CONTENTS

Vol. 10, Part 1

Page

Frontispiece: Dr. E. P. Phillips.

1. Edwin Percy Phillips, M.A., D.Sc. (1884-1967). M.D. GUNN...............00seeeeeee 1

Damihe SouthvAtncicants pectesiote Vy rica sep SIC LICKer Ensen 5

3. The Genus Talinum (Portulacaceae) in Southern Africa. H.R. TOLKEN................. 19

4. A New Species of Raphia from Northern Zululand and Southern Mozambique. A. A.

OBERMEYER’ And IRS Gs IS TREY acces cpsceistcay sos scree cr euca chavo ever asks Io role MIAO alate ne iote oisn ohne te relays 29

5. The Disc in the Southern African Species of Maerua. D.J. B. KILLICK................. 39

6. A Note on Erythrophleum R.Br. in South Africa. J. H. ROSS......... 0.200000. eee neces 43

7. Stapelieae from South Tropical Africa, V. L.C. LEACH............... 00.00 ceeceeees 45

8. New and Interesting Records of African Plants—

Mizoacede: Res GySTREY: syesicuasisisieleysiovs io 5 Ore os svate vo lewee reie eis verale Fores See relen oookeao anaes 55

Amaryllidaceae:, T.:\©5 VERDOORN iin tata eisterepicre-s oc snersrs 18 ayers aexonsiave avoir averse Seis 56

Capparacedeu Lek hOLKEN SD» spe ake 1 Okan ener enrennerninnieniieieiiiie 59

Compositaci Dy By baLricksand' Cs Gaull CLAASSENEE EEE ennai 68

Grassulaceae: DAU RETO Kersey areveseen coe reteei cra to erator ee a vayieitenelateieee oil oneonoonelc te tataieliosisielore 70

(Camron, \W/G MMAR. ooo0c0ccc000000000000 SOE EEG Ra Dees De ROO Ea amare On 70

Graminedes BADE WINDER a GayANDERSONGEEE EEE eee ee ena 72

Wiliacede:. -As, A: (OBERMEYER sais cershersie een oe loners orec oie ROE RAO leita actor ere 74

Malvaceae, (OA IEEISTNER’: 0: cisioveravaxcteieloneyaics oe elticaanc ie enecararcuen ree era aleel ater re ciate te oils 74

Selaginaceaes D); J). BAKA hI GK 6 tyararcreave (acre ere te eires Sierevieieyevsievaes Sv oTer ey oooh ouereralaneneforahevons 74

Sterculiaceae.7 Ii GC VERDOORN:<auteateer cit ieee OG Ge cree iterate eee eee 77

Miliacede.; «le iCWERDOORNis(aicis &2h< recent ysiey alors ero le ie Fe eel aie cleo ers NISC AIOE ee ecco 81

9. New and Interesting Records of South African Fungi, VI. G. C. A. VAN DER WESTHUIZEN

andi:Ke Ts WAND WARMELO’ frcetortetcnere oraicret ole ieietiosslensvsite melas srermionre ortyaioe crater 83

10. An Ordination of the Vegetation of Ntshongweni, Natal. J. W. Morris................ 89

11. An Investigation of the Plant Ecology of the Hawaan Forest, Natal, using an Ordination

Techniques “Hadi MOL roe secre ae ciericiatee eres ce RRS cher el CE OTe ainiceinerere 121

1 Lolo) <a: -\14 (-\) eae RPI OMA ClO D BROOD occ Gro Lea Ole UMC on noi CLG Oana Once GRT Cd oeainG 129

Pe ae LIA

Vol. 10, Part 2

Edited

L. E. Codd, D.Sc.

Botanical Research Institute

Navorsingsinstituut vir Plantkunde

Department of Agricultural Technical Services

Departement van Landbou-Tegniese Dienste

South Africa/Suid-Afrika

1971

12.

13.

14,

CONTENDS

Vol. 10, Part 2

Frontispiece: Dr. I. B. Pole Evans.

Illtyd Buller Pole Evans, M.A. (Cantab.), D.Sc. (Wales), LI.D. (Rand) (1879—1968).

M: (Ds UGUNN, cs. hess and oe, jiertoen nice baa ae meee ED ie tena saa ate

Cultural Characters and Carpophore Construction of some Poroid Hymenomycetes.

G. C. A. VAN DER WESTHUIZEN on. cee eee sore Sear Peer Rate ae ees | Ss

Somatic Nuclear Division in Stemphylium botryosum. K. T. vAN WARMELO o.oo

Conidial Nucleation in Stemphylium botryosum. K. T. VAN WARMELO onc cc seu coe

Some Members of the Sphaeropsidales from South Africa. KAREL CEJP

A New Species of Gonatobotryum from South Africa. K. T. vAN WARMELO o.oo

The Acacia Species with Glandular Glutinous Pods in Southern Africa. J. H. Ross

Two New Ornithogalum Species from South West Africa. A. A. OBERMEYER... .....

A Note on the Acacia giraffae X A. haematoxylon Hybrid. J. H. Ross

New and Interesting Records of African Plants —

Asclepiadaceae. R. A. Dyer _..... Sip) cit hia axtoy ries a hae Be aches

Celastraceae. L. E. Copp wn. ue ae 2 ee

Gramineae. W. Gtess

Inliaceaes Dy SHARD Ya wercuiee mee inert Gee con cman Teen, cate egy pee a Sees

MMpastrevecn@, Io AL, Idan mel ID), dio 1, LSU oo amp ame om amm ou am

Zamiaceae. R. A. Dyer _...... Ie wadlesso SO | HOA yas eerie Metical cent,” cach etka sd

Notes on Brachystelma. R.A. DYER wa une nee Esc) Pace top ei Ia epee erin, nh etesaye oasnte a

A Further New Species of Cycad from the Transvaal. R. A. DYER oe oe

Aker Tepe we Sotin@an “Woven, Uo Ub URGES Gon ain Gam ame umm amp unm oe one

BookoReview: 7) cis. ton cence Gene aa anit ceeelitees Retro ne aenelhas Bry ates ee ea aise

Page

BOTHALIA

Vol. 10, Part 3

Edited

L. E. Codd, D.Sc.

Botanical Research Institute

Navorsingsinstituut vir Plantkunde

Department of Agricultural Technical Services

Departement van Landbou-Tegniese Dienste

South Africa/Suid-Afrika

1971

10.

CONTENTS

Vol. 10, Part 3

Frontispiece: Johan Graham Anderson

Johan Graham Anderson (1926-1970). B. DE WINTER ce cece ccc cone sssse tants snent canes nee

Cotton Staining caused by Crebrothecium ese (Eremothecium ashbyi) in South

PAE ri Ca Wisse Heat OS IMIAIRIASA'S» ycceey eae ome eee Geer pereey tee See ee eee eee ee

New and Interesting Records of South African Fungi, Part VII. W. F. O. Marasas

and G. C. A. vAN DER WESTHUIZEN

New Combinations in the Genera BIEL es Kluyveromyces, Lodderomyces and

Wingea. J. P. vAN pER WALT .....

Acacia brevispica and A. schweinfurthii. J. H. ROSS occ cece ce cee cee sees nets cette ne

A Variant of Acacia karroo from Sekukuniland. J. H. Ross

. New and Interesting Records of African Plants —

Asclepiadaceae. R. A. Dyer

Flacourtiaceae. E. THom

Thymelaeaceae. J. H. Ross ww...

Vitaceaes i Jiae)s (Mis AWAING DER) (VIER WIE; curse es re er ee

. Principal Components Analysis of Acacia burkei and A. nigrescens in Natal. J. H.

Ross and J. W. Morais .....

. The Rate of Forest Tree Growth and a Forest Ordination at Xumeni, Natal

E. J. Motu and D. B. Woops

The Flora of the Mariepskop Complex. H. P. vAN peR ScHisrrF and E. ScHOONRAAD

Book Review

BOTHALIA

Vol. 10, Part 4

Edited

L. E. Codd, D.Sc.

Botanical Research Institute

Navorsingsinstituut vir Plantkunde

Department of Agricultural Technical Services

Departement van Landbou-tegniese Dienste

South Africa/Suid-Afrika

1972

CONTENTS

—— § 3 ——

Wolk 10), Rare 4!

Page

Frontispiece: Marguerite Gertrud Anna Henrici

Marguerite Gertrud Anna Henrici. M. D. GUNN...... 0... 2.00 eee eee 503

The Genus Pithomyces in South Africa. W. F. O. MARASAS and INGRID H. SCHUMANN.. . 509

Studies of Wood-rotting Fungi. II. Basidiomycetes from the Wood-preservative Field

Exposure Test Plot at Kruisfontein. G. C. A. VAN DER WESTHUIZEN..........-.00-000 Sl 7/

A New Species of Encephalartos from Swaziland. R. A. DYER....................-0-- 539

Notes on Acacia Species in Southern Africa. II. J. H. Ross..................0.000: 547

Contribution to the Caryological Study of the African Grass Aristida rhiniochloa

Hochst., Based on Specimens from the Southern Hemisphere. PrERRE BOURRIEL,

ALAIN GESLoT, MONIQUE GORLIER and BERNARD DE WINTER...........-+0-++ 00000 555

New and Interesting Records of African Plants—

Gelastraceae! WEA COpD ess iccira sapere ots sus i salensiave see teye soneser are altutacctn rcawe eelelvenewe ou siteyone 565

Cucurbitacea es, Ts EU: ROSS ss 2 cet arcs dest svete ahepateay auee ney ces ev cy ues ceeded abo fo el raitexereventa av 568

Flacourtiaceae: (Ds, JAB! Ie @ Ke eresare oss suai ees ssee sede vel eneptcelie eee caval ein es heveecerneren 568

Miyntaceae Re. GeiSDREV cassis era ten wera io temee RH eMaIe ma Salen acue Tere e Gaerne SP aeeu ene 569

Sterculiaceaes Ta€ VERDOORNis cr cir cients ree oe use VAT ee Somes eee Uo ouae Ser oe Sore 571

Vegetative Multiplication of Strelitzia reginae and its Allies. R. A. DYER.............. SS)

An Apparatus for Facilitating the Manual Tabulation of Phytosociological Data.

P. J. MULLER, M. J. A. WERGER, B. J. CoerzeeE, D. E>wArpbs and N. G. JARMAN........ 579

Species—Area Relationship and Plot Size: With Some Examples from South African

Vevetation.- Migs Ac: WERGER ser sciciceihn a etecierore unter tere ae or ieee ne oem once spveetetene 583

. The Current Status of Mistbelt Mixed Podocarpus Forest in Natal. E. J. MOLL......... 595

A Phytosociological Study of the Cape Fynbos and Other Vegetation at Jonkershoek,

Stellenbosch. M. J. A. WERGER, F. J. KRUGER and H. C. TAYLOR.................. 599

A Preliminary Account of the Dune Communities at Pennington Park, Mtunzini, Natal.

| Sip Rul") Co} by Omen me Pe cgmnareer tone ications etcics c. cee een ear orcas Pipa Oto too ORO a Ich oo nen oe ner Aetna one 615

The Distribution, Abundance and Utilization of the Lala Palm, Hyphaene natalensis,

in Tongaland; Natal MEO USMORi esos. cetatiaee chy om aekecicte rote ar vareae nine wv are cavoerencners 627

Notes on the vegetation of the Cape Flats. H. C. TAYLOR................ esse eeeee 637

BookJReviewSix. Gos.cc ees orien WRC rit ere ta GiSTaT RCI TeRe es SiR tEL Goal eo avapenaue enauaraianeherouen ions 647

illips

Percy Ph

PLATE 1.—Edwin

Bothalia i0, 1: 1-3

Edwin Percy Phillips, M.A., D.Sc. (1884-1967)

M. D. Gunn

Edwin Percy Phillips, who was Chief of the Division of Botany and Plant

Pathology, Pretoria (the present Botanical Research Institute), from 1939 to 1944,

died in Cape Town on 12th April, 1967. The son of Ralph Edward Phillips, a Cape

Town businessman, and Edith Minnie Phillips, née Crowder, he was born on 18th

February, 1884, at York House, Hall Street, Sea Point. There were nine children

in the family, five sons and four daughters. His grandfather, Samuel Phillips, was

a well-known merchant and owned the fine property “‘ Bellwood House’’, with its

adjoining estate, below Lions Head, Sea Point, in the early 1870's.

After attending the South African College School he entered the South African

College (later to become the University of Cape Town), where he came under the

stimulating influence of the inspired teacher and botanist, Prof. H. H. W. Pearson.

In 1905 he obtained the B.A. degree with first class honours in botany and, in 1908,

his M.A.

In 1907 he was appointed as assistant in the herbarium of the South African

Museum, of which Prof. Pearson was then Honorary Curator, and applied himself

with vigour to studying and collecting the local flora. The result was that, in 1910,

he was granted ten months leave of absence by the Trustees of the Museum, to

proceed to the Herbarium of the Royal Botanic Gardens, Kew. Here he dealt with

the family Proteaceae, a group of which he had first hand field knowledge, in colla-

boration with the Kew botanists Otto Stapf and John Hutchinson for the “ Flora

Capensis ’’.

When Prof. Pearson resigned as honorary curator of the S.A. Museum herbarium

in 1911 to become custodian of the Bolus Herbarium at the South African College,

E. P. Phillips was appointed as assistant in charge of the Museum herbarium.

In 1912 he married Edith Isabel Dawson of Sea Point. The union was to prove

a happy one. Two daughters were born of the marriage and he was always a devoted

husband and father.

On his return to South Africa he worked for some years on collections sent in

by Mme. A. Dieterlen, wife of a French missionary stationed at Leribe in northern

Lesotho. In order to gain a better knowledge of the flora in situ he visited the

territory in 1913. His field studies, together with the taxonomic work entailed,

resulted in a paper entitled “A contribution to the flora of the Leribe Plateau and

environs: with a discussion on the relationships of the flora of Basutoland, the

Kalahari, and the south-eastern regions’’, for which he was awarded the degree

of D.Sc. by the University of the Cape of Good Hope in 1915.

His collecting localities were mainly in the western Cape where, as a keen

mountaineer and member of the Cape Mountain Club (of which he was for some time

Secretary), he actively explored and collected in the rich flora of the Cape mountains.

Dr. Phillips was a member of the Percy Staden Memorial Expedition, which explored

the Kamiesburg, Gifberg and Olifants River mountains during September 1911.

In May, 1918, Dr. Phillips joined the Government Service as Curator of the

National Herbarium of the Division of Botany, a section of the Department of

Agriculture, Pretoria. At that time the herbarium was a relatively small one, started

in 1903 by Joseph Burtt Davy and, on his retirement in 1913, incorporated with the

Plant Pathology Section into the Division of Botany under Dr. I. B. Pole Evans.

When Dr. Phillips took charge of the National Herbarium, a Botanical Survey

of the Union of South Africa was being conducted and this led to the rapid growth

of the Herbarium which was also supplemented by donations of important private

collections. Shortly after his appointment, three scientific publications were started

by the Division. The first of these was the series of Botanical Survey Memoirs,

which appeared initially in 1919 and of which he was author of Memoirs 9, 10 and

25. He was responsible for the major part of the text in the first 20 volumes of

the illustrated serial, The Flowering Plants of South Africa, issued in 1920. Bothalia,

a record of the contributions from the National Herbarium, Pretoria, appeared in

1921, and was the venue for many of his botanical papers.

During twenty-one years as botanist, and Chief from 1939-1944 of the Division

of Botany and Plant Pathology, he published some 200 scientific papers and several

major works, including ““ The Genera of South African Flowering Plants ’’, ““ South

African Grasses ’’, ‘““ The Weeds of South Africa’’ and after his retirement he was

re-employed to complete a second edition of the “‘ Genera’’. His last contribution

to South African botany was spontaneously and gratuitously given as a tribute to

a former friend and colleague, C. A. Smith, when he undertook the gigantic task

of editing ““Common Names of South African Plants ”’.

His activities on behalf of science were not entirely confined to botany: he

was Fellow of the Royal Society of South Africa and of the Linnean Society of

London, a Member of the South African Association for the Advancement of Science

and the South African Biological Society. He served on the Council of the South

African Association for the Advancement of Science for many years and was an

efficient secretary and recorder of numerous congresses; in 1930 he was President

of Section C and delivered an address entitled ‘““‘The development of botanical

science in South Africa’’.

In 1935 he was awarded the South Africa Medal and Grant by the Council

of the Association, becoming President in 1942; his address on this occasion was

““The advancement of science’’. He was an active member of the South African

Biological Society holding the position of Honorary Secretary for a record period

of 25 years, being elected President and awarded the Senior Captain Scott Medal

in 1925.

When the centenary meeting of the British Association for the Advancement

of Science took place in London in 1931, he was one of the delegates chosen to

represent South Africa. He visited the U.S.A. and Canada in 1934 under the Visitors’

Grant Committee of the Carnegie Corporation of New York. A Report on his visit,

‘** Herbaria and botanical institutions in the United States of America and Canada

in relation to similar institutions in South Africa’’, was published in the South

African Biological Pamphlet 8, 1935.

* Life and Living, A Story for Children’’. This small book published in 1933

was for the teaching of biology in schools.

Apart from his official and scientific interests, he devoted time to the welfare

of the Public Service and the Public Servants Association in particular. He became

a member of the Executive Committee of the Association in 1920 and served

continuously until 1944, when he was appointed Research Officer to the Public

Service Commission. He was Chairman of the Public Servants Association from

1935-36, and President from 1936-38, and Chairman of the Committee appointed

by the Public Service Commission to investigate a medical benefit scheme for public

servants. In local affairs he was a member and later Chairman of the Pretoria

Technical College for many years, and in 1940 was President of the Association

of Technical Colleges.

In 1946 he was appointed Scientific Liaison Officer for the Council for Scientific

and Industrial Research and stationed at Washington, D.C. Shortly before his

return to South Africa in 1948 he sustained the loss of his wife.

As recreation, when nearing his fifties, he joined a local Bowling Club, became

President and had the rare distinction of honorary life membership in recognition

of his services to the Club.

He married again in 1949, his second wife being Miss Susan Kriel of Pretoria

with whom he had worked for many years in the interests of the Public Service.

In 1950 he commemorated her name in the Composite genus Susanna. She

predeceased him in 1965. He spent his remaining days in the home of his younger

married daughter in Cape Town. His health rapidly deteriorated and he suffered

several slight strokes before his death.

All his long life he had enjoyed good health and had a great capacity for work

without fatigue. He had been associated with Dr. Marloth in his early days at the

Cape and also later when Marloth was a committee member of the Botanical Survey.

He wrote two years before his death: ‘‘ I owe much to Marloth. He was a tireless

worker and one day he said to me: Phillips, if you want to succeed in science, you

must work twenty-four hours a day’’. Marloth’s words must have stimulated him

throughout his working days. E. P. Phillips was conscientious, methodical and all

his dealings had the stamp of integrity.

- : 5 nt

“2

f ; :

i : ab) oS 7 o L

b en on ma

; i Fp ft

t : 4 : p 7 hai

: :

: J ;

’ '

{

—

4 é a:

; ;

7 7 (

1 a

‘

: i :

” i 7

j i t : acne

y J i zs : i ” as rh ‘) 1

Bothalia 10, 1: 5-17

The South African Species of Myrica

D. J. B. Killick

ABSTRACT

The South African species of Myrica are revised, the 19 species previously recognized being reduced

to 9. One variety is elevated to specific rank, viz. M. conifera Burm.f. var. integra A. Chey. becomes

M. integra (A. Chey.) Killick.

The only comprehensive revision of Myrica ever undertaken is that by Chevalier

in his Monogr. Myric. (1901). Since that time a mass of material of Myrica has

accumulated in various herbaria and identification using Chevalier’s species-criteria

has become increasingly difficult. Adamson (1950) writes: “‘A genus much in need

of revision. There is much uniformity in floral structure and the specific distinctions

are based almost wholly upon vegetative characters. Several species exhibit variation

in leaf characters in relation to age, habitat or as the result of fires or other forms of

interference. Hybridisation almost certainly occurs whenever species are associated ”’.

It is clear, therefore, that a modern revision has become both necessary and urgent.

According to Chevalier the South African Myricaceae belong to the section

Cerophora and the subsection Africanae. Chevalier described ten new varieties and

five new species from South Africa, viz. M. incisa, M. dregeana, M. myrtifolia, M.

glabrissima and M. elliptica, very often basing his new taxa on single specimens.

