginisque pilosulis’; and p. 33: ‘Panicum numidianum

var. B. Culmo magis ramoso, vaginis rhachi et ramis

racemi hirsutis’. However, later authors occasionally

cite the Nees varieties in their diagnoses of new

species, such as Festuca longipes Stapf, based on F.

costata ‘var. fascicularis’ of Nees, and Bromus

firmior Stapf, based on B. speciosus ‘var. firmior’ of

Nees. The course which allows the most consistent

treatment of Nees's infraspecific categories appears

to be to ignore them for formal nomenclatural pur-

poses.

Nees’s herbarium has been scattered, and I have

not been able to find any specimens annotated by

Nees. Nees identified his material only by collector

and locality, so that the identification of isotypes

may also be difficult. Generally I have accepted the

locality and collector date given on the sheets at

Kew, although much of it was added in a hand differ-

ent to that of the collector.

I have lectotypified isotype material, and where

there are several syntypes, I selected the best ma-

terial as the lectotype. These lectotypifications are

given below:

Bromus firmior (Nees) Stapf in F.C. 7: 733

(1900). Bromus speciosus Nees var. firmior Nees, FI.

Afr. Austr. 454 (1841). Type: Cape, in the Storm-

berge and the Witteberg, 5000-6000 ft, Drege s.n.

(SAM, lecto.!; TCD!).

Cynosurus odoratus Lehm ex Nees, FI. Afr.

Austr. 439 (1841). Type: Cape Province, ? Swellen-

dam Distr., Mundt (SAM, lecto.!). Nees cited a

Mundt collection from the Swellendam District as

type. In SAM are two Mundt collections, Zeyher

128 : ‘probabiliter in sylvaticus ad Plettenbergs bay

lectus’, and Zeyher 25 : ‘verisimiliter in distr. George

locis sylvaticus lectum’. These labels may be wrong,

as this species has not since been recorded from for-

ests, but the collections are probably isotype ma-

terial.

Festuca africana (Hack.) W. D. Clayton in Kew

Bull. 40: 727 (1985). Brachelytrum africanum Hack,

in Bull. Herb. Boissier 3: 382 (Aug. 1895). Type:

Transvaal, Houtbosch, Rehmann 5732 (K, lecto.!).

Festuca caprina Nees, FI. Afr. Austr. 443

(1891). Type: Cape Province, Los Tafelberg near

Queenstown, Dec., Drege s.n. (K, lecto.!).

Festuca costata Nees, FI. Afr. Austr. 447 (1841).

Types: Cape Province, Katberg, Drege s.n. (K,

lecto.!); Windvogelberg, Drege s.n. (SAM!); Win-

terberg near Phillipstown, Zeyher s.n. (K!); at Her-

renhuter Mission, Ecklon & Zeyher s.n. (not found).

Festuca vulpioides Steud., Syn. PI. Glum. 1: 305

(1859). Vulpia megastachya Nees, FI. Afr. Austr.

441 (1841), non Festuca megastachys Hegetscher &

Heer (1840). Type: South Africa, without precise

locality, Drege s.n. (K, lecto.!; TCD!).

Poa binata Nees, FI. Afr. Austr. 578 (1841).

Types: Cape Province, Queenstown Division, Los

Tafelberg, Drege s.n. (K, lecto.!; E!; TCD!); in

mountains between Klipplaatsrivier and Katrivier,

Drege s.n. (not found).

Puccinellia angusta (Nees) C.A. Sm. & C. E.

Hubb. in Kew Bull. 1929: 85 (1929). Sclerochloa

angusta Nees, FI. Afr. Austr. 381 (1841). Type:

Cape Province, Uitenhage, along the Swartkops

River, December, Ecklon s.n. (SAM, lecto.!; K!).

ACKNOWLEDGEMENTS

The work for this paper was done at the Royal Bo-

tanic Gardens, Kew. I am grateful to Dr Tom Cope,

Dr W. D. Clayton and Mr S. Renvoize for much

stimulating discussion, and to the Directors of PRE,

BOL, STE, and NBG for the loan of their material

of the Poeae.

REFERENCES

LINDER, H. P. 1985. Conspectus of the African species of Res-

tionaceae. Bothalia 15: 387-503.

NEES AB ESENBECK, C. G. 1841. Florae Africae Australioris.

Glogau.

SAINT-YVES, A. 1929. Contribution a l’etude des Festuca (sub-

gen. Eu-Festuca) de l’Afrique australe et de l’Oceanie.

Candollea 4: 65-129.

STAFLEU, F. A. 1983. International Code of Botanical Nomen-

clature. Utrecht: Bohn, Scheltema & Holkema.

TERRELL, E. E. 1967. Meadow fescue: Festuca elatior L. or F.

pratensis Hudson? Brittonia 19: 129-132.

H.P. LINDER

62

PORINACEAE

A NEW SPECIES OF PORINA ON LIMESTONE

Bothalia 16,1 (1986)

Porina balanina Brusse, sp. nov.

Thallus endolithicus. Tunica thallina perithecii

hemispherica, fusca, ad basem constricta, 300-700

pm crassa, 1 ,0-1,5 mm lata, superficiaris crystallis

grandibus 50-250 pm crassis (oxalas calcii) inclusis.

Algae sphaericae vel ellipsoideae, 7-24 pm di-

ametro, ad Trentepohliam pertinentes. Perithecia

hyalina, in textura thallina omnino immersa,

600-800 pm diametro, globosa. Parietes hyalini vel

straminei, in hydroxido kalii crocescentes, verticale

et periclinate prosoplectenchymati, 25-30 /.an crassi,

ostiolum versus incrassati et paraplectenchymati,

cellulis 3-6 pm diametro, badiis, in acido nitrico

rosescentibus. Area circa ostiolum applanata vel

concava, sed non peltulata. Hymenium hyalinum,

circa 500 pm altum, guttulis inspersis. Subhymenium

hyalinum, circa 100 pm crassum. Paraphyses sim-

plices, perlongae, flaccidae, graciles, 1,0-1, 5 pm

crassae, ecapitatae, septatae, septis 7-11 pm distan-

tibus, in gelatina inclusae, gelatina hyalina, J-. Asci

cylindrici, 8-spori, parietibus, apicum inclusis, tenui-

bus, J-. Ascosporae oblique uniseriatae, hyalinae,

J-, longe ellipsoideae, triseptatae, 14-19 x 5-7 ^m,

halone tenue 0,8-1 ,3 pm crassa circumdatae, parieti-

bus tenuibus. Pycnidia globosa, hyalina, 160-180 //m

diametro, in textura thallina superficiari immersa,

parietibus hyalinis sed apices versus brunneolis. Pyc-

nidiosporae bacillares, hyalinae, 3, 5-4, 2 x 1,0-1, 2

pm.

TYPUS. — Cape, 3419 (Caledon): Gansbaai area,

Byneskranskop near Strandskloof overlooking the

Uilkraals River Valley, on a sandy limestone kranz

(-CB). F. Brusse 3835 , 1981.05.14 (PRE, holo.).

Fig. 20.

Thallus endolithic. Thalline covering of the peri-

thecium hemispherical, fuscous, constricted at base,

300-700 pm thick, 1,0-1 ,5 mm broad, superficial,

embedded with large crystals 50-250 )<m thick (cal-

cium oxalate). Algae spherical to ellipsoid, 7-24 pm

across, Trentepohlia. Perithecia hyaline, completely

immersed in thalline tissue, globose, 600-800 jUm in

diameter. Walls hyaline to stramineous, K+, dingy

orange, vertically and periclinally prosoplectenchy-

matous, 25-30 pm thick, thickened in the ostiolar re-

gion and paraplectenchymatous, cells 3-6 pm in di-

ameter, brown, becoming pinkish in concentrated

nitric acid. Ostiolar region flat or concave (i.e. with a

thickened ring around it) but not peltulate. Hyme-

nium hyaline, about 500 pm high, inspersed with ‘oil

droplets’. Subhymenium hyaline, about 100 )rm

thick. Paraphyses simple, very long and slender,

flaccid, 1 ,0—1 ,5 pm thick, septate, septa 7-11 /.an

apart, ecapitate, embedded in gel, gel hyaline, J-.

Asci cylindrical, 8-spored, with walls, apices in-

cluded, thin, J-. Ascospores hyaline, long ellipsoid,

triseptate, 14-19 x 5-7 ^m, obliquely uniseriate,

thinly halonate, halo 0,8-1 ,3 pm thick, walls thin.

Pycnidia hyaline, globose, 160-180 jum in diameter,

immersed in superficial thalline tissue, walls hyaline

but brownish above. Pycnidiospores hyaline rods,

3, 5-4, 2 x 1,0-1, 2 /urn. Chemistry: no lichen sub-

stances detected by TLC.

This new species is most similar to Porina corru-

gata Mull. Arg., also on limestone, from the Mel-

bourne-Adelaide area of Australia, with a Mediter-

ranean climate similar to that of the south-western

Cape. Porina balanina is, however, more robust,

with larger perithecia and associated thalline tissue.

FIG. 20. — Porina balanina

Brusse, Brusse 3792. Scale

in mm.

Bothalia 16,1 (1986)

63

FIG. 21. — Scale margins in different species otRiccia and in Ricciocarpus. 1, 2, R. villosa denticulate at apex;

3, 4, R. okahandjana with crenate margins; 5, 6, undescribed Riccia species with multicellular append-

ages; 7, R. rhodesiae with crenate margins; 8, Ricciocarpus natans, tip of scale, margins dentate. (1,3,5,

SEM micrographs; 2, 4, 6- 8, LM (light microscope) photographs by S. M. Perold). Scale bar = 100 pm.

64

Bothalia 16,1 (1986)

More importantly, the ascospores are not as delicate

and fusiform as in Porina corrugata , being narrowly

ellipsoid, with sturdier, but still evenly thin, walls

and septa.

The walls of the perithecium are pale yellowish or

straw-coloured and immediately become dingy

orange in a 2 molar solution of potassium hydroxide

(K). The pigment does not diffuse and no crystals

develop in association with this colour in K.

The new species is known from several limestone

localities in the Gansbaai-Bredasdorp area of the

south-western Cape, at altitudes lower than 100 m.

CAPE. — 3419 (Caledon): Gansbaai area, Byneskranskop near

Strandskloof overlooking the Uilkraals River Valley, on a sandy

limestone kranz (-CB), F. Brusse 3835, 1981.05.14 (PRE). 3420

(Bredasdorp): Bredasdorp District near Mierkraal, Die Poort

(Eleuningsnes River Valley) on a S-facing sandy limestone kranz

(-CA), F. Brusse 3815, 1981.05.14 (PRE; LD); Cape Agulhas,

limestone ridge behind the town, on limestone on steep S slope

(-CC), F. Brusse 3792, 1981.05.13 (PRE).

F. BRUSSE

RICCIACEAE

PTERORICCIA SCHUST., SHOULD IT BE UPHELD?

The family Ricciaceae Dum. is distinguished from

other families in the order Marchantiales by the sim-

ply constructed sporophyte embedded within the

thallus tissue. Until recently it has been generally ac-

cepted that the family comprises only two genera:

Ricciocarpus Corda and Riccia (Mich.) L.

Ricciocarpus is monotypic, with the species R. na-

tans (L.) Corda, and is separated from Riccia by its

long pendant scales with serrate margins and the

presence of oil cells in the scales and the thallus. Ric-

cia, on the other hand, is a large genus, now divided

into several subgenera and probably comprises as

many as 200 species worldwide.

A third genus, Pteroriccia, based on Riccia villosa

Steph., was recently recognized by Schuster (1984a,

1984b). His decision was made after only examining

Arnell’s (1963) text and figures, which, by his own

admission, are not good. Schuster’s reasons for sep-

arating this taxon from the other Riccia species are

as follows:

(a) the distinctive dorsal epithelium of loose, erect

cellular ‘filaments’ (‘superficies dorsalis thalli velu-

tina ob filamenta cellulae erecta isolata’); and

(b) the large erect and imbricate ventral scales

with serrate margins (Fig. 21.1, 21.2) (‘squamae ven-

trales ingenter, usque ad 1.5 mm long.; erectae (sic-

catae super thallum incurvatae), apicibus eorum ser-

ratis’).

V

According to Schuster, such scales do not occur in

any Riccia species, but they are found in the aquatic

form of Ricciocarpus natans (Fig. 21.8), to which

there is, however, no close affinity. He is, therefore,

of the opinion that this new genus is at least as differ-

ent from Riccia as is Ricciocarpus , which is generally

recognized as a distinct genus.

The following comments can be made on the

characters used by Schuster to separate his new

genus:

(a) the epithelium of R. albomarginata Bisch. was

already described by Sim (1926) as consisting of free

‘pillars of long empty cells about five cells deep’. In

his key, Arnell (1963) uses the ‘velvet-like’ dorsal

surface ‘caused by free cell pillars’ as a character for

R. albomarginata and R. concava Bisch. He also re-

fers to the free cell pillars in his description of R.

villosa although he does not place it in his key with

the aforementioned two species. A total of eight

southern African species with loose dorsal pillars are

now known and are grouped together in the section

Pilifer Volk (1983).

(b) the scale margins in some southern African

Riccia species vary from crenate, as in R. okahand-

jana S. Arnell (Fig. 21.3, 21.4) and R. rhodesiae S.

Arnell (Fig. 21.7), to serrate in R. villosa, to having

long multi-cellular appendages (Fig. 21.5 & 6) in an

undescribed species collected at Victoria West and

Springbok and on the Kamiesberg.

CONCLUSION

The loose dorsal ‘filaments’ are not peculiar to R.

villosa; furthermore scale margins vary considerably

in some species of Riccia. Therefore, a species ex-

hibiting the characters mentioned by Schuster need

not be placed in a separate genus and it is proposed

that Pteroriccia be regarded as synonymous with

Riccia.

REFERENCES

ARNELL, S. 1963. Hepaticae of South Africa, pp. 411. Stock-

holm: Swed. Nat. Sci. Council.

SCHUSTER, R.M. 1984a. Diagnoses of some new taxa of Hepa-

ticae. Phvtologia 56, 2: 65-74.

SCHUSTER, ' R.M. (ed.) 1984b. New Manual of Bryology. 2

Vols. pp. 1295. The Hattori Bot. Lab.

SIM.T.R. 1926. The bryophyta of South Africa. Trans. R. Soc. S.

Afr. 15: 1-475.

VOLK, O.H. 1983. Vorschlag fur eine Neugliederung der Gat-

tung Riccia L. Mitt. bot. StSamml., Munch. 19: 453-465.

VOLK, O.H. & PEROLD, S.M. 1984. Studies in the liverwort

genus Riccia (Marchantiales) from the south-west Cape. Bo-

thalia 15: 117-124.

S. M. PEROLD

Bothalia 16,1: 65-69 (1986)

The shape and ultrastructure of the caryopsis of Pentameris and Pseu-

dopentameris species (Arundinoideae, Poaceae)

N. P. BARKER*

Keywords: caryopsis, leaf anatomy, Pentameris , Pseudopentameris, spikelet morphology

ABSTRACT

Pentameris Beauv. and Pseudopentameris Conert differ in their leaf anatomy, spikelet morphology and ovary

structure. Caryopses of species of both genera were examined under the scanning electron microscope (SEM) and

these species are grouped according to caryopsis features. These groups are compared to the groupings suggested

by spikelet morphology and leaf anatomy.

INTRODUCTION

1 Present classification (based largely on spikelet

morphology)

The genera Pentameris Beauv. and Pseudopenta-

meris Conert are members of the subfamily Arundi-

noideae, which contains diverse tribes not particu-

larly closely related. Some tribes are included be-

cause they do not fit clearly into other subfamilies

(Renvoize 1981). Both the genera studied here were

formerly treated in the tribe Danthonieae, but are

now assigned to the Arundineae.

1.1 Pentameris

This genus, originally created by Palisot de Beau-

vois (1812), contains five described species. These

are P. longiglumis (Nees) Steud., P. macrocalycina

(Steud.) Schweick., P. dregeana Stapf, P. obtusifolia

(Hochst.) Schweick. and P. thuarii Beauv. with var.

burchellii Stapf. There is a further rare undescribed

species that belongs to Pentameris (if the present

generic limits are accepted). This species is referred

to below as Pentameris sp. nov.

The number of veins in the glumes distinguish

Pentameris (one or occasionally three) from Pseudo-

pentameris (five or more).

1.2 Pseudopentameris

The genus Pseudopentameris was described by

Conert (1971) as a segregate of Danthonia DC. The

two species of this genus, P. macrantha (Schrad.)

Conert and P. brachyphylla (Stapf) Conert, were re-

moved from Danthonia because they differ con-

spicuously from the other species in their 2-flowered

spikelets which are over 40 mm long. Conert (1971)

claims that Pseudopentameris can be easily separ-

ated from both Danthonia and Pentameris by the

many-nerved glumes, the structure of the caryopsis

and the leaf anatomy.

2 Leaf anatomy

Ellis (1985a) has examined the leaf anatomy of the

above eight species in detail, but his findings on Pen-

tameris dregeana have not yet been published.

* Botanical Research Institute, Department of Agriculture and

Water Supply, Private Bag X101, Pretoria 0001, South Africa.

2.1 Pentameris

Ellis (1985d) states that Pentameris macrocaly-

cina, P. obtusifolia and P. longiglumis form a natural

group within the present genus boundary. However,

P. thuarii is quite different in its anatomy from the

other species, and is considered by Ellis (1985c) to

show similarities to certain species of Pentaschistis.

Pentameris dregeana was divided by Ellis (pers.

comm.) into three forms: a ‘typical’ form, a ‘hairy’

or pubescent form and an ‘atypical’ form. The typ-

ical forms may be incorrectly identified specimens,

as many of them also show features similar to Pen-

taschistis species. Pentameris sp. nov. is similar in its

leaf anatomy to the Pseudopentameris species, and

Ellis (1985b) postulates that the taxon may be of hy-

brid origin, with Pentameris longiglumis and a

species of Pseudopentameris as parent species.

2.2 Pseudopentameris

Ellis (1985a) considers Pseudopentameris to be

clearly distinct from other supposedly closely related

genera such as Pentameris and Merxmuellera Con-

ert. He found only slight anatomical differences be-

tween the two Pseudopentameris species.

3 Ovary and caryopsis

Chippendall (1955) and Conert (1971) state that

ovary and caryopsis structures are useful in distin-

guishing between the two genera, but caryopsis

shape and coat sculpturing have seldom been used in

grass taxonomy.

Pentameris has deciduous hairs on the top of the

ovary (Chippendall 1955). The caryopsis, described

by Stapf (1899) from P. thuarii, is globose-ellipsoid,

truncate at the top, and has a linear hilum as long as

half the grain or longer.

Pseudopentameris has a glabrous ovary with the

stigmatic hairs joined over the top (Chippendall

1955). The caryopsis of P. macrantha was described

by Stapf (1899) as linear-oblong, and with the hilum

longer than half the grain. The genus Pentaschistis,

which is closely related to Pentameris and Pseudo-

pentameris, has glabrous ovaries and Stapf (1899)

described the grain as oblong, semiterete to subte-

rete with an obscure hilum one quarter to one third

the length of the grain.

66

Bothalia 16,1 (1986)

FIG. 1. — 1 & 2, Pentameris sp. nov. (Hilton-T aylor 352): 1, whole caryopsis, narrowly elliptic; scale bar = 500 p; 2, stylar end of

the caryopsis showing reticulate sculpturing; scale bar = 200p. 3 & 4, P. obtusifolia (Andreae 706): 3, close up of colliculate

sculpturing; scale bar = lOOp; 4, whole caryopsis, elliptic; scale bar = 500p. 5 & 6, Pseudopentameris macrantha (Ellis 2327):

5, close up of reticulate sculpturing; scale bar = 50p; 6, whole caryopsis, narrowly elliptic; scale bar = lOOOp.

The present paper describes the shape of the cary-

opsis and the coat sculpturing of some of the species,

and emphasizes the differences and similarities be-

tween the species and the genera. The species are

grouped on the basis of these similarities, and this

classification is compared to the groupings based on

spikelet morphology and to the groups suggested by

Ellis (1985a-d) which are based on leaf anatomy.

METHODS AND MATERIAL

Caryopses were obtained from herbarium speci-

mens housed in the National Herbarium, Pretoria

(PRE), and were prepared for the SEM. They were

mounted on stubs, placed in an Eiko Engineering IB

2 ion coater and coated with gold at a current of 7

micro-amps for 7 minutes. The stubs were examined

using an ISI-SX-25 Scanning Electron Microscope.

Photomicrographs were taken on Ilford FP4 120

film.

Specimens examined:

Pentameris dregeana

CAPE. — 3322(Oudtshoorn): Prince Albert (-AC), Bond 1253.

Pentameris macrocalycina

CAPE. — 3419 (Caledon): Hermanus (-AC), Barker N. P. 64.

3322 (Oudtshoorn): Blesberg (-BC), Bond 848.

Pentameris obtusifolia

CAPE. — 3319 (Worcester): Wemmershoek (-CC), Andreae

706.

Bothalia 16,1 (1986)

67

FIG. 2. — 7 & 8, Pentameris macrocalycina (Bond 848): 7, whole caryopsis, elliptic; scale bar = 500p; 8, colliculate to tuberculate

sculpturing near stylar end of caryopsis; scale bar = 50p. 9 & 10. P. dregeana (Bond 1253): 9, rugose sculpturing at approx, a

third of the length from the base; scale bar = 50p; 10, whole caryopsis, cuneiform; scale bar = 500p. 11 & 12, P. thuarii var.

thuarii ( Barker , W. F. 7653): 11, whole caryopsis, globose; scale bar = 500p; 12, colliculate sculpturing of the surface; scale

bar = 50p. 13 & 14. P. thuarii var. burchellii (Van Breda 337): 13, whole caryopsis, globose-truncate; scale bar = 500p; 14,

colliculate to tuberculate sculpturing close to stylar end of caryopsis; scale bar = 100p.

68

Bothalia 16,1 (1986)

Pentameris thuarii

CAPE. — 3322 (Oudtshoorn): Jakkalsvlei, Outeniqua Pass

(-CD), Barker, W. F. 7653. 3320 (Montagu): Grootvadersbosch

(-DD), Barker, IV. F. 8812.

Pentameris thuarii var. burchellii

CAPE. — 3319 (Worcester): Brandwagt (-CB), Van Breda 337.

Pentameris sp. nov. (sensu Ellis 1985b)

CAPE. — 3418 (Simonstown): Kogelberg, Platberg (-DD), Hil-

ton-Taylor 352.

Pseudopentameris macrantha

CAPE. — 3418 (Simonstown): Cape of Good Hope Nature Re-

serve (-AD), Ellis 2327. 3420 (Bredasdorp): De Hoop Nature

Reserve (-AD), Barker, N. P. 48.

Note: No caryopsis material of Pentameris longiglu-

mis or Pseudopentameris brachyphylla was available

in PRE.

RESULTS AND DISCUSSION

Caryopsis shape *

Three general types of caryopsis shapes occur in the

species examined:

1) cuneiform, as shown by Pentameris dregeana

(Fig. 2.10);

2) globose, as shown by Pentameris thuarii (Fig.

2.11 & 13);

3) elliptic to narrowly elliptic, as shown by Penta-

meris macrocalycina (Fig. 2.7), P. obtusifolia (Fig.

1.4), P. sp. nov. (Fig. 1.1) and Pseudopentameris

macrantha (Fig. 1.6). These features are summa-

rized in Table 1.

Caryopsis coat sculpturing

Three types of caryopsis coat sculpturing were ob-

served:

1) colliculate sculpturing as shown by Pentameris

macrocalycina (Fig. 2.8), P. obtusifolia (Fig. 1.3)

and P. thuarii (Fig. 2.12, 2.14).

2) reticulate sculpturing, as found in Pseudopenta-

meris macrantha (Fig. 1.5) and Pentameris sp. nov.

(Fig. 1.2).

3) rugose sculpturing as shown by Pentameris dre-

geana (Fig. 2.9).

Caryopsis hairs

All the Pentameris species, including Pentameris

sp. nov., possessed hairs at the stylar end of the cary-

opsis. None of the Pseudopentameris specimens

examined had these hairs. This confirms Chippin-

dall’s (1955) observation that only Pentameris

species possess such hairs.

Pentameris can therefore be separated from Pseu-

dopentameris by the presence of hairs on the stylar

end of the caryopsis. Within Pentameris the species

can be grouped on combinations of the other charac-

ters observed: Pentameris macrocalycina and P. ob-

tusifolia agree in general caryopsis shape, and both

have a colliculate surface. The surface of P. macro-

calycina, however, becomes more tuberculate near

the stylar end of the caryopsis, as does the surface of

P. thuarii. P. thuarii, however, differs markedly in

caryopsis shape, and should therefore perhaps be

separated from this group. P. dregeana with its ru-

gose caryopsis sculpturing pattern and cuneiform

shape is also different from the other two species of

Pentameris that were examined.

The caryopsis sculpturing of the only species of

Pseudopentameris observed is similar to that of Pen-

tameris sp. nov., and suggests that the new taxon

could be placed in Pseudopentameris. However, this

is contradicted by the presence of hairs on the stylar

end of the caryopsis and the presence of only one

vein in the glume, which places this taxon in Penta-

meris.

The generic and subgeneric boundaries for the

two genera, according to the groupings based on dif-

fering sets of characters, are shown in Fig. 3. Fig. 3a

shows the boundaries based largely on floral charac-

ters, with particular reference to the number of veins

in the glume (one, or occasionally three in Pentame-

ris, five or more in Pseudopentameris) . Fig. 3b shows

these boundaries as they may be interpreted from

the leaf anatomy data of Ellis (1985a-d and pers.

comm.). Fig. 3c shows the boundaries based on the

caryopsis morphology. Pentameris thuarii and P.

dregeana are somewhat differently placed in the

three classifications. Pentameris sp. nov. is treated

differently in all three groupings. This may indicate

that Ellis’s hypothesis of a hybrid origin for the new

taxon could be correct.

TABLE 1. — The caryopsis characters of each species examined are shown. Characters of species which were not examined are tenta-

tively indicated by question marks

Bothalia 16,1 (1986)

69

PENTAMERIS | i PSEUOOPENTAMERIS I

Po

PI

Psb

PENTAMERIS

PSEUDOPENT AMERIS

FIG. 3. — The generic and subgeneric boundaries of Pentameris and Pseudopentameris. The sets indicate possible subgeneric limits,

the lines beneath indicating the generic limits, a, presently accepted limits, based largely on floral characters; b, divisions

based on leaf anatomy; c, boundaries based on shape and surface features of caryopsis. Note the dubious position of Penta-

meris thuarii, P. dregeana and P. sp. nov. Pt, Pentameris thuarii ; Pd, P. dregeana: Pm, P. macrocalycina ; Po, P. obtusifolia-,

PI, P. longiglumis: Pnov, P. sp. nov.; Psb, Pseudopentameris brachyphylla and Psm, P. macrantha.

Although no caryopsis material was available for

Pentameris longiglumis or Pseudopentameris brachy-

phylla, it is possible to make predictions about car-

yopsis structure and shape for these taxa if a correla-

tion between the caryopsis features and leaf anat-

omy is assumed. Ellis (1985d) states that Pentameris

macrocalycina, P. obtusifolia and P. longiglumis are

similar in anatomy, and are not closely related to any

other taxon. P. macrocalycina and P. obtusifolia also

show similarities in caryopsis shape and coat sculp-

turing. It can therefore be predicted that P. longiglu-

mis will show similar caryopsis features, namely an

elliptic shape and colliculate sculpturing. Similar logic

applies to Pseudopentameris brachyphylla. The leaf

anatomy of this species is very similar to the anat-

omy of P. macrantha and Pentameris sp. nov. (Ellis

1985a & b). It can therefore be predicted that the

caryopsis of Pseudopentameris brachyphylla will be

narrowly elliptic, and exhibit reticulate sculpturing.

Both predictions are shown in Table 1 as question

marks.

CONCLUSION

Caryopsis shape, coat sculpturing and presence of

hairs on the stylar end of the caryopsis appear to be

important characters in the classification of Penta-

meris and Pseudopentameris and may even be of

value to the taxonomy of the entire tribe. Caryopsis

characters are correlated with leaf anatomy in these

two genera.

ACKNOWLEDGEMENTS

I would like to thank Mrs S. Perold for help and

instruction in the use of the SEM, Dr R. P. Ellis, Dr

G. E. Gibbs Russell, Mr E. R. Robinson and Miss

L. Smook for their help and contributions and Mr C.

Hilton-Taylor for collecting specimens of Pentameris

sp. nov. Thanks are also extended to Mrs

Romanowski for the preparation of the photo-

graphs. Research was carried out in part fulfilment

for an M.Sc. dissertation. University of the Wit-

watersrand.

UITTREKSEL

Pentameris Beauv. en Pseudopentameris Conert

verskil in hulle blaaranatomie, die morfologie van die

blompakkies en die struktuur van die vrugbeginsels.

Vruggies van spesies van albei genusse is onder die

skandeerelektronmikroskoop bestudeer en die spesies

is volgens vrugkenmerke in groepe geplaas. Hierdie

groepe word vergelyk met groeperings wat op blom-

pakkiemorfologie en blaaranatomie berus.

REFERENCES

CHIPPENDALL, L. K. A. 1955. In D. Meredith, The grasses

and pastures of South Africa. Johannesburg; CNA.

CONERT, H. J. 1971. The genus Danthonia in Africa. Mitt. bot.

StSamml., Munch. 10: 299-308.

ELLIS, R. P. 1985a. Leaf anatomy of South African Danthonieae

(Poaceae). X. Pseudopentameris. Bothalia 15: 561-566.

ELLIS, R. P. 1985b. Leaf anatomy of South African Danthonieae

(Poaceae). XI. Pentameris longiglumis and Pentameris sp.

nov. Bothalia 15: 567-571.

ELLIS, R. P. 1985c. Leaf anatomy of South African Danthonieae

(Poaceae). XII. Pentameris thuarii. Bothalia 15: 573-578.

ELLIS, R. P. 1985d. Leaf anatomy of South African Danthonieae

(Poaceae). XIII. Pentameris macrocalycina and Pentameris

obtusifolia. Bothalia 15: 579-585.

PALISOT DE BEAUVOIS, A. M. F. J. 1812. Essai dune nou-

velle agrostographie. Paris.

RENVOIZE, S.A. 1981. The subfamily Arundinoideae and its

position in relation to a general classification of the Grami-

neae. Kew Bull. 36: 85-102.

SCHWEICKERDT, H. G. 1938. Descriptions and notes on

South African grasses. Feddes Reprium 43: 88-92.

STAPF, O. 1899. Gramineae. Flora Capensis 7: 310-750.

Bothalia 16,1: 71-75 (1986)

William Keit and the Durban Botanic Garden

d. p. McCracken*

Keywords: biography, Durban Botanic Garden, William Keit

ABSTRACT

William Keit was born in Saxony in 1841 and in early life travelled across Europe working in many famous

nurseries and gardens. In 1872 on the recommendation of the director of Kew Gardens, Keit emigrated to Natal to

become curator of the Durban Botanic Garden. So dilapidated was this garden that Keit was faced with the task of

virtually re-establishing it. Though he was largely successful in this endeavour, as he was in fortifying the link

between Natal and Kew, Keit could not solve the problems of a severe drought, a labour shortage and a scarcity of

funds. In 1881 he resigned his position leaving a solid foundation on which the renowned botanist, John Medley

Wood was to build. Keit in later life ran a successful nursery in Durban and for 30 years was curator of the Parks

and Gardens Department, in which capacity he did more than anyone else to beautify Durban.

The early Botanic Garden

In April 1848 the Natal Agricultural and Horticul-

tural Society was founded in Durban with the princi-

pal aim of establishing a botanic garden to serve the

young colony of Natal (Natal Agricultural and Hor-

ticultural Society 1848). It was not until June 1851,

however, that a permanent site of some 50 acres on

the lower slopes of the Berea was taken over by the

Horticultural Society (Durban City Estates Depart-

ment n.d.). Though in the next 20 years the garden

spearheaded crop experimentation and seed distri-

bution, it was never properly developed into a bo-

tanic garden. It remained largely virgin bush, un-

fenced save for a small allotment area.

Because the curator’s salary was so low, the So-

ciety had difficulty in finding a suitable man for the

post. Between 1851 and 1872 there were seven cura-

tors (Strey n.d.). The best known and the longest

serving of these was a Scot named Mark Johnston

McKen who was curator from 1851 to 1853 and again

from 1860 to 1872 ( Natal Colonist 23 April 1872; Na-

tal Mercury 25 April 1872). McKen did much for the

economic development of Natal but failed to lay out

a proper botanic garden. When he died in March

1872, leaving his wife and six children destitute, the

Horticultural Society resolved to employ a curator

who, unlike McKen, would devote his whole ener-

gies to the garden. As they could find no one suit-

able in the colony they wrote to Sir Joseph Hooker,

the director of the Royal Botanic Gardens at Kew,

requesting him to recommend a replacement for

McKen ( Natal Government Notice No. 191; 1872).

Hooker had recently been impressed by a young

German gardener named Julius Wilhelm Keit, or

William Keit as he called himself in the English-

speaking world, who at the time was employed at the

Royal Botanic Gardens, Glasnevin, outside Dublin

(Strey 1972; 1974).

Keit’s early career

William Keit was born in Dresden in the state of

Saxony on 1 May 1841. He was the son of a master

* University of Durban-Westville, Private Bag X54001, Durban

4000, South Africa.

soap-maker (Pers. comm.: Frl. Klara Keit, Mann-

heim, W. Germany). His father died when he was

young and Keit inherited a small sum which was ad-

ministered until he came of age by his uncle and

guardian, Hermann Steinmetz. As a young man,

Keit led a somewhat carefree if lonely life. Having

been sacked from his first job, he moved across

Europe working in gardens and nurseries in Munich,

Basle, possibly Paris, in Linden’s nursery in Brus-

sels, the Exhibition Palace winter gardens in Dublin

and in the gardens of Blyth Hall in Nottinghamshire,

England. Finally, in June 1868, he was appointed to

Glasnevin (Keit n.d. (a)).

Though Keit considered Dublin the dirtiest city he

had ever seen, he was fond of Glasnevin and had a

great respect for its director, Dr David Moore. In

September 1869 he accompanied Moore’s two chil-

dren to a boarding school in Hanover. By 1872 Keit

FIG. 1. — William Keit.

72

Bothalia 16,1 (1986)

was foreman-in-charge of the propagating houses at

Glasnevin (Keit 1867-1869).

When Sir Joseph Hooker received the request

from Natal he wrote to Moore asking him his opin-

ion of Keit’s suitability for the job. In his reply

Moore described Keit as ‘an excellent practical gar-

dener and very ingenious at applying various meth-

ods for propagating plants’ (Moore 1872). A few

days later Keit wrote to Hooker expressing his will-

ingness to take the Natal post (Keit 1872a). Hooker

duly appointed him.

Six years previously, Keit’s cousin Heinrich

Hoehne had emigrated to the Cape and Keit had

greatly envied him this experience (Keit 1866). Now

on 8 September 1872 Keit wrote home excitedly,

‘The job in Natal is, according to all accounts, very

challenging and I hope I will do justice to it.’

Keit at the Durban Botanic Garden

Keit left Dublin on 25 September 1872, travelled

to see his family in Dresden where he collected a

small proportion of his legacy, and returned to Eng-

land to take ship for South Africa. Hooker pre-

sented him with a Wardian case of plants to take to

his new garden. Unfortunately, the glass panels on

the side of the case were broken on what proved a

traumatic voyage and the plants suffered badly. Keit

arrived in Natal on 14 December and settled into a

‘pretty house’ at the foot of the Berea beside the gar-

den. He officially started work on 1 January 1873.

He hid from his benefactor. Dr Moore, his disap-

pointment at the terrible state of the garden and

wrote of the great heat, and the beauty of the

bougainvillea in full bloom on Christmas day. To

Hooker and to his family he was more candid and

spoke of his concern about the neglect of the

grounds. The garden did have a small and practically

empty greenhouse, which had been erected in 1870,

and two summer-houses which afforded panoramic

views of the town, the bay and the sea. There was

little else to commend the garden. Most of the paths

had been washed away and not rebuilt, and what few

flowerbeds existed were overgrown. Worse still, in

Keit’s eyes, was the fact that there was ‘no systema-

tic arrangement, the plants having been planted

where there was space and, what I regret most, there

are no names on them.’ He observed, ‘it will take me

a long time and all my energies to bring something

like order in this place and to rename the plants’

(Keit 1872b; 1873a; 1873b).

Sir Joseph Hooker soon realized the impossible

position Keit was in and offered to try and find him

another position elsewhere (Keit 1874). But by then

Keit had become engaged to Louisa Currie, the

daughter of a prominent Durban citizen and future

mayor of the town. The couple were married on 9

September 1874 and were to have eight children.

Though this marriage served to hold Keit in Natal he

did not dismiss the possibility of emigrating once

again. In September 1875 he appealed to Hooker,

‘should I fail then I trust you will not forsake me’

(Keit 1875).

William Keit did much for the Durban Botanic

Garden. By 1874 he had increased the area under

cultivation from approximately 4| to 9 ha. He fenced

the lower part of the garden, laid out flowerbeds,

rebuilt the paths, sowed grass on the banks to pre-

vent soil erosion and persuaded the town sanitary

department to give him large quantities of manure to

improve the quality of the rather poor red soil. He

established a proper nursery in 1875 and two years

later a pinetum which contained 29 varieties of con-

ifer. He also experimented with wood, zinc and cast-

iron labels to see which could best withstand the de-

structive combination of white ants and weather. In

his annual curator’s report for 1877 Keit noted: ‘In

the absence of any systematic arrangement of plants

in the Gardens, it has been my aim, when making

fresh plantations to plant individuals of the same or-

der in groups, having regard to effect and situation.’

By the end of 1875 the garden contained 670 plant

species. Keit’s own interests lay in sea algae, palms,

of which he grew 10 species, and cycads. He listed

the cycads in his garden as: Cycas media, Encepha-

lartos altensteinii, E. caffra, E. ghellincki, E. hilde-

brandtii, E. natalensis, E. villosus and Macrozamia

tenuifolia (Durban Botanic Garden 1873-1880).

Before he left Glasnevin Keit had written to

Hooker: ‘My botanical knowledge is more general

than particular. I can arrange plants botanically after

knowing their names, but I could not undertake to

know or describe new plants scientifically.’ In the

Durban Botanic Garden Keit’s taxonomic know-

ledge increased. Many years after he had resigned

the curatorship, it gave Keit quiet satisfaction to see

that a plant which he had correctly identified was in-

correctly identified by his eminent successor, John

Medley Wood (Keit n.d. (b)). Before he died in

1875, the naturalist Thomas Baines was in occasional

contact with Keit (1873c).