The next treatment of Myrica was by Hutchinson in Flora Capensis (1925). Hut-

chinson accepted Chevalier’s delimitation of species, except that he sunk M. natalensis

C.DC. under M. conifera Burm.f. Also, he omitted M. pilulifera Rendle which, at

that time, was considered to be entirely a tropical species.

In the present revision, the 17 species recognized by Chevalier plus the two species

subsequently described by Burtt Davy, M. mossii and M. rogersii, are reduced to

nine species.

In spite of Adamson’s remarks concerning the uniformity of floral structure in

Myrica, all the species were carefully dissected, but unfortunately no significant dif-

ferences between species were revealed. This meant that as in the past the key had to

be based almost entirely on vegetative characters.

Type and other specimens were kindly lent or photographs thereof given to the

author by the following institutions: BOL, CN, G, M, NBG, P and SAM. Material

at the BM, K and LINN was consulted when the author was stationed at Kew between

1954 and 1957.

MYRICA

Myrica L., Sv. Pl. ed. 1, 1024 (1753); Gen. Pl. ed. 5: 449 (1754); Benth. & Hook.f.,

Gen. Pl. 3, 1: 400 (1880); C.DC. in DC., Prodr. 16, 2: 147 (1864); A. Chev., Monogr.

Myric. 139 (1901); Hutch. in Fl. Trop. Afr. 6, 2: 307 (1917); FI. Cap. 5, 2: 561

(1925); Adamson in Fl. Cape Penins. 311 (1950); Phill., Gen. ed. 2: 245 (1951).

Type species: M. gale L.

Gale Duham., Trait. Arb. Arbust. 1: 253 (1755); A. Chev., l.c. 92 (1901).

Trees or shrubs, frequently aromatic. Leaves alternate, penninerved, entire,

serrate, dentate or pinnately lobed; stipules 0. Flowers usually dioecious in bracteate

spikes; male spikes axillary, solitary, usually densely flowered; female spikes axillary,

longer or shorter than the male spikes. Perianth 0. Male flowers subtended by a

solitary bract; stamens 2-many; filaments short, more or less cohering; anthers erect,

oblong, with 2 parallel longitudinally dehiscing cells, sometimes papillose. Female

flowers subtended by a solitary bract and with a whorl of hypogynous scales. Ovary

sessile, 1-celled; style short, with 2 spreading or ascending branches, filiform or some-

times flattened; ovule 1, erect from the base of the cell, orthotropous. Drupe small,

globose or ovoid, usually warted and covered with white wax; endocarp hard. Seed

erect; testa membranous; albumen 0; embryo straight with planoconvex fleshy

cotyledons and a short radicle.

A genus of about 56 species, distributed throughout the Northern Hemisphere,

6 in the Mascarenes, 11 in Tropical Africa (fide Hutchinson, /.c. 562) and 9 in South

Africa.

The name Myrica is derived from the Greek, muron, a scent.

Leaves narrowly attenuate at base:

Leaves 5-10 cm long:

Leaves reticulate and not usually gland-dotted..........................005- 1. M. integra

Leaves smooth (not reticulate), but gland-dotted.....................00.eeeees 2. M. serrata

Leaves 3-5 cm long:

Leaves) usually pinnately lobed (@ak-like)i)55.5...-.42sc00e+sssesceseee- 3. M. quercifolia

Leaves elliptic to obovate and margins repand-serrate.................... 4. M. diversifolia

Leaves broadly cuneate to round at base:

Leaves distinctly petiolate and blade 3-2-6:5 cm long.....................5. 5. M. pilulifera

Leaves shortly or scarcely petiolate and blade 0-6-4:5 cm long:

Leaves narrowly elliptic or cordate and less than 3 cm long:

Leaves narrowlyAelliptice sa. corn cee ee OR SE Oe ae ee 6. M. brevifolia

Teeavies nCondateas acccsae hae ae OR Oe eee ae 7. M. cordifolia

Leaves elliptic-ovate to broadly ovate and usually longer than 3 cm:

Male spikes robust and female spikes up to 8 cm long............... 8. M. kraussiana

Male spikes not robust and female spikes up to 4 cm long.................. 9. M. humilis

1. Myrica integra (A. Chey.) Killick, stat. nov. Type: Clanwilliam, Olifantsrivier,

Blackfontein, Schlechter 8026 (P, holo.!; PRE!).

M. conifera Burm.f. var. integra A. Chev., Monogr. Myric. 148 (1901).

M. linearis sensu A. Chev., l.c. 148 (1901); Hutch. in Fl. Cap. 5, 2: 568-569 (1925).

Tree or shrub, 2-3 m high. Bark grey or black. Ultimate branches glabrous to

pubescent. Leaves narrowly elliptic, 6-8 cm long, 0:8-1-5 cm wide, apex acute, base

narrowly cuneate, margin usually entire, sometimes remotely serrate, midrib prominent

below lateral veins fairly conspicuous, 20-25 at obtuse angle to midrib, surface finely

reticulate, eglandular, glabrous, coriaceous; petiole 0:5-1-2 cm long. Flowers dioe-

cious. Male spikes axillary, solitary, 0-8-1-3 cm long; rhachis not visible; bracts

solitary, broadly ovate to broadly trullate, 1-1-5 mm long, 1-5-2 mm wide, deeply

concave on inner surface, ridged, ciliate; stamens 4; anthers 1-2 mm long, papillose.

Female spikes 0-8-1:2 cm long; rhachis scarcely visible; bracts solitary, ovate to

broadly ovate-trullate, 1-2—1-5 mm long, 1-4-3 mm wide, ciliate; hypogynous scales 4,

roughly ovate, 0:4-0-8 mm diam., keeled, fleshly; style branches 2, filiform or flattened,

1-8 mm long. Fruits globose, +3 mm diam., warted, covered with white wax.

EIGaaleale

This species is restricted to an area between Clanwilliam and Stellenbosch in the

S.W. Cape, where it is found on riverbanks.

Cape.—Ceres: Mitchells Pass, Esterhuysen 15220; Wittelskloof, Esterhuysen 14735; 15220 (BOL).

Clanwilliam: rocky stream flowing into Olifants River, Esterhuysen 14960 (BOL); bank of Jan Diesels

River, Pillans 9883; Schlechter 916, 8026. Paarl: between Paarl and Lady Grey railway bridge, Drege

s.n.; Donkerkloof, Esterhuysen 15177 (BOL); near river Dutoitskloof, Marloth 604; Wemmershoek,

Smuts 1124. Stellenbosch: Diep Gat Ravine, Pillans 18276 (BOL). Worcester: Bains Kloof,

Compton 13881 (BOL, NBG); Lewis 1895 (SAM).

M. integra is very closely allied to M. serrata and can really only be distinguished

by the conspicuous reticulations on both surfaces of the leaf. Also, it is not usually

gland-dotted whereas M. serrata invariably is.

Up till now this species has been called M. /inearis, but it is impossible to establish

the identity of M. linearis with any certainty. C. de Candolle in his original description

of M. linearis in DC., Prodr. 16, 2: 154 (1864) states “‘Arbuscula patria ignota (in herb.

Candolle et Kew)’’. A search for the type(s) in both the herbaria mentioned failed to

reveal their presence in these institutions. At Kew there are three specimens in the type

cover of M. linearis, viz. Zeyher s.n., 76-6 and Marloth 604—the two former from

Clanwilliam and the last-named from Dutoitskloof. For obvious reasons these speci-

mens cannot be regarded as type material.

One way out of this predicament would have been to select a neo-type for M. /inearis.

This might have been justified if the description of M. /inearis exactly fitted the accepted

concept of the species. However, it does not: while the description emphasizes the

prominent secondary nerves of the leaves, which is a diagnostic feature of the species,

it falls short in describing the petioles as short and the branches as glabrous.

In view of the above, it was decided to reject the name M. /inearis and to adopt

the combination M. integra (A. Chev.) Killick (=M. conifera Burm.f. var. integra

A. Chev.

2. Myrica serrata Lam., Encycl. 2: 593 (1786); Killick in Bothalia 8, 2: 175

(1964). Type: Africa, Sonnerat s.n. (or 708?) (P, holo.; PRE, photo.).

M. aethiopica L., Mant. Alt. 298 (1771), nom. illegit., pro parte, quoad spec. in

Herb. Linn. tantum. Type: C.B.S. (LINN 1169-4). MM. banksifolia Wendland, Coll.

Pl. 1: 70, t.24 (1808). Type: t.24. M. natalensis C.DC. in DC., Prodr. 16, 2: 148

(1864); A. Chev., Monogr. Myric. 149-150 (1901). Type: Natal, Gueinzius (G, holo.!;

PRE, photo.). M. conifera auct. non Burm.f., Prodr. Cape. 31 (1768); A. Chev.,

Vics 144; Hutch. in Fl. Trop: Afr: 6, 2: 314 (1917); Fl. Cap. 5, 2: 571-572 (1925);

Adamson in Fl. Cape Penins. 313-314 (1950). —-var. banksifclia (Wendl.) A. Chev.,

yes 1A7, var. glabra A. Chev., l.c. 147-148. Syntypes: several, from Port Natal

and Cape, Drege s.n. (G), Drege no. f. (G), Verreaux s.n. (G); Zeyher 306, 3847 (P),

Boivin s.n. (P). var. tomentosa A. Chev., l.c. 147. Type: Africa merid., Wallich

s.n. (G, holo.!; PRE, photo.). M. mossii Burtt Davy, in J.S. Afr. Bot. 4: 123 (1938).

Syntypes: Krugersdorp, Witpoortjie Kloof, Moss 6653, 6828 (BM!; K!).

Tree or shrub, up to 6m high. Bark brown to black. Ultimate branches glabrous

to tomentulose. Leaves narrowly elliptic, 4-5-10 cm long, 0-6-1-8 mm wide, apex

acute, base long attenuate into petiole; petiole 1 cm long; margin entire or remotely

serrate with teeth sometimes recurved, midrib prominent especially below, 12-18

subimmersed veins, glabrous to pubescent, lower surface usually conspicuously gland-

dotted. Flowers monoecious or dioecious. Male spikes 0-8-3 cm long; rhachis not

or completely visible, glabrous or puberulous; bracts solitary, variable in shape, trullate

to very broadly ovate, concave on inner surface, ridged, membranous at edge, ciliate,

gland-dotted; stamens 4, sometimes branched at two levels; anthers 1-3 mm long,

papillose or epapillose. Female spikes 1-2-2-6 mm long; rhachis visible, tomentulose,

glandular; bracts ovate, 1-5 mm long, 1-8 mm wide, ciliate, gland-dotted; hypogynous

scales 3-5, ovate-triangular, 0-5 mm diam., fleshy, ciliate; style branches filiform to

slightly flattened, 2 mm long. Fruits globose, c.2 mm diam., warted, covered with

Wihiteswaxee: FIG) li; 25 PLATE, I.

aw,

NaN

So!

ALEIDA VAN DER MERWE

Fic. 1.—Leaves of: 1, Myrica integra; 2, M. serrata; 3, M. quercifolia; 4, M. diversifolia; 5, M.

pilulifera; 6, M. brevifolia; 7, M. cordifolia; 8, M. kraussiana; 9, M. humilis.

A streambank species recorded in all provinces of the Republic except the Orange

Free State (though probably occurring there) and extending into Lesotho, South West

Africa and tropical Africa.

Cape.—Albany: Beggars Bush, Archibald 5967; Howieson’s Poort, Britten 968; outskirts of

Grahamstown, Britten 2176; Signal Hill, Grahamstown, Ga/pin 2921; nature reserve, Grahamstown,

Martin 4639; Slaaikraal, Mauve & Wells 20. Alexandria: Zuurberg Range, Archibald 3931; Kanitra

River Valley, Archibald 5767. Caledon: Kleinmond, Isaac s.n.; Houwhoek, Schlechter s.n. Ceres:

Zandfontein, Compton 19447 (SAM). Clanwilliam: Brakfontyn, Zeyher sub SAM 20279 (SAM).

Humansdorp: Assegaaibosch, Esterhuysen 6723; Slang River, Phillips 3434. Kentani: along streams,

Pegler 883. King William’s Town: Hogsback mountains, Rattray 302 (BOL). Knysna: Harkerville

Plantation, Keet 468. Komga: along streams, Flanagan 339. Ladismith: Buffelskloof, Esterhuysen

18562; 18564 (BOL). Laingsburg: Witteberg Kloof, Compton 3010 (BOL). Lusikisiki: Ntabene,

Fraser sub Schonland 3747a; Egossa, Sim 2520 (BOL). Mossel Bay: Mossel River, Potts 1658 (SAM).

Mount Currie: Glengarry area, Killick 2229. Paarl: Dutoitskloof, Esterhuysen 11539; Donkerkloof,

Great Drakenstein, Esterhuysen 15177; base of Klein Drakenstein near Salem, Galpin 11040; French

Hoek Forest Reserve, Leighton s.n. Peninsula: Brightwater swamp, Compton s.n.; Buffelsbay, Compton

13070 (NBG); Clifton Marloth 3432. Port Elizabeth: Van Staadens, Drege sub Marloth 5594;

Winterhoek Mountains, Weimarck 939; SimQ.11. Riversdale: Muir 2819. Somerset East: Zuurberg,

Holland 153. Stellenbosch: Swartboskloof, Van der Merwe 24.24. Stockenstroom: Acocks 11108.

Stutterheim: Fort Cunynghame, Sim 2029. Swellendam: Zuurbraak, Schlechter 2128. Tulbagh:

near Tulbagh waterfall, Isaac s.n.; Ecklon 42. Uitenhage: Zwartkops River, Zeyher 128; 3876

(BOL). Umzimkulu: Clydesdale, Tyson 2556 (SAM). Uniondale: Joubertina, Esterhuysen 6919.

Victoria East: Hogsback, Rattray 303; 1279. Willowmore: Taylor 396 (NBG). Worcester: Hex

River, near De Doorns, Bolus 11948.

NATAL.—Bergville: Mahai River Valley, Galpin 9496; Mont-aux-Sources, Hutchinson, Verdoorn

& Forbes 144: Cathedral Peak Forest Research Station, Killick 1650. Eshowe: Umbhlatuzi Swamp,

Kotze 45. Estcourt: stony dolerite koppie, Acocks 10493. Hlabisa: St. Lucia Estuary, Landsdell

3722; Pole Evans 3644. Inanda: Wood 985 (SAM). Ingwavuma: Mangusi Forest, Boocock (F.D.

Herb. 5316). Mtunzini: Ngoye Forest, Ward 3468; Wells & Edwards 68. Nkandla: Nsuzi River

Valley, Codd 1423. Pietermaritzburg: Table Mountain, Killick 374. Port Shepstone: Shelly Bay,

Mogg 12742. Ubombo: Mbaswana Forest, Boocock (F.D. Herb. 5723); source of Sordwana River,

Michelmore 38; Lake Sibayi area, Tinley 203. Umzinto: Dumisa, Rudatis 399. Utrecht: Spieshoek,

Smuts 1340. Vryheid: Dumuka Mountain, Gerstner 45711; Nhlazatsche, Ward 3428.

TRANSVAAL.—Barberton: Louws Creek, Thorncroft 2046; 4353 (BOL). Bronkhorstspruit: “* Spits-

kop”, Killick 3460. Krugersdorp: near Hekpoort, Phillips 364; Witpoortjie Kloof, Killick 3459;

Mogg 21342; Gladysvale, Rodin 3913. Lydenburg: Wilms 5835; Galpin 12179: 22 miles S.E. of

Lydenburg on Nelspruit road, Marais 323. Pilgrim’s Rest: Bushbuckridge River, Smuts 96; 46 miles

from Acornhoek on Graskop road, Story 3989. Pretoria: Debbe’s Ravine, Mogg 15052a; Willows,

Repton 1695; Garsfontein, Story 1229. Rustenburg: “Ananda,” Rose Innes 208; 236; Sutton 888.

White River: 15 miles east of Skukuza on lower Sabi Road, Codd & De Winter 5050; Pretoriuskop,

Van der Schijff. Waterberg: Hangklip, Maguire 1423 (NBG).

LesoTHO.—Leribe: Dieterlen 7043 (SAM).

SoutH West ArricA.—Okavango: Popa Falls, near Andara, Maguire 1679. Caprivi Strip: Singa-

lamwe, Killick & Leistner 3234; 3235.

Until recently this species was known as M. conifera Burm.f. See Killick in

Bothalia 8: 175 (1964) for reasons why this name must be rejected in favour of M.

serrata Lam.

At first it was thought that M. mossii Burtt Davy, described from the Transvaal,

could possibly be treated as a variety of M. serrata, because of its entire leaves.

However, natural populations were discovered which showed all gradations from

entire to serrate margins (see Plate 1).

3. Myrica quercifolia L., Sp. Pl. ed. 1: 1025 (1753), Lam., Encycl. 2: 593 (1786);

Willd., Enum. Pl. Hort. Berol. 2: 1012 (1809); Thunb., Fl. Cap. ed Schult. 159 (1823);

Drege, Zwei Pfl. Documente 98, 106, 132 (1843); C.DC. in DC., Prodr. 16, 2: 148

(1864); A. Chev., Monogr. Myric. 161 (1901); Marloth, Fl. S. Afr. 1: t.23, fig. A,

1 & 2 (1913); Hutch. in Fl. Cap. 5, 2: 570-571 (1925); Adamson in Fl. Cape Penins.

313 (1950). Syntypes: ‘“‘Aethiopica”’, LINN 1169.6 (lecto.!); LINN 1169.5!

M. hirsuta Mill., Gard. Dict. ed. 8, No. 6 (1768). Type: apparently not in existence.

M. ilicifolia Burm.f., Fl. Ind. et Prodr. Fl. Cap. 31 (err. typ. 27) (1768). Type: Burmann

s.n. (G, holo.!; PRE, photo.). M. /aciniata Willd., Enum. Hort. Berol. 2; 1012 (1809).

Type: not seen. M. zeyheri C.DC. in DC., Prodr. 16, 2: 149 (1864). Type: mountain

ridges, near Caledon, Zeyher 3878 (K, holo.!; SAM!; PRE, photo.). M. incisa A.

Chev., I.c. 150 (1901). Type: Cape, Burmann (G, holo. F PRE, photo.). L. quercifolia

L. var. hirsuta (Mill.) A. Chev., l.c. 163 (1901). var. ilicifolia (Mill.) A. Chev.,

l.c. 165, Pi. 8, D9, 11, 12 (1901). ———var. /atifolia A. Chev., l|.c. 166 (1901). Syntypes:

Roxburgh s.n. (G), Drege, Myric. 3 (G!; PRE, photo.). ——var. microphylla A.

Chev., l.c. 163, fig. 20B (1901). Syntypes: prope Devils Peak, Cape Town, Wilms

3634 (P, K}), Boivin 536 (P). ———var. multiformis A. Chev., l.c. 163, fig. 20A, C (1901).

Syntypes: dunes pres du Cap Recief, a 500 pieds d’alt., Zeyher ISSa, Lehmann 1832,

Boivin s.n. (all P).

Le pe? we WT UAT eetiara ates ee i (HEE HEE ETAL HEL SE atten i

| 3) yh EES 4 i 6 i ; i i es jo ae ia Sate

is ; } fee ‘ Ve Pi 2 P hie S

j elie ladigs Hi eee

PLATE | .—Myrica serrata. The range in leaf variation (right to left: entire to serrate) from a population

of this species on the farm “* Spitzkop ’’, Bronkhorstspruit District, Transvaal (Killick 3460).

Low, spreading shrub, 15-60 cm high. Bark grey to black. Ultimate branches

glabrous to tomentulose. Leaves spathulate-obovate in outline, usually pinnatisect or

coarsely repand-dentate, 3-5 cm long, 0-5—2-:5 cm wide, apex acute to obtuse, base

long-attenuate, midrib fairly prominent, lateral veins immersed, coriaceous, lower

surface and sometimes upper surface conspicuously gland-dotted, usually glabrous.

Flowers dioecious. Male spikes axillary, solitary, 0:5—0-8 cm long; rhachis not visible;

bracts very broadly ovate, 1-3 mm long, 1-3-1-8 mm wide, membranous at margin,

ciliolate, gland-dotted; stamens 2-4; anthers 1 mm long, papillose. Female spikes

axillary, solitary, 0-5-I-5 mm long; rhachis scarcely to clearly visible, densely gland-

dotted; bracts very broadly ovate, 1-7 mm long, 1-3 mm wide, ciliolate, gland-dotted;

hypogynous scales 2 or 4, rhombate, fleshy, ciliolate; style branches filiform, 2-3 mm

long. Fruits globose, 3-4 mm diam., warted, covered with white wax. Fic. 1: 3.

An extremely variable species occurring between Malmesbury and the mouth of

the Kei River in the eastern Cape.

Cape.—Albany: 9 miles from Kaffirdrift on road to Grahamstown, Story 1285. Bredasdorp: valley

side at base of mountain, Gi/pin 11233; near Strandkloof, Maguire 63 (NBG); Zoetendalsvlei, in

dunes, Smith 3097; Brandontein, Smith 3122; Ratelsrivier, Van Breda 921. Caledon: near Dasbos,

Grobler 1044; Hermanus location, Grobler 1066; near Bath, Marloth 7668; Genadendal, Roser (15413).

Humansdorp: Slang River, Phillips 3338. Komga: near the Kei mouth at Redwalls, Flanagan 2582.

Malmesbury: Hopefield, Marloth 482; Mamre, Pillans 9253 (BOL). Paarl: Top of Franch Hoek

Pass, Leighton s.n. Peninsula: Kirstenbosch, Compton 8079; Esterhuysen 11774; Devils Peak, Ester-

huysen 7975; Claremont, Hutchinson 1; near Brightwater, Leighton 444; Cape Flats, Strey 655;

Wynberg, Zeyler 1553; Table Mountain Zeyler 1553. Port Elizabeth: Anthony 44; near Port

Elizabeth, Fries, Norlindh & Weimarck 317; Greenbushes, Long 690; Theesecombe, near Port

Elizabeth, Long 996; towards Witteklip, Rodin 1017; 6 miles west of Walmer, Story 2730; on

the downs by Port Elizabeth, Zeyher 749; Cape Recife, Zeyher 1553. Riversdale: Muir 2820.

Stellenbosch: Brackenfel, Hafstrom & Acocks 376. Helderburg, Parker 4102; 4103(NBG). Uitenhage:

Thornhill, Compton 23409 (NBG); Zeyher 749 (SAM).

4. Myrica diversifolia Adamson in J. S. Afr. Bot. 10: 128 (1944). Type: Cape

Peninsula, Modderdam, Adamson 3368, holo. (no material traced); Smitswinkei,

Isaac s.n. in BOL 22583 (BOL, lecto.!; PRE, photo.); paratypes several, including:

Klaasjagersberg, Adamson 3325, 3345; Contour Path, Kirstenbosch, Levyns 1085 etc.

(all CT!.)

Erect shrub, up to | cm high. Bark grey to black. U/dtimate branches glabrous

to tomentulose. Leaves elliptic or obovate, 3-5 cm long, 1-5-2 cm wide, apex acute to

somewhat round, base cuneate, margin repand-serrate in upper half with serrations

often recurved, glabrous or pubescent, gland-dotted, midrib and lateral veins (8) fairly

prominent on lower surface, reticulate. Flowers dioecious. Male spikes axillary,

solitary, 1-2 cm long, rhachis scarcely to not visible, gland-dotted, puberulous: bracts

solitary, broadly obtrullate-obovate 1-8 mm long, 1-9 mm wide, ciliolate, gland-dotted;

stamens 2; anthers 1-5 mm long, papillose. Female spikes axillary, solitary, 2 cm

long; rhachis visible, puberulous, gland-dotted; bracts broadly ovate, 1-8 mm long,

1-6 mm wide, ciliolate; hypogynous scales 3, c. ovate, 0-5 mm diam., fleshy, ciliolate;

style branches 1-2 mm long, flattened, thick. Fruits globose, 3 mm diam., warted,

covered with white wax. Fic. 1: 4.

Cape.—Peninsula: Klaasjagersberg, Adamson 3325; 3345 (C.T.); Smitswinkel Flats, Adamson

3321 (C.T.); Kirstenbosch, Compton 14663; Table Mountain, Compton s.n.; Ascension Buttress,

Esterhuysen 11616 (NBG); Silverstream Buttress, Esterhuysen 7652; Grootkop, N.E. slopes, Ester-

huysen 11405 (BOL); Nursery Buttress Esterhuysen 11861 (BOL); Isaac s.n. (BOL. 22583); Smitswinkel,

Salter 8758 (BOL); Killick 3828; 3829: 3830; 3831; Modderdam, Killick 3832; Contour Path,

Kirstenbosch, Levyns 1085 (C.T.); Kalk Bay, Levyns 1086; 1087; 1088 (C.T.); Silvermine Valley,

Levyns 1094 (C.T.); Salter 8756 (BOL).

Restricted entirely to the Cape Peninsula. It is possible that this species is a

hybrid between M. kraussiana and M. quercifolia. In fact, Adamson, the author of

M. diversifolia, suggested this to the present author in a personal communication (8th

January, 1963). However, in Fl. Cape Peninsula, p. 313, Adamson had previously

written that “‘ probable hybrids with M. humilis, M. quercifolia and M. zeyheri have

been observed ’’. Considerable field work is necessary to elucidate this problem finally.

5. Myrica pilulifera Rendle in Trans. Linn. Soc. ser. 2, 4: 43 (1894); A. Chevy.,

Monogr. Myric. 142 (1901); Engl. Bot. Jahrb. 45: 279; figs. G-H (1911); Hutch. in

Fl. Trop. Afr. 6, 2: 311-312 (1917). Type: Malawi, Mount Milanji, Whyte s.n. (BM,

holo.!; K!, PRE, photo of iso.).

M. pilulifera Rendle var. puberula Rendle in J. Bot. 41: 86 (1903). Type: Malawi,

Buchanan 939 (BM, holo.!; K!, PRE, photo. of iso.). MM. rogersii Burtt Davy, Fl

Transyv. 2: 433 (1932). Type: Lydenburg, Sabie, Rogers 23083 (K, holo.!; PRE!),

Rogers 20319 (PRE, para.!).

33261—2

y Mh Wi

“ppb ty Uys ‘

Le HY IMMPAIEAANG A Tai

My} HL COHE Ps UE

ALEIDA v.D. MERWE

Fic. 2.—Mpyrica pilulifera. a, twig with male spikes, natural size; b, male bract, x 10; c, stamens,

x 20; d, female spike, x 2; e, female bract, x 12; f, pistil surrounded by hypogynous scales,

x 10; g, female hypogynous scales, x 10. (d-g, Scheepers 677).

Tree or shrub up to 30 m high with girth up to 30 cm. Bark dark brown to black.

Branchlets often prominently lenticellate, glabrous or tomentulose. Leaves narrowly

elliptic to elliptic-obovate, 3-2—6-:5 cm long, 1-5—2-8 cm wide, apex acute to round,

apiculate. base cuneate, margin crenato-serrate in upper 4-3, midrib prominent parti-

cularly on lower surface and sometimes pubescent, lateral nerves subimmersed, 12-15,

forming obtuse angle to midrib and usually dividing before reaching margin, puberulous

when young, becoming giabrous, coriaceous rarely conspicuously gland-dotted; petioles

8-10 mm long. Flowers dioecious. Male spikes axillary, solitary, 8-25 mm long;

rhachis not or scarcely visible, glabrous or pubescent, gland-dotted; bracts approxi-

mately obtrullate, 1-5—-2-6 mm long, 2-2-5 mm wide, ciliate, ridged, concave on inner

surface, not imbricate; stamens 5-8 in two series; anthers 0-8 mm long, minutely

papillose. Female spikes axillary, solitary, 1-5—4-5 cm long; rhachis visible, pubescent;

bracts 1, caducous, ovate, 1-2—2:8 mm long, 0-8-2 00 mm wide, ciliate; hypogynous

scales 4-6, roughly ovate, 0-5-1 mm diam., unequal, fleshy, keeled, ciliate, gland-

dotted; style branches 2 (rarely 3), 1-1-5 mm long, flattened. Fruits ellipsoid-globose,

4-5 mm diam., warted, usually wax-covered. Fics. 1:5; 2.

Found usually at high altitudes in Rhodesia, Malawi, Swaziland and South Africa

as far south as the eastern Cape.

Cape.—Tsolo: Ntywenka, Miller B/692. Umtata: south of All Saints Nek, Acocks 131812; Baziya

Forest Station, Killick & Marais 2067.

NATAL.—Bergville: Upper Sinyati River Valley, Edwards 850; Ntonjelane, Mnweni area, Esterhuysen

14513; Mnweni area, Esterhuysen 18653; Cathedral Peak Forest Research Station, Killick 1709; 1817.

Estcourt: Cathkin area, Esterhuysen 7959 (NBG); Ntabamhlope, West 205. Ngotshe: Ngome,

Gerstner 4365; 4868; 4869. Nkandla: Nkandla Forest, Gerstner 4604. Pietermaritzburg: Little

Zwartkop, Doidge sn. Utrecht: Kaffir Drift, Thode A. 246. Vryheid: hill on east side of Zungeni

Peak. Acocks 11501; 11561.

TRANSVAAL.—Barberton: Barberton, Rogers 24097; 24897; 8 miles from Havelock mine on Bar-

berton road, Wells 2016; kloof on Piggs Peak road, West 3008; 3017. Letaba: south face of Piesangs-

kop, Scheepers 433; above Grootbos Govt. Forest Reserve, Scheepers 677. Lydenburg: Op-de-Berg,

Keet 1127; 19-3 miles south-east of Lydenburg on Nelspruit road, Marais 322. Nelspruit: Kaapsche

Hoop, Strey 3601. Pietersburg: Blaauwberg, Esterhuysen 20496; 21496 (BOL); Rooikoppies, near

Politsi siding, Galpin 9403; Wolkberg, Gerstner 5615; the Downs, Renny DE. 14. Pilgrim’s Rest:

4 miles west of Graskop, Codd 3316; Lulu Mountains, Mogg 16932; 46 miles from Acornhoek on

Graskop road, Story 3993; Mariepskop, Van der Schijff 4757; 5566; 5614. Rustenburg: in the

“ kloof ’’, Galpin 11640; 3 miles south of Breedts Nek, Story 958. Soutpansberg: Entabeni Forest

Reserve, Codd 3054; south of Franz Hoek Peak, Galpin s.n. (PRE No. 14902).

SWAZILAND.—Mbabane: Gobolo, Compton s.n.; near Umsindusi Bush, Dlamini s.n. Piggs Peak:

Havelock, Compton 29138; Miller 6042.

The var. puberula has not been upheld, because the species is very variable as

regards pubescence: populations have been studied in the field which contain both

glabrous and puberulous forms.

M. rogersii is quite clearly a synonym of M. pilulifera. The latter was probably

overlooked by Burtt Davy, because at that time it was known only from tropical

Africa.

6. Myrica brevifolia E. Mey. ex C.DC. in DC., Prodr. 16, 2: 150 (1864); A. Chev.,

Monogr. Myric. 158 (1901); Hutch. in Fl. Cap. 5, 2: 567 (1925). Type: Queenstown,

Winterberg Range, Zeyher 5 (K, holo.!; SAM!; PRE, photo. of iso.).

Dwarf shrub with erect stems up to | m high arising from underground rootstock.

Bark dark grey to black. Ultimate branches puberulous to tomentulose. Leaves

narrowly elliptic, occasionally obovate, 1-5-3 cm long, 0-5—1-3 cm wide, apex acute,

apiculate, base cuneate to round, margin with 1-5 serrations in upper 4, midrib distinct,

lateral veins 4-8, subimmersed or immersed, forming obtuse angle to midrib, glabrous

or pubescent, conspicuously glandular, coriaceous, petiole 2-4 mm long. Flowers

dioecious. Male spikes axillary, solitary, 7-12 mm long; rhachis not visible, puberulous ;

bracts more or less imbricate, trullate-triangular, 1-6-1-7 mm long, 1-8-3 mm wide

ciliate, deeply concave on inner surface; stamens usually 2 but occasionally 4; anthers

0-5 mm long, epapillose. Female spikes axillary, solitary, 0-5—1-2 cm long; rhachis

not or scarcely visible, pubescent; bracts 1, ovate, 1-6 mm long, 1-5 mm wide, ciliolate;

hypogynous scales 4, roughly ovate, 0-5-O0-8 mm diam., unequal, fleshy, keeled, style

branches 2, | mm long, filiform (but slightly flattened). Fruits globose, 2-3 mm diam.,

warted, wax-covered. Fic. |: 6.

A dwarf species with an underground rootstock which is confined to hills or

mountains in Natal and the eastern Cape. Hutchinson in Fl. Cap. 5: 567 (1925) cites

a specimen from the Transvaal, viz. Worsdel! s.n. from Belford, but this specimen

cannot be traced.

Care.—Adelaide: Winterberg, Ecklon & Zeyher 13:6. Albany: Tunnel Hill near Grahamstown,

Davies s.n.; Grahamstown Nature Reserve, Story 2819; Wells 3867; 3869. Herschel: Sterkspruit,

Hepburn 33 (GRA). Humasndorp: MacPeak, Taylor 928 (NBG). Keiskammahoek: Cata, Acocks

15729; Hogsback Mountain, Rattray 330 (BOL); 408 (GRA). Komga: among rocks near Komga,

Flanagan 93. Queenstown: Zeyher 20271 (SAM). Somerset East: summit of Boschberg, Macowan

1925. Stockenstroom: Katberg Pass, Acocks 12135. Stutterheim: summit of Dohne Peak, Galpin 2458.

NaTAL.—Bergville: Mnweni, Esterhuysen 15547. Pietermaritzburg: Little Zwartkop, Doidge s.n.

Underberg: top of rocky hills, Himeville, Bews s.n. Vryheid: Dumuka Mountain, Gerstner 4643.

A note on the type specimen at Kew by Hutchinson reads: “ Someone has altered

Zeyher to Drege in the Prodomus (Kew copy), but Zeyher’s specimen must be the type,

because it is female and De Candolle only describes the female’. The reference in

the Prodromus does not actually cite No. 5, but does refer to ““ Zeyher in H. Kew”.

7. Myrica cordifolia L., Sp. Pl. 1025 (1753); A. Chev., Monogr. Myric. 168 (1901);

Hutch. in Fl. Cap. 5, 2: 563-564 (1925); Adamson & Salter, Fl. Cape Penins. 312

(1950). Syntypes: Cape, LINN 1169.7 (lecto.!); LINN 1169.8!

M. cordifolia L. var. microphylla A. Chev., Monogr. Myric. 170 (1901); Hutch.,

l.c. 564 (1925). Type: Doornhoogde in der Kapflaéche, Ecklon & Zeyher 78-4 (G:

in Herb. Delessert!; P: in Herb. Drake; PRE!). . elliptica A. Chev., l.c. 166. t.8,

figs. 1-8 & 10 (1901). Type: Cape, Burmann (G, holo!; PRE, photo.).

A much-branched prostrate or erect shrub c. | m high (teste Muir 176, 3 m high).

Ultimate branches puberulous to tomentulose. Leaves usually imbricate, sessile, broadly

ovate to orbicular, 0:6—2-1 cm long, 0:4—1-5 cm wide, apex acute to round, mucronate,

base cordate (rarely round or broadly cuneate), margin repand-dentate, glabrous,

coriaceous, conspicuously gland-dotted especially below, midrib distinct, prominent

below, lateral nerves 4-7 at obtuse angle to midrib, practically immersed. Flowers

dioecious. Male spikes axillary, solitary, 2-S mm long; rhachis glabrous or puberulous,

slightly to fairly visible; bracts solitary, broadly ovate (broadly)-trullate, | mm long,

1-2-1-4 mm wide, concave on inner surface, ciliate glandular; stamens 2; anthers

0-5-1 mm long. Female spikes 1 cm long; rhachis glabrous, visible; bracts solitary

ovate, 1-3-1-7 mm long, 1-1-3 mm wide, ciliate; hypogynous scales 4, roughly ovate,

0:-4-0:7 mm diam., fleshy, ciliate, style branches 2, filiform, 0:6 mm long. Fruits

globose, 0:5—O-8 mm diam., warted, densely covered with wax. Fic. 1:7.

A common sand-dune shrub extending from the Cape Peninsula to near the mouth

of the Kei River in the eastern Cape. The fruits are sometimes melted down and the

wax is used as polish or for making candles. The plant is variously known as the

Waxberry (Wasbessie), Candle Berry, Vegetable Wax or Glashout.

Cape.—Alexandria: frequent on secondary dunes, Archibald 5022. Bathurst: Kasonga Mouth,

Britten 2298; Port Alfred, Hutton 450; sand slopes of foreshore, Port Alfred, Tyson s.n. Bredasdorp:

Papkuilsfontein; Rycroft 1839 (NBG); Buffelsjacht, Van Breda 943. Caledon: Mossel River,

Hermanus Pole Evans 476; Betty’s Bay, Van Rensburg 2151. East London: Nahoon River Mouth,

Galpin 5679; East London, Sims.n. Humansdorp: Phillips sn. Komga: near Kei Mouth, Flanagan

1055. Peninsula: Houtbay, Goulimis 27256 (BOL); Cape Flats, Isaac 27254 (BOL); Schusters Bay,

Isaac 27233 (BOL); Cape Flats, Marloth 8943; Robben Island, Walgate 633 (NBG); Uitvlugt, Wolley

Dod 2626 (BOL); Houtbay, Schlechter 965. Port Elizabeth: Humewood, Long 1373; Sim 1976.

Riversdale: Still Bay, Muir 176; 5336; Van Zinderen Bakker 299. Simonstown: Fishhoek, near

Kalkbay, Peans sub Marloth 13511. Somerset West: sand dunes at Strand, Parker 3660; 3574 (BOL):

Somerset Strand dunes, Strey 714. Uitenhage: Britten 2084.

M. cordifolia is probably the most clear-cut of the South African species of Myrica:

because of its characteristic leaf shape it cannot easily be confused with other species.

The type of M. elliptica, Burmann s.n., is matched by no other specimen I have

seen. It has unequally round bases to the leaves, but otherwise resembles the eastern

Cape forms of M. cordifolia, e.g. Flanagan 1055 ete. Occasionally specimens of M.

cordifolia have the odd leaf with an unequally round instead of a cordate base. These

facts, in my opinion, justify its attachment to M. cordifolia rather than its retention

as a distinct species.

8. Myrica kraussiana Buching. ex Meisn. in Flora 6: 89 (1845); A. Chev., Monogr.

Myric. 152-153 (1901); Hutch. in Fl. Cap. 6, 2: 566-567 (1925); Adamson in FI.

Cape Penins. 314 (1950). Type: Cape Peninsula, between rocks on summit of Steenberg

Mountain, Krauss 1564, holo. (no specimens traced); summit of Steenberg Mountain,

Killick 3456 (PRE, neo.).

M. humilis sensu C.DC. in DC., Prodr. 16, 2: 150 (1864); A. Chev., l.c. 158 (1901);

Hutch. in Fl. Cap. 5, 2: 565-566 (1925); Salter in Fl. Cape Penins. 312 (1950), non

Cham. & Schlechtd. in Linnaea 6: 535 (1831); M. kraussiana Buching. ex Meisn. var.

latifolia A. Chev., |.c. 153 (1901). Type: Cape, Harvey (CN, holo.!; PRE, photo.).

Low shrub. Bark brown to black. Ultimate branches tomentulose to tomentose.

Leaves elliptic, 1-5—5-5 cm long, 0-8-3-5 cm wide, apex acute to round, apiculate,

base round (rarely somewhat cordate), margin entire to crenatoserrate in upper half,

coriaceous, glabrous to densely pubescent, midrib and lateral veins (9-14) prominent

on lower surface which is conspicuously reticulate and gland-dotted; petiole 1-3 mm

long. Flowers dioecious. Male spikes axillary, solitary, 1-5-2 cm long; bracts imbri-

cate, trullate-ovate-spathulate-geniculate, large, 3-5-4-5 mm long, 2-5-4 mm wide,

concave on inner surface, ciliate, gland-dotted; stamens 4-5; anthers 2-5 mm long,

finely and shortly puberulous. Female spikes axillary, solitary, 2-8 cm long; rhachis

visible, tomentulose; bracts solitary, narrowly ovate, 3-4-8 mm long, 1-7 mm wide,

ciliate, gland-dotted; hypogynous scales 5, ovate, 0-5 mm diam., keeled, fleshy; style

branches 2, flattened, 2mm long. Fruits subglobose, 2-3 mm diam., warted. Fic. 1: 8.

Chiefly confined to the Cape Peninsula, but extending eastwards as far as Zitzi-

kamma.

Cape.—Caledon: Langkloofberg, Esterhuysen 9143 (BOL); Hottentots Holland, Stokoe 8949 (BOL).