Two problems which seriously affected Keit in his

botanical studies were weak eyesight and the lack of

a microscope. None the less he has the credit for

publicizing Agapanthus campanulatus ‘mooreanus’ a

dwarf flower sometimes referred to as Keit’s blue

lily. Another lily which carries his name is Littonia

modesta var. keitii (Nelson 1984).

The role of the Durban Botanic Garden as an ex-

perimental station for various crops continued under

Keit. He made full use of his new nursery. By the

mid-1870's it contained: arrowroot, breadfruit,

chillies, china grass, the quinine-yielding plant Cin-

chona succirubra, clove, cocoa, ginger, groundnut,

indigo, khus khus, lemon grass, maize, mulberry,

New Zealand flax, nutmeg, prickly comfrey, sorrel,

tomatoes, turmeric, vines and yams. In addition,

Keit experimented with several plantation crops. He

brought in new varieties of coffee, sugar cane, tea

and tobacco from Kew and distributed the seed to

planters. New strains of sugar cane were especially

in demand and although it was not Keit who intro-

duced the famous uba cane he did introduce 60 new

varieties into Natal from Kew and Mauritius. He

also did much to encourage the growing of sugar

cane. Two exotics with which Keit experimented

were the opium poppy, Papaver somniferum, and

the rubber plants, Ceara and Hevea braziliensis (Mc-

Cracken 1985).

Bothalia 16,1 (1986)

73

Unfortunately many of the plantation species Keit

promoted proved unsuitable for the Natal climate,

being susceptible to disease or parasites, or too

exotic in the eyes of the planters. It was in the dis-

tributing of trees that Keit made his greatest contri-

bution. By 1877 the demand for forest trees in Natal

was very great. Keit observed that even on the coast,

trees were becoming more scarce and valuable every

year. For firewood he recommended the planting of

the silver oak, Grevillea robusta (Durban Botanic

Garden 1877). He provided these on request; he also

provided over 10 000 plants of Eucalyptus globulus

which were planted in an around Durban. These

were especially used to reclaim swampy ground near

the town.

In a period of seven years Keit was to distribute

locally over 25 000 plants and 2 000 packets of seed.

In this undertaking he was assisted by the Natal

Government Railway Company which carried the

packages from the botanic garden free of charge.

Keit was disappointed in the lack of economic pro-

gress in the colony. He once wrote; ‘Natal is a

strange place. Everything under the sun grows here

except rare tropical plants and plants from cold

countries. Even so we have no industry which is in-

digenous, and so hardly anything is fully utilized’

(Keit 1887).

As well as distributing plants and seeds within the

colony Keit sent Wardian cases of indigenous Natal

plants overseas, often using the official government

mail bag for smaller packages. He despatched up-

wards of 40 Wardian cases and packages a year. He

exchanged plants with the botanic gardens at Ade-

laide, Brussels, Calcutta, Cape Town, Glasnevin

(Dublin), Grahamstown, Hamburg, Howrah, Kew,

Lucknow, Madras, Melbourne, Pamplemousses

(Mauritius), Pretoria and Sydney. In addition, Keit

exchanged plants with five other botanical or horti-

cultural societies and 19 nurseries or individuals

overseas. One Wardian case of Encephalartos nata-

lensis which he sent to Adelaide suprisingly survived

one and a half years in a bonded store in Melbourne

and was nearly three years in a dormant state before

being planted out.

Problems facing Keit

Despite the sterling efforts of Keit in establishing

a proper botanic garden in Durban and in winning

international recognition for it, there were funda-

mental problems over which he had no control and

which from 1876 onwards gradually undermined his

work. The first of these was drought. The annual

rainfall, which in 1875 had been 1 372 mm, de-

creased to 889 mm in 1876, and stayed at this level

for 1877, before plummeting to a meagre 771 mm in

1878. With only a 13,6 hi water tank in the garden

and all the watering having to be done with buckets,

many plants, especially conifers and palms, soon

died (Durban Botanic Garden 1878).

The second problem facing Keit was one generally

experienced in Natal, that of a shortage of labour.

Until 1877 he had the use of six short-sentence con-

victs to augment his African labour-force which fluc-

tuated from four to seven in total. After 1877 he had

the use of only three convicts, with the result that in

1879 when his African labourers left because of the

Anglo-Zulu war he began employing Indian labour.

So drastic was his labour crisis that Keit was often

forced to do heavy manual labour himself to keep

the large area of the garden under control. Though

between 1875 and 1879 he had the assistance of a

young German named Paul Hansch, Keit was tied to

the garden by the necessity of taking twice-daily

readings of the government meteorological instru-

ments which had been installed in the garden in Feb-

ruary 1873. As a result, Keit was prevented from

going on plant-collecting expeditions which meant

he had to rely on the public and on his African staff

to collect on his behalf. This made him very de-

pressed. As early as September 1873 he sent a cycad

cone to Kew apologizing that he had nothing else:

‘Our garden,’ he explained, ‘yields nothing you

might care to have’ (Keit 1873d).

Yet another problem facing Keit was that of the

financial viability of the garden. Income rarely ex-

ceeded £450 per annum, the government grant being

£350. Public subscriptions to the garden declined

from £62 in 1873 to a mere £17 in 1877 due mainly to

the fact that Keit became increasingly embarrassed

about demanding money as he had few exotics to

give subscribers in return for their subscriptions.

Sales of plants in the same period, however, rose

from £6 to £92 and would have been even greater

had Keit been able to explore Natal and Zululand

for plants. As far as expenditure was concerned,

wages, rations and his own salary of £150 accounted

for over 70 %. Money spent on building and main-

tenance rarely rose above £50 or 11 % of expendi-

ture.

This fatal combination of drought, labour short-

age and lack of funds seriously affected the garden.

By 1880, due to lack of maintenance, the greenhouse

and summer-houses had collapsed. Even the garden

seats were dangerous to sit on because they were so

ant-eaten. The previous year Mary Elizabeth Barber

visited the Durban Botanic Garden. She wrote after-

wards: ‘The gardens are “out of sight and out of

mind’’; no one appears to either think or care about

them; they have a weedy and neglected appearance.

The conservatory was empty ... In a pond we sud-

denly came upon that superb Indian water lily (Nym-

phaea rubra). It was growing side by side with our

own lovely blue water lily (Nymphaea stellata), and a

very handsome contrast they formed; the Curator

(Keit) very kindly gathered us a bunch of the for-

mer; I begged him not to pick so many, he declared

that we were only too welcome to them, that no one

ever came to look at them in that solitary place. We

took them down to the town, where they were much

admired by people who told us that they had no idea

that they were growing there’ (Barber 1963).

The work done by Keit in saving the garden was

ignored by the public who increasingly made him the

scapegoat for the inadequacies of the by now near-

moribund Horticultural Society. Keit could have

gained popular support by turning the garden into a

pleasure park but he refused to do this. The fact that

Keit was foreign and was careful to register at least

some of his children as German citizens did not en-

74

Bothalia 16,1 (1986)

dear him to some of his fellow townsfolk. Others

found his quiet manner and firm personality objec-

tionable. His successor was later to describe Keit as

‘peculiar’ (Wood 1882).

At the end of 1881 following a re-organisation of

the horticultural society Keit resigned as curator.

Writing to his cousin he merely commented: ‘I could

not come to terms with the rules of the new Garden’s

administration' (Keit 1881).

Kelt’s later career

William Keit did not pass into obscurity in 1881.

In 1876 he had bought a 2 ha plot adjacent to Berea

Road. On this ground he built a home. With money

left by his father-in-law he was also able to establish

a nursery and small dairy there. He traded with

overseas nurseries and, in particular, supplied large

quantities of ferns of the genus Polystichum to Ger-

many. In 1883 he was appointed curator of Durban’s

parks and gardens at a salary of £10 a month. With a

small labour-force, one mule and a lawnmower he

set about beautifying the town. He planted many

trees in the streets, bamboos along Berea Road and

palms along the Victoria Embankment Esplanade

(Henderson 1904). His greatest achievement was

properly laying out and planting Albert, Bulwer and

Victoria Parks in the years between 1883 and 1885.

He also established a municipal nursery at Congella.

He died on 27 August 1916 when the First World

War was at its height and anti-German feeling was

prevalent in Natal (Bruss 1981). Nonetheless, Keit’s

contribution to Durban was not forgotten later, and

a new road through his former property connecting

Berea and Moore roads was named Keits Avenue.

Keit’s legacy

Keit’s successor was a local naturalist and farmer

called John Medley Wood. Wood greatly benefited

from the dissolution of the old Horticultural Society

in 1883 and the establishment of a new Botanic So-

ciety to run the garden (Natal Colony 1883). He also

benefited from additional government money which

facilitated the employment of a Kew-trained gar-

dener called James Wylie, and the establishment of

an herbarium (Schrire 1983). The construction of a

50 000 gallon reservoir by the Town Council in the

grounds of the garden further eased the problems

facing the Durban curator.

By the 1890s Wood, who was an excellent taxo-

nomist, had developed the Durban Garden into the

finest botanic garden in Africa. But his success was

built on the foundations laid by Keit. Despite their

dilapidated state, in 1881 Wood inherited a garden

which was properly laid out and labelled, and which

contained a large collection of plant species. Wood

also inherited the goodwill which the senior staff at

Kew had borne towards Keit; for the next 33 years

Wood carried on a fruitful correspondence and ex-

change of plants with them.

Even after Keit’s departure from the garden, Kew

stood by him and defended him against attack

(Wood 1882). In the nine years that Keit had been

curator of the Durban Botanic Garden they could

see no evidence to contradict the assessment of him

made by Dr Moore of Glasnevin: ‘I have never

known a man whose moral conduct stood higher in

every respect than his does, besides he is a shrewd,

sensible man’ (Moore 1872). Keit was a consistent

man. Seventeen years earlier his employer in Basle

had commented in a reference for him: ‘I was im-

pressed by his modesty and politeness and his

exemplary loyalty to his profession’ (Keit 1864).

UITTREKSEL

William Keit is in 1841 in Sakse/Duitsland gebore.

In sy jong dae het hy dear Europa gereis en in talle

bekende kwekerye gewerk. In 1872 het hy, op aanbe-

veling van die direkteur van Kew Gardens, na Natal

geemigreer om kurator van die Durbanse Botaniese

Tuin te word. Die tuin was so vervalle dat Keit dit

feitlik van die grond af weer moes opbou. Alhoewel

hy grootliks in hierdie poging geslaag het, asook in

die versterking van bande tussen Natal en Kew, kon

hy nie die probleme van ’n ernstige droogte, ’n ar-

beidstekort en ’n tekort aan fondse oorkom nie. Toe

hy in 1881 uit sy pos bedank het, het hy 'n stewige

fondament nagelaat waarop die beroemde plantkun-

dige John Medley Wood voortgebou het. Later het

Keit ’n suksesvolle kwekery in Durban bedryf en 30

jaar lank was hy kurator van die Parke- en Tuinede-

partement. In hierdie hoedanigheid het hy meer ge-

doen as enigiemand anders om die stad Durban te

verfraai.

REFERENCES

BARBER, M. E. 1963. Wanderings in South Africa by sea and

land, 1879. IV. S. Afr. Library Quart. Bull. 18: 3-17.

BRUSS, G. G. 1981. The impact of the first world war on the Ger-

man communities in Natal. M. A. thesis, University of Na-

tal, pp. 161-193.

DURBAN BOTANIC GARDENS 1873-1880. Curator’s reports

for the years 1873-1880. Natal Government Gazette, Natal

Archives, Pietermaritzburg.

DURBAN BOTANIC GARDENS 1877. Curator's report for the

year 1877. Natal Government Gazette, Natal Archives,

Pietermaritzburg.

DURBAN BOTANIC GARDENS 1878. Curator's reports for

the year 1878. Natal Government Gazette, Natal Archives,

Pietermaritzburg.

DURBAN CITY ESTATES DEPARTMENT n.d. File reference

TC15/5J636C.

DURBAN TOWN COUNCIL 1883-1889. Council minutes vol.

14: 14 September 1883, 30 September 1884, 6 October

1884 and vol. 17: 7 January 1889.

HENDERSON, W. P. M. 1904. Durban-, fifty years’s municipal

history, p. 125. Durban.

KEIT, W. 1864. Unpublished letter, 19 February 1864, referred

to by G. Juiz, Basle.

KEIT, W. 1866. Unpublished letter to Hermann Steinmetz, 15

June 1866. Property of Mrs K. Plekker.

KEIT, W. 1867-1869. Unpublished letters to Hermann Stein-

metz: 6 January 1867, 3 September 1868, 11 August 1869.

Property of Mrs K. Plekker.

KEIT, W. 1872a. Unpublished transcript of letter to J. D.

Hooker, Kew, 29 August 1872, in Strey Papers, pp. 776-

777, Killie Campbell Africana Library, Durban.

KEIT, W. 1872b. Unpublished transcript of letter to J. D.

Hooker. Kew, 19 December 1872, in Strey Papers, p. 778,

Killie Campbell Africana Library, Durban.

KEIT, W. 1873a. Unpublished letter to D. Moore, Glasnevin, 10

January 1873, Glasnevin Botanic Gardens, Dublin.

KEIT, W. 1873b. Unpublished letter to Hermann Steinmetz, 10

January 1873. Property of Mrs K. Plekker.

Bothalia 16,1 (1986)

75

KEIT, W. 1873c. Unpublished transcript of letters to J. D.

Hooker, Kew, for example, letter written 16 July 1873, in

Strey Papers, p. 783, Killie Campbell Africana Library,

Durban.

KEIT, W. 1873d. Unpublished transcript of letters to J. D.

Hooker, Kew, 23 September 1873, and to W. T. Thiselton-

Dyer, Kew, 10 January 1873, in Strey Papers, pp. 784-785

and 801 respectively, Killie Campbell Africana Library,

Durban.

KEIT, W. 1874. Unpublished transcript of letter to J. D. Hooker,

Kew, 6 December 1874, in Strey Papers, p. 791, Killie

Campbell Africana Library, Durban.

KEIT, W. 1875. Unpublished transcript of letter to J. D. Hooker,

Kew, 2 September 1875, in Strey Papers, p. 796, Killie

Campbell Africana Library, Durban.

KEIT, W. 1881. Unpublished letter to Wilhelm Steinmetz, 10 Au-

gust 1881. Property of Mrs K. Plekker.

KEIT, W. 1887. Unpublished letter to Wilhelm Steinmetz, 2

April 1887. Property of Mrs K. Plekker.

KEIT, W. n.d.(a). Unpublished letters to family in Dresden.

Property of Mrs K. Plekker.

KEIT, W. n.d.(b). Unpublished letter to Wilhelm Steinmetz,

written about 1895. Property of Mrs K. Plekker.

KEW, ROYAL GARDENS 1881. Report on the progress and

conditions of the Royal Gardens at Kew, during the year

1880 , p. 29. London.

McCRACKEN, DONAL P. 1985. Natal’s botanic gardens: their

role in the colonial economy and society. Unpublished

paper presented to the biennial conference of the South

African Historical Society, University of Cape Town,

1985.

MOORE, D. 1872. Unpublished letter to J. D. Hooker, Kew, 27

August 1872, Glasnevin Botanic Gardens, Dublin.

NATAL AGRICULTURAL AND HORTICULTURAL SO-

CIETY 1848. Minute book, vol. 1, 18 April 1848, Durban

Local History Museum, H1130A.

NATAL COLONY 1883. Law number 21, to incorporate the

Durban Botanic Society.

NELSON, E. CHARLES 1984. An Irish flower garden (illus-

trated by Wendy Walsh). Kilkenny, pp. 83-84 and 183.

SCHRIRE, D. B. 1983. Centenary of the Natal Herbarium,

1882-1982. Bothalia 14: 223-236.

STREY, R. G. 1972. The second curator of the botanic gardens,

Durban, arrived 100 years ago. Natal Wildlife, November

1972, p. 18.

STREY, R. G. 1974. William Keit. The Berea Pictorial 1,4, June

1974, p. 12.

STREY, R. G. n.d. Notes on curators, Natal Herbarium, Dur-

ban, unpublished.

WOOD, JOHN MEDLEY 1882. Unpublished transcript of letter

to W. T. Thiselton-Dyer, Kew, 8 May 1882, in Strey

Papers, p. 1700, Killie Campbell Africana Library, Durban.

Bothalia 16,1: 77-82 (1986)

The plant ecology of the farm Groothoek, Thabazimbi District. III. An

annotated checklist

R. H. WESTFALL*, N. VAN ROOYEN** and G. K. THERON**

Keywords: checklist, growth forms, savanna, utilization classes, vegetation

ABSTRACT

A checklist of 332 plant species representing 226 genera and 84 families is given for the farm Groothoek, Thaba-

zimbi District, Transvaal. The annotations are growth form, utilization class and voucher specimen.

INTRODUCTION

The plant ecology of the farm Groothoek, which

covers 4 000 ha in the Thabazimbi District, was stu-

died in order to supply data on the Sour Bushveld

(Acocks 1975) for the natural resource inventory of

the Department of Agriculture and Water Supply.

This veld type is found mainly in the Waterberg of

the Transvaal and little is known about it.

A phytosociological classification of the vegeta-

tion in a separate publication (Westfall et al. 1985)

describes the habitats of most of the plant species

and these are therefore not repeated here. The

checklist serves as a permanent record of the various

plant species encountered during the study, and the

utilization class annotations are a preliminary classi-

fication of plant species response to management in

the study area.

STUDY AREA

The study area has been described in a companion

paper (Westfall et al. 1985).

METHODS

A reconnaissance of the study area was under-

taken from October to December 1979 to carry out

ground control of aerial photograph interpretation

and to become familiar with the terrain, vegetation

and plant species. Fertile herbarium specimens were

collected during this period. Sampling of the vegeta-

tion began in January 1980 and was completed in

May of the same year. Species collected were num-

bered and later identified by the staff of the National

Herbarium. The specimens are housed in the Na-

tional Herbarium (PRE) of the Botanical Research

Institute, Pretoria.

* Botanical Research Institue, Department of Agriculture and

Water Supply, Private Bag X101, Pretoria 0001, South Africa.

** Department of Botany, University of Pretoria, Pretoria 0002,

South Africa.

Nomenclature and arrangement follow Gibbs

Russell et al. (1984). Collecting numbers (all by R.

H. Westfall) are indicated in parentheses. All plant

names mentioned in Westfall et al. (1985) are reflec-

ted in this list, even in cases where nomenclature or

taxonomy have since been changed. Taxa without

collecting numbers represent sight records.

Utilization class annotations are according to

Foran et al. (1978):

1) Decreaser species — those with a high basal

cover in veld which is in good condition. When veld

deteriorates, the basal cover of these species de-

creases.

2) Increaser I species — those with a low basal

cover in veld which is in good condition. When veld

is under-utilized, the basal cover of these species in-

creases.

3) Increaser II species — those with a low basal

cover in veld which is in good condition. When veld

is over-utilized, the basal cover of these species in-

creases.

4) Increaser III species — those with a low basal

cover in veld which is in good condition. When veld

is selectively utilized, the basal cover of these species

increases.

The classification of the plant species into the util-

ization classes is based on the experience of G. du

Plooy, Extension Officer, Department of Agricul-

ture and Water Supply, Thabazimbi (pers. comm.)

as well as experience gained during field work. The

classification is preliminary and subject to modifica-

tion as more experience is gained.

RESULTS

A total of 332 plant species representing 226 gen-

era and 84 families, was recorded in the study area

(Table 1). An analysis of Table 1 shows that six fami-

lies (7,14% of the total number of families) belong

to Pteridophyta, two families (2,38%) to Gymnos-

permae, eight families (10,71%) to Monocotyle-

dones and 67 families (79,77%) to Dicotyledones.

The Poaceae with 30 genera (13,22% of all genera)

and 50 species (15,02% of all species) is the largest

family in the study area.

78

Bothalia 16,1 (1986)

TABLE 1. — Plant families recorded on the farm Groothoek,

Thabazimbi District, showing number of genera and

species recorded in each family

CHECKLIST

PTERIDOPH YT A

Selaginellaceae

Selaginella dregei (C. Presl) Hieron.

Schizaeaceae

Mohria caffrorum (L.) Desv. ; Increaser I.

Cyatheaceae

Cyathea dregei Kunze (Alsophila dregei (Kunze) Tryon);

Increaser 1(1018).

Dennstaedtiaceae

Pteridium aquilinum (L.) Kuhn ; Increaser I.

Adiantaceae

Cheilanthes hirta Swartz ; Increaser I (905, 1 004).

Cheilanthes viridis (Forssk.) Swartz ( Pellaea viridis (Forssk.)

Prantl var. glauca (Sim) Sim); Increaser I (1 002).

Pellaea calomelanos (Swartz) Link ; Increaser I.

Aspleniaceae

Asplenium splendens Kunze ; Increaser I (945).

GYMNOSPERMAE

Podocarpaceae

Podocarpus latifolius (Thunb.) R. Br. ex Mirb. ; tree; Increas-

er I.

Cupressaceae

Widdringtonia nodiflora (L.)Powrie; shrub; Increaser I (926).

ANGIOSPERMAE

Monocotyledones

Poaceae

Urelytrum agropyroides (Hack.) Hack. (U. squarrosum Hack.);

Increaser II.

Elionurus muticus (Spreng.) Kunth ; Increaser III.

Miscanthus junceus (Stapf) Gibbs Russell ( Miscanthidium

junceum (Stapf) Stapf); not recorded in sample quadrats;

Increaser I (1 006).

Schizachyrium sanguineum (Retz.) Alston ; Increaser II.

Andropogon appendiculatus Nees; Increaser II (821).

Andropogon schirensis.4. Rich. ; Increaser II (913).

Cymbopogon excavatus (Hochst.) Stapf ex Burtt Davy ; In-

creaser I.

Cymbopogon validus ( Stapf) Stapf ex Burtt Davy ; Increaser

II (901, 924, 1021).

Cymbopogon sp. (817).

Hyparrhenia hirta (L.) Stapf; Increaser II (8 1 8).

Monocymbium ceresiiforme (Nees) Stapf; Decreaser (830).

Trachypogon spicatus (L. f) Kuntze; Increaser I (917).

Heteropogon contortus (L.) Roem. & Schult. ; Decreaser.

Diheteropogon amplectens (Nees) Clayton ; Increaser II.

Bothalia 16,1 (1986)

79

Themeda triandra Forssk. ; Decrease!.

Digitaria diagonalis (Nees) Stapf; Increase! II (982).

Digitaria eriantha Steud. ‘subsp. eriantha’; Decrease! (499).

Digitaria eriantha Steud. ‘subsp. transvaalensis’; Decreaser (876).

Digitaria monodactyla (Nees) Stapf; Increaser II (955).

Alloteropsis semialata (R. Br.) Hitchc. ; Increaser I (819).

Brachiaria nigropedata ( Munro ) Stap f ; Decreaser (878).

Paspalum scrobiculatum L. ; Decreaser (838).

Panicum maximum Jacq. ; Decreaser.

Panicum natalense Hochst. ; Decreaser.

Setaria lindenbergiana (Nees) Stapf; Decreaser (981).

Setaria megaphylla (Steud.) Dur. & Schinz ; Decreaser (993).

Setaria sphacelata (Schumach.) Moss var. sphacelata (S. peren-

nisHack.); Decreaser (837, 1 026, 1040).

Rhynchelytrum nerviglume (Franck.) Chiov. (R. setifolium

(Stapf) Chiov.); Decreaser.

Rhynchelytrum repens (Willd.) C.E. Hubb. ; Decreaser.

Tristachya biseriata Stapf ; Increaser II.

Tristachya rehmannii Hack. -, Increaser II (849).

Loudetia simplex (Nees) C.E. Hubb. ; Increaser III (912).

Aristida aequiglumis Hack. ; Increaser II (859).

Aristida canescens Henr. ; Increaser II (954).

Aristida diffusa Trin. subsp. burkei (Stapf) Meld. -, Increaser

II (974).

Aristida junciformis Trin. & Rupr. subsp. junciformis; In-

creaser II (820, 956).

Perotis patens Gand. ; Increaser II.

Sporobolus africanus (Poir.) Robyns & Toumay, Decreaser

(1044).

Eragrostis capensis (Thunb.) Trin. -, Decreaser (827).

Eragrostis curvula (Schrad.) Nees; Increaser II (880).

Eragrostis gummiflua Nees; Increaser II (835).

Eragrostis lappula Nees var. divaricata Stapf ; Increaser 11(822).

Eragrostis pallens Hack. ; Increaser II.

Eragrostis racemosa (Thunb.) Steud. ; Increaser II.

Microchloa caffra Nees; Increaser II.

Cynodon dactylon (L.) Pers. ; Increaser II (961).

Pogonarthria squarrosa (Roem.& Schult.) Pilg. ; Increaser II.

Trichoneura grandiglumis (Nees) Ekman; Increaser II.

Enneapogon pretoriensis Stent; Increaser II (975).

Cyperaceae

Cyperus albostriatus Schrad. ; sedge; Increaser II (949).

Cyperus denudatus L. f. ; sedge; Increaser II (812).

Cyperus leptocladus Kunf/i ; sedge; Increaser II (894).

Cyperus margaritaceus Vahl; sedge; Increaser II (888).

Cyperus obtusiflorus Vahl; sedge; not recorded in sample

quadrats; Increaser II (782).

Cyperus mpesths Kunth ; sedge; Increaser II (775).

Cyperus sphaerospermus Schrad . ; sedge; not recorded in

sample quadrats; Increaser II (76 1).

Mariscus rehmannianus C.B. Cl. ; sedge; Increaser II (497).

Bulbostylis boeckeleriana (Schweinf.) Beetle; sedge; In-

creaser II (81 1).

Bulbostylis burchellii (Fical. & Hiem) C.B. Cl.; sedge; In-

creaser II (842).

Rhynchospora brownii Roem. & Schult. (R. rugosa (Vahl)

Gate); sedge; not recorded in sample quadrats; In-

creaser II (762).

Coleochloa setifera (Ridley) Gilly ; sedge; Increaser II (904).

Carex spicato-paniculata C.B. Cl.; sedge; Increaser II (952,

1019).

Xyridaceae

Xyris congensis Buettn. ; forb; not recorded in sample qua-

drats; Increaser II (773).

Commelinaceae

Commelina africana L. var. lancispatha C.B. Cl.; forb; In-

creaser 11 (931,967).

Commelina erecta L. ; forb; Increaser II (530).

Commelina undulata R. Br. ; forb; not recorded in sample

quadrats; Increaser II (789).

Cyanotis pachyrrhiza Oberm. ; forb; not recorded in sample

quadrats; Increaser II (774).

Liliaceae

Littoniamodesta//oofc. ; geophyte; Increaser II (887).

Anthericum galpinii Bak. var. norlindhii (Weim.) Oberm. ;

geophyte; Increaser II (1 013).

Eriospermum sp.; geophyte; Increaser II (857).

Aloe transvaalensis Kuntze; succulent; Increaser II.

Scilla nervosa (Burch.) Jessop; geophyte; Increaser II (941).

Protasparagus buchananii (Bak.) Oberm. ( Asparagus buchan-

anii Bak.); suffrutex; Increaser II.

Protasparagus setaceus (Kunth) Oberm. ( Asparagus setaceus

Kunth); suffrutex; Increaser II (1 048).

Protasparagus suaveolens (Burch.) Oberm. ( Asparagus suave-

olens Burch.); suffrutex; Increaser II.

Protasparagus virgatus (Bak.) Oberm. (Asparagus virgatus

Bak.); climber; Increaser II (950).

Myrsiphyllum asparagoides/XJ Willd. (Asparagus asparagoides

(L.) Wight); climber; Increaser II (1 008).

Amaryllidaceae

Scadoxus puniceus (L.) Friis & Nordal; geophyte; not record-

ed in sample quadrats; Increaser II (755).

Brunsvigia sp., cf. B. radulosa Herb. ; geophyte; Increaser II

(927).

Hypoxidaceae

Hypoxis angustifolia Lam. ; geophyte; Increaser II (723).

Hypoxis obtusa Burch. ; geophyte; Increaser II (85 1).

Hypoxis rigidula Bak. ; geophyte; Increaser II (863, 1 028).

Velloziaceae

Xerophy ta retinervis Bak. ; forb; Increaser II.

Iridaceae

Aristea woodii N.E. Br. ; geophyte; Increaser II (939).

Tritonia nelsonii ZtaAr. ; shrub; Increaser II (983).

Babiana hypogea Hurc/i. ; geophyte; Increaser II (1 039).

Dicotyledones

Ulmaceae

Celtis africana Burm.f ; tree; Increaser I.

Moraceae

Ficus ingens (Miq.) Miq. ; tree; Increaser I.

Ficus sur Forssk. (F. capensis Thunb.); tree; Increaser I

(713).

Ficus thonningii Blume (F. burkei (Miq.) Miq.); tree; In-

crease! II (1 058).

Proteaceae

Faurea saligna Harv. ; tree; Increaser II (770).

Protea caffra Meisn. ; tree or shrub; Increaser II.

Protea gaguedi J.F. Gmel. ; shrub; Increaser II.

Protea roupelliae Meisn. ; tree or shrub; Increaser II.

Loranthaceae

Tapinanthus natalitius (Meisn.) Danser subsp. zeyheri (Harv.)

Wiens; parasitic shrub; not recorded in sample quadrats

(715).

Santalaceae

Osyris lanceolata Hochst. & Steud. ; shrub; Increaser I & II

(1066).

Thesium racemosum Bemh. ; forb; Increaser II (836).

Olacaceae

Ximenia americana L. ; shrub; not recorded in sample qua-

drats; Increaser I & II (793).

Amaranthaceae

Kyphocarpa angustifolia (Moq.) Lopr. (Cyphocarpa angusti-

folia Cops.); forb; Increaser II (893, 979).

Cyathula cylindrica Moq. ; forb; Increaser II (1 007).

Achyranthes aspera L. ; forb; Increaser II (786).

Achyranthes sicula (L.) All. ; forb; Increaser II (1 016).

Brayulinea densa (Willd.) Small; forb; Increaser II (962).

Aizoaceae

Limeum viscosum (Gay) Fenzl subsp. viscosum var. glomera-

tum (Eckl. <6 Zeyh.) Friedr.; forb; Increaser II (882).

Psammotropha mucronata (Thunb.) Fenzl var. mucronata;

forb; Increaser II (930).

Portulacaceae

Talinum caffrum (Thunb.) Eckl. & Zeyh.; forb; Increaser II

(490).

Portulaca kermesina N.E. Br. ; forb; Increaser II (91 1).

Caryophyllaceae

Dianthus mooiensis 1. Williams subsp. kirkii (Burtt Davy)

Hooper; forb; Increaser II (779).

80

Ranunculaceae

Clematis sp.; climber; Increaser II (1 042).

Clematopsis scabiosifolia (DC.) Hutch. ; forb; not recorded

in sample quadrats; Increaser II (790).

Annonaceae

Hexalobus monopetalus (A. Rich.) Engl. & Diels; shrub; In-

creaser II (896).

Capparaceae

Cleome maculata (Sond.) Szyszyl. ; forb; Increaser II (874).

Crassulaceae

Crassula sarcocaulis Eckl. & Zeyh. subsp. sarcocaulis; succu-

lent forb; Increaser II (1 035).

Pittosporaceae

Pittosporum viridiflorum Sims; tree; Increaser II (1 052).

Myrothamnaceae

Myrothamnus flabellifolia (Sond.) Welw.; shrublet; Increaser

II.

Rosaceae

Parinari capensis Harv. subsp. capensis; shrublet; Increaser II

(834).

Fabaceae

Subfamily Mimosoideae

Acacia ataxacantha DC. ; semi-scandent shrub; Increaser I.

Acacia caffra (Thunb.) Willd. ; tree; Increaser II (1 Oil).

Acacia karroo Hayne; tree; Increaser II.

Elephantorrhiza burkei Benth. ; shrub; Increaser II (899, 900).

Elephantorrhiza obliqua Burtt Davy var. glabra Phill. ; shrub;

Increaser II (749).

Subfamily Caesalpinioideae

Burkea africana //ooA:. ; tree; Increaser II (722).

Cassia comosa (E.Mey.) Vogel var. capricomia Steyaert ; forb;

Increaser II (781).

Subfamily Papilionoideae

Calpurnia sp., cf. C. aurea (Ait.) Benth. ', tree; Increaser II

(1073).

Lotononis sp.; forb; Increaser II (815).

Pearsonia cajanifolia (Harv.) Polhill subsp. cryptantha (Bak.)

Polhill; forb; Increaser II (1 033).

Pearsonia sessilifolia (Harv.) Duemmer subsp. sessilifolia; forb;

Increaser II (850).

Argyrolobium transvaalense Sc/rinz; shrub; Increaser II (861).

Indigofera comosa N.E. Br. ; shrublet; Increaser II (1 023).

Indigofera egens N.E. Br. ; forb; Increaser II (868).

Indigofera fHipes Benth. ex Harv. ; forb; Increaser II (960).

Indigofera hedy antha Eckl. & Zeyh. ; forb; Increaser II (920).

Indigofera hilaris Eckl. & Zeyh. ; forb; not recorded in sample

quadrats; Increaser II (746).

Indigofera spicata Forssk. ; forb; Increaser II (847).

Psoralea polystictafiewf/i.; shrub; Increaser II (1 047).

Tephrosia elongata E. Mey. var. elongata; forb; Increaser II

(918).

Tephrosia longipes Meisn. var. lurida (Sond. J.B. Gillett; forb;

Increaser II (879).

MunduleasericeafHWM.) A. Chev.; shrub; Increaser II (734).

Zomia milneana MoWew&r. ; forb; Increaser II (854, 1 038).

Abrus laevigatus E. Mey. ; twining undershrub; Increaser II

(992, 1061).

Glycine wightii (Wight & Am.) Verde, subsp. wightii var. longi-

cauda (Schweinf.) Verde. ; twining forb; Increaser II (988).

Erythrina lysistemon Hutch. ; tree or shrub; Increaser (989).

Rhynchosia monophylla Schltr . ; twining forb; Increaser II

(853, 925).

Rhynchosia spectabilis Schinz; undershrub; Increaser II

(980).

Rhynchosia totta (Thunb.) DC. ; twining forb; Increaser II

(966).

SphenostylisangustifoliaSo/id. ; shrublet; Increaser II (1031).

Oxalidaceae

Oxalis depressa Eckl. & Zeyh. ; forb; Increaser II (111).

Erythroxylaceae

Erythroxylum emarginatum Thonn. ; shrub; Increaser I (890).

Rutaceae

Calodendrum capense (L.f.) Thunb.; tree; not recorded in

sample quadrats; Increaser I (808).

Bothalia 16,1 (1986)

Vepris lanceolata (Lam.) G. Don. ; tree or shrub; Increaser I

(1 063).

Simaroubaceae

Kirkia wilmsii Engl ; tree; Increaser I (994).

Malpighiaceae

Sphedamnocarpus pruriens ( Juss .) Szyszyl. var. pruriens;

climber; Increaser II (1 074).

Polygalaceae

Polygala amatymbica Eckl. & Zeyh. ; forb; Increaser II (745).

Polygala hottentotta Presl; forb; Increaser II (928).

Polygala uncinata E. Mey. ex Meisn. ; forb; not recorded in

sample quadrats; Increaser II (802).

Euphorbiaceae

Pseudolachnostylis maprouneifolia Pax; tree or shrub; In-

creaser II (886).

Phyllanthus incurvus Thunb. ; forb; Increaser II (504).

Phyllanthus parvulus Sond. ; forb; Increaser II (881).

Croton gratissimus Burch, subsp. subgratissimus (Prain) Burtt

Davy ; tree; Increaser II (889).

Acalypha angustata Sond. var. glabra Sond. ; forb; Increaser

11(914).

Clutia pulchella L.; shrub; Increaser I & II (892, 935, 948).

Euphorbia ingens E. Mey. ex Boiss. ; tree ; Increaser I.

Euphorbia schinzii Pax; succulent herb; Increaser II.

Buxaceae

Buxus macowanii Oliv.; shrub; Increaser I (987, 1 067).

Anacardiaceae

Lannea discolor (Sond.) Engl. ; shrub; Increaser II (877).

Lannea edulis (Sond.) Engl. ; shrublet; Increaser II (875).

Ozoroa paniculosa (Sond.) R. & A. Fernandes; tree or shrub;

Increaser II (797).

Rhus dentata Thunb. ; shrub; Increaser I & II (1 037).

Rhus dura Schonl. ; shrub; Increaser I (873).

Rhus leptodictya Diels; tree or shrub; Increaser I & II.

Rhus pyroidesf?«rc/j. ; shrub; Increaser II (957, 1 054).

Celastraceae

Maytenus tenuispina (Sond.) Marais; shrub; Increaser II (856,

883).

Maytenus undata (Thunb.) Blakelock; tree; Increaser I & II

(971, 1059).

Pterocelastrus echinatus N.E. Br.; tree or shrub; Increaser II

(951).

Pterocelastrus rostratus (Thunb.) Walp.; shrub; Increaser I

(1010).

Icacinaceae

Apodytes dimidiata E. Mey. ex Am. subsp. dimidiata; tree or

shrub; Increaser I (895, 907, 970).

Sapindaceae

Pappea capensis Eckl. & Zeyh. ; tree; Increaser II (1 053).

Rhamnaceae

Ziziphus mucronata Willd. subsp. mucronata; tree or shrub;

Increaser II (963).

Vitaceae

Rhoicissus digitata (L.f.) Gilg. & Brandt; climber or shrub;

Increaser I & II (869).

Rhoicissus tridentata (L.f.) Wild & Drum. ; shrub; Increaser

11(1025).

Tiliaceae

Grewia monticola Sond. ; shrub; Increaser II (1 07 1).

Grewia occidentalis L. ; shrub; Increaser I & II (809).

Grewia rogersii Burtt Davy & Greenway ; shrub; not recorded

in sample quadrats; Increaser II (738).

Triumfetta rhomboidea/ac<7. ; shrublet; Increaser II.

Triumfetta sonderi Fical. & Hiem; forb; Increaser II (852).

Malvaceae

Pavonia columella Cav. ; forb; Increaser II (1 020).

Sterculiaceae

Dombeya rotundifolia (Hochst.) Planch, var. rotundifolia;

tree; Increaser I & II.

Hermannia depressa N.E. Br. ; forb; Increaser II (1 072).

Waltheria indica L.; forb; Increaser II (485).

Ochnaceae

Ochna pulchra Hook. ; tree or shrub; Increaser I & II.