Knysna: Lottering Bush, Zitzikamma, Galpin 4581. Peninsula: above Skeleton Gorge, Esterhuysen

12015; 12304; Devils Peak, upper slopes, Esterhuysen 12941; rocky plateau on Table Mountain,

Esterhuysen 17563; Groenekloof, Galpin 4583; summit of Table Mountain, Galpin 4584; summit

of Steenberg, Killick 3456; Taylor 3293; Table Mountain, Marloth 1939. Swellendam: slopes of

Langebergen, Esterhuysen 10481 (BOL); Zuurbraak Mtn., Galpin 4582.

Up till now this species has been known as M. humilis Cham. & Schlechtd.—

probably because the type was never consulted in previous revisions of the genus.

The type/s of M. humilis is Bergius, Mund & Maire. At first it was assumed that this

represented one specimen, but on looking through Chamisso and Schlechtendal’s

descriptions in Linnaea, it became apparent that these authors frequently cited more

than one specimen under a collective heading. For example, collectors who were

not at the Cape at the same time, are often linked together, e.g. Bergius, Ecklon &

Zeyher. After failing to locate a Bergius, Mund and Maire specimen, an isotype of

Mund & Maire s.n. was received from Geneva (Herb. Boiss.) annotated as M. humilis

and bearing a type label. Much to the author’s surprise this specimen did not represent

our present concept of M. humilis, but was actually M. burmannii. Moreover, it fitted

the original description of M. humilis better than what we presently call M. humilis.

This means that M. burmannii is a synonym of M. humilis and that what we have been

calling M. burmannii must now be known as M. humilis. It also means that another

name must be found for M. humilis auct. The only name available is Myrica kraussiana

Buching. ex Meisn. The type is Krauss 1564, but this specimen cannot be traced in

any European herbarium. In the original description of M. kraussiana the type locality

is given as “inter rupes in summitate montium Steenberge, Cap. Sept. Alt. 3,000’ ”.

In May, 1962, the author visited this mountain and the only Myrica found growing

there was a somewhat depauperate form of what we have been calling M. humilis.

In view of this, it seems reasonable to adopt the name M. kraussiana. It should be

mentioned also that M. kraussiana var. latifolia clearly equals M. humilis auct.

9. Myrica humilis Cham. & Schlechtdl. in Linnaea 6: 535 (1831); C.DC. in DC.,

Prodr. 16, 2: 150 (1864); A. Chev., Monogr. Myric. 158 (1901); Hutch. in FI. Cap. 5,

2: 565-566 (1925). Type: Cape, ‘‘ Bergius, Mund & Maire’? (Mund & Maire, G,

iso.!, PRE, photo.).

M. burmannii E. Mey. ex C.DC. in DC., Prodr. 16, 2: 149 (1864); A. Chev., l.c.

154 (1901); Hutch., l.c. 564-565 (1925); Adamson in Fl. Cape Penins. 312 (1950).

Type: Kleyn Rivier, Caledon Division, Zeyher 3875 (K, holo.!; PRE!, SAM!). ™.

brevifolia E. Mey. ex C.DC. var. subintegra A. Chev. l.c. 160 (1901). Type: Burmann

65 (G, holo.; PRE, photo.). M. dregeana A. Chev., l.c. 155 (1901); Hutch. l.c. 565

(1925). Type: Wan Stadens Berg, Uitenhage Division, Ecklon & Zeyher 4-7 (G,

holo.!, K!, PRE!, SAM!). MM. glabrissima A. Chey., |.c. 156-157 (1901). Hutch.

l.c., 569 (1925). Type: Long Kloof, about the source of Keurboom’s River in a rocky

kloof, Mar., George Division, Burchell 5081 (P, holo.; K!, PRE, photo. of iso.).

M. myrtifolia A. Chev., l.c. 155, t.7, fig. A (1901). Type: Cape, specimen ex Herb.

Pet. Thouars (P, holo.!; PRE, photo.).

Shrub, 4-1 m high. Bark pale grey to black. Ultimate branches glabrous to

tomentulose. Leaves elliptic-ovate-broadly ovate (rarely orbicular), 2-5-4:5 cm long,

0-8-2-5 cm wide, apex acute to obtuse, apiculate, base round (occasionally somewhat

cuneate), margin entire or with few serrations in upper quarter, midrib fairly prominent,

lateral veins 5—10 subimmersed, lower surface reticulate, glabrous, occasionally pubescent

on midrib. Flowers dioecious. Male spikes solitary, axillary, 1-2 cm long; rhachis

glabrous to puberulous, scarcely visible; bracts broadly trullate-ovate ridged, concave

on inner surface, ciliate, gland-dotted, 1-8—2-8 mm long, 2: 3-2-5 mm wide; stamens 4;

anthers 1-8 mm long, epapillose. Female spikes solitary, axillary, 2-5-4 cm long;

rhachis visible, glabrous to tomentulose; bracts broadly ovate, 1-3-1-7 mm long,

1-3-1-75 mm wide, ciliate, gland-dotted; hypogynous scales 4, ovate, 0-5 mm diam.,

fleshy, ciliate; style branches flattened, thick, long. Fruit globose, 4-5 mm diam.,

warted, covered with white wax. Fic. 1: 9.

A coastal species occurring between the districts of Bredasdorp and Albany.

Cape.—Albany: near Grahamstown, Noe/ 1323. Bredasdorp: summit of Potberg, Pillans 9315.

Caledon: Hottentots Holland Mtns. Stokoe 7981 (BOL). George: near Touw River, Burchell 5739;

near George, Michell 16094; near George, Schlechter 5776; Outeniqua Pass, Van Breda 1134. Humans-

dorp: Flats, Ratelsbosch, Fourcades 29; Thode A. 1013; Assegaai Bosch, Thode A. 2571. Knysna:

Lottering Bush, Zitzikama, Galpin 4581; Duthie 734 (BOL). Melkhoutkraal, Keet 555; 556; 3070;

3071. Oudtshoorn: top of Robinson Pass, Acocks 20583. Port Elizabeth: Van Stadensberg, Ecklon

& Zeyher 4-7 (BOL). Riversdale: Aasvoélbergnek, Horn s.n.; Langeberg above Plattekloof, Muir

387; Driefontein, Albertinia, Muir 959; Langeberg above Nivo, Muir 2638; stony flats near Albertinia,

Muir 4531. Swellendam: slopes in Tradouw Pass, Marloth 12163. Uniondale: Kouga Mts., Ester-

huysen 10784; 10821 (BOL). Uitenhage: Zuurberg Mountain, Fries, Norlindh & Weimarck 579.

See discussion under M. kraussiana for reasons why this species, known up to

now as M. burmannii, must be called M. humilis.

M. humilis is an extremely variable species as regards leaf shape, size and pube-

scence. Leaf shape varies from elliptic-ovate to broadly ovate (rarely orbicular) and

size from 2:5-4:5 cm long and 0-8—-2-5 cm wide. Two species which have been sunk

under M. humilis represent the extremes in degree of pubescence of this species: M.

glabrissima is completely glabrous, whereas M. dregeana has tomentulose ultimate

branches with hairs on the midrib of the leaves. The rolled edges of the leaves of

M. dregeana, as represented by the type, Ecklon & Zeyher 4-7, can probably be attributed

to the way in which the specimen was dried rather than to an inherent character of

the plant.

M. myrtifola can be linked with M. humilis via Michell 16094 from George, which

serves as a good intermediate.

Species dubia:—

Myrica ovata Wendl. f. in Bartling & Wendland, Beit. zur Botanik, 2: 3 (1825).

The identity of this species is uncertain. The type, Hesse s.n., could not be traced:

the author tried B, GOET, HAN and S without success.

Lt? eras wee er om :

aii AY irene bret nig: Ate pay

op BP. carigh elke Ghee ser ee

al eere e~. tidied Be ph is 2a rin tacoma y

i; deni ead aeidieal ye) aoe Gatw aid: “_ “a eS

z aa!

== oyntremnt pata hat te rie ie fa) ae 9 ir TO

lean ne doy |

ae =f ae -

as ee

o>. GE eee Bnew: 4) 4 1-aeee .ui i al

- Shi y 09a’ Oe mie ) wrt

Se Seen £ Sas 1 Ati

are he - Ss

a a ee

~~ wa a

ete fa a 2 a\y2u

° = ; ro Fist Ves

fsa

ce Se

Bothalia 10, 1: 19-28

The Genus Talinum (Portulacaceae) in Southern

Africa

H. R. Tolken

ABSTRACT

A revision of the five species of Talinum indigenous to South Africa has been undertaken. A

key to the species is provided.

INTRODUCTION

During a visit to South West Africa observations were made on plants of Talinum,

but it was very difficult to find correct names for them as seeds were not always

available and seed structure is the major key character used. Indeed, the seed characters

are very reliable in this genus and the specific patterns of the papillae on the seed

membrane can be seen at an early developmental stage. However, specimens without

seeds and more specifically male plants of T. crispatulum could not be identified. This

difficulty, particularly with regard to the four species with yellow flowers which are

emphasized in this work, as well as the different interpretations of the species in the

literature, indicated the need of taxonomic clarification of the species of this genus.

Further observations in the northern Cape Province convinced me that species

can be recognized in the field even without flowers, which open only for a short time in

the afternoon. The leaves, although they are very variable, show a few characteristics

which may help in identification. For instance, the leaves of T. crispatulum have

typical crisped leaf margins and are never revolute (see Fig. 1: 3), a combination of

characters not found in any of the other species, although young leaves of 7. arnotii

often have a crisped leaf margin. Occasionally specimens of the other three species,

especially T. arnotii, do not show the revolute leaf margin when grown under particularly

humid conditions. The leaves of T. tenuissimum and T. caffrum are usually linear,

but in young plants of T. caffrum rather broad leaves are produced and these are so

similar to those of T. arnotii that flowers and fruits are needed for identification. The

leaves of 7. tenuissimum are always linear and the whole plant is smaller and more

delicate than those of T. caffrum. Diagnostic characters of these four yellow-flowered

species are summarized in Table | to illustrate the differences and variation found in

this complex. The table will also facilitate the identification of specimens without

fruits and/or flowers.

When working on this revision, it was realized that many of the syntypes of species

described by Dinter no longer exist at Berlin Herbarium where, according to Lanjouw

& Stafleau (1954) in the index of collectors in Index Herbariorum, the original sets

of Dinter specimens were deposited. This, and the occurrence of mixed collections

found on certain sheets of syntype material seen, made it necessary to select lectotypes.

In this connection, specimens from Berlin should get preference as the original set was

deposited there and, in addition, Dinter’s own herbarium is now in Berlin Herbarium,

having been acquired at a more recent date.

uWIn[Iy oy} WoIJ 1noO

suneiper oeijided

o}eSUOTS YIM WIOFUSI

(SI-) ZI-OT

y10Ys AIOA IO jyuosge

SOpoulwie}s SNOIOTUNU

Zuo] wu (¢-) p-€

saloa}oe1q Jo sued (Z) T

Suo[ wo (¢-Z-) ¢-I-T

PoroMmoy-(€) Z 10 T

peoiq wo (¢-]) I-9-0

Suo] WO (€-) ¢-7-T-I

91P|

-ooour] A[peorg Ajyensn

sjunjd ajouaf

ed s]IOS Apues

CeCe . (S¢-) O£€-0Z

sete eee ee suo] WUL S-b

$g]09}0e.1qg Jo sired (€) Z

-_: BuO] wd (S-) €-Z

oS paroMmoy-(S) 7-7

P9AINIeI TOASU

apriq jeo, ‘podsiso

peoiq

wo (8-0-) 9-0-4-0

suo] wo (¢-7-) $-I-T

31e[090UR|

Deis: Arey, 0} osoyided

UMOIQ BUIUIOSIG 314M

sjunjd ajou

wnjnjodsisd wh

wnjnjpdsisd “J, pue wnuissinual 7

oi bbe Nees [TOS Apues

wIn{Iy oy} Woy

jno Zuljeipes oeyjided

9}e3U0[9 YIM WAOJUSI

UdT[OMS “SUOT WU Z-T

sews eae “(pI-) OI-8

"++ *-Buo] uu (¢-) p-€

So[09}

-ovIq Jo sired Z uso

sete ee suo] wo ¢-I-S-0

ra paremoy-(p) €-1

SUIAINIOI

opriq jeay ‘aynjoAor

peroiq

wo (¢-0-) €-0-T-0

BuO] Wd (S—) p-Z (-S - 1)

Se0c0c 00000 * TeOUI]

o00000 0000000 snoiqe|3

uMO.Ig-YsIPpar

wnuissinua]

“WNAL{9I * J, “11JOUAD

sdo1o

-jno yoor uo Ay[ensn

U99M}9q IPI

-jided yuIM sosprit 917

-us0u0d YM [evoroyds

6000000000000 -“Op-Cz

BUISI]S MOQ

UdTJOMS “SUO] WUT S—¢

ee eee see ‘(Qs-) SE-ST

"+" Zuoy wu (¢T—) 8-9

$9]09}98.1g Jo I1vd Q IO T

suo] wd (¢-7-) ¢-I-S$-0

eee eee " poroMoy-(Z) I

SUIAINIOI

Spr[q Jeol ‘ojnjOAoI

peoirg wo

(Z-I-) 9-0-€-0 (-Z-0)

Buo] Wd (§-) p-Z (-S- T)

9yeuold

NU “SUOTGO 0} IOUT]

0000000000000 snoiqe3

uUnAf{oD *T,

**sjtos AsAvD 0} Apues

SMO O1I]UI9

-u0d uajjo ‘ov[ided

QS1VO9 O} OUY ‘WIOJIUET

9000000006 SO erg

BUISI]S MOl[9q US|

-[OMS ‘SUO] WU f-¢ (—Z)

cooccoon coe) S€-07

“++ **Zuo] wut (,-) 9-7

s9[o9}0e1q Jo ed Q 10 |

Zuo] wo (¢-4-) $-€-Z

Heres ess DazaMOp-¢-]

SUIAINIII

opeiq jeoy ‘sunoA

usyM podsiio ‘ajyn[oAos

prog

wo (€-) ZI (s-0)

Suc] wo (¢-) --~ (I)

ondiy[a

A[PeOI1g 0} 9o}e[OI0URT]

snoiqe[s SUIWIOS

-oq ‘asoyided <Ajouy

UMOIQ-USIp

-pol SUTUIOS9q 3}I4M

MJOUID *T,

[e197]

ERT A Cee

SIOMOL.

jo ao1ped

505090 gjounpeg

** Q0U989Sa1OYUT

*“UISIVUL JeaT

SOAROT

H0000 soyourig

siaqny Jo Ysoy

UNUIDT, UIIMJOqG SooUdIOyIp Jo Areuung— | alavL

The author wishes to acknowledge with thanks the loan of specimens from the

following herbaria: Albany Museum Herbarium; Herbarium Berlin-Dahlem; Royal

Botanic Gardens, Kew; McGregor Museum Herbarium; National Botanic Gardens

and South African Museum Herbaria at Kirstenbosch; Natal Herbarium; and Stellen-

bosch Herbarium.

TALINUM

Talinum Adanson, Fam. 2: 245 (1763); Benth. & Hook.f., Gen. Pl. 1: 157 (1862);

Oliver in Fl. Trop. Afr. 1: 149 (1868); Sonder in Fl. Cap. 2: 385 (1862); von Poellnitz

in Fedde Rep. 35: 1 (1934); Wild in Fl. Zamb. 1, 2: 369 (1961). Type species: T.

triangulare (Jacq.) Willd.

Shrublets with annual branches from a perennial base, usually tuberous. Leaves

linear to broadly elliptic, obovate, petiolate, succulent, alternate often irregularly

spaced; stipules linear, setaceous, usually keeled, with membranous margin, 1-3 mm

long, usually caducous. Inflorescence in panicles or axillary, cymose. Sepals 2,

narrowly ovate to broadly-ovate, keeled, slightly hooded at the apex, green, usually

with membranous margin. Petals (4) 5 (—7), ovate, pointed or mucronate, often faintly

keeled at the apex. Stamens 10-30 (-50); filaments usually connate at the base. Ovary

superior, one-chambered with three carpels, with 10-40 ovules on a free central

placenta; style 1-3 mm long or absent; stigmas 3, papillose. Fruit a capsule, ovoid to

conical, shiny yellow, dehiscing by 3 valves. Seeds spherical to reniform, often laterally

compressed, with more or less distinct patterns of papillae, dark brown or black.

Species of Talinum occur in most parts of Africa, parts of Asia and North and

South America. Five species are indigenous in South Africa and occur only in the

summer rainfall areas. JT. paniculatum (Jacq.) Gaertn. from North America is often

cultivated and has been recorded a few times as a garden escape, but it does not seem to

spread as a weed.

In the African species the pedicels are always swollen below the fruit and are more

or less recurved when fruiting.

Leaves obovate, obtuse or rounded at the apex; inflorescence terminal, paniculate, with pink

IKON MES grorcdsn oid atalehore ov corce a cond OMe ERO OSI IA Geto OO Coun nicer E eeieior or anteniare 1. T. portulacifolium

Leaves linear, broadly elliptic to ovate, tapering towards the apex; inflorescence axillary, cymose,

with yellow flowers:

Leaf margin crisped, never revolute; leaves irregularly alternate with internodes rarely longer

than 1 cm, varying in length on the same branch; plant dioecious:

Leaves 1-1-5 (-2) cm long, 4-6 (-8) mm broad; anthers with pollen; ovary without ovules

4. T. crispatulum (male)

Leaves 1-5-2-5 (—3) cm long, 6-10 (-15) mm broad; anthers without pollen; ovary with

O17 ONES so. cos atond oS hc BOOS POO Oe reictrn SORT EA ears 4. T. crispatulum (female)

Leaf margin rarely crisped when young, revolute; leaves usually alternate with internodes usually

longer than 1 cm and not varying in length on the same branch; plant monoecious:

Pedicel thread-like, with (1) 2 pairs of bracteoles; calyx 3-4 mm long; stamens 8-14; seeds

C=, Gil) joer eHoswls, 2-3 marin SOs 5 coos onooccnocddscoducuscooNd 5. T. tenuissimum

Pedicel not thread-like, with 0 or 1 pair of bracteoles; calyx 5-15 mm long; stamens 20-50;

seeds 20-40 per capsule, c. 1 mm long:

Inflorescence 1-flowered; seeds with concentric ridges perpendicular, elongate papillae

between the ridges; leaves linear to oblong................2.eeeeeeeeee 2. T. caffrum

Inflorescence 1 or 2 (3)-flowered; seeds without ridges, papillose (sometimes arranged

in concentric rows, but without elongate papillae between them); leaves narrowly

ovatemtombroadlyaellipuicheemeeaeeree toes carer tecrerrac 3. T. arnotii

1. T. portulacifolium (Forsk.) Aschers. ex Schweinf. in Bull. Herb. Boiss. 4, App.

2: 172 (1896); Wild in Fl. Zamb. 1, 2: 372 (1961). Type: Arabia.

Orygia portulacifolia Forsk., Fl. Aegypt.-Arab. 103 (1775).

Portulaca cuneifolia Vahl, Symb. Bot. 1: 33 (1790), nom. illegit. Type: the same

as for O. portulacifolia.

co LLETT~

Fic. 1.—1, Talinum arnotii, branch of broad leaved form, = 1 (Té/ken 1281); la, branch of narrow

leaved form, < 1 (Télken 1284). 2, T. caffrum, branch with flower and fruits, x 1 (Té/ken 1293).

3, T. crispatulum, branch of female plant, < 1 (Té/ken 1282); 3a, branch of male plant, < 1

(Tolken 1283).

Talinum cuneifolium Willd., Sp. Pl. ed. 4, 2: 864 (1800), nom. illegit.; Burtt Davy,

Fl. Transv. 1: 166 (1926); von Poellnitz in Fedde Rep. 35: 14 (1934), partly, excl.

specimens from S.W.A.

Shrubs with annual glabrous branches up to | m high from a perennial base.

Leaves obovate, obtuse or rounded and mucronate at the apex, cuneate at the base,

2-6 (-10) cm long, 1-3 (-4) cm broad; petiole 1-3 mm long. Inflorescence terminal,

paniculate, many-flowered, often with leaf-like bracts 5-15 mm long on the main axis,

usually with one pair of pointed membranous bracteoles below each flower. Sepals

broadly ovate, apiculate, 3-4 mm long, slightly keeled at the apex. Petals obovate,

pointed at the apex, pink or mauve. Stamens 25-30 (—35) with terete filaments not

connate at the base. Ovary spherical with 30-40 ovules; style 2-3 mm long, dividing

into 3 stigmas 1-2 mm long, papillose. Capsule ovate, 5-8 mm long, 5-6 mm broad,

shiny yellow. Seeds ovate to reniform, laterally compressed, up to | mm long, with

indistinct elongate papillae radiating out from the hilum, black.

T. portulacifolium is a shade-loving plant in the northern and eastern Transvaal,

but also occurs in most parts of Africa, Arabia and India.

TRANSVAAL.—Letaba: Rooirotse, Van der Schijff 3322; Gravelotte, Van der Merwe 2323. Messina:

Messina, Rogers s.n. in BOL 19280 (BOL). Potgietersrus: Soutpan, Bremekamp & Schweickerdt 281.

Sibasa: Klopperfontein, Codd 5417; Van der Schijff 3563. Waterberg: 42 miles north of Vaalwater,

Meeuse & Strey 10454.

The type specimen of this species has not been seen, but Forskal’s original descrip-

tion is so complete that there is no doubt as to the species concerned.

2. T. caffrum (Thunb.) Eckl. & Zeyh., Enum. 282, No. 1802 (1836); Sonder in

Fl. Cap. 2: 385 (1862); von Poellnitz in Fedde Rep. 35: 12 (1934), partly, excl. specimens

from S.W.A.; Wild in Fl. Zamb. 1, 2: 370 (1961). Type: Cape, Thunberg s.n. (UPS,

2 sheets; PRE, photo.!).

Portulaca caffra Thunb., Prod. 85 (1800).

Talinum minus Eckl. & Zeyh., Enum. 282, No. 1803 (1836). Type: Katriviersberg,

Ecklon & Zeyher s.n. (SAM!). T. caffrum var. minus (Eckl. & Zeyh.) Sonder in FI.

Cap. 2: 385 (1862). 7. esculentum Dinter & Schellenberg in Dinter, Veget. Feldkost

D.S.W.A. 12 (1912); Fedde Rep. 23: 369 (1927); von Poellnitz in Ber. Bot. Ges.

S1= 118 (1933); Bedde Rep: 35: 17 (1934); Podlech im Prodr. Fl. S.W.A. 29: 12

(1967). Type: Otavipforte, Dinter 730a (B, lecto.!).

Claytonia caffra (Thunb.) Kuntze, Rev. Gen. 1: 57 (1891).

Plant erect or decumbent; tubers branched, wita white flesh. Branches glabrous

or almost so, 15-25 (-40) cm long. Leaves linear to oblong, 2-4 (-8) cm long,

(2) 3-6 (-10) mm broad, with margins revolute or recurving; first leaves on branch

short and broadly oblong; petiole 1-3 mm long. Jnflorescence axillary, cymose,

1 (2)-flowered. Peduncle 0-5—1-5 (—2-5) cm long, often keeled or laterally compressed,

ending in two keeled bracts with membranous margins. Pedicel 1-1-5 (—2) cm long,

swollen, often ridged below the ovary; bracteoles usually absent, rarely one lateral

flower with one pair of bracteoles. Sepals lanceolate, pointed, 5-8 (-15) mm long,

keeled, slightly hooded, with membranous margins. Petals 5, obovate to elliptic,

yellow. Stamens 25-30 (—50), with filaments becoming broader towards the base,

fused into a ring c. 1 mm high. Ovary conical, with 25-35 (—40) ovules; style 2-3 (—4)

mm long, dividing into 3 stigmas 2 mm long, papillose. Capsule ovoid, 6-10 mm

long, 6-7 mm broad, shiny, yellow. Seeds spherical to comma-shaped, hardly laterally

compressed, c. | mm long, with ridges with perpendicular grooves between the ridges,

dark brown or black. Fic. 1: 2.

___ This species occurs mainly on rocky outcrops and has mainly an eastern distribution

in Africa from the eastern Cape to Kenya, but also extends across Rhodesia and Zambia

into north-eastern South West Africa.

Cape.—Albany: Glen Boyd, Linstaedt 47. Aliwal North: Elandshoek, F. Bolus 156. Beaufort

West: 20 miles east of Beaufort West, Adamson D 176. Bedford: Eastpoort, Burtt Davy 12219;

Bedford, Glane 130 (GRA). Cathcart: Klipplaatrivier, Ecklon & Zeyher sn. (SAM). Cradock:

Mortimer, Kensit sn. (BOL). De Aar: De Aar, Lander sn. (GRA). Fort Beaufort: Alice, Barker

2325 (NBG); Kroomie, Marloth 9388. Graaff-Reinet: Ryneveld Pass, Bolus 372 (BOL, GRA).

Kentani: Maquanauli, Pegler 639. Kimberley: on road to Boshof, Brueckner 29; 30 miles west of

Kimberley, Té/ken & Schlieben 1162. Komga: near Komga, Flanagan 1074. Kuruman: 4 miles

south of Olifantshoek, Té/ken & Schlieben 1181. Middelburg: Grootfontein, Theron 360. Port St.

Johns: Port St. Johns, Watt & Brandwijk 1670. Prieska: Prieska, Bryant in STE 18287 (STE).

Queenstown: Bongola Poort, Galpin 1984. Seymour: Katriviersberg, Ecklon & Zeyher sn. (SAM).

Stockenstroom: Stockenstroom, Dyer 712. Tarkastad: between Cradock and Tarkastad, Adamson

D 230. Vryburg: Pitsani, Cole in NBG 43756 (NBG). Warrenton: Warrenton, Adams 128.

ORANGE FREE STATE.—Bloemfontein: Bloemfontein, Pole Evans H 19634. Boshof: between Boshof

and Windsorton, Burtt Davy 12447. Fauresmith: Fauresmith, Smith 5594; Verdoorn 1057; Henrici

1953. Jacobsdal: Mauretzfontein, Esterhuysen 761 (BOL). Kroonstad: Kroonstad, Wagener sub

Marloth 13529.

TRANSVAAL.—Barberton: near Edwin Bray Battery, Galpin 1178. Bloemhof: Kameelpan, J. J.

Theron S 469. Brits: Crocodile River, Zeyher s.n. (BOL, SAM). Cullinao: Rooikop, Smuts and

Gillett 2003; 2115. Klerksdorp: Klerksdorp, Lucy 2 (GRA). Letaba: Tzaneen, Rogers 12411.

Lichtenburg: Grasfontein, Sutton 385. Lydenburg: Sekukunie, Barnard 188. Marico: Zeerust,

Thode A 1408. Piet Retief: Piet Retief, Leipoldts.n. Pilgrim’s Rest: Erasmus Pass, Killick & Strey

2519; Mariepskop, Van der Schijff 6374. Potchefstroom: Boskop, Louw 654; Potchefstroom, G. C.

Theron 2; Liebenberg 1073. Potgietersrus: Potgietersrus, Bolus s.n. (BOL). Pretoria: Koedoespoort,

Smith 1531; Brooklyn, Mogg 16046; Doornpoort, Pole Evans H 13262. Rustenburg: Rustenburg,

Pegler 2001. Schweizer Renecke: Burtt Davy 12776. Sibasa: Punda Milia, Van der Schijff 3595;

Lang in TRV 32200 (BOL). Vereeniging: Vereeniging Estates, Phillips s.n. Warmbaths: Warmbaths,

Burtt Davy 5341. Waterberg: Mosdene, Galpin R 236. White River: 5 miles north of Malelane,

Codd 5230: near Crocodile River Bridge, Van der Schijff 3276.

NATAL.—Camperdown: Camperdown, Franks sub Wood 11945 (BOL, NH). Eshowe: Umblatuz.

Valley, Gerstner in NH 22548 (NH). Estcourt: Estcourt, Pentz 494; West 502; Acocks 11393 (NH)i

Hlabisa: Hluhluwe Game Reserve, Ward 1755: 15 miles from Mtubatuba to Nongoma, Strey 5455.

Louwsburg: near Magut, Gerstner 2424. Lower Umfolozi: 11 miles south-west of Empangeni, Codd

1880. Umzinto: Dumisa, Rudatis 2013 (STE).

LesorHo.—Leribe: Thaba Phafiva, Dieterlen 701. Maseru: Maseru, Jacot-Guillarmod 3212.

SWAZILAND.—Lubombo: Big Bend, Compton 30992 (NBG).

SoutH West ArricA.—Grootfontein: Aris-Aukas, Dinter 730 (B, SAM); Otavipforte, Dinter 730a

(B). Otjiwarongo, Pillans in BOL 27947 (BOL).

This species has distinctive seeds, but is very variable in vegetative characters,

especially the shape and size of the leaves. The broadly oblong leaves of young plants

might be confused with those of 7. arnotii. T. caffrum, can, however, be recognized

by the presence of more than one flower per inflorescence on a short peduncle, while,

in the more tropical areas where it sometimes produces more than one flower, the

sepals are longer than 8 mm (see Fic. 1: 2).

Two sheets of this species are present in the Thunberg Herbarium (UPS) and

both bear flowers and fruit.

The identity of 7. esculentum Dinter & Schellenberg (1912) can be recognized

from the original description on the basis of the characteristic linear leaves and the

white flesh of the tuber, the combination of which applies only to this species. However,

no specimens were cited together with this description. Of the specimens cited by

Dinter (1927) under this species, two species are in Berlin Herbarium of which Dinter

730a is selected as lectotype. This specimen was collected in 1908, added to which

it bears a note recording the white flesh of the tuber, thus providing evidence that it

is probably one of the specimens on which the original description was based. The

second specimen Dinter 730 in Berlin Herbarium, belongs to the same species, but on

the sheet of this number in SAM is a mixed collection and seeds of T.esculentum and

T. tenuissimum are present in the capsule. TJ. esculentum is a synonym of T. caffrum,

but t.42 in Neue und wenig bekannte Pflanzen Deutsch-Siidwest-Afrikas appears to be

a narrow-leaved form of 7. arnotii. Several lateral flowers on a long peduncle and the

recurved, narrowly ovate leaves are rarely found in T. caffrum. From the few specimens

seen, it appears that 7. caffrum is not as common in South West Africa as Dinter

(1912) estimated.

3. T. arnotii Hook.f. in Bot. Mag. t. 6220 (1876); von Poellnitz in Fedde Rep.

65581934) Waldtine kine Zambs1he2372 (196) 5" hype: Bot. Mag. t) 62202)

caffrum sensu Dinter, Veget. Feldkost D.S.W.A. 12 (1912); Fedde Rep. 23: 369 (1927);

sensu von Poellnitz in Fedde Rep. 35: 13 (1934), partly, including specimens from

S.W.A.; sensu Podlech in Prodr. Fl. S.W.A. 29: 11 (1967). 7. cuneifolium sensu von

Poellnitz in Fedde Rep. 35: 15 (1934), partly, including specimens from S.W.A.;

sensu Podlech in Prodr. Fl. S.W.A. 29: 12 (1967).

Shrublet, semi-erect, rarely decumbent; tubers long, branched, usually reddish-

brown inside. Branches slightly hairy, becoming glabrous, succulent, 30-40 (-60) cm

long. Leaves narrowly or broadly ovate or elliptic, mucronate, 2-4 (-5) cm long,

0-8-1-5 (2-5) cm broad, with margins often irregularly recurved, appearing wavy;

petiole 2-3 (-4) mm long. Inflorescence axillary, cymose, 1—3-flowered. Peduncle

2-3 (-4) cm long ending in two keeled bracts with membranous margins. Pedicel

0-5-1 cm long, swollen below the ovary, usually with lateral flowers with one pair of

bracteoles. Sepals 5-6 (7) mm long, ovate, keeled and slightly hooded at the apex,

with membranous margin. Petals (4) 5, ovate, mucronate, faintly keeled at the apex.

Stamens 20-30 (—50); filaments slightly broadened towards the base, fused into a ring

c. 1 mm high. Ovary conical with 25-40 ovules; style 2-3 (-4) mm long dividing

into 3 stigmas 2-3 mm long, papillose. Capsule ovate, 8-9 mm long, 5-6 mm broad,

shiny yellow. Seeds reniform, laterally compressed, 0:5-1-5 mm long, with fine

elongate papillae or coarse papillae arranged in circular rows, dark brown or black.

Fic. 1:1, la.

This species occurs usually in sandy, but often also in clay soils, and is distributed

in and around the Kalahari basin and well into the centre of South West Africa.

Cape.—Barkly West: Boetsap, Brueckner 1263. Hay: La Dauphine, Acocks 227; Foot of Wol-

kaarskop, Esterhuysen 2451 (BOL); Dronfield, Wilman in KMG 1934(KMG). Kimberley; Spytfontein,

Schweickerdt 1119; near Rivermead Station, Leistner 2017; Kimberley, Compton 23945 (NBG); 25

miles west of Kimberley, Talken & Schlieben 1160. Kuruman: 4 miles south of Olifantshoek, Télken

& Schlieben 1182. Postmasburg: Klipbok, Repton 4798. Prieska: Prieska, Bryant 567; J. 15.

Vryburg: near Vryburg, Rodin 3525; 4 miles south of Vryburg, Brueckner 1114 (KMG). Warrenton:

Warrenton, Pole Evans H. 11628.

ORANGE FREE STATE.—Fauresmith: Roodepoortje, Henrici 3033. Jacobsdal: Mauretzfontein,

Esterhuysen 762 (BOL).

TRANSVAAL.—Groblersdal: Marble Hall, Marais 1099. Pietersburg: north of Blauwberg, Obermeyer,

Schweickerdt & Verdoorn 93. Potgietersrus: Steilloop, Té/ken 1126b. Thabazimbi : Near Northam,

Télken 1280; near Makoppa, Té/ken 1281; 1284. Soutpansberg: between Salt Pan and Waterpoort,

Obermeyer, Schweickerdt & Verdoorn 260.

SoutH West ArricA.—Bethanien: Aris, Range 1280 (SAM). Gobabis: 80 miles west of Gobabis,

De Winter 2514: Sandfontein, Gillman 69 (SAM); Sturmfeld, Té/ken 1301. Grootfontein: Groot-

fontein, Schoenfelder 310; Borle 49. Kaokoveld: 4 miles north of Ohopoho, De Winter & Leistner

5259; Kaross-Kamanjab, Thorne in SAM 35750 (SAM). Karibib: Okomitundo, Seyde/ 1270. Keet-

manshoop: Kraikluft, Pearson 8208 (BOL); 8269 (BOL, GRA). Okahandja: Okahandja, Dinter

1851 (SAM). Okavango: Runtu, De Winter 4043; Nama Pan, Story 5256; Gautscha Pan, Maquire

2133. Otjiwarongo: Quickborn, Bradfield 110. Waterberg, Liebenberg 4752. Outjo: Outjo, Barnard

in SAM 33224 (SAM). Rehoboth: Rehoboth Fleck 182 (Z); between Rehoboth and Uhlenhorst,

Wilman in KMG 463: Biillspoort, Herre s.n. (BOL). Warmbad: near Dabaigabis, Pearson 4433

(BOL). Windhoek: 4 miles north of Windhoek, Codd 5787; Neudamm, Van Vuuren 1014; Gameros,

Dinter 2671b (SAM).

T. arnotii is a very variable species. The leaves vary greatly in shape and size

and are crisped along the margin when young. Among specimens of this species seen,

two types of seeds were observed. Firstly, plants originating mainly from the northern

Cape and southern and central South West Africa produce seeds with fine elongate

papillae radiating out from the hilum. These seeds are very similar to those of T.

tenuissimum and T. crispatulum, but are never longer than 1:5 mm. The second form

occurs in the northern parts of the Cape Province, Transvaal, Botswana and South

West Africa and also in Rhodesia and Zambia. The papillae of the seed membrane

are coarse, about as long as broad and arranged in concentric rows. In the northern

parts of the distribution of this form, the papillae become less clearly arranged in

concentric rows and the seeds are usually less than | mm long. In the Transvaal the

two forms were found together at two localities (near Thabazimbi Té/ken 1284, 1281;

near Makapanstad, Tod/ken 1278, 1277). The corresponding plants can usually also

be distinguished. The plants with finer papillae occur on the more clayey soils and

produce narrower leaves and normally one, rarely many, flowers on a long peduncle

(see Fic. 1: la). The plants with coarse seed papillae occur on sandy soils and have

broad leaves usually irregularly reflexed and usually with three flowers per inflorescence

(see Fic. 1:1). However, intermediates to all these characters, except the seed

characters, were found in both localities and also in herbarium material. Specimens

seen from north-eastern South West Africa and northern Cape also indicate that the

two forms apparently occur together in these areas without any definite intermediates.

In addition, specimens from the eastern central South West Africa have seeds with fine

papillae approaching those of the second type. Specimens with definite intermediate

seed characters have not been seen, but in view of insufficient material seen from

Botswana, the complex is not thoroughly understood and can thus not be fully

evaluated.

4. T. crispatulum Dinter in Fedde Rep. 23: 369 (1927); von Poellnitz in Ber.

Bot. Dt. Ges. 51: 117 (1933); Fedde Rep. 35: 18 (1934); Podlech in Prodr. Fl. S.W.A.

29: 12 (1967); Wild in Fl. Zamb. 1, 2: 372 (1961), as T. crispatulatum, partly. Type:

Grootfontein: Palmenhain, Dinter 2357 (SAM, lecto.!).

Plant dioecious, prostrate or decumbent; tubers branching from the top, with

white flesh. Male plants with branches 25-40 cm long, little branched, papillose to

hairy when young, becoming almost glabrous; internodes rarely longer than 1 cm,

varying greatly in length on the same branch. Leaves narrowly ovate, 1-1-5 (—2) cm

long, 4-6 (-8) mm broad, with margins crisped, never revolute; petiole 1-2 (-3) mm

long. Inflorescence axillary, 2—4 (—S)-flowered. Peduncle 2—3 (—S) cm long, thread-like,

ending in two keeled bracts with membranous margin. Pedice/ 1-1-5 cm long, of the

central flower with 0 or | pair of bracteoles, of the lateral flowers with (1) 2 (3) pairs

of bracteoles. Sepals 2-4 (-5) mm long, lanceolate, keeled and slightly hooded at the

apex, with membranous margin. Petals 5, lanceolate, tapering to the base, yellow.

Stamens 20-30 (—35); filaments broadened towards the base, fused into a ring c. | mm

high. Ovary almost spherical with no ovules developed; style absent; three stigmas

c. | mm long, stiff erect, not papillose. Female plants with branches 20-30 cm long,

branched, papillose to hairy when young becoming almost glabrous; internodes rarely

longer than | cm, varying in length on the same branch. Leaves lanceolate, 1 -5—2-5 (—3)

cm long, 0-6-1 (-1:5) cm broad, with margins crisped, never revolute; petiole 1-3 (4)

mm long. Jnflorescence axillary, 1-2 (3)-flowered. Peduncle 1-1-5 (—2-5 cm long,)

thread-like, ending in two keeled bracts with membranous margin. Pedicel 0-5—-1-5

cm long, swollen below the ovary, of lateral flowers with 1 (2) pairs of bracteoles.

Sepals 3-4 (-5) mm long, ovate, keeled and slightly hooded at the apex, with mem-

branous margin. Petals 5, ovate, pointed, sometimes uneven, yellow. Staminodes

25-30, fine; filaments broadened towards the base, usually fused into a ring. Ovary

almost spherical with 10-12 ovules; style absent or less than 1 mm long; stigmas 3,

2 (-3) mm long, papillose-plumose. Capsule conical, pointed, 5-6 mm long, 3-4 mm

broad shiny yellow. Seeds reniform, laterally compressed, 2-2-5 mm long, with fine

elongate papillae radiating out from the hilum, dark brown or black. Fic. 1: 3, 3a.

This species is common in sandy soils in and around the Kalahari basin.

Capre.—Barkly West: Newlands, Esterhuysen in NBG 5474 (NBG). Hay: Rietkloof, Acocks 8552

(BOL, GRA); Bermolli, Wilman s.n. (GRA, KMG). Kuruman: Kuruman, Esterhuysen 759; 2436

(BOL); Batharos, Si/k 228 (KMG); 4 miles south of Olifantshoek, Tol/ken & Schlieben 1190; 1191;

Cotton End, Swan in KMG 5071 (KMG). Postmasburg: Klipbok, Repton 4794. Vryburg: between

Armadillo Creek and Vergelegen, Burtt Davy 11166; Armadillo Creek, Burtt Davy 11180.

TRANSVAAL.—Potgietersrus: Steilloop, To/ken 1126a. Thabazimbi: near Makoppa, Télken 1282;

1283.