Bothalia 16,1 (1986)

81

Clusiaceae

Hypericum aethiopicum Thunb. subsp. sonderi (Bred.) N.K.B.

Robson ; forb; Increaser II (841).

Hypericum lalandii Choisy ; forb; Increaser II (831).

Elatinaceae

Bergia decumbens Planch, ex Harv. ; forb; Increaser II (816,

826).

Cactaceae

Opuntia sp.; succulent shrub; exotic; Increaser II.

Thymelaeaceae

Gnidia caffra Meisn. ; forb; not recorded in sample quadrats;

Increaser II (798).

Passerina montana Thoday; ericoid shrub; Increaser I (756).

Combretaceae

Combretum apiculatum Sond. ; tree; Increaser II.

Combretum moggii Exell; shrub; Increaser I (891, 998).

Combretum molle R. Br. ex G. Don; tree; Increaser II.

Combretum zeyheri Sond. ; tree; Increaser II.

Terminalia sericea Burch, ex DC. ; tree; Increaser II.

Myrtaceae

Syzygium guineense (Willd.)DC. ; tree; not recorded in sample

quadrats; Increaser I & II (729).

Heteropyxis natalensis Harv. ; tree; Increaser I & II (848).

Melastomataceae

Dissotis debilis (Sond.) Triana var. debilis forma debilis; forb;

not recorded in sample quadrats; Increaser I & II (765).

Araliaceae

Cussoniapaniculata Eckl. & Zeyh. ; tree; Increaser II (1 029).

Cussonia spicata Thunb. ; tree or shrub; Increaser I & II (978).

Apiaceae

Heteromorpha arborescens (Spreng.) Cham. & Schlechtd.;

shrub; Increaser II (1 068).

Myrsinaceae

Myrsine africanai. ; shrub; Increaser I & II (946).

Sapqtaceae

Bequaertiodendron magalismontanum (Sond.) Heine & J.H.

Hemsl. ; tree or shrub; Increaser I & II.

Mimusops zeyheri Sond. ; tree or shrub; Increaser II (910).

Ebenaceae

Euclea crispa (Thunb.) Guerke var. crispa; tree or shrub; In-

creaser I & II (518).

Euclea linearis Zeyh. ex Hiem ; shrub; Increaser II.

Euclea natalensis A. DC. \ shrub; Increaser II (959, 1 049).

Diospyros lyciodes Desf. subsp. guerkei (Kuntze) De Winter,

shrub; Increaser II (958, 1 050).

Diospyros whyteana (Hiem) F. White', shrub; Increaser II

(944).

Oleaceae

Schrebera alata (Hochst.) Welw. ; tree; Increaser II (972).

Olea europaea L. subsp. africana (Mill.) P.S. Green', tree or

shrub; Increaser II (999, 1 065).

Jasminum multipartitum Hochst. ; climbing shrub; Increaser

I (754).

Loganiaceae

Strychnos cocculoides Bak. ; tree; Increaser II.

Strychnos madagascariensis Poir. ; tree; Increaser II (973).

Strychnos pungens Soler. ; tree; Increaser II (872).

Nuxia congesta R. Br. ex Fresen. ; tree; Increaser II (909,

997).

Buddleja salviifolia (L.) Lam. ; tree; Increaser I (1 017).

Gentianaceae

Chironia purpurascens (E. Mey.) Benth. & Hook. f. subsp.

humilis (Gilg) Verdoom; forb; Increaser II (829).

Apocynaceae

Acokanthera oppositifolia (Lam.) Codd; tree or shrub; In-

creaser II (1 057).

Landolphia capensis Oliv. ; shrub; Increaser II (866).

Diplorhynchus condylocarpon (Mull. Arg.) Pichon; tree or

shrub; Increaser II (544).

Periplocaceae

Cryptolepis oblongifolia (Meisn.) Schltr. ; shrub; Increaser II.

Asclepiadaceae

Pachycarpus schinzianus (Schltr.) N.E. Br. ; forb; Increaser II

(846).

Asclepias fruticosaZ,.; forb; Increaser II (711).

Asclepias sp.; forb; Increaser II (828).

Secamone alp in ii Schultes', shrubby climber; Increaser II

(953).

Secamone filiformis (L.f.) J.H. Ross; shrubby climber; In-

creaser I (991).

Convolvulaceae

Ipomoea bathycolpos Hallier f. var. bathycolpos; forb; In-

creaser II (858).

Boraginaceae

Ehretia rigida (Thunb.) Druce; shrub; Increaser II (1 069).

Verbenaceae

Lantana rugosa Thunb. ; shrub; Increaser I & II (509).

Lippia javanica (Burm. f.) Spreng.; shrublet; Increaser II

(840, 1041).

V itex pooara Corbishley ; tree or shrub; Increaser II (885).

Vitex rehmannii Guerke; tree or shrub; Increaser II (538).

Clerodendrum glabrum E. Mey. ; tree; Increaser II (1 070).

Lamiaceae

Leonotis leonurus (L.) R. Br.; forb; Increaser II.

Stachys natalensis Hochst. var. galpinii (Briq.) Codd; forb;

not recorded in sample quadrats; Increaser II (740).

Stachys natalensis Hochst. var. natalensis; forb; Increaser II

(862, 969).

Plectranthus fruticosus L ’Herit. ; forb; Increaser II (947).

Plectranthus verticillatus (L.f.) Druce; forb; Increaser II

(990).

Plectranthus sp.; undershrub; Increaser II (906).

Becium obovatum (E. Mey. ex Benth.) N.E. Br.; forb; In-

creaser II (902).

Solanaceae

Solanum giganteum Jacq. ; forb; Increaser II (1 009).

Solanum panduraeforme E. Mey. ; forb; Increaser II (1 051).

Solanum rigescens Jacq. ; forb; not recorded in sample qua-

drats; Increaser II (771).

Scrophulariaceae

Nemesia fruticans (Thunb.) Kuntze; forb; Increaser II (929).

Sutera palustris ///era; forb; Increaser II (1 022).

Selaginaceae

Selago capitellata Schltr. ; forb; Increaser II (919).

Striga bilabiata (Thunb.) Kuntze; parasitic forb (968).

Striga gesnerioides (Willd.) Vatke ex Engl. ; parasitic forb

(984).

Acanthaceae

Chaetacanthus costatus Nees; shrublet; Increaser II (933).

Ruellia cordata Thunb. ; forb; Increaser II (860).

Barleria bremekampii Oberm. ; shrublet; Increaser II (1 005).

Blepharis subvolubilis C.B. Cl. var. subvolubilis ; forb; In-

creaser II (976).

Crossandra greenstockii S. Moore; shrublet; not recorded in

sample quadrats; Increaser II (731).

Hypoestes verticillaris (L.f.) R. Br. ex C.B. Cl. ; forb; Increas-

er II (995).

Rubiaceae

Oldenlandia herbacea (L.) Roxb. ; forb; Increaser II (763).

Tarenna supra-axilaris (Hemsl.) Brem. subsp. barbertonensis

(Brem.) Bridson ( T . barbertonensis (Brem.) Brem.); shrub;

Increaser 11(1 000).

Gardenia volkensii K. Schum. subsp. volkensii (G. spatulifolia

Stapf & Hutch.); shrub; Increaser I & II (884, 1 046).

Rothmannia capensis Thunb. ; tree; Increaser I (1 051).

Tricalysia lanceolata (Sond.) Burtt Davy; tree; Increaser I

(1 060).

Pentanisia angustifolia (Hochst.) Hochst. ; forb; Increaser II

(943).

Vangueria infausta Burch, subsp. infausta; shrub; Increaser

II (772).

Pygmaeothamnus zeyheri (Sond.) Robyns var. zeyheri; peren-

nial shrublet with woody rootstock; Increaser II (832,

897).

Tapiphyllum parvifolium (Sond.) Robyns; shrub; Increaser

11(871).

Canthium gilfillanii (N.E. Br.) O.B. Miller; shrub not record-

ed in sample quadrats; Increaser I (717).

82

Bothalia 16,1 (1986)

Canthium huillense Hiem ; shrub; Increase! II (996).

Pachystigma triflorum Robyns ; shrub; Increaser I & 11(870).

Fadogia monticola Robyns', shrublet; Increaser II (799).

Anthospermum rigidum Eckl, & Zeyh.; forb; Increaser II

(855).

Richardia brasiliensis Gomez ; procumbent forb; Increaser II

(843).

Richardia scabra L. (Borreria scabra (Schumach. & Thonn.)

K. Schum.); forb; Increaser II (1 043).

Dipsacaceae

Scabiosa columbaria L. ; forb; Increaser II (825).

Campanulaceae

Wahlenbergia caledonica Sond. ; forb; Increaser II (922).

Wahlenbergia lycopodioides Schltr. & V. Brehm.\ forb; In-

creaser II (938).

Lightfootia paniculata Sond. ; forb; not recorded in sample

quadrats; Increaser II (750).

Lobeliaceae

Lobelia aquaemontis E. Wimm. ; forb; Increaser II (940).

Lobelia decipiens Sond. ; forb; Increaser II (814).

Asteraceae

Vernoniagalpinii Klatt\ shrublet; Increaser II (977, 1 012).

Vemonia natalensis Sch. Bip.; shrublet; Increaser II (865,

1 027).

Vernonia oligocephala (DC.) Sch. Bip. ex Walp.; shrublet;

Increaser II (844).

Vernonia staehelinoides/forv.; forb; Increaser 11(965, 1 032).

Felicia muricata (Thunb.) Nees subsp. strictifolia Grow, forb;

Increaser II (803).

Conyza scabrida DC. ; forb; Increaser 11 (1 055).

Brachylaena rotundata S. Moore; shrub; Increaser II (903).

Gnaphalium filagopsis Hilliard & Burtt ( Amphidoxa filaginea

Fical. & Hiern); forb; Increaser II (824).

Helichrysum caespititium (DC.) Harv.; forb; Increaser II

(751).

Helichrysum cephaloideum DC. ; forb; Increaser II (937).

Helichrysum kraussii Sch. Bip. ; shrublet; Increaser II.

Helichrysum mimetes S. Moore; forb; Increaser II (923).

Helichrysum nudifolium (L.) Less. var. nu difolium; forb; In-

creaser II (839).

Helichrysum uninervium 5nrff Davy, forb; Increaser II (778).

Helichrysum sp.; forb; Increaser II (921).

Stoebe vulgaris Levyns; ericoid shrublet; Increaser III (724).

Geigeria burkei Harv. subsp. burkei var. burkei; shrublet; In-

creaser II (833).

Geigeria burkei Harv. subsp. burkei var. zeyheri (Harv.)

Merxm.; shrublet; Increaser II (934).

Geigeria elongata Alston ; forb; Increaser II (898).

Bidens pilosa L. ; exotic forb; Increaser II.

Schkuhria pinnata (Lam.) Cabr. ; forb; exotic; Increaser II

(964).

Tagetes minuta Z,. ; forb; exotic; Increaser II.

Brachymeris bolusii Z/wtcA. ; shrublet; Increaser II (1 014).

Lopholaena coriifolia (Sond.) Phill. & CA. Sm.; shrub; In-

creaser II (1 036).

Cineraria lobata L ’H&rit. ; forb; Increaser II (932).

Senecio barbertonicus Klatt; shrub; Increaser II (1 003).

Senecio conrathii W.ZT. Br. ; forb; Increaser II (823).

Senecio erubescens Ait. var. crepidifoliusDC. ; forb; Increas-

er II (813).

Senecio oxyriifoliusDC. ; forb; Increaser II (1 024).

Senecio pleistocephalusS.Afocwe; shrub; Increaser 11(1 056).

Senecio ruwenzoriensis S. Moore; forb; Increaser II (1 034).

Senecio venosus Harv. ; forb; Increaser II (864).

Euryops pedunculatus V.£’. Br. ; shrub; Increaser II (916).

Osteospermum jucundum (Phill.) T. Norl. ; forb; Increaser II

(942).

Ursinia nana DC. ; forb; Increaser II (936).

Gazania krebsiana Less, subsp. serrulata (DC.) Roessl. ; forb;

Increaser II (804).

Berkheya carlinopsis Welw. ex O. Hoffm. subsp. magalismon-

tana (H. Bol.) Roessl. ; forb; Increaser II (908).

Dicoma anomala Sond. subsp. anomala; forb; Increaser II

(915).

Gerbera am bigua (Cass.) Sch. Bip. ; forb; Increaser 11(1 030).

ACKNOWLEDGEMENTS

The authors thank Dr J. C. Scheepers for com-

ments and the staff of the National Herbarium for

identification of the taxa.

UITTREKSEL

’n Kontrolelys van 332 plantspesies, wat 226 genus-

se en 84 families verteenwoordig, word gegee vir die

plaas Groothoek, distrik Thabazimbi, Transvaal.

Groeivorm, benuttingsklas en kontrole-eksemplaar

word aangetoon.

REFERENCES

ACOCKS, J. P. H. 1975. Veld types of South Africa. 2nd edn,

Mem. bot. Surv. S. Afr. No. 40.

DE DELLA TORRE, C. G. & HARMS, H. 1958. Genera

Siphonogamarum. Leipzig: Engelmann.

DYER, R.A. 1976. The genera of southern African flowering

plants. Vol. 2. Gymnosperms and monocotyledons. Pretoria:

Government Printer.

FORAN, B. D., TAINTON, N. M. & BOOYSEN, P. DE. V.

1978. The development of a method for assessing veld condi-

tion in three grassveld types in Natal. Proc. Grassld Soc. sth.

Afr. 13: 27-33.

GIBBS RUSSELL, G. E., GERMISHUIZEN, G., HERMAN,

P., OLIVIER, P., PEROLD, S. M., REID, C., RETIEF,

E., SMOOK, L., VAN ROOY, J. & WELMAN, G. 1984.

List of species of southern Africa plants. Mem. bot. Surv. S.

Afr. No. 48.

WESTFALL, R. H., VAN ROOYEN, N. & THERON, G. K.

1985. The plant ecology of the farm Groothoek, Thabazimbi

District. II. Classification. Bothalia 15: 655-688.

Bothalia 16,1: 83-85 (1986)

A checklist of Pteridophyta of the north-eastern Orange Free State

J. P. ROUX*

Keywords: checklist, north-east Orange Free State, Pteridophyta

ABSTRACT

A checklist of Pteridophyta recorded within a defined area of the north-eastern Orange Free State is presented.

The geology, climate and vegetation of this area are discussed in brief.

INTRODUCTION

The 63 species of pteridophytes mentioned in this

checklist occur in an area within the north-eastern

Orange Free State. The area comprises the following

quarter degree square grids:

2828 (Bethlehem): -AD, -BC, -CB, -DA, -DB;

2829 (Harrismith): -AA, -AB, -AC, -AD, -BA,

-BB, -BC, -CA.

A single voucher specimen is cited for each quar-

ter degree square in which the respective species has

been recorded to date. All the cited vouchers are

currently housed in the Drakensberg National Bo-

tanic Garden Herbarium, unless otherwise stated.

TOPOGRAPHY AND GEOLOGY

The study area borders on Natal and Lesotho. To

the south of it lies Lesotho which forms the water-

shed between the Orange and Vaal River systems

and the eastern escarpment which forms the water-

shed between the Vaal and Tugela River systems of

Natal.

Included in the area are the steep mountainous

parts with the highest point at c. 3 281 m near Mont-

aux-Sources. Although the elevation of most of the

area is much lower (c. 1 500-1 700 m) the topogra-

phy is much dissected along the foothills of the high

Drakensberg and the Natal border. Further inland

the topography is less dissected but a few prominent

outliers such as Platberg (2 394 m), Bakerskop

(2 037 m) and Rensburgkop (2 235 m) arise almost

directly from the surrounding almost flat landscape.

King (1982) suggested that these steep-sided residu-

als of the Clarens Formation and black Drakensberg

basalt, occurring largely along the present-day es-

carpment, already formed a watershed in Early Ce-

nozoic time.

As in the High Drakensberg, most of the higher

outliers are overlain by basaltic lava to a thickness of

c. 914 m which forms a steep upper catchment of

well grassed hills and valleys. The geology of the

area and surroundings has been mapped and de-

scribed by Van Eeden (1937), Visser (1955), Killick

(1963), Van der Eyck (1967), Spies (1969) and King

(l.c.).

* Drakensberg National Botanic Garden, P.O. Box 157, Harri-

smith 9880, RSA.

CLIMATE

The climate within the region can be described as

temperate during the summer months but severe

during the winter. Temperatures fluctuate consider-

ably and are influenced by factors such as topogra-

phy, altitude and wind. During the summer (Octo-

ber-March) day temperatures may reach 30°C while

the night temperatures may often reach a maximum

of 13°C. Winter temperatures are more severe and

the day temperatures frequently do not reach the

15°C mark. On many occasions the maximum tem-

perature barely reaches 5°C. Night temperatures

may drop as low as -15°C. It can be accepted that at

higher elevations the temperature may be consider-

ably lower and that frost is an almost daily occur-

rence during the months April to August.

The predominant winds blow either from the west

or the east. The westerly winds may often reach gale

force during the period August to September and

regularly cause dust storms. These winds are dry and

warm, whereas the easterly winds usually cause a

considerable, usually rapid, drop in temperature

along the escarpment and higher-lying areas.

Rain usually falls either in the form of gentle soak-

ing rain or severe thunderstorms, often associated

with hail. The main rainy season is from September

to March. The average precipitation at Harrismith is

about 777 mm per annum while at Phuthaditjhaba

(Witsieshoek) it is 811 mm per annum (Weather Bu-

reau, 1965). At higher elevations the precipitation

can exceed 1 000 mm (Killick 1978). Here snowfalls

are experienced annually between the period July to

September; lower-lying areas experience only spo-

radic snowfalls.

VEGETATION

Acocks (1975), recognized three major vegetation

types within the study area. Highveld Sourveld (no.

44) occupies the lower-lying areas whereas the High-

veld Sourveld to Cymbopogon-Themeda Veld Tran-

sition (no. 56) occurs along the Natal escarpment.

Themeda-Festuca Alpine Veld (no. 58) is largely

confined to the High Drakensberg along the Lesotho

border.

Published works on the vegetation of the north-

eastern Free State and surroundings are limited to

species lists of the Golden Gate Highlands National

Park by Roberts (1969), Markotter’s (1930) compila-

tion of Thode’s plant collections between 1891 and

84

Bothalia 16,1 (1986)

1914 at Phuthaditjhaba (Witsieshoek), Oliviershoek

Pass and Koolhoek, and a plantsociological study on

the pioneer vegetation in the north-eastern Free

State carried out by Stam (1973).

Vegetation surveys of the higher lying regions of

the Natal Drakensberg bordering on the study area

were carried out by Staples & Hudson (1938), Kil-

lick (1963) and Edwards (1967).

Most of the area is covered by grassland. The deep

valleys and protected gorges at lower elevations,

especially on the moist eastern slopes, are forested

and are mostly dominated by Leucosidea sericea and

mixed Leucosidea-Buddlejci scrub forest. The more

sheltered forests are dominated by Podocarpus lati-

folius (Thunb.) R. Br. ex Mirb.

The abundance of bulbous plants together with

the fire-resistance of most plants occurring in the

grassveld areas suggest that fire has been an import-

ant factor since very early times (Bayer 1955).

UITTREKSEL

’n Kontrolelys van Pteridophyta, wat binne ’n om-

skrewe gebied in die noordoostelike Oranje-Vrystaat

waargeneem is, word gegee. Die geologie, klimaat en

plantegroei van die gebied word kortliks bespreek.

REFERENCES

ACOCKS, J. P. H. 1975. Veld types of South Africa. Mem. bot.

Surv. S. Afr. No. 40.

BAYER, A. W. 1959. The ecology of grasslands. In D. Meredith,

The grasses and pastures of southern Africa, edn 2: 539-550.

Parow: Cape Times.

EDWARDS, D. 1967. A plant ecological survey of the Tugela

River basin, Natal. Mem. bot. Surv. S. Afr. No. 36.

KILLICK, D. J. B. 1963. An account of the plant ecology of the

Cathedral Peak of the Natal Drakensberg. Mem. bot. Surv.

S. Afr. No. 34.

KILLICK, D. J. B. 1978. Further data on the climate of the Al-

pine vegetation belt of eastern Lesotho. Bothalia 12:

567-572.

KING, L. 1982. The Natal Monocline: explaining the origin and

scenery of Natal, South Africa, edn 2. Pietermaritzburg: Uni-

versity of Natal Press.

MARKOTTER, E. I. 1930. ’n Plantegeografiese skets en die

flora van Witsieshoek, O.V.S., Oliviershoek, Natal; en

Koolhoek, O.V.S. Ann. Univ. Stell. 8, A,l: 1-50.

ROBERTS, B. R. 1969. The vegetation of the Golden Gate

Highlands National Park. Koedoe: 15-28.

SPIES, J. J. 1969. Die geologie en geomorfologiese geskiedenis

van Golden Gate Hoogland Nasionale Park. Koedoe:

184-198.

STAM, R. D. 1973. Pioneer vegetaties uit den bergen van de Oos-

telijke Oranje Vrijstaat (Zuid- Africa). Unpublished survey.

STAPLES, R. R. & HUDSON, W. K. 1938. An ecological survey

of the mountain areas of Basutoland. Litchworth: The

Golden City Press.

WEATHER BUREAU 1965. Climate of South Africa, part 9.

W.B.20. Department of Transport, Republic of South

Africa.

CHECKLIST

L Y CO POD I ACE AE

Lycopodium L.

clavatum L. 2829 (Harrismith): (-AD), Roux 846 ; (— CA),

Roux 1078.

saururusitfm. 2829 (Harrismith): (- AC), Roux 987; (-AD),

Roux 852: (-CA), Roux 1283.

verticillatum L.f. 2829 (Harrismith): (— CA), Roux 836.

SELAGINELLACEAE

Selaginella Beauv.

caffrorum (Mildej Hieron. 2828 (Bethlehem): (-DA), Roux

1014. 2829 (Harrismith): (-AC), Roux 1549: (-CA),

Roux 879: (-CB ),Roux 928.

imbricata (Forssk.) Spring ex Decne. 2828 (Bethlehem):

{-DA), Roux 1030: (-DB), Roux 906.

mittenii Bak. 2829 (Harrismith): {-AC), Roux 868: (-AD),

Roux 884.

ISOETACEAE

Isoetes L.

transvaalensis Jermy & Schelpe 2828 (Bethlehem): (-DB),

Roux 955. 2829 (Harrismith): (- AC), Roux 1277.

EQUISETACEAE

Equisetum L.

ramosissimum Desf 2828 (Bethlehem): (-D A), Roux 1023

2829 (Harrismith): (- AC), Roux 1551 .

OPHIOGLOSSACEAE

Ophioglossum L.

lancifolium Presl 2828 (Bethlehem): (-DB), Roux 954.

polyphyllum A. Br. 2828 (Bethlehem): (-BC), Roux 939:

(-BD), Roux 950. 2829 (Harrismith): (-AA), Roux

1035: (-AC), Van der Zeyde s.n. ; (- AD), Roux 1060.

OSMUNDACEAE

Osmunda L.

regalisT. 2829 (Harrismith): (-CA ), Roux 1285.

SCHIZAEACEAE

Mohria Swartz

caffrorum (L.) Desv. 2828 (Bethlehem): (-BC), Roux 947:

(-DB), Roux 912. 2829 (Harrismith): (-AA), Roux

1501: {- AC), Roux 807: (-AD ), Roux 861.

hirsuta J.P. Roux 2828 (Bethlehem): (-DB), Roux 1214.

2829 (Harrismith): (-CA), Roux 1068.

GLEICHENIACEAE

Gleichenia J.E. Sm.

polypodioides (L.) J.E. Sm. 2829 (Harrismith): (-AA),

Roux 1505: (-CA), Roux 1073.

CYATHEACEAE

Cyathea J.E. Sm.

diegeiKunze 2829 (Harrismith): {-AD), Roux 862: (-CA),

Roux 1287.

HYMENOPHYLLACEAE

Trichomanes L.

pyxidiferum var. melanotrichum (Schlechtd.) Schelpe 2829

(Harrismith): {- AD), Roux 886.

Hymenophyllum J.E. Sm.

tunbridgense (L.) J.E. Sm. 2829 (Harrismith): {-AC), Roux

986: {-AD), Roux 851.

MARSILEACEAE

Marsilea L.

burchellii (Kunze) A. Br. 2829 (Harrismith): (- AC), Roux

989.

macrocarpa Presl 2829 (Harrismith): (-AC), Jacobz 4719.

DENNSTAEDTIACEAE

Pteridium Gled. ex Scop.

aquilinum (L.) Kuhn subsp. aquilinum 2829 (Harrismith):

(- AC), Roux 1530: (-CA), Roux 1284.

ADIANTACEAE/PTERIDACEAE

Pityrogramma Link

calomelanos (Swartz) Link var. aureoflava (Hook.) Weath. ex

Bailey 2829 (Harrismith): (- AC), Roux 1280.

Bothalia 16,1 (1986)

85

Anogramma Link

leptophylla (L.) Link 2829 (Harrismith): (- AC), Roux 1074.

Adiantum L.

capillus-veneris L. 2828 (Bethlehem): (— BC), Roux 935 ;

(-DA), Roux 1029. 2829 (Harrismith): (-AC), Roux

1547 ; (-CA), Roux 837.

poiretii Wikstr. 2829 (Harrismith): (-AC), Roux 1543;

(-CA), Roux 1526.

Pteris L.

cretica L. 2828 (Bethlehem): (—DA), Roux 1027. 2829

(Harrismith): (-CA ), Roux 1527.

Cheilanthes Swartz

capensis (Thunb.) Swartz 2829 (Harrismith): (-AC), Roux

927.

eckloniana ( Kunze ) Mett. 2828 (Bethlehem): (-BC), Roux

940; (-BD), Roux 951 ; (-DA), Roux 1015 ; (-DB),

Roux 920. 2829 (Harrismith): (—AC), Van der Zeyde

s.n.

hirta Swartz 2828 (Bethlehem): (-BC), Roux 933; (-DA),

Roux 1016. 2829 (Harrismith): (-AC), Roux 795 ;

(-CA), Roux 1525.

multifida (Swartz) Swartz 2829 (Harrismith): (- AC), Roux

822.

quadripinnata ( Forssk .) Kuhn 2828 (Bethlehem): (-BC),

Roux 932; (-DA), Roux 1018. 2829 (Harrismith):

(-AC), Roux 812.

viridis (Forssk.) Swartz var. viridis 2829 (Harrismith): (- AC),

Van der Zeyde s.n.

viridis var. glauca (Sim) Schelpe & N.C. Anthony 2828 (Beth-

lehem): (-DA), Roux 1020; (-DB), Roux 1236. 2829

(Harrismith): (- AC), Roux 863.

Pellaea Link

calomelanos (Swartz) Link 2828 (Bethlehem): (-DA), Roux

1017. 2829 (Harrismith): (-AC), Roux 869; (-CA),

Van der Zeyde s.n.

POLYPODIACEAE

Polypodium L.

polypodioides (L.) Hitchc. subsp. ecklonii (Kunze) Schelpe

2829 (Harrismith): (-CA), Roux 878.

vulgar e Z. 2829 (Harrismith): (- AC), Roux 820; (-AD),

Roux 890.

Pleopeltis H.B.K. ex Willd.

macrocarpa (Bory ex Willd.) Kaulf. 2828 (Bethlehem):

(— BC), Roux 946 ; (-DA ), Roux 1025. 2829 (Harrismith):

(-AC ),Jacobz 4716.

schraderi (Mett.) Tardieu 2828 (Bethlehem): (-DA), Roux

1024. 2829 (Harrismith): (-AA), Roux 1506; (-AC),

Roux 821 ; (-DA), Roux 853.

ASPLENIACEAE

Asplenium L.

adiantum-nigrum L. 2828 (Bethlehem): (-BC), Roux 929;

(-DA), Roux 1028. 2829 (Harrismith): (-AC), Roux

797.

aethiopicum (Burm. f.) Becherer 2828 (Bethlehem): (-BD),

Roux 952; (-DB ), Roux 918; 2829 (Harrismith): (-AC),

Roux 1545; (- AD), Roux 854; (-CA), Roux 829.

monanthesZ. 2829 (Harrismith): (- AC), Roux 796; (-CA),

Roux 877.

platyneuron (L.) Oakes 2828 (Bethlehem): (-BC), Roux

930. 2829 (Harrismith): (—AC), Roux 785; (- AD), Roux

860.

splendens Kunze subsp. drakensbergense Braithwaite 2828

(Bethlehem): (-DB), Roux 1532. 2829 (Harrismith):

(- AC), .Roux 921.

stoloniferum Bory 2828 (Bethlehem): (-DA), Roux 1219.

2829 (Harrismith): (- AC), Roux 818; (-CA), Roux 833.

theciferum (H.B.K.) Mett. var. concinnum (Schrad.) Schelpe

2829 (Harrismith): (- AD), Roux 889; (-CA), Roux 830.

trichomanesZ. 2828 (Bethlehem): (-BC), Roux 934; (-D A),

Roux 1026; (— DB), Roux 1683. 2829 (Harrismith):

(-AC), Roux 785.

varians Wall, ex Hook. & Grev. subsp. fimbriatum (Kunze)

Schelpe 2828 (Bethlehem): (-DA), Roux 1218. 2829

(Harrismith): (- AD), Roux 888; (-CA ), Roux 1215.

Ceterach DC.

cordatum (Thunb.) Desv. 2828 (Bethlehem): (-BC), Roux

941. 2829 (Harrismith): (- AC), Roux 1554.

ATHYRIACEAE

Cy stopteris Berwfc.

fragilis (L.) Bemh. 2828 (Bethlehem): (-DA), Roux 1019;

(-DB), Roux 922. 2829 (Harrismith): (-AC), Roux

1544; (- AD), Roux 855.

Athyrium Roth

scandicinum (Willd.) Presl 2829 (Harrismith): (-AC), Roux

1531.

LOMARIOPSIDACEAE

Elaphoglossum Schott

acrostichoides (Hook & Grev.) Schelpe 2828 (Bethlehem):

(-B C), Roux 936. 2829 (Harrismith): (- AC), Roux 799;

(- AD), Roux 847; (-CA), Roux 838.

drakensbergense Schelpe 2828 (Bethlehem): (-DB), Roux

971. 2829 (Harrismith): (- AC), Roux 824.

spathulatum (Bory) T. Moore 2829 (Harrismith): (-AD),

Roux 1062; (-CA), Roux 882.

ASPIDIACEAE/DRYOPTERIDACEAE

Woodsia R. Br.

montevidensis (Spreng.) Hieron. var. burgessiana (Gerr. ex

Hook. & Bak.) Schelpe 2828 (Bethlehem): (-DB), Roux

956. 2829 (Harrismith): (-AA), Roux 1503; (-AD),

Roux 856.

Dryopteris Adans.

dracomontana Schelpe & N.C. Anthony 2828 (Bethlehem):

(-DB), Roux 904. 2829 (Harrismith): (- AC), Roux 1040.

pentheri (Krasser) C. Chr. 2828 (Bethlehem): (-DA), Roux

1217. 2829 (Harrismith): (- AC), Roux 1546; (-AD),

Roux 1063; (-CA), Roux 875.

Polystichum Roth

alticola Schelpe & N.C. Anthony 2829 (Harrismith): (-AC),

Roux 782; (- AD), Roux 892; (-CA), Roux 876.

dracomontanum Schelpe & N.C. Anthony 2829 (Harrismith):

(-AA), Roux 1502; (-AD), Roux 896.

pungens (Kaulf.) Presl 2829 (Harrismith): (-AC), Roux

987.

luctuosum (Kunze) T. Moore 2828 (Bethlehem): (-DB),

Roux 1228. 2829 (Harrismith): (-AC), Roux 987.

transvaalense N.C. Anthony 2828 (Bethlehem): (-BC),

Roux 937.

BLECHNACEAE

Blechnum Z.

giganteum (Kaulf.) Schlechtd. 2829 (Harrismith): (-CA),

Roux 883.

inflexum (Kunze) Kuhn 2829 (Harrismith): (-AC), Roux

793; (-AD), Roux 859; (-C A), Roux 831.

punctulatum Swartz 2828 (Bethlehem): (-B C), Roux 948.

2829 (Harrismith): (-AA), Roux 1504; (-AC), Roux

1548; (-CA), Roux 827.

Bothalia 16,1: 87-91 (1986)

Miscellaneous notes

VARIOUS AUTHORS

CHROMOSOME STUDIES ON AFRICAN PLANTS. 1.

The use of cytogenetics in plant taxonomy is im-

portant for the determination of phylogenetic re-

lationships. The aim of this series is to publish mis-

cellaneous information regarding chromosome stu-

dies on African plants. Chromosome numbers deter-

mined by meiotic studies are given as the haploid

number (n), in contrast to numbers determined by

mitotic studies which are given as the somatic chro-

mosome number (2n). The herbarium voucher

specimen number and locality of the collected ma-

terial (according to the Degree Reference System —

Edwards & Leistner 1971; Leistner & Morris 1976),

is included.

POACEAE

The grass genera mentioned in this paper, appear

in the same order as that described by Gibbs Russell

et at. (1984). The subdivision of the plants into tribes

is according to Dyer (1976). All voucher specimens

are housed in the National Herbarium (PRE).

Andropogoneae

Arthraxon lanceolatus (Roxb.) Hochst. var. lanceo-

latus : n = 18.

TRANSVAAL. — 2430 (Pilgrim’s Rest): Blyderivierspoort

Nature Reserve (-DB), Spies 1423.

Andropogon eucomus Nees: n = 20.

TRANSVAAL. — 2530 (Lydenburg): 13 km from Boshoek to

Buffelsvlei (-AC), Spies 1525.

Andropogon lacunosus J. G. Anders.: n = 9.

TRANSVAAL. — 2530 (Lydenburg): 18 km from Dullstroom

to Goede Hoop (-AC), Spies 1468, 1475.

Trachypogon spicatus (L. f.) Kuntze: n = 10.

TRANSVAAL. — 2530 (Lydenburg): 18 km from Dullstroom

to Goede Hoop (-AC), Spies 1437.

Heteropogon contortus (L.) Roem. & Schult.: n =

20, n = 23.

TRANSVAAL. — 2529 (Witbank): 67 km from Lydenburg to

Roossenekal (-BB), Spies 1626 (n = 20). 2530 (Lydenburg):

Goede Hoop (-AC), Spies 1578 (n = 23).

Paniceae

Alloteropsis semialata (R. Br.) Hitchc. sub£p. ecklo-

niana (Nees) Gibbs Russell: n = 9.

TRANSVAAL. — 2430 (Pilgrim’s Rest): Blyderivierspoort

Nature Reserve (-DB), Spies 1420.

Brachiaria nigropedata (Fical. & Hiern) Stapf: n =

9.

TRANSVAAL. — 2430 (Pilgrim's Rest): 23 km from Boshoek

to Diepgeset (-CD), Spies 1546.

Setaria nigrirostris (Nees) Dur. & Schinz: n = 18.

TRANSVAAL. — 2430 (Pilgrim’s Rest): Blyderivierspoort

Nature Reserve (-DB), Spies 1422.

Setaria pallide-fusca (Schumach.) Stapf & C.E.

Hubb.: n = 18.

TRANSVAAL. — 2530 (Lydenburg): 31 km from Lydenburg

to Roossenekal (-AB), Spies 1609.

Rhynchelytrum nerviglume (Franch.) Chiov.: n =

18.

TRANSVAAL. — 2530 (Lydenburg): 15 km from Dullstroom

to Goede Hoop (-AC), Spies 1454.

Tricholaena monachne (Trin.) Stapf & C.E. Hubb.

var. monachne : n = 18.

TRANSVAAL. — 2530 (Lydenburg): 15 km from Boshoek to

Diepgeset (-AC), Spies 1541.

Anthephora pubescens Nees: n = 18.

CAPE PROVINCE. — 2624 (Vryburg): Armoedsvlakte Ex-

perimental Farm (-DC), Spies 742, 743, 744 & 745.

Arundinelleae

Arundinella nepalensis Trin.: n = 20.

TRANSVAAL. — 2530 (Lydenburg): Goede Hoop (-AC),

Spies 1583.

Aristideae

Aristida canescens Henr. subsp. canescens : n = 22, n

= 55/2.

TRANSVAAL. — 2530 (Lydenburg): 6 km from Goede Hoop

to Roossenekal (-AA), Spies 1497 (n = 22); 24 km from Dull-

stroom to Goede Hoop (-AC), Spies 1484 (n = 22); Frischge-

waagd (-AC), Spies 1569 (n = 55/2).

Chlorideae

Microchloa caffra Nees: n = —50.

TRANSVAAL. — 2530 (Lydenburg): 36 km from Lydenburg

to Roossenekal (-AA), Spies 1603.

Harpochloa falx (L. f.) Kuntze: n = 20.

TRANSVAAL. — 2530 (Lydenburg): 18 km from Dullstroom

to Goede Hoop (-AC), Spies 1440.

Ctenium concinnum Nees: n = —50.

TRANSVAAL. — 2530 (Lydenburg): 6 km from Goede Hoop

to Lydenburg (-AA), Spies 1495.

Eragrostideae

Eragrostis barbinodis Hack.: n = 10.

TRANSVAAL. — 2430 (Pilgrim’s Rest): 23 km from Boshoek

to Diepgeset (-CD), Spies 1549.

Eragrostis capensis (Thunb.) Trin.: n = 20.

TRANSVAAL. — 2530 (Lydenburg): 15 km from Dullstroom

to Goede Hoop (-AC), Spies 1451.

Eragrostis curvula (Schrad.) Nees: n = 20.

TRANSVAAL. — 2530 (Lydenburg): 18 km from Dullstroom

to Goede Hoop (-AC), Spies 1442.

CAPE PROVINCE. — 3227 (Stutterheim): 5 km from Kei

Road to Macleantown (-DA), Spies 1684.

Eragrostis gummiflua Nees: n = 30.

TRANSVAAL. — 2530 (Lydenburg): 31 km from Lydenburg

to Roossenekal (-AB), Spies 1613.

Eragrostis micrantha Hack.: n = 20.

TRANSVAAL. — 2529 (Witbank): 67 km from Lydenburg to

Roossenekal (-BB), Spies 1627.

Eragrostis pseudosclerantha Chiov.: n = 20.

TRANSVAAL. — 2530 (Lydenburg): 13 km from Lydenburg

to Roossenekal (-AB), Spies 1512 ; 13 km from Boshoek to

Buffelshoek (-AC), Spies 1532.

Eragrostis sclerantha Nees subsp. sclerantha : n = 20.