SoutH West ArFricA.—Gobabis: Gobabis, Dinter in SAM 74148 (SAM); Sandfontein, Wilman

in SAM 27105 (SAM); Uitsig, Merxmiiller & Giess 1105; Sturmfeld, 7é/ken 1302. Grootfontein,

Palmenhain, Dinter 2357 (SAM). Okavango: east of Karakuwise, Maguire 2065 (NBG); Gautscha

Pan, Maguire 2133 (NBG). Otjiwarongo: Quickborn, Bradfield 361.

I do not agree with Podlech that the author of this species should be “‘ Dinter ex

von Poellnitz ’’ as the crisped leaves and smaller flowers mentioned in Dinter’s diagnosis

plus the type cited amount to more than a nomen nudum (see Int. Code of Bot. Nomen-

clature, Article 32, 1966).

Wild (1961) uses the name T. crispatulatum consistently, but there seems to be no

reason to change the epithet from the originally published version, T. crispatulum.

5. T. tenuissimum Dinter, Neue Pl. D.S.W.A. 55 (1914); Fedde Rep. 23: 369

(1927); von Poellnitz in Fedde Rep. 35: 18 (1934); Podlech in Prodr. Fl. S.W.A.

29: 12 (1967), partly, excluding specimens cited. Type: Kalkfontein—Gobabis, Dinter

2725 (B, lecto.!; SAM!). T. transvaalense von Poellnitz in Fedde Rep. 35: 34 (1934).

Type: Transvaal, Sibasa, Baiandbai, Lang in TRV 32199. T. dinteri von Poellnitz

in Fedde Rep. 35: 32 (1934), nomen subnudum. Type: Jakkalskuppe, Dinter 4204

(B, holo.!). 7. crispatulatum sensu Wild in Fl. Zamb. 1, 2: 372 (1961), partly.

Plant semi-erect or decumbent; tubers rarely longer than 5 cm, rarely branched,

with reddish-brown flesh. Branches glabrous, 15-20 (—30) cm long. Leaves linear,

2-4 (-5) cm long, 2-4 mm broad, margins revolute; petiole 1-2 (-3) mm long.

Inflorescence axillary, 1-2 (3)-flowered. Peduncle 0:4-1-5 mm long, thread-like, ending

in two keeled bracts with membranous margin. Pedicel 1-2-5 cm long, swollen below

the ovary, of central and lateral flowers with (0) 1 or 2 pairs of bracteoles often uneven

long. Sepals 3-4 mm long, keeled and slightly hooded at the apex. Petals 5, oblong,

obtuse, mucronate, hardly tapering towards the base, yellow. Stamens 8-14; filaments

slightly broadened towards the base, not fused into a complete ring. Ovary almost

spherical with 10-12 ovules; style c. 1 mm long, swollen, dividing into three spreading

stigmas c. 1 mm long, papillose. Capsule conical, pointed or almost beaked, 6-7 mm

long, 3-4 mm broad, shiny yellow. Seeds reniform to comma-shaped, much laterally

compressed, 2-2-5 mm long with fine elongate papillae radiating out from the hilum,

dark brown to black.

T. tenuissimum occurs on sandy Kalahari soils in South Africa, but its distribution

extends into Mozambique, Rhodesia, Zambia and Tanzania.

Cape.—Barkly West: Newlands, Esterhuysen 760 (BOL). Gordonia: Tweegesigspan, Liebenberg

7049. Hay: Tygerkop, Wilman in KMG 2347 (KMG). Kimberley: 4-5 miles from Kimberley to

Boshof, Esterhuysen 763 (BOL). Kuruman: near Kuruman Esterhuysen 2437 (BOL).

TRANSVAAL.—Brits: Welgefonden, Obermeyer in TRV 34762. Pietersburg: Swerwerskraal, /rvine

110; Herre in BOL 27950 (BOL). Sibasa: 28 miles west of Punda Milia, Lang in TRV 32199. Sout-

pansberg: north of Salt Pan, Schlieben 9216; Obermeyer, Schweickerdt & Verdoorn 193.

33261—3

SouTtH West AFricA.—Bethanien: Kuibis, Dinter 1194 (SAM). Gobabis: Gobabis—Kalkfontein,

Dinter 2725 (B, SAM); Gobabis—Kehopo, Dinter in SAM 74153; Babi-Babi, Wilman in SAM 27106a

(SAM). Lideritz: Jakkalskuppe, Dinter 4204 (B). Okahandja: Okakeva, Dinter 3315 (SAM).

Okavango: Kapupahedi Camp on Omuramba Omatako, De Winter & Marais 4632; Gautscha Pan,

Maguire 2195 (NBG). Otjiwarongo: Quickborn, Bradfield 320B. Rehoboth: Klein Nauas, Dinter

1937 (SAM). Windhoek: Gameros, Dinter 2671a (B, SAM).

Wild (1961) includes this species under T. crispatulatum and his Tab. 71, Al depicts

a plant of typical T. tenuissimum. Although the seeds of the two species are very

similar, T. tenuissimum has flowers with 8-14 stamens and linear leaves with revolute

margins. In cultivation (Dinter in SAM 74152), the leaves become slightly broader,

but they retain their linear shape. In T. crispatulum the flowers always produce more

than 20 stamens and the narrowly ovate leaves have crisped, never revolute margins.

T. dinteri von Poellnitz, a nomen subnudum based on Dinter 4204, is very similar

to T. tenuissimum in that it produces linear leaves, more than one flower per peduncle,

c. 12 stamens and 10 ovules. Unusual for 7. tenuissimum is, however, the single pair

of bracteoles on the pedicel, but this character is apparently variable as the absence

of additional pairs of bracteoles can also be observed in the specimen Dinter 1794

(SAM), a specimen apparently growing under similar unfavourable conditions. This

latter specimen, collected on sand dunes in the Bethanien district, provides a link

between the distribution as well as the morphology of Dinter 4204 and the Kalahari

population of T. tenuissimum.

The specimen Pearson 4162 (K!) collected at Jakkalskuppe confirms the identi-

fication of this plant as T. tenuissimum as it bears the typical seeds. This specimen

is inscribed T. lissospermum C. A. Smith MS.

INDEX

Claytonia caffra (Thunb.) Kuntze, 23 cuneifolium Willd., 23

Portulaca caffra Thunb., 23 cuneifolium sensu Podlech, 25

cuneifolia Vahl, 21 dinteri von Poellnitz, 27

Orygia portulacifolia Forsk., 21 esculentum Dinter & Schellenberg, 23

Talinum Adans., 19 minus Eckl. & Zeyh., 23

arnotii Hook.f., 25 paniculatum (Jacq.) Gaertn., 21

caffrum (Thunb.) Eckl, & Zeyh., 23 portulacifolium (Forsk.) Aschers. ex Schweinf.,

var. minus (Eckl. & Zeyh.) Sond., 23 21

caffrum sensu Dinter, 25 tenuissimum Dinter, 27

crispatulum Dinter, 26 transvaalensis von Poellnitz, 27

crispatulatum sensu Wild, 27 triangulare (Jacq.) Willd.. 21

Bothalia 10, 1: 29-37

A New Species of Raphia from Northern Zululand

and Southern Mozambique

A. A. Obermeyer and R. G. Strey

ABSTRACT

The Raphia Palm from northern Zululand and southern Mozambique has been known for many

years as Raphia vinifera Beauv., but an investigation of flowering material revealed that it was distinct

from that species and represented a new species, which is here described as R. australis Oberm. & Strey.

In addition to the description, notes on the morphology, history, preservation and uses of the plant are

given.

DESCRIPTION

Raphia australis Oberm. & Strey, sp. nov., R. vinifera Beauv. affinis, sed inflorescentia

terminali erecta fructu ellipsoideo squamis convexis leviter sulcatis differt.

Caudex simplex. Spadix erectus terminalis spicis fertilibus patentibus tortilibus

floribus femineis in dimidio inferiore. Flos masculus staminibus 6 liberis conniventibus.

Flos femineus annulo staminodio dentibus inaequalibus aliquot antheris minutis. Fruc-

tus ellipsoideus squamis convexis 6-orthostichis leviter sulcatis.

Type: Natal, Ingwavuma District, Kosi Bay area, west of Lake Amanzimnyana,

6:3 miles east of Maputa, edge of dense forest which is inundated in the rainy season,

November 1967, Strey 7785 (PRE, holo.; NH).

A large unbranched stout tree up to 16 m tall. Stem up to 10 m tall, covered with

the old persistent adpressed erect leafbases. Leaves rosulate, younger erect, outer

somewhat spreading, pinnate, up to c. 10 m long; petiole unarmed concave and clasping

below; the rhachis narrows into a canaliculate stalk for about two thirds of its length

with the two spiny margins unequal in height gradually approaching one another until

they merge (the lower ridge disappearing); at this point the leaf bends outwards; the

pinnae are inserted on the outside of the marginal ridges, linear, 45-65 cm long, 5 cm

broad, folded below, pale green above, waxy below, the margins bearing small sharp

ascending pale spinules about 4 mm long, midrib raised with a double row of spinules

in the lower half which merge to form a single row above. Inflorescence apical, erect,

exserted centrally above the crown of leaves, conical, up to 3 m tall, its lowest flowering

branches in the axils of the upper leaves; the subsequent leaves (c. 13) subtending the

flowering branches, gradually reduced, the uppermost c. 50 cm long; peduncle 11 cm

in diam. at the base, covered by 2-ranked, clasping imbricate tubular hard bracts; the

basal bract of each branch differing from the following ones in shape; it is 2-pronged

like a lobster-claw, flattened and bi-carinate, up to 40 cm long and 6 cm broad; it recurs

on all the following side-branches becoming progressively smaller higher up; the sub-

sequent bracts tubular with a truncate apex which forms an acute point, the points

arranged alternately from side to side; branches exserted, laterally compressed, sub-4-

ranked with 2 side-branches close together on each side of the rhachis, erect at first,

patent in fruit, bracts similar to those covering the peduncle but reduced in size: the

side branches bearing the numerous abbreviated fertile spikes in close proximity to one

another; their arrangement is also sub-4-ranked with two rows of spikes close together on

each side of the laterally compressed rhachis; fertile spikes spreading and twisting, with

the unisexual flowers 2-ranked, the female occupying the lower part of the spike, the

male the upper part; the apical spikes with male flowers only; the lower spikes up to

40 cm long becoming shorter towards the apex; fertile bracts similar in shape and form

to those of the branches but smaller. Male flowers surrounded by a bi-carinate bracteole;

hiding inside it is a small tubular, truncate calyx and exserted from it a corolla consisting

of 3 hard linear-falcate segments which are nearly free to the base, 10 mm long, 2 mm

broad, with a triangular, slightly swollen apex; stamens 6, free, filaments connivent,

swollen, 4 mm long, 3-angled, apiculate, anthers with parallel linear locules 6 mm long.

Female flowers surrounded by a bi-carinate bracteole and a second one inside it which is

short, tubular and truncate, calyx and corolla enlarging after fruit has set, exserted from

the bracteoles; calyx tubular, splitting irregularly into 3 truncate lobes when the ovary

enlarges, 8 mm long; corolla segments 3, hard, beaked; staminodial ring adnate to the

corolla, flattened, with unequal blunt teeth, some bearing minute anthers; ovary 3-celled,

uni-ovulate, with the minute scales fimbriate; stigmas 3, sessile, apical, tongue-shaped,

erect, at anthesis exposed through a small apical aperture of the corolla before the latter

enlarges. Fruit one-seeded, ellipsoid, 6-9 cm long and 3-5 cm in diam. hard, shiny

yellow-brown, shortly beaked, the reflexed convex scales arranged in 6 vertical rows,

median groove shallow, edges with a short golden fringe of hairs.

R. australis is found in swamp forests near the coast in northern Zululand and

southern Mozambique.

NATAL.—Ingwavuma: Kosi Bay area, west of lake Amanzimnyana, 6:3 miles east of Maputa,

Strey 7785; Strey 8200; De Winter & Vahrmeijer 8617; Cult: Mtunzini, Garland in PRE 30368;

Durban Botanic Station, Strey 7373; 7800; 8060; 8263.

Dr. Gomes e Sousa in a personal communication, 19th February 1969, reported

seeing it in fruit on the Boboli River near Marracuene, 40 km north of Lourenco

Marques. It is called “‘ Imali’’ in Ronga.

In the past 40 years much has been written on the Raphia Palm from Zululand but,

owing to its size, no complete collections of the inflorescence (only an occasional nut)

or photographs of flowering plants have reached herbaria. It was only when two palms

in the garden of the Botanic Station in Durban, which were raised from seed collected

by Dr. V. E. Wager at Mtunzini, developed erect inflorescences, that it was realized that

we were dealing with an undescribed species. Before this, the Raphia Palm from

Zululand was referred to R. vinifera, but this species bears pendulous inflorescences in

the axils of the leaves and its fruits too, have a different shape.

MORPHOLOGICAL NOTES

Roots.—(a) Pneumatophores. Around its base the palm forms negatively

geotropic roots up to 10 cm high, exserted above the swampy ground which is

flooded seasonally. (b) Epigeal rootlets. In the axils of the lower leaves rootlets

are developed, which penetrate into the fibrous mass seeking moisture and nutrients.

The thickness of this root-mantle around the lower part of the stem may be up to

5 cm in thickness.

Leaves.—About two new leaves are formed annually as observed on the palm

growing at the Botanic Station, Durban. The dying old leaves gradually fall down

backwards leaving only a short clasping leaf-base c. 60 cm wide below. The rhachis

is canaliculate for about two thirds of the way up and the rainwater thus runs into the

axils, which are filled with humus and harbour epiphytes such as the ferns Stenochlaena

tenuifolium, Lygodim microphyllum and Psilotum nudum.

Sill

Inflorescence.—The lowest branch of the inflorescence measured up to 3 m long and

had a circumference of 50 cm at the base. The terminal shaft was estimated to be c.

5 m tall.

Fruits.—About 8-10,000 fruits are developed on an inflorescence. The weight

of 50 fruits averaged about 3lb. Therefore a fruiting inflorescence may weigh between

five and six hundred pounds.

Seed.—The yellow spongy integument under the scales is sweet and sticky and

attracts the Palm-nut vultures. The structure of the seed and embryo is typical of the

family. The small anatropous embryo has the funiculus situated in a groove of the

pericarp where the two aborted locules meet. The endosperm is ruminate and the

cotyledon forms a haustorium, which extracts the food for the embryo.

Rate of Growth.—The Raphia Palm is said to reach a height of about 40-50 feet in

20-40 years from seed and then flowers and dies after setting fruit. The specimens

growing at the Botanic Station in Durban flowered after 24 years. Regeneration from

seed is excellent, but nearly always confined to the immediate neighbourhood of the old

palms. It does not sucker.

HISTORICAL NOTES

Some of the early history of the Mtunzini Grove is related in an interesting article

by Mr. W. M. Austen, a ranger of the Natal Parks, Game & Preservation Board, in

The Ostrich, September, 1953, where he described his observations made on Palm-nut

Vultures that nest in the Raphia Palm and feed on the fruits. He remembered seeing

the Mtunzini Grove in 1915; it was established by the late Mr. C. C. Foxon, who was

Magistrate and Native Commissioner at Mtunzini and who had collected the seeds

near the Maputa village in Tongaland. It occurs there naturally in the fresh-water

swamp-forests at the southern end of the five Kosi lakes, either as isolated trees or forming

small groups of about 20 individuals. This swamp-forest type of vegetation is also

found further northwards in Mozambique and Portuguese botanists believe that it

constitutes the climax vegetation of this area.

PRESERVATION

The palm grove near the Mtunzini Railway Station has been declared a national

monument. Solitary individuals or small groups are however found in the neighbour-

hood; in one grove about thirty individuals were counted. Further north in Tongaland

they are not protected.

USES

The Natives use the old leaf-midribs for hut building and rafts. At the bus terminal

at Maputa is a fenced-in market place where market-stalls and communal shelters have

been constructed from the leaf-midribs. Rafts are also made for crossing the Sihadla

River (see photos). The young leaves are not stripped for raffia as in R. farinifera, nor

is the sap tapped for making wine as in R. vinifera and other species.

REFERENCES

AITKEN R. D. & GALE G. W., (1912). Botanical Survey of Natal and Zululand. Mem. Bot. Sury.

S. Afr. No. 2:17, t. 1. hay

AUSTEN, W. M., (1953). Palm-nut Vultures (Gyophierax angolensis) in Raphia Palms at Mtunzini,

Zululand. The Ostrich, September 1953.

RusseELL, T. A., (1965). The Raphia Palms of West Africa. Kew Bull. 19:173.

Fic. 1.—Raphia australis. 1, flowering spike showing female flowers in lower half and male flowers

above, x 4; 2, young female and male flowers, x 4; 3, male flower showing bract (6), bracteole (br)-

calyx (ca) and corolla (co), x 2; 4, longit. section of male flower showing 2 petals and 3 stamens

x 2; 5, female flower showing bract (5), bracteole (6), inner bracteole (7), calyx (8), corolla and

staminodial ring (9) and ovary (10), all x 2; 11, longitudinal-section of a young fruit with embryo, x4;

12, cross-section of a young fruit, « 4; 13, cross-section of ruminate endosperm, x 4; 14, fruit, x

4; 15, 16, 17, 18, cross-sections of the leaf-rhachis at various points from base to top, (the pinnae

cut off), <x 4; 19, seed, showing two lateral sterile locules and the groove containing the funiculus,

x 4; 20, young seedling, cross-section of haustorium and endosperm, x 4; 21, seedling showing

remains of yellow integument and pericarp, x 4. 1-10, Strey 7373; 11-13, 19, Strey 7785;

14, Garland in PRE 30368; 15-18, Strey 7373; 20-21, Strey 7785.

PLATE 1.—Raphia australis. 1, part of flowering inflorescence, branch of second order. Natal Her-

barium garden, 30th May 1967 (Strey 7493). 2, same as 1, but lower down showing part bearing

female flowers and above male. 3, ripe fruit from Mtunzini Grove, 27th February 1967 (Garland

in PRE 30368). 4a, ripe fruits showing scales; 4b, inner yellow sticky, sweetish integument; 4c,

seed showing 2 aborted carpels and funicle in vertical furrow; 4d, seed cut open showing hard

endosperm and resinous intrusions, Lake Amanzimnyana, 20th November 1967 (Strey 7785

3879); 5, seedlings showing haustorium, root and shoot, from Lake Amanzimnyana. Photos:

1-3, H. J. Schlieben; 4,5, R.G. Strey.

AN... Sephroradoonty’

PLATE 1

PLATE 2.—Raphia australis in flower in the PLATE 3.—Lower part of trunk of palm in

Natal Herbarium garden Durban, May Plate 2 with leaf-bases removed to

1967. Photo: E. J. Moll. show epigeal roots. March 1967.

Photo: R. G. Strey.

PLATE 4.—Flowering tree at Lake Aman- PLATE 5.—Single leaf of a tree from Lake

zimnyana near Maputa. October Amanzimnyana. October 1967.

1967. Photo: B. de Winter. Photo: B. de Winter.

Si1/

PLATE 6.—Grove of Raphia palms growing on the west side of Lake Amanzimnyana. November

1967. Photo: E. J. Moll.

PLATE 7.—Raft built by the Tongas from PLATE 8.—Bus terminal and market place,

the midribs of the Raphia leaves. Maputa, showing communal buildings

Sihadhla River. Photo: E. J. Moll. and stalls made of midribs of Raphia

leaves. Photo: E. J. Moll.

7 : a : i o y

< \ i 7 - D3 : mn

7 =

‘ p -

; wer _

a ox i

mT {

Bothalia 10, 1: 39-42

The Disc in the Southern African Species of

Maerua

D. J. B. Killick

ABSTRACT

The six types of disc present in the Southern African species of Maerua (Capparaceae) are described

and illustrated. A key to the species, based largely on the disc character, is presented.

While working on the genus Maerua for the Flora of Southern Africa, the author

was struck by the variety of disc forms present in the Southern African species of the

genus. Discs of various kinds are present in many genera of the Capparaceae. They

are Well reviewed by Pax in Engler and Prantl’s Naturlichen Pflanzenfamilien, Teil 3,

Abt. 2:215-217 (1891) and Pax and Hoffmann, op.cit., Band 17b:156-157 (1936). In

Maerua the inner margin of the receptacle is produced into a disc, which in many

species projects beyond the receptacle as a corona, annulus, lobes or teeth. In this

paper the author has followed Wild in Flora Zambesiaca 1, 1:216 (1960) and Elffers

et al. in Flora of Tropical East Africa (1964) in treating the projecting part of the recep-

tacle as the disc. Strictly speaking, the inner margin or lining of the receptacle should

also be regarded as part of the disc.