TRANSVAAL. — 2530 (Lydenburg): 36 km from Lydenburg

to Roossenekal (-AA), Spies 1593.

Eragrostis superba Peyr. : n = 20.

TRANSVAAL. — 2530 (Lydenburg): 10 km from Boshoek to

Buffelsvlei (-AC), Spies 1520.

Eleusine indica (L.) Gaertn. subsp. africana (K.-

O’Byrne) S. M. Phillips: n = 18.

TRANSVAAL. — 2430 (Pilgrim’s Rest): 38 km from Lyden-

burg to Pilgrim’s Rest (-DC), Spies 1560.

Stiburus alopecuroides (Hack.) Stapf: n = 10.

TRANSVAAL. — 2530 (Lydenburg): Goede Hoop (-AC),

Spies 1580.

Poeae

Festuca caprina Nees var. caprina: n = 28.

TRANSVAAL. — 2530 (Lydenburg): 18 km from Dullstroom

to Goede Hoop (-AC), Spies 1441.

Festuca scabra Vahl: n = 35, n = 63/2.

TRANSVAAL. — 2530 (Lydenburg): 15 km from Dullstroom

to Goede Hoop (-AC), Spies 1460 (n = 35); Long Tom Pass

(-CA), Spies 1436 (n = 63/2).

Lolium multiflorum Lam.: n = 7.

TRANSVAAL. — 2530 (Lydenburg): 31 km from Lydenburg

to Roossenekal (-AB), Spies 1617.

Bromus unioloides H.B.K.: n = 14, n = 28.

TRANSVAAL. — 2430 (Pilgrim’s Rest): Blyderivierspoort

Nature Reserve (-DB), Spies 1425 (n = 14). 2530 (Lydenburg):

31 km from Lydenburg to Roossenekal (-AB), Spies 1616 (n =

28).

Agrostideae

Agrostis barbuligera Stapf var. barbuligera : n = 21.

TRANSVAAL. — 2530 (Lydenburg): 18 km from Dullstroom

to Goede Hoop (-AC), Spies 1469.

Agrostis eriantha Hack. var. eriantha: n = 21.

Bothalia 16,1 (1986)

TRANSVAAL. — 2530 (Lydenburg): Frischgewaagd (-AC),

Spies 1574.

Aveneae

Holcus lanatus L.: n = 14.

TRANSVAAL. — 2530 (Lydenburg): 5 km from Belfast to

Dullstroom (-CA), Spies 1561.

DISCUSSION

Normal meiosis was observed in all specimens ex-

cept Spies 1578, an aneuploid specimen of Heteropo-

gon contortus (2n = 46), in which a number of univa-

lents, anaphase I bridges and chromosome laggards

during anaphase I were observed.

As a result of this study, and with reference to

published chromosome numbers (Darlington & Wy-

lie 1955; Ornduff 1967-1969; Fedorov 1969; Moore,

R. J. 1970-1977; Moore, D. M. 1982; Goldblatt 1981

& 1984), it was inferred that in most of the investi-

gated species the following basic chromosome num-

bers occur: x = 7 in the Poeae, Agrostideae and

Aveneae, x = 9 in the Paniceae, x = 10 in the An-

dropogoneae, Arundinelleae, Chlorideae and Era-

grostideae, and x = 11 in the Aristideae. Excep-

tions were Arthraxon lanceolatus (2n = 36) and An-

dropogon lacunosus (2n = 18) in the Andropogo-

neae and Eleusine indica (2n = 36) in the

Eragrostideae.

REFERENCES

DARLINGTON, C. D. & WYLIE, A. P. 1955. Chromosome at-

las of flowering plants. London: Allen & Unwin.

DYER, R. A. 1976. The genera of southern African flowering

plants. Vol. 2. Pretoria: Department of Agricultural Techni-

cal Services.

EDWARDS, D. & LEISTNER, O. A. 1971. A degree system for

citing biological records in Southern Africa. Mitt. bot.

StSamml., Munch. 10: 501-509.

FEDOROV, A. A. 1969. Chromosome numbers of flowering

plants. Leningrad: Academy of Science, USSR.

GIBBS RUSSELL, G. E. et al. 1984. List of species of southern

African plants. Mem. bot. Surv. S. Afr. No. 48.

GOLDBLATT, P. 1981. Index to plant chromosome numbers

1975-1978. Monogr. Syst. Bot. 5: 1-553.

GOLDBLATT, P. 1984. Index to plant chromosome numbers

1979-1981. Monogr. Syst. Bot. 8: 1-427.

LEISTNER, O. A. & MORRIS, J. W. 1976. Southern African

place names. Ann. Cape Prov. Mus. 12: 1-565.

MOORE, D. M. 1982. Flora Europaea — checklist and chromo-

some index. Cambridge: University Press.

MOORE, R. J. 1970-1977. Index to plant chromosome numbers

for 1968-1974. Regnum Veg. 68; 77; 84; 91; 96.

MOORE, R. J. 1973. Index to plant chromosome numbers for

1967-1971. Utrecht: Oosthoek Uitgevers.

ORNDUFF, R. 1967-1969. Index to plant chromosome numbers

for 1965-1967. Regnum Veg. 50; 55; 59.

J. J. SPIES and H. DU PLESSIS

A BRIEF NOTE ON TCD AND DBN AND THE HERBARIA OF SONDER,

HOOKER AND HARVEY

Recently I had the opportunity to spend two days

at the herbarium of the School of Botany, Trinity

College, Dublin (TCD) and one day at the National

Botanic Gardens of Glasnevin in Dublin (DBN). As

the collections were so much richer than I had ex-

pected, I write this note, based on my very inade-

quate information, to alert other botanists to the im-

portance of the collections housed at TCD and

DBN.

Although there had been a small collection of

plants at TCD, the herbarium was really developed

by W. H. Harvey, from 1845 onwards. Harvey held

the post of curator from 1845, and in 1848 was ap-

pointed as Professor of Botany of the Royal Dublin

Society, and in 1856 Professor of Botany at Trinity

College. He had worked in the Cape for about four

years between 1835 and 1842, and started the Flora

Capensis in collaboration with Sonder. They pro-

Bothalia 16,1 (1986)

89

duced the first three volumes between 1860 and

1865. To this end Harvey received numerous collec-

tions from amateur and professional collectors in the

Cape Province and Natal, mainly between 1850 and

1870. He published many new species, the holotypes

of which are at TCD. In addition, the relevant sec-

tions of the herbarium were loaned to Kew during

the preparation of the second part of the Flora Ca-

pensis under Thiselton-Dyer (between 1896 and

1913) and so contain iso-, syn-, and some holotypes

of the very numerous new species published during

the preparation of that great work.

After Harvey’s death in 1866 the College did not

spend any more money on the herbarium, although

further gifts of material were received from all over

the world. Much of this material was not mounted or

incorporated, and was kept in boxes. Some of this

material was donated to the National Botanic Gar-

dens of Glasnevin. Unfortunately they are as short-

staffed as the University, and consequently much of

this material is still in the original parcels. It does

contain at least some material from Harvey’s herbar-

ium, and might contain historically important collec-

tions.

As a sample of what TCD contains, I surveyed

two subtribes of the Orchidaceae. The following col-

lectors were represented in the Disinae: Harvey

(about half of the collections, mostly around the Cape

Peninsula), Ecklon & Zeyher (material from the

Uitenhage area), Drege, Hutton, Fannin (some

watercolours, with vouchers collected by G. F. F.

(George F. Fannin), Gerrard & McKen, MacOwan,

Bowker, Barber (they often collected together,

signed as F. W. B. and J. H. B.), Hallack, Saunders,

Sanderson, Holland, Cooper, Plant, Krauss, Ver-

reaux (acquired from the Delessert herbarium in

1844), W. S. M. D’urban and Brownlee. This collec-

tion includes the types of 17 names, of which several

are holotypes.

In the Coryciinae (Orchidaceae) there are even

more types. In several other groups there are inter-

esting collections: a large set of Ericaceae from the

Cedarberg collected by Wallich and a set of Cypera-

ceae from around Graaff-Reinet by H. Bolus. I am

sure that more detailed research will expose much

more material of great interest.

DBN houses, amongst others, a collection by

McNab, that includes a set of all plants that flowered

at Kew in the first decade of the nineteenth century.

This is being catalogued by Dr Nelson, the botanist

at Glasnevin. This collection may provide types for

names published in the Hortus Kewensis.

Until 1870 there were three large herbaria of

South African plants: those of Sonder, Harvey and

Hooker. Sonder’s herbarium included a fairly com-

plete set of Ecklon & Zeyher collections, as well as

other collectors. The Cape plants in this herbarium

were sold to Stockholm and to Melbourne, and it is

not always clear where the types are. Hooker’s her-

barium is at Kew, and forms the core of the modern

collections there. Harvey’s herbarium is still pre-

served intact. As very little material has been added

to it since 1868, it constitutes a time-capsule of the

material available to botanists in the mid-nineteenth

century, and so can be invaluable in illuminating the

species concepts of authors such as Harvey and Son-

der.

H.P. LINDER

BASELINE DATA FOR THE VEGETATION OF TWO PROTECTED PLOTS AT THE MATIMBA

POWER STATION. ELLISRAS, NW TRANSVAAL

In December 1983 an approximately five hectare

study site was enclosed with security fencing for joint

research by the Botanical Research Institute and ES-

COM. The study site is situated in the north-western

corner of the Matimba Power Station terrain, about

15 km west of Ellisras. The site is representative of

Acocks’s (1975) Mixed Bushveld with a mean annual

rainfall of 487 mm.

In February 1984 two structurally representative

plots of the vegetation in the study site were selected

for long-term monitoring of the floristic composi-

tion. Plot no. 1-01/84 represents a closed woodland

formation (Edwards 1983) and plot no. 2-01/84 re-

presents an open shrubland formation (Edwards^

1983). Plot numbers were allocated by the Secretar-

iat of the National Working Group for Vegetation

Ecology. In each of the formation types one 10 m x

10 m plot was permanently demarcated by fencing

standards situated at each corner. Rectangularity

was obtained with optical squares. Each plot was

subdivided into a grid of 100 lmr subquadrats, num-

bered as indicated in Fig. 1. The presence of all plant

species identifiable during sampling, rooted within

each subquadrat, was recorded. Voucher specimens

were collected outside each plot for all species re-

corded. Vegetation structure was recorded accord-

ing to Edwards (1983). Species abundance was cal-

culated as percentage frequency in 100 subquadrats

using 10% class intervals. The data were loaded onto

the Burroughs B7900 computer of the Department

of Agriculture and Water Supply via a Sharp PC1500

portable computer using the PHYTOCAP program

(Westfall 1985), and ordered according to frequency

and occurrence using the PHYTOTAB program

package (Westfall et al. 1982). The results of the

floristic baseline data are given in Table 1 and struc-

ture is illustrated by means of layer diagrams accord-

ing to Ito (1979). Fig. 2.

90

Bothalia 16,1 (1986)

TABLE 1. — Baseline data for the two protected plots at the Matimba Power Station, Ellisras, in the form of a two-way matrix, showing

floristic composition and percentage frequency. Numbers in brackets are voucher specimens collected by Westfall (PRE)

Percentage frequency *

Floristic composition in protected plot no.

1-01/84 2-01/84

Euclea undulata Thunb. var. myrtina (Burch.) Hiem (1617) 8

Indigofera nebrowniana J.B. Gillet (1616) 7

Barleria mackenii Hook. (1631) 5

Cenchrus cUiaris L. (1629) 5

Talinum crispatulum Dinter ex V. Poelln. (1626, 1632) 5

Grewia flava DC. (1628) 4

Boscia foetida Schinz subsp. rehmanniana (Pestal) Toelken (1627) 3

Combretum apiculatum Sond. subsp. apiculatum (1621) 3

Commelina benghalensis L. (1640) 3

Grewia subspathulata N.E. Br. (1637) 3

Hibiscus micranthus L. f. (1641) 3

Schmidtia pappophoroides Steud. (1643) 3

Talinum amotii Hook. f. (1636) 3

Boscia albitrunca (Burch.) Gilg & Ben. (1647) 2

Cyphostemma puberulum (C.A Sm.) Wild & Drum. (1639) 2

Hoffmannseggia burchellii (DC.) Benth. ex Oliv. subsp. rubroviolacea (Bak. f.) Brummitt & J.H. Ross (1625) 2

Monechma divaricatum (Nees) C.B. CL (1638) 2

Omithogalum seineri (EngL & Krause) Oberm. (1634) 2

Spirostachys africana Sond. (1630) 2

Gisekia phamaceoides L. (1654) 9

Commelina undulata R. Br. (1660) 7

Acanthosicyos naudiniana (Sond.) C. Jeffrey (1652) 6

Aristida stipitata Hack, subsp. graciliflora (Pilg.) Meld. (1669) 6

Rhynchelytrum repens (Willd.) C.E. Hubb. (1653) 6

Geome rubella Burch. (1657) 5

Indigofera daleoides Benth. ex Harv. var. daleoides (1655) 5

Limeum fenestratum (Fenzl) Heimerl (1649) 5

Phyllanthus burchellii Muell. Arg. (1650) 5

Tephrosia purpurea (L.) Pers. subsp. leptostachya (DC.) Brummitt (1656, 1665) 5

Acacia tortUis (Forssk.) Hayne subsp. heteracantha (Burch.) Brenan (1668) 3

Cyperus margaritaceus Vahl (1662) 3

Felicia mossamedensis (Hiern) Mendon$a (1671) 3

Hibiscus meeusei Ex ell (1659) 3

Pogonarthria squarrosa (Roem. & Schult.) Pilg. (1667) 3

Tribulus terrestris L. (1651) 3

Cassia sp., cf. C. absus L. (1664) 2

Waltheria indica L. (1648) 2

Ziziphus mucronata Willd. subsp. mucronata (1673) 2

Acacia erioloba E. Mey. (1635) 3 2

Aristida congesta Roem. & Schult. subsp. congesta (1619) 3 8

Dichrostachys cinerea (L.) Wight & Arn. subsp. africana Brenan & Brummitt var. africana (1644) 3 4

Eragrostis lehmarmiana Nees var. chaunantha (Pilg.) De Winter (1618) 5 8

Grewia retinervis Burret (1633) 5 2

Panicum maximum Jacq. (1666) 8 5

Mariscus chersinus N.E. Br. (1646, 1663) 2 4

Ruellia patula Jacq. (1672) 6 3

Solanum panduriforme E. Mey. (1645, 1661) 2 6

Tragus berteronianus Schult (1623) 4 3

Urochloa brachyura (Hack.) Stapf (1624) 4 8

* blank = no occurrence; 1=1- 10%; 2=11- 20%; 3 = 21 - 30%; 4 = 31 -40%; 5 = 41 — 50%; 6=51- 60%;

7 = 61 - 70%; 8 = 71 - 80%; 9 = 81 - 90%; 0= 91 - 100%

ujOL

Bothalia 16,1 (1986)

91

1 1 1 1 1 1 1 1 1

10 20 30 40 50 60 70 80 90 100

1

i I I I I L

10 m

FIG. 1. — A diagrammatic representation of the layout of one of

the plots showing numbering of subquadrats.

Changes in floristic composition over time will be

shown by means of a temporal classification for each

plot. The baseline data show species at a given time.

When sampling over a period of time, the species

that are present, or have disappeared, or occur

throughout that period, can be effectively depicted

in the form of a two-way matrix. The repeatability of

quantitative, quadrat-derived frequency data to-

gether with the concise yet complete depiction of re-

sults in a two-way matrix makes this a particularly

reliable and sensitive method for determining vege-

tation change. The authors thank ESCOM for sup-

port, the staff of the National Herbarium, Pretoria,

for specimen identification and Mr C. W. Ries for

technical assistance.

REFERENCES

ACOCKS, J. P. H. 1975. Veld types of South Africa. Mem. bot.

Surv. S. Afr. No. 40.

EDWARDS, D. 1983. A broad-scale structural classification of

vegetation for practical purposes. Bothalia 14:705-712.

ITO, K. 1979. A tentative study of stratification diagrams. In A.

Miyawaki & S. Okuda, Vegetation und Landschaft Japans.

Yokohama: The Yokohama Phytosociological Society.

WESTFALL, R.H., DEDNAM, G., VAN ROOYEN, N. &

THERON, G.K. 1982. PHYTOTAB — A program pack-

age for Braun-Blanquet tables. Vegetatio 49: 35-37.

WESTFALL, R.H. 1985. PHYTOCAP. A field data capture pro-

gram for the PHYTOTAB program package. Bothalia

15:749-750.

M. D. PANAGOS, R. H. WESTFALL and J. C. SCHEEPERS

20

30

Percentage cover

(a)

I"*

20 30

Percentage cover

an

FIG. 2. — The structure of the protected plots at the Matimba Power Station, Ellisras, illustrated by means of layer diagrams

showing height classes and percentage cover: a, Plot no. 1-01/84; b, Plot no. 2-01/84.

92

Bothalia 16,1 (1986)

PLATE. 1. — Kniphofia splendida E. A. Bruce.

Bothalia 16, h 93-96 (1986)

OBITUARY

CYTHNA LINDENBERG LETTY (1895-1985)

With the death of Cythna Letty (Mrs. Forssman)

(Fig. 1) in Pretoria at the age of 90 on 3 May 1985,

the career of one of South Africa’s most distin-

guished and prolific botanical arists came to an end.

She was born in Standerton, Transvaal, on 1 January

1895, the first of five children of the marriage be-

tween Walter Edward Letty of Greenwich, England,

and Josina Christina Lindenberg of Worcester, Cape

Province, who had previously borne six children dur-

ing her earlier marriage to David Johannes de Vaal

Leibbrandt. Figs 2 & 3.

FIG. 1. — Cythna Lindenberg Letty (Mrs Forssman), c. 1945.

In 1899 the family moved to Estcourt in northern

Natal where Cythna attended her first of 13 schools

and, in 1904, back to Standerton. The delicate and

fleeting wild flowers of the veld had fascinated

Cythna from her early childhood and it was during

her second stay at Standerton that her mother, an

able and talented woman, gave Cythna and her

younger sister their first lessons in watercolour

painting. In 1910 they moved to Heidelberg, Trans-

vaal, where her mother completed a second volume

of wild flower paintings. Included are several species

which at that time were new to science.

In 1914 Cythna spent her last year of schooling at

the Girl’s High School, Pretoria, after which she

worked as a governess for a short time on the farm

Kameeldrift, north-east of Pretoria. One of the

paintings she made here, in 1915, is reproduced in

Fig. 4. She then trained as a nurse for a year (quali-

fying as a mid-wife) and as a clerk with the Railways

before moving to Cape Town from 1920 to 1924 to

assist her brother-in-law, Dr Bosenberg.

Returning to Pretoria in 1925 she received her first

appointment in which she could exercise her artistic

talent, at the Onderstepoort Veterinary Laborato-

ries. This was before the time of colour photography

and she was required to record, for future reference,

the appearance of post mortem specimens typical of

various animal diseases and abnormalities. ‘Interest-

ing, if gory, work’, she recalls. ‘Once, while I was

busy drawing an opened-up carcase of a sheep. Dr.

Gilles de Kock hurried in with a large bottle of disin-

fectant to counteract the stench, expecting to find

me in a dead faint.’