In the 11 species of Maerua in Southern Africa there are six types of disc. These will

be discussed in turn and are illustrated in Fig. 1.

1. Fimbrillate disc

This type of disc consist of a row of regular fimbrillae not more than | mm long.

The disc is somewhat fleshy and the fimbrillae are mostly erect, but some are incurved.

Actually the disc is very shortly coronate or annular. This disc is present only in M.

brevipetiolata Killick, a new species from Swaziland described on page 65 of this number

of Bothalia.

2. Four-lobed laciniate disc

This type consists of dorso-ventrally compressed, unequally laciniate lobes situated

opposite the four sepals. The lobes are connected basally. This type of disc is repre-

sented in M. rosmarinioides (Sond.) Gilg & Ben., M. juncea Pax (both subsp. juncea

and subsp. crustata (Wild) Wild and M. nervosa (Hochst.) Oliv.

3. Many-lobed laciniate disc

The individual lobes of this type are laterally compressed, incurved and laciniate.

In side view the lobes resemble cockcombs. Only M. gilgii Schinz possesses this type

of disc.

4. Prominently coronate disc

In this type the disc consists of a prominent corona usually exceeding 2 mm in

length with several types of margin. In M. edulis (Gilg & Ben.) De Wolf [=Courbonia

glauca (Klotzsch) Gilg & Ben.] the margin is undulate or denticulate, while in M. schinzii

Pax and M. angolensis DC., it is unequally laciniate.

Fic. 1.—Disc structure in the Southern African species of Maerua. 1, Maerua brevipetiolata, (Compton

30088). 2, M. rosmarinoides, (Tinley 587). 3, M. juncea subsp. crustata, (Theron 2073). 4, M.

nervosa, (Forbes & Obermeyer 33624). 5, M. gilgii, (Hardy 596); Sa, single lobe of disc, x 10. 6, M.

edulis, (Obermeyer, Schweickerdt & Verdoorn 328); 6a, (Vahrmeijer 1525). 7, M. schinzii, (Keet

1534). 8, M. angolensis, (Van der Schijff 5799). 9. M. cafra, (Louw 1022). 10, M. racemulosa,

(Killick 118); 10a, (Borquins.n.). 11, M. parvifolia, (Lang in TRV 31580); 11a, (Van der Schijff 87).

All x 4, except 5a.

ALEIDA VAN DER MERWE

FIG. 1

The main difference between this type of disc and that of (2) above is that the latter

is 4-lobed. It is interesting to note that the species with a 4-lobed disc are petaloid,

whereas those with a coronate disc are apetalous.

5. Entire or scalloped disc

Here the disc is not or scarcely produced beyond the receptacle and the margin is

entire or scalloped with the concavities opposite the petal bases. M. cafra (DC.) Pax

possesses a scalloped disc, M. racemulosa (A.DC.) Gilg & Ben. an entire or scalloped

disc and M. parvifolia Pax an entire or very rarely a minutely denticulate disc.

6. Denticulate disc

As mentioned under (5), this disc is sometimes possessed by M. parvifolia.

The disc has been used as a character for distinguishing genera, e.g. Pax, and Pax

and Hoffmann (/.c.), but apparently rarely for distinguishing species. It was found

possible, as an academic exercise rather than for practical purposes, to construct a key

to the species of Maerua using the disc character. However, since some species have

the same type of disc, additional floral characters had to be used.

Disc not produced: entire or scalloped:

Sepalsilessithankonmmsloneanenane cece eee Ueno eecaneGricn ace micniceee ier: M. racemulosa

Sepals more than 6 mm long:

SepalsnGsidiirmmbl Om Bi. ceyerca ces rerecawstorste wea mene secu gears omen eect eo atane he Cte ne eA M. parvifolia

Sepalsmi2Q=l7imimelongaceycmree sve cenee ee eae ie cen teak ces teem Ieee eee eae M. cafra

Disc produced into corona (or annulus), lobes or minute teeth:

Disc coronate (or annular):

ID FeO ECs Lior 00W) Lolo o nicreveen niie cece ian cececeniaoeeoie renin o miciectcatoa cies ccc oc maa c M. brevipetiolata

Disc exceeding 1 mm long:

Marin Of Glee wineliate Or CMW. o5550000000060000c0000000000000000006 M. edulis

Margin of disc laciniate:

Receptacle infundibular or cylindric, 0-5-1-2 cm long, usually puberulous.... M. schinzii

Receptacle cylindric, 1-0-1-6 cm long, usually glabrous................... M. angolensis

Disc lobed or denticulate:

Disc lobed:

LOWES menny, ikicmllhy Comyoresssel., oooccoccscv900000000000000cC0000000C0000 M.. gilgii

Lobes 4, dorso-ventrally compressed:

DiseJess: ‘thant vh ammplon gee severe ececcesie cere ates ciate aiansvenstecsensenaecsie eyes M. nervosa

Disc 1 mm or more long:

Receptacle cy limdricin ne ecco enc gots Soros yensononc neds te ee eua (a cierer ae eayre rave iaenneks M. rosmarinoides

RGSS] stavitivralowW EP, ooccodbagovccdo00 Do s0NEs000000000G0 000000000006 M. juncea

ID Rosca ott Clie acta no cdiolcoroicicctamelan cic corer o anc ncecr eet ctad cic M. parvifolia

Bothalia 10, 1: 43-44

A Note on Erythrophleum R. Br. in South Africa

J. H. Ross

ABSTRACT

The Natal specimens of Erythrophleum have in the past been variously referred either to E. /asianthum

Corbishley or to E. suaveolens (Guill. & Perr.) Brenan (=E. guineense G. Don). It was found that all

specimens are referable to E. lasianthum and that E. suaveolens is absent from southern Africa. E.

guineense G. Don var. swaziense Burtt Davy was found to be a synonym of E. lasianthum. E. lasianthum

and E. africanum (Benth.) Harms are the only two species encountered in southern Africa. A synopsis

of the differences between these two species is given.

Whilst preparing the Caesalpinioideae for the revision of The Flora of Natal and

Zululand (Bews, 1921) irregularities in the naming of specimens of Erythrophleum

became apparent. The Natal specimens, although fairly uniform, have in the past

been variously referred to EF. /asianthum Corbishley or to E. suaveolens (Guill. & Perr.)

Brenan (—E£. guineense G. Don). Consequently it was necessary to establish the identity

of the Natal specimens.

The stamen filaments in the Natal specimens are woolly tomentose to near the apex

and cannot therefore be referred to E. suaveolens, which has glabrous stamen filaments.

The Natal material is therefore all referable to E. lasianthum.

Burtt Davy in Fl. Transyv. 2:330 (1932) based his E. guineense var. swaziense on a

specimen collected by Nicholson in Swaziland (without a precise locality). This variety

differed from typical E. guineense “in the much smaller and relatively broader leaflets,

which are more obtuse and rounded at base, and less acuminate at apex.’ Following

the placing of E. guineense as a synonym of E. suaveolens by Brenan in Taxon 9:194

(1960), E. guinzense var. swaziense has been regarded as a synonym E. suaveolens

[De Winter et al. in Sixty Six Transvaal Trees:170 (1966) ].

The type of var. swaziense is a fruiting specimen. However, four other specimens

from Swaziland, all from the Stegi district, have been examined. Of these, one collected

by the Assistant Commissioner H 30333 (PRE) in Nov. 1924, which is vegetatively indis-

tinguishable from the type of var. swaziense, is in flower. The stamen filaments are

woolly tomentose to near the apex thus proving the specimen to be referable to E.

lasianthum. Since no other species of Erythrophleum is present in Natal, Swaziland or

in southern Mozambique whilst none is present in the Transvaal, it is assumed that

Burtt Davy’s type specimen is also referable to E. /asianthum. There is certainly no

distinguishing character to enable any other conclusion to be reached.

Gomes e Sousa in Dendrologia Mozambique 1 :244 (1966) records E. lasianthum

from south of the Limpopo River, but I have seen no specimen from Mozambique.

E. lasianthum, which is only recorded from Natal (Zululand), Swaziland and southern

Mozambique, is apparently geographically isolated from any other species of Erythro-

phleum.

It seems necessary to clarify the various references in literature relating to E.

lasianthum.

33261—4

E. lasianthum Corbish/ey in Kew Bull.:27 (1922). Type: Natal, Ingwavuma, Nov.

1919, District Magistrate sub PRE H 1228 (K, holo.; PRE).

E. guineense G. Don var. swaziense Burtt Davy in Fl. Transv. 2:330 (1932); v. Breiten-

bach in Indig. Trees of S. Afr. 3:319 (1965). Type: Swaziland, without precise locality

or date, B. Nicholson s.n. (K, holo.; PRE sub H 30335, iso.). E. guineense sensu Henkel

in Woody Pl. of Natal and Zululand: 236 (1934). E. suaveolens sensu Compton in

Annotated Check List of the Flora of Swaziland, J.S. Afr. Bot., Suppl. 6:46 (1966);

sensu De Winter ef al. in Sixty Six Tvl. Trees:170 (1966); sensu Moll in Forest Trees of

Natal :69 (1967).

The only other species of Erythrophleum encountered in the area delimited for the

Flora of Southern Africa is E. africanum (Benth.) Harms, which occurs in South West

Africa. As in E. /asianthum, the stamen filaments in E. africanum are woolly tomentose

to near the apex although on occasional specimens (not in our area) the filaments are

subglabrous. However, E. africanum and E. lasianthum differ in a number of characters

and are readily distinguishable (see Table 1). Furthermore, there is a large geographical

discontinuity between the species, E. /asianthum having a very restricted distribution in

relation to E. africanum which is widespread in Africa.

TABLE 1.—Synopsis of the differences between Erythrophleum africanum and E. lasianthum

E. africanum E. lasianthum

pinnae 2-5 pairs pinnae 2-4 pairs

leaflets (6—) 8-17 leaflets 4-13

leaflets narrowly elliptic to elliptic or with ovate leaflets ovate, ovate-elliptic, -_ symmetric

tendency, often somewhat asymmetric

leaflets 1-2-6 x 0-9-3 cm in South West Africa leaflets 1-8-6-5 x 1-3-5 cm

leaflets obtuse or sometimes rounded apically, leaflets usually with +. pronounced acumen

not acuminate, -_ emarginate apically, emarginate

leaflets usually appressed-pubescent ab- and leaflets glabrous, midrib rarely slightly pubes-

adaxially, often + glabrous above, or glabrous cent abaxially

above and below except for pubescence on

midrib abaxially

leaflets coriaceous, venation conspicuous ab- leaflets thin, venation relatively inconspicuous

and adaxially apart from midrib

Petiolule pubescent, up to 4 mm long Petiolule glabrous, up to 7 mm long

Rachides pubescent Rachides glabrous

The differences between EF. suaveolens and E. africanum were well amplified by

Brenan in FI. Trop. E. Afr. Legum.—Caesalpinioideae 18-21 (1967). As mentioned by

Brenan E&. africanum is “a distinctly variable species’’. E. suaveolens is readily dis-

tinguished from E. /asianthum in having glabrous or occasionally subglabrous stamen

filaments. Certain vegetative specimens of EF. /asianthum closely resemble some of

those of E. suaveolens, but the leaflets in the former are usually smaller. However,

because of the geographical discontinuity between the two species, it is unlikely that

difficulty will be experienced in naming specimens.

I am grateful to Mr. E. G. H. Oliver, South African liaison botanist at the Royal

Botanic Gardens, Kew, England, for information concerning type specimens.

Bothalia 10, 1: 45-54

Stapelieae from South Tropical Africa, V*

L. C. Leach

ABSTRACT

The generic position of Huernia verekeri Stent and some related species is discussed. H. verekeri

Stent var. pauciflora Leach is described, also several putative hybrids of H.verekeri. Three new combina-

tions are made, viz. H. tanganyikensis (Bruce & Bally) Leach (Duvalia tanganyikensis Bruce & Bally),

H. procumbens (R. A. Dyer) Leach (Duvalia procumbens R. A. Dyer) and H. andreaeana (Rauh) Leach

(Duvalia andreaeana Rauh).

The discovery of a creeping variety of Huernia verekeri Stent, which appeared to be

very close to Duvalia Haworth, led the writer to examine carefully the generic position

of the group of related taxa which included Duvalia tanganyikensis Bruce & Bally,

D. procumbens R. A. Dyer, D. andreaeana Rauh and Huernia schneideriana Berger;

the conclusions reached, from the study of a relatively large quantity of material as well

as of the published descriptions and figures, are set out below.

Huber in Prodr. Fl. S.W. Afr. 114: 38 (1967) suggests that H. verekeri should be

considered to be a Duvalia rather than a Huernia and includes it in his keys to both

genera. There seemed, at first sight, no reason why the new variety should not be

placed in Duvalia and yet the typical variety appeared to be correctly placed in Huernia

as did also the closely related H. schneideriana.

However, although many of the generic distinctions in the Stapelieae are somewhat

arbitrary, it is considered that, in the two genera concerned, ther are good diagnostic

criteria on which they may be satisfactorily separated. An analysis now follows,

with the characters arranged in order of relative reliability, the above-mentioned some-

what controversial species being temporarily omitted.

DUVALIA HUERNIA

(1) Corona stipitate Corona sessile

with the outer corona resting on the with the outer corona, when present,

rim or sides of the annulus (i.e. closing closely appressed to the base of the

what corresponds to the tube in tube

Huernia) No exception known

No exception known

(2) Stem teeth with a pair of denticles Stem teeth without denticles

(stipules) at the base of the leaf No exception known

One exception: D. sulcata N. E. Br.

(3) Corolla lobes replicate, at least to some Corolla lobes not replicate, usually to

extent (occasionally only slightly folded some extent channelled or canaliculate

towards the apex) towards the apex, sometimes flat or

Exceptions: D. sulcata with slightly convex

convex lobes and possibly some speci- No exception known

mens of D. polita var. parviflora

‘*L_IV published in J. S. Afr. Bot.

(4) Corolla lobes usually ciliate, at least to Corolla lobes not ciliate

some extent No exception known

Exceptions: D. elegans, D. pubescens

and possibly D. radiata

No. 4 is not particularly useful, as the cilia are frequently lost when specimens are

preserved, nevertheless when they are present then Duvalia seems definitely to be

indicated.

Omitted from the above analyses as not being diagnostic in the present context are :—

Raised annulus:—occurring in both genera

Intermediate lobes:—occurring in both genera to some extent although usually

much more prominent in Huernia

Corona shape:—the outline of the outer corona and the form of the inner lobes are

closely matched in both genera

Pollinia:—do not appear to display any generic morphological correlation

Note:—No trace of denticles was found in any of the Huernia species examined, par-

ticular attention being paid to species considered to be closely related to, or with

stems somewhat similar to those of D. tanganyikensis etc.

D. sulcata: despite appearing twice as an exception appears to be correctly placed in

Duvyalia on its remaining morphological criteria, and moreover, fails entirely to

qualify for consideration for any other genus.

On the evidence of the foregoing, it is considered that, although closely related,

separation of the two genera is not only amply justified, but that they may possibly

prove to be the most distinctly differentiated in the whole tribe.

It remains now to consider the position of the taxa excluded from the foregoing

analyses.

D. tanganyikensis, D. procumbens and D. andreaeana: In all three species the

coronas are sessile, the stem teeth devoid of denticles and the corolla lobes neither

replicate nor ciliate.

From this it appears that the only way in which they approach Duvyalia is in the shape

of the corolla with its raised annulus; at first sight the flowers do have a Duvalia-like

appearance, particularly those of D. procumbens, but this seems to be quite superficial.

The shallowness of the tubes contribute to this appearance but cannot be considered to

be at all significant, as may be demonstrated by comparison with the almost flat corolla

of H. marnieriana Lavranos which is undoubtedly correctly generically placed; further-

more the corolla lobes of all three are either channelled or have a tendency to become

canaliculate.

It appears, therefore, that these taxa should be transferred to Huernia with the

diagnostic characters of which they entirely agree, while their stems are also comparable

with those of such species as H. aspera N.E. Br., H. erinaceae Bally and H. repens

Lavranos.

H. verekeri and H. schneideriana: On analysis these both fail to comply with any of

the requirements of Duvalia as set out above while agreeing in all respects with those of

Huernia; there seems every reason, therefore, why these should be retained in Huernia.

This proposed rearrangement of three species, it should be noted, requires no amend-

ment to the existing generic circumscription of Huernia, moreover “corona stipitate ”’ vs.

‘** corona sessile’ may be used as a key character for the separation of the genera,

alternatively ‘‘stem teeth with denticles”’ vs. ““stem teeth without denticles” may serve

with almost equal accuracy.

Phytogeographic considerations also lend some support to the presently proposed

transfers as, although Huernia “ blankets ” a vast area, Duvalia is, with the exception of

D. politaN. E. Br. and D. sulcataN. E. Br. (Arabia), restricted to the southern and south-

western portions of Africa. D. polita is now known to be distributed from Angola to

Mozambique and from Natal to Zambia, its territory including practically the whole of

that of H. verekeri, so that we have the interesting situation where two species, one from

each of these closely related genera, occupy the same area and habitats and yet, despite

the hybridising propensities of H. verekeri when in contact with other Huernia species,

not one case of suspected hybridisation with D. polita has been reported from this most

extensive area.

Via

H.PROCUMBENS 12286A

D. RADIATA 12048 D.SULCATA 12210 D.Pouta 14083

Fic. 1.—A comparison of some characteristic Huernia and Duvalia coronas with corolla sections.

Drawn from specimens in the author’s spirit collection, as follows:— 12257 Huernia zebrina,

Moamba, Mozambique; 12226 H. oculata, Usakos, S.W. Africa (Hardy s.n.); 11889 H. verekeri var.

pauciflora, Mambone, Mozambique; 12136 H. tanganyikensis, Mt. Longido, Tanzania (comm. R.D.

Bayliss); 12220 H. marnieriana, Arabia (Lavranos); 12286A H. procumbens, Nuanetsi, Rhodesia;

12048 Duvalia radiata, Ladismith (Bayliss 1912); 12210 D. suleata, Aden (Lavranos 1068); 14083

D. polita var. parviflora, Pietersburg (Plowes 2635). All x 4 approx.

On making a study of material and especially of living plants of Huernia verekeri Stent,

one is immediately struck by the remarkable uniformity of floral characters displayed

by this species. Despite its wide distribution only slight variations have been noted.

these being restricted to the precise outlines of the corona and the dimensions and depth

of colour of the corolla.

In all species of the genus known to the author the variation in the shape of the outer

corona is such as to render that organ virtually valueless for diagnostic purposes, at oF

below specific level. H. verekeri is no exception in this respect and the lobed form o1

which White & Sloane based their variety stevensonii has been found to occur in widely

scattered individuals, sometimes as an individual aberrant among flowers with the more

usual subcircular corona; it has not been possible, therefore, to uphold this variety.

Size of flowers, at least within this species, seems to be almost, if not entirely, a

matter of nourishment, as may be seen from the comparison between wild and cultivated

specimens of Wild & Drummond 6921.

Generally the species is of very scattered occurrence but, in favourable seasons, large

concentrations of plants have been found in the Sabi Valley of Rhodesia, particularly

in the vicinity of Birchenough Bridge.

Near the mouth of the Save (Sabi) River in Mozambique a vegetative variant occurs,

which, despite its entirely different stems and habit, is considered to be best treated as

being only of varietal rank, especially as it seems possible that the one small population

found may be of clonal origin.

Huernia verekeri Stent in Kew Bull. 1933:145 (1933); Phillips, Flow. Pl. S. Afr.

15: t.591 (1935); White & Sloane, Stap. 3:848 (1937); Luckhoff, Stap. S. Afr. 199 (1952);

Jacobsen, Handb. Succ. Pl. 2:630 (1960); Huber, in Prodr. Fl. S. W. Afr. 114:38 (1967).

Type: Rhodesia, Sabi Valley, L.S.A. Vereker 5427 (K. holo.; PRE!).

H. verekeri var. stevensonii White & Sloan, |. c. 850, 1145 (1937); Jacobs., !. c. (1960).

Type: Rhodesia, Nyamandhlovu District (No specimen appears to have been preserved).