In 1927 she was transferred to the then Division of

Plant Industry under Dr. I. B. Pole Evans where, in

the National Herbarium, she started to contribute

her superb paintings of African plants which, for

many years were the mainstay of the journal Flow-

ering Plants of Africa. Some years later Dr. George

H. M. Lawrence, director of the then Rachel Mc-

Masters Miller Hunt Botanical Library at the Carne-

gie Institute, Pittsburgh, USA, commented: ‘It is

FIG. 2. — Cythna with her

mother and father, Johan-

nesburg 1886.

94

Bothalia 16,1 (1986)

FIG. 3. — Cythna standing at extreme right, front row, with her family, Estcourt 1900.

clear that she stands with the very few among the

world’s top botanical artists.’

She resigned in 1938 to marry Oscar William Alric

Forssman whose grandfather, Chevalier Oscar Wil-

FIG. 4. — An early study of a Transvaal orchid by

Cythna in 1915.

helm Alric Forssman had come from Sweden to

South Africa in 1844. One son, Bruce, was born in

1940. In 1945 she returned to her post in the Na-

tional Herbarium and continued working until her

final retirement in 1968. During her service she com-

pleted over 740 plates for Flowering Plants of Africa

as well as contributing paintings and black and white

illustrations to several other publications of the BRI:

Botanical Survey Memoir No. 26 (Trees and Shrubs

of the Kruger National Park), Weeds of South Africa

and the Institute’s journal Bothalia. In 1962 one of

her great ambitions was realised when her book Wild

Flowers of the Transvaal appeared, with 145 plates in

colour and the text written partly by herself but

largely by other staff members of the BRI. The

translation by her cousin Anna Rothman, Veld-

blomme van Transvaal, was the first major illus-

trated botanical work to appear in Afrikaans.

Bothalia 16,1 (1986)

95

FIG. 6. — Cythna Letty, 1978.

When South Africa changed to the decimal system

it was decided that the coinage would depict indigen-

ous birds, animals and plants. Cythna was asked to

submit several floral motifs and her designs for the

50c ( Strelitzia , Zantedeschia and Agapanthus, repre-

senting the orange, white and blue of the South Afri-

can flag), 20c (Protea cynaroides and P. repens ) and

10c (Aloe aculeata) were accepted (Fig. 5). Minting

began in December 1964 and the coins were first re-

leased in 1965. About the same time she also drew

the design of Gloriosa virescens for the Rhodesian

sixpence, which was minted in 1964.

After her retirement she undertook a revision of

Zantedeschia, a genus in which she had been inter-

ested for many years and for which she had as-

sembled the necessary background information. It

was published in Bothalia 11: 5-26 (1973), freely il-

lustrated with her own inimitable paintings. Com-

missions for her paintings, which were then in great

demand, kept her occupied but she found time to

prepare a number of drawings of indigenous trees

which were published in two small books, Trees of

South Africa in 1975 and More Trees of South Africa

in 1980.

A project which had been in the back of her mind

for some years materialised in her 86th year when 23

paintings of what she called ‘strange little flowers’

appeared, together with a selection of her poetry

which she had jotted down over the years, under the

title Children of the Hours (1981). Professor Ridley

Beeton of the University of South Africa, in an in-

troduction to the book, commented on her ‘desire to

use words to say things not beyond what she said in

her paintings but to explore new routes to percep-

tion.’ She once said in an interview with a reporter

that poetry had been an extension of her paintings, ‘I

could put only part of my love for flowers into my

drawings.’

Volume 30 (1945) of Flowering Plants of Africa

was dedicated to her ‘in grateful recognition of her

signal service to South African botany and the art of

botanical illustration’, but general acknowledgement

of her merit came to her only late in life. In 1966 she

visited Europe and the United States for the first

time when some of her paintings were included in an

international exhibition of botanical art held at the

Hunt Botanical Institute, Pittsburgh. She repeated

the visit in 1970 when she received the Grenfell Sil-

ver Medal from the Royal Horticultural Society for

an exhibit in London of her paintings of Transvaal

wild flowers. In 1974 the University of the Wit-

watersrand bestowed on her an honorary Ll.D. de-

gree in appreciation of ‘a lifetime devoted to superb

craftsmanship that has assisted our scientific advance

96

Bothalia 16,1 (1986)

and given joy to so many here and abroad’. In the

same year the Johannesburg newspaper The Star

named her ‘Woman of the Year’. She was made a

fellow of the American Cactus Society in 1978 and,

in 1981, the South African Association of Botanists

awarded her their Certificate of Merit for her contri-

butions towards the advancement of botany in South

Africa.

She collected about 500 herbarium specimens, in-

cluding some of the earliest to be recorded from the

Kruger National Park during a visit at the invitation

of the then Warden, Col. Stevenson Hamilton. Her

name is commemorated in the botanical names Aloe

lettyae Reynolds and Crassula lettyae Phillips and in

the Cythna Letty Nature Reserve in the mountains

near Barberton. Although physically frail in the last

years of her life her spirit was unquenchable and,

even in her last few months, she was planning for the

future, including exploratory enquiries regarding the

possibility of reproducing the two volumes of her

mother’s paintings. Perhaps her life is epitomised in

her verse:

I hitched my wagon to a daisy

Direction vague and destination hazy

But,

Could any star have guided

me

more exactly

to where I most dearly loved to be?

L.E. CODD

Bothalia 16,1: 97-109 (1986)

E. A. C. L. E. (Ted) Schelpe (1924-1985) — a biography

E. G. H. OLIVER*

Keywords: biography, collecting expeditions, plant collections, publications, Schelpe E.A.C.L.E.

ABSTRACT

Prof. E.A.C.L.E. Schelpe was born in Durban on 27 July 1924 and died in Cape Town on 12 October 1985. He

studied at the University of Natal and at Oxford, England. He was awarded an M.Sc. (S. Afr.) for a thesis on the

ecology of the Natal Drakensberg and a D. Phil. (Oxon.) for a thesis on the ecology of bryophytes. For a brief

period he was Curator of the Fielding Herbarium, Oxford. In 1953 he was appointed Lecturer in Botany at the

University of Cape Town, until in 1973 he was awarded a full professorship (ad hominem) and the title of Director

of the Bolus Herbarium. Here he established a school of taxonomy and promoted 22 theses. His main fields of

research were the taxonomy and phytogeography of Pteridophyta (especially African groups) and of Orchidaceae.

He has 112 publications to his credit and collected over 7 000 numbers in various regions of Africa, in Europe and

the Himalayas. He was a keen gardener and was active in several societies promoting horticulture, orchidology and

nature conservation. He was a member of several scientific committees and was repeatedly honoured for his work.

Three children were born from his marriage to Sybella Gray, also a botanist.

CONTENTS

Parents, early youth and school days 97

University of Natal and first employment.... 98

Oxford 99

Cape Town: academic and family life 99

Plantsman and teacher 101

Health and last years 103

Grants, honours, committees 103

Fields of research and publications 103

Collecting expeditions and collections 104

List of publications 106

Conference proceedings 108

Theses of post-graduate students 108

Plants named after Schelpe 108

Acknowledgements 108

Uittreksel 108

PARENTS, EARLY YOUTH AND SCHOOL DAYS

Shipwrecks and disasters at sea have been very

much part of the history of southern Africa. One

such event had a profound effect on the history and

development of botany, not only on this subconti-

nent but in Africa as a whole.

Edmund and Martha Schelpe, refugees from Bel-

gium during the First World War, were en route

from England to start a new life in Australia when

their ship caught fire near Durban. All the passen-

gers were landed in Durban to await further arrange-

ments for their journey. The couple liked the city

and its climate so much that they decided not to go

on to Australia but rather to settle in Durban. Their

only child was born there on 27th July 1924 and

christened Edmund Andre Charles Louis Eloi —

‘Ted’ as he later became known.

The Schelpe parents came from the Brugge area

of Flanders and had both Flemish and French as

home languages. He was a musician and soon ob-

tained a post as organist at the Roman Catholic Ca-

thedral in Durban and later opened his own school

of music in that city. Ted’s mother, through her

background and training in the traditional art of

lace-making, taught at the Durban Technical Col-

* Botanical Research Unit, Department of Agriculture and

Water Supply, P. O. Box 471, Stellenbosch 7600.

lege. In Durban with its subtropical climate she

found scope for her interest in plants, awakened in

her by her own father’s enthusiasm for gardening. It

was in this environment that young Ted grew up, at-

tending firstly Marist Brothers’s College and then

Durban Boys High where he matriculated at the

early age of 17.

Ted always remembered the excitement of seeing

and collecting his first wild orchid plant at the top of

Bain’s Kloof Pass while on his way to Cape Town by

bus with his father in 1936 when he was only 12 years

old. In the same year he wrote a school essay on the

perennial theme of ‘What do you wish to be when

you grow up?’. In that essay he made it quite clear

that he was going to be a ‘Professor of Botany’.

On visits to the Cape Ted spent much of his time

in the municipal botanical gardens where he met the

horticultural staff and watched the repotting of

greenhouse plants, especially orchids. He also met a

Mr Duncan of Jutas, the publishing firm, who was a

keen grower of orchids. He spent many hours with

this enthusiast at his home chatting about orchids

‘over ginger beer and biscuits’. This contact was a

significant one because Duncan noted Ted’s remark-

able memory for plants and later wrote to his father

saying that the boy should be given every encourage-

ment to take up botany as a profession.

Prof. Michael Webb of Stellenbosch and Ted

Schelpe were contemporaries in their early school

days in Durban, both attending Marist Brothers.

They lived close to each other and often found them-

selves walking to school together. One interest they

shared was stamp collecting and this brought Mi-

chael to the Schelpe home. He remembers the house

in Currie Road standing in large, impressive

grounds. A fine jacaranda tree in the front garden

had numerous exotic orchids attached to it and of

these Ted was very proud. Beyond it stood a superb

brick and glass conservatory with a spray irrigation

system which his parents had built for his orchid col-

lection — and this all while he was still in his early

teens. During this period Michael Webb remembers

Ted as being a very self-assured and friendly boy,

interested mainly in his hobbies: plants and, to a

lesser extent, stamps.

98

UNIVERSITY OF NATAL AND FIRST EMPLOYMENT

No wonder then that Ted Schelpe went to Pieter-

maritzburg to enrol at the then Natal University Col-

lege for a degree in botany. He arrived there in 1941,

just after Prof. Adolf Bayer had taken over the de-

partment from Prof. John Bews who had become

Vice-Principal. In 1943 Ted obtained his B.Sc. de-

gree with distinction in botany, the other major be-

ing chemistry. Fig. 1.

FIG. 1. — B.Sc. graduation, Pietermaritzburg, 1943.

Michael Webb met up with Ted again at univer-

sity, arriving a year later, so that Ted demonstrated

to him in his first year. As a demonstrator he was

very helpful but meticulous about details and neat-

ness of anatomical and morphological drawings. Mi-

chael also remembers well Ted's pet hate at univer-

sity — Prof. Bayer’s habit of referring to him as

‘EAGLE’ Schelpe! Michael Webb and several other

students remember with great pleasure the excur-

sions which Ted organized for botany and zoology

students. (See also the paragraph Collecting expedi-

tions and collections below). In December 1943 he

took them on a two-week excursion down to Port St

Johns, and in December 1944 he organized a major

expedition to his favourite stamping ground in the

Drakensberg, the Cathedral Peak area. It was ob-

vious to the other students that Ted knew the area

and its plants extremely well.

Prof. Olive Hilliard remembers that Ted often

rounded up the students on a Saturday, or even a

Sunday, and took them out to Town Bush Valley or

Bothalia 16,1 (1986)

Chase Valley on the municipal bus to teach them

something about the local plants.

Much of the material he collected in Natal was de-

posited in the Natal University Herbarium which he

was paid to look after during his student days. Con-

sequently numerous labels and species covers are

written in his hand. Many of his specimens in the

herbarium are labelled ‘cultivated in Durban’, which

gives an indication of the large and varied living col-

lection he must have had at home. Ted Schelpe was

accepted by the other students as a leader and they

regarded him as an authority on a wide range of sub-

jects. Even in those early student days his know-

ledge of plants and their names appears to have been

encyclopaedic.

All of those involved with Ted in his student days

have vivid memories of his favourite pastime: yodel-

ling. In the Drakensberg on excursions or in the de-

partment while working, he would break into yodel-

ling, which he apparently performed very well. At

sports functions in particular, there would be stamp-

ing and clapping and cries of ‘Schelpe, Schelpe’. This

was the signal for him to leap up and yodel. At one

of the swimming galas held at the baths at Alexandra

Park he chose to stand on the end of the high diving

board, the better to be heard. Needless to say, some-

one crept up behind him and he did an involuntary

dive amid mighty cheers and laughter.

During 1944, towards the end of the Second

World War, Ted enrolled with the army and was

posted to the Aviation Medicine Research Section of

the South African Medical Corps in Johannesburg as

a laboratory technician. Here he found himself

doing numerous uninspiring blood counts. So when

volunteers were asked to ‘feed’ the experimental

bedbugs, Ted was prepared to do anything for a

change. The trouble came when all his bedbugs died

of overdoses of his blood, so back it was to the

bloodcounts. His apparent immunity to bedbugs

stood him in good stead during the expedition to the

Himalayas: when use had to be made of local accom-

modation he was the only member of the party who

slept in peace.

After demobilization at the end of 1945, he went

back to university to complete his thesis for the

M.Sc. degree. With the strong ecological bias at

Pietermaritzburg, and his great interest in the Dra-

kensberg, it is not surprising that he followed an eco-

logical line of research. His dissertation was entitled

The plant ecology of the Cathedral Peak area of the

Natal Drakensberg. The study began in July 1942

while he was still an undergraduate and continued

during several subsequent visits to the area in 1943

and 1944. He chose this area because it was, at that

time, biotically one of the least disturbed regions of

the range. The greater part of the thesis was written

while he was serving in the Medical Corps. He ob-

tained his degree, which was conferred by the Uni-

versity of South Africa (Natal was only a College at

the time), in 1946. The thesis remains unpublished.

For the first part of 1947 Ted worked at the Royal

Natal National Park as horticulturist preparing the

place for the visit of King George VI later that year.

He was responsible for laying out the gardens

Bothalia 16,1 (1986)

99

around the hotel which was owned by the Zunckel

family. While there, he built up a fine collection of

the fauna and flora of the park and prepared an ex-

cellent exhibit. This was so admired by Queen Eliza-

beth that she asked to see the young man, who un-

prepared for the occasion, was ushered into her

presence in boots and khaki workclothes. Ted was

most impressed with the way in which she put him at

ease.

OXFORD

Late in 1947 he set sail for England. He stopped

over in Cape Town and made his first ascent of

Table Mountain in the company of Jan Graaff with

whom he later teamed up in an expedition to the Hi-

malayas. In Michaelmas term, Ted entered Wadham

College to work in the Botany School for the D.Phil.

degree under the supervision of the Sherardian Pro-

fessor, the late T. G. B. Osborn. With his ecological

training at Natal University College as background

he chose to work on the ecology of lower plants. In

Trinity Term (June) 1951 he successfully submitted a

thesis entitled The ecology of Bryophytes on arable

land in the Oxford District. It is surprising that he

chose bryophytes as he had not previously shown

any special interest in that group. This gave rise to

only one short publication, on the techniques for the

experimental culture of bryophytes, but this thesis

was highly regarded by the late E. F. Warburg, the

taxonomist in the Botany School and one of the

leading bryologists of his day. A copy of his thesis is

still on the open shelves in the Botany School library

and shows signs of having been handled frequently.

Ted Schelpe’s urge to organize collecting expedi-

tions had not been left behind in Natal, and in 1947

he began planning for a university expedition to

Africa (see also Collecting expeditions and collec-

tions below). He had decided to have a Cambridge

botanist in the team and eventually Frank White was

provisionally chosen. He was summoned to meet

Ted at the Royal Geographical Society’^ headquar-

ters in Kensington Gore, and so began a life-long

friendship. They had received a grant from the Uni-

versity Exploration Society, but on return to Britain

found that the expedition was very much in the red.

To make up the shortfall they gave numerous lec-

tures, showing the films that had been taken. They

also had interviews with the BBC in the very early

days of television. Thus they were able to meet their

debts and turn the trip into a financial success.

After completing his D.Phil., Ted held a tempo-

rary post for a short period as curator of the Fielding

Herbarium at Oxford. During his tenure he worked

on the ancient herbaria and some recent collections

of South American plants. In 1952 he became in-

volved in another collecting expedition, this time a

small private one to the Himalayas.

From his expeditions, Ted brought back living

plants, mainly orchids, for cultivation in the Oxford

Botanic Garden. Some of them still survive. Among

them was the attractive epiphytic orchid, Aerangis

rhodosticta (from Ethiopia). This was exhibited at

the Royal Horticultural Society on 17 November

1953 and subsequently became widely grown. The

terrestrial orchid, Eulophia welwitschii, which he

collected at the Ngong Hills in Kenya, received a

unanimous Award of Merit at the R.H.S. on 22 May

1951.

Ted came to Oxford with considerable horticultu-

ral skills. His techniques for growing orchids were

adopted at the Botanic Garden with conspicuous

success. It is said that during one year the orchids he

had coaxed into bloom could be seen adorning vari-

ous young ladies at the Commemoration Balls. The

story is possibly apocryphal, but it caused him great

amusement.

Ted gave tutorials to undergraduates in his rooms

at Banbury Road. Many are the times that a fellow

lodger heard him droning on about ‘a drupe being a

true fruit . . .’ Like most Oxford students he en-

joyed the camaraderie of groups at the local pubs,

particularly the Abingdon Arms, where he is known

to have done the flower arrangements for the hostess

on a number of occasions. He developed a certain

sartorial elegance and always referred to his NBPS

— navy blue pinstripe suit! This was in strong con-

trast to his later years when the safari suit (Fig. 2) or

sports jacket and baggy grey flannels were very

much in evidence. However, one bit of Oxford garb

which stayed with him was his academic gown which

he referred to as his ‘Basuto blanket’ at graduation

ceremonies.

Ted is remembered in Oxford with affection as a

‘character’ with a refreshingly original approach to

life, a robust sense of humour and infectious laugh-

ter. He often took the lead in organizing get-to-

gethers and parties and his hospitality was prover-

bial. At his rooms in the evenings on most nights of

the week one could meet interesting people from all

walks of academic and non-academic life. There

were nurses, dons, Chinese scholars, archaeologists

and even some of the 1951/52 Springbok Rugby

Team, Stephen Fry, Ben Myburg and ‘Chum’

Ochse. Ted would get out his ukelele and they raised

the roof with Sarie Marais and other songs, ducking

out to the Pheasant nearby for jugs of ale to keep the

voices lubricated. When he returned to South Africa

his closest friends thought Oxford might never be

the same again.

CAPE TOWN: ACADEMIC AND FAMILY LIFE

When he joined the Botany Department at the

University of Cape Town in February 1953, Prof.

William Edwyn Isaac had just taken over the De-

partment from Prof. Robert Adamson. Ted took

over from Audrey Rose-Innes lecturing in general

botany, including taxonomy. In 1954 he was pro-

moted to senior lecturer with the sole responsibility

of Plant Taxonomy. When Dr Louisa Bolus retired

at the end of 1955, after 45 years as Honorary Cura-

tor of the Bolus Herbarium, Ted was appointed to

the first post of Curator of the herbarium in 1956. In

1968 he was promoted to Associate Professor and in

1973 he was awarded a full professorship (ad homi-

nem) and the title of Director of the Bolus Herbar-

ium.

Prof. Schelpe expanded and enriched South Afri-

can botany by his establishment of a strong plant tax-

onomy teaching and research school centred in the

100

Bothalia 16,1 (1986)