Plowes 2472

Nyanyadzi, Sabi Valley,

Huernia verckeri Rintiadio

D.c.H.Plowes

Fic. 2.— Huernia verekeri var. verekeri. 1, section of corolla with corona; 2, portion of flower from

above; 3, corona, side view; 4, corona, from above. All Plowes 2472.

The relationship of this species appears to be broadly with the taxa of the H.

macrocarpa Sprenger affinity, such as H. oculata Hook. f. and H. schneideriana, while

there is also a close relationship between the new variety and H. procumbens (the transfer

of which from Duvalia is effected elsewhere in this paper).

From H. oculata and H. schneideriana this widespread taxon may immediately be

distinguished by its narrowly attenuate corolla lobes and its entirely different coloration;

in addition the long, attenuate sepals usually much exceed the corolla sinuses and,

together with the attenuate lobes, impart a rather spidery appearance which it shares

with H. procumbens, and which is quite unlike that of any of its other relatives.

The typical variety is almost certainly nearest in overall characters to H. schneider-

iana which also has similar short acute hairs clothing the limb and lobes. The possibility

that this latter species is of hybrid origin is discussed elsewhere in this paper.

Var. pauciflora seems to be closest, in many ways, to H. procumbens from which it is

easily distinguished by its more open, larger tube and lack of a prominent annulus.

var. verekeri

Plant succulent, dwarf, caespitose, glabrous, usually very floriferous. Stems up to

10 cm high, 1-25 cm diam. excluding the teeth, 5—7, usually 6-angled, sharply sulcate

between the strongly toothed angles; teeth spreading, deltoid acuminate, up to 15 mm

high, laterally much compressed when young. Jnflorescence from near the base of the

younger stems, flowers opening more or less successively, produced from a stout cuspi-

date peduncle. Pedicel c. 10 mm long, glabrous. Sepals 5-8 mm long, narrowly ovate

attenuate, usually much exceeding the sinuses of the corolla. Corolla c. 3-5 cm diam.,

outside glabrous, whitish at the base becoming pale green above, usually flushed pink;

tube more or less hemispherical, c. 3 mm deep, inside whitish at the base becoming

suffused with maroon above, deepening to the dark maroon, subcircular or obtusely

pentagonal annular rim, 6-5-8 mm diam., sharply dividing the tube from the limb;

limb and lobes pale greenish yellow, covered with short, stout, acute, maroon hairs;

lobes horizontally spreading, c. 5 mm wide at the base, 15 mm long, attenuate, markedly

convex on the upper surface; intermediate lobes small, strongly reflexed, so that they are

usually visible only from beneath. Outer corona c. 3 mm diam. subcircular or obscurely

lobed (rarely distinctly 5-lobed), closely appressed to the base of the tube, glabrous,

ivory. Jnner corona lobes closely incumbent on the anthers, but shorter than them, with

a flattened, broadly rounded or somewhat sharply ovate, spreading dorsal gibbosity

which tapers sharply into the somewhat obtuse, raggedly pubescent apex, usually slightly

suffused with pink. Pollinia yellow-brown (appearing dark brown in their cells),

attached to the winged, dark brown carrier by translucent red-orange connectives.

BoTSWANA.—Ngamiland, between Nokareng and Aha hills, Wild & Drummond 6921 (SRGH), idem

cult. SRGH (K; PRE; SRGH), cult. Nelspruit (NBG).

S. W. ArricA.—Okavango, Andara, near Caprivi Strip, Giess 9576 (HSWA; M).

ZAMBIA.—Feira: near Feira, Fanshawe 9424 (SRGH).

RuopesiA.—N. Darwin: Mt. Darwin, Bingham s.n. cult. SRGH 3176 (K; LISC; SRGH). Sipolilo:

5 miles west of Kanyemba Miiller & Kerr 362 (SRGH). Binga: Mwenda Research Station, cult. SRGH,

Grosvenor 123 (SRGH): ibid. Mitchell 910 (SRGH). W. Nyamandhlovu: Tjolotjo Rd., 50 miles north-

west of Bulawayo, Plowes s.n. (PRE); Rochester Farm, cult. Nelspruit, Leach & Bullock 13199 (HSWA;

M; NBG; NDO; Z); Farm Burford; Paterson in Herb. Bul. Mus. 32 (SRGH). Plumtree: Nata River,

Davies in SRGH 23215 (SRGH); 6 miles west and 6 miles north-west of Plumtree, Bullock 141; 147

(SRGH). Wankie: Deka Rd., Levy s.n. (PRE); Lukosi River, Geise s.n. (PRE); Wankie, Paterson in

Herb. Bul. Mus. 50 (SRGH). E. Chipinga: Sabi Valley, Vereker in SRGH 5427 (PRE); ibid. Vereker

s.n. (PRE); Thorncroft s.n. in PRE 19574 (PRE); near Birchenough Bridge, Obermeyer in TRV 37493

(PRE), ibid. Leach 5579 (PRE; SRGH). Leach 9972 (SRGH). Melsetter: Hot Springs, Vereker in

Herb. Eyles 7631 (SRGH); Biriwiri, Plowes 2560 (PRE; SRGH); Umvumvumyu River, cult. Nelspruit,

Plowes sub Leach 12313, (K; SRGH); + 10 miles north-east of Birchenough Bridge, Leach & Miiller

13140 (SRGH). Umtali: Maranke Reserve, Robisnon in SRGH 41637 (SRGH). S. Bikita: Bikita, Wild

4422 (PRE; SRGH); Moodie’s Pass, Leach 9768 (PRE; SRGH); near Birchenough Bridge, Stock in

SRGH 20039 (SRGH); ibid. Hall 1138 (NBG). Gwanda: Mwewe River, Bullock 105 (PRE; SRGH);

Gwanda, Paterson in Herb. Bul. Mus. 302-307 (G; K; PRE; Z). Ndanga: Gurudzi River, 9 miles

north of Chipinda Pools, Bullock 134 (SRGH). Nuanetsi: Lundi River, Taylor in NBG 563/51 (NBG),

ibid. cult. Nelspruit, Cannell sub Leach 13799 (K; LISC). Some of the Chipinga District records are

not precisely located by the collectors and have been allocated to that district on the score of probability

Mataw1.—S. Province, Mpatamanga Gorge, Leach 5637; 10591 (SRGH).

MOZAMBIQUE.—Tete: Mesuza, Chase 2820A (SRGH).

var. pauciflora Leach, var. nov.

A varietate typica caulibus repentibus multo longioribus, angulis obtusis, dentibus

parvissimis distantioribus; floribus paucioribus facile distinguenda.

Prate 1.—Huernia verekeri. Plant from Nyamandhlovu District, Rhodesia,

showing the variation in the shape of the tube (Plowes 1901). Photo: D. C. H.

Plowes.

PLATE 2.—Huernia verekeri. Comparison of wild and cultivated (right) specimens.

Botswana (Wild & Drummond 6921).

Type: L. C. Leach & R. D. Bayliss 11889 (SRGH, holo.; K; LISC; PRE).

MOZAMBIQUE.—Sul do Save: Save River near Mambone, alt.c. 3 m, cult. Nelspruit, Leach & Bayliss

11889 (K; LISC; SRGH), idem cult. PRE (PRE).

Var. pauciflora was found growing under heavy shade in a thicket to the south of

Mambone, near the mouth of the Save River, on slightly rising ground in an area of

mangroves and swampy tidal inlets, and is known only from this locality (similar plants

have recently been reported from the lower Sabi Valley, but as no material exists it has

not been possible to confirm this report).

It was at first thought that the elongated creeping stems of these plants might be

developed only under the heavy shade conditions under which they were discovered but

when placed in cultivation at Nelspruit under varying degrees of shade, they have retained

their characteristic habit (eventually not thriving unless heavily shaded), as have also

those culitvated at both the Botanical Research Institute, Pretoria, and the National

Botanic Gardens, Kirstenbosch. Plants of the typical variety have likewise retained

their normal floriferous, caespitose habit in cultivation.

Var. pauciflora differs from the typical variety in being a sparingly branched, creeping

plant with very much longer stems with rounded angles and much smaller more distantly

spaced teeth; flowers are borne very much more sparingly towards the base of the

younger parts; in these respects the plant is very different from the floriferous, strongly

toothed, erect, dwarf clumps of var. verekeri.

PLATE 3.—Huernia verekeri var. pauciflora. Cult. Nelspruit (Leach & Bayliss 11889).

It has become apparent that H. verekeri tends to hybridize freely with its neighbours

and, from a study of such putative hybrids, it appears that the dominant characters of

H. verkeri are:—

1. Stems with a tendency to a greater number of angles and longer, more acumi-

nate teeth.

Tube shallow, more or less hemispherical.

Limb and lobes of the corolla clothed with short, acute, maroon hairs.

Outer corona tending to be less deeply lobed.

Inner corona lobes with an enlarged dorsal gibbosity.

Shape of pollinia and particularly of the connectives.

ON A gS

The general form of the inner corona, the papillation and the coloration appear to

be mainly inherited from the other parent.

Some field records of putative hybrids follow :—

H. verekeri < H. hislopii subsp. hislopii

Assumed crosses of this parentage show little variation, so much so, that but for the

accident of a similar hybrid occurring spontaneously in the Salisbury garden of an

observant collector/gardener (Mr. J. A. Whellan), where only H. verekeri and H. hislopii

were previously cultivated, this particular cross might easily have been published as a

distinct species. The accumulated evidence however, leaves little doubt regarding its

parentage.

RHobDESIA.—Gwanda: near Tuli, cult. Nelspruit, O. West sub Leach 11681A (BOL; G; HSWA; K;

LISC; M; NBG; NDO; PRE; SRGH; Z). Buhera: Matendera Ruins, cult. Umtali, Walters sub

Plowes 2636 (SRGH). Umtali: Maranke Reserve, Plowes 2577 (PRE; SRGH).

H. verkeri x H. hislopii subsp. robusta

Very similar to crosses with the typical subsp., but the corona leaning more towards

H. hislopii and the papillae larger and more dense.

RHODESIA.—Nyamandhlovu: Rochester Farm, Leach & Bullock 13200 (SRGH).

H. verekeri x H. longituba subsp. cashelensis

These assumed hybrids, unlike the foregoing, show considerable variation, as is

perhaps to be expected, as H. Jongituba is itself rather variable. Some examples bear a

remarkable resemblance in corolla shape to H. schneideriana, particularly is this evident

in Leach 5405 and Plowes 2470, both from the same vicinity.

RHODESIA.—Melsetter: Mutambara, Leach 5405 (PRE; SRGH); ibid. cult. Nelspruit, Plowes 2470

(PRE; SRGH),.

Huernia schneideriana Berger in Monatsschr. Kakt., 23: 177 (1913); White & Sloane,

l.c. 844 (1937); Jacobsen, I.c. 629 (1960). Type: Tanzania (“ Nyassaland ’’), Rungwe

District, Kyimbila (“ Kimbila am Nordend des Nyassa-Sees”’), Stolz 1407 (Bf).

Neotyre: Hort. bot. Monacensis, fl. 31. 10. 1927 (M!).

The type of H. schneideriana was a living plant collected by Adolf Stolz at Kyimbila

(c. 9°20’S, 33°35’E), which was sent to Berger by the Dahlem gardener Strauss. It seems

fairly certain that if herbarium material was prepared by Berger it would have been

destroyed during World War II.

Unfortunately only one herbarium specimen seems to have survived. This is

preserved at Miinchen (M), and is said to have been prepared from the original imported

plant and has therefore, been selected as the neotype.

The species is believed to have attained an extensive distribution in cultivation

(presumably all clonal divisions of the original gathering) but it is now, apparently,

rare in collections.

It has not been possible, in the absence of material, either to confirm or confound

the Mozambique localities tentatively mentioned by White & Sloane, (I.c.), but that of

Mangulane in the Loureng¢o Marques District seems particularly doubtful.

Plant tufted, stems c. 5 cm high, 1-1-2 cm diam., generally 6-angled, with relatively

small acute teeth about 3 mm high, 16 mm apart along the angles. Flowers few, from

near the base of the stems, borne on short glabrous pedicels. Sepals glabrous, c. 3-5 mm

long, 1-25 mm wide. Corolla shallowly campanulate, c. 2-5 cm across the points of the

lobes; tube more or less hemispherical, glabrous, blackish purple inside; /Jobes c. 7 mm

long, deltate, acute; both limb and lobes clothed with short, acute, blackish maroon

hairs. Outer corona c. 3 mm diam., dark purple, scarcely lobed, slightly undulate on

the margin. Jnner corona with lobes closely incumbent on the anthers but shorter than

them with a broadly rounded, more or less flat, spreading dorsal projection.

The above description is based partly on the original description by Berger and

partly on the Miinchen material.

In view of the fact that H. schneideriana has apparently been collected only once

and remembering its apparently vigorous qualities, it is thought that the possibility of

hybrid origin should be considered. If this possibility is accepted, then consideration

of the dominant characters of H. verekeri in putative hybrids of that species would

suggest it as one of the possible parents, since most, if not all, these dominant characters

are possessed by H. schneideriana; while its other characters would seem to indicate

that the other parent should be sought amongst the dark-flowered species such as H.

aspera and H. keniensis.

PLate 4.—1, Huernia yerekeri ~ H. longituba subsp. cashelensis. Melsetter District, Rhodesia (Leach

5405); 2, H. schneideriana. Cult. ‘‘ Les Cedres”. Photo: M. J. Marnier-Lapostolle.

What amounts to almost certain confirmation of this hybrid origin theory has

recently been received from Professor G. Reese in a paper entitled, “‘ Untersuchungen

iiber die Chromosamenzahlen der Stapelieae ’ [G. Reese u. H. Kressel, in Port. Act.

Biol. Ser. A, 10: 33 (1967)], in which a specimen identified as H. schneideriana has been

found to be a triploid; in the same article H. verekeri is listed as a diploid and a variety of

one of the other suggested possible parents is reported as a tetraploid.

The probability of H. schneideriana being of hybrid origin is, therefore, of a very

high order indeed, while that H. verekeri should be one of the parents seems, on the

morphological evidence, to be of only a slightly lesser order of probability.

The geographical isolation of H. schneideriana from the suggested parents may be

more apparent than real, as neither northern Mozambique nor southern Tanzania has

been intensively botanised, so that it is quite possible that both may occur, without

however having been recorded in herbarium collections.

Huernia tanganyikensis (Bruce & Bally) Leach, comb. nov.

Duvalia tanganyikensis Bruce & Bally in Cact. & Succ. J. Amer. 13: 179, tt. 114, 115

(1941); Bally in J. E. Afr. Nat. Hist. Soc. 16: 160, t. 53 (1942); Bruce in Flow. Pl. Afr.

28: t. 1082 (1950); Jacobsen, Handb. Succ. Pl. 1: 363 (1960); Rauh in Kakt. u. and.

Sukk. 12: 114 (1961). Type: Tanzania, N. Prov., Mt. Longido, Bally S. 19.

TANZANIA.—N. Prov., Mt. Longido at c. 1,500 m alt. on gneiss, ““mat forming cf. H. aspera’’, Bally

S. 19 (PRE, photos!); ibid. comm. Bayliss, cult, NBG 234/59 (PRE); Mbulu: Lake Manyara National

Park, 3,500 ft. alt., cult. PRE, fl. 4. iv.1956, Greenway & Kanuri 11414 (PRE).

Huernia procumbens (R. A. Dyer) Leach, comb. nov.

Duvalia procumbens R. A. Dyer in Flow. Pl. Afr. 31: t. 1218 (1956); Rauh, l.c.

(1961); Letty, Wild Flowers Transv. 268, t. 1331 (1962).

Type: S. Africa, Transvaal, Pafuri, Van der Schijff 3618 (PRE!).

TRANSVAAL.—Soutpansberg: 32 miles north of Punda Maria, Codd 5389 (PRE); 8 miles west of Pafuri,

cult. Nelspruit, Leach, H. H. & D. C. Mockford 12277 (G; PRE; SRGH); Kruger National Park, Van

der Schijff 3618 (PRE).

RHODESIA.—Nuanetsi: PesuRi ver Gorge, + 12 miles west of Pafuri, cult. Nelspruit, Leach, H. H. &

D. C. Mockford 12286A (K; LISC; PRE; SRGH; ZSS).

There is considerable variation in the inner corona of this species, the lobes of which

may be either widely spreading with the inner face somewhat channelled, or strictly

erect and more or less triangular in cross-section; the latter being, apparently, the more

common.

Huernia andreaeana (Rauh) Leach, comb. nov.

Duvalia andreaeana Rauh, |.c. (1961). Type: Tanzania, between Mombasa and Voi,

Rauh Ke 867 (HEID, holo.; PRE, photos!).

ACKNOWLEDGEMENTS

The author is much indebted to and wishes to thank the following :—

Messrs. R. D. A. Bayliss, D. S. Hardy and J. J. Lavranos for living plants of Sta-

pelieae.

Dr. L. E. Codd, Chief, Botanical Research Institute, Pretoria (PRE), for the

facilities of the herbartum and much other assistance.

Mr. R. B. Drummond, Chief Botanist, Government Herbarium, Salisbury (SRGH)

for the herbarium facilities and Mr. Th. Miiller, Curator of the Botanic Gardens, for his

help in respect of the cultivation of plants from Rhodesia.

The directors of the following herbaria: Royal Botanic Gardens, Kew (K);

Botanische Staatssammlung, Miinchen (M); National Botanic Gardens, Kirstenbosch

(NBG); Institut fiir Systematische Botanik der Universitat, Ziirich (Z), for the loan of

material and for information regarding the history of H. schneideriana.

M. J. Marnier-Lapostolle for a photograph and a live plant of H. schneideriana.

Mr. and Mrs. H. Mockford for hospitality when the Pafuri area was visited. Mr.

D. C. H. Plowes for the line drawing of H. verekeri accompanying this article, for

photographs and living plants of that species and of some putative hybrids.

Bothalia 10, 1: 55-82

New and Interesting Records of African Plants

Various Authors

AIZOACEAE

A NOTE ON THE IDENTITY OF MESEMBRYANTHEMUM BRACHYPHYLLUM WELW.

In 1873 a paper by Welwitsch on “An undescribed species of Mesembryanthemum

from the south of Portugal ’’ was published posthumously in J. Bot. 11: 289. In an

explanatory note the editor, Trimen, said that Welwitsch had been struck by “‘ some

remarkable instances of connection between the vegetation of the southern province

of Portugal and the Cape of Good Hope, and had selected this new Mesembryanthemum

(M. brachyphyllum) as the most striking example’. The paper includes a full description

in Latin of the species, together with a tab. and reference to type material.

Berger in his Mesembrianthemen und Portulacaceen 292 (1908) included the

species in a list of “* doubtful and missing species ”, while Jacobsen in his Handbook

of Succulent Plants 3: 1268 (1960) mentioned that “it is a very ill-defined species ”’.

Being curious about the identity of this Mesembryanthemum growing in Europe,

the present author in 1965 wrote to Dr. O. A. Leistner, the Institute’s liaison officer

at Kew, and asked him to check on the identity of M. brachyphyllum. Dr. Leistner

replied that the Welwitsch specimen of M. brachyphyllum at Kew had been identified

by N. E. Brown as Lampranthus glaucus (L.) N. E. Br., a Cape species, and that he

fully agreed with this determination. It is clear, therefore, that M. brachyphyllum

must be regarded as a synonym of L. g/aucus. It is interesting to note that Welwitsch

commented that M. brachyphyllum was very closely related to M. glaucum L. on which

L. glaucus was based.

It is indeed ironical that the plant from Portugal chosen by Welwitsch to illustrate

the similarity between the floras of southern Portugal and the Cape should eventually

prove to be a native of the Cape.

The synonymy and literature references can be summarized as follows :—

Lampranthus glaucus (L.) N. £. Br. in Gard. Chron. 87: 212 (1930); Bolus in

Fl. Cape Penins. 385 (1950).

Mesembryanthemum glaucum L., Sp. Pl. 486 (1753); Sond. in Fl. Cape. 2: 417 (1862).

M. brachyphyllum Welw. in J. Bot. 11: 289, t.136 (1873); Berger, Mesembrianthemen

und Portulacaceen 292 (1908); Jacobsen, Handbook of Succulent Plants 3: 1268

(1960).

It should perhaps be mentioned that M. brachyphyllum Pax ex Schltr. & Diels

taken up in the Index Kewensis is a nomen nudum and the species is apparently unrelated

to the Welwitsch species. In Schultze’s Aus Namaland und Kalahari, p. 693 (1907),

Schlechter and Diels mention that a Hermann specimen in Herb. Berlin was annotated

by Pax as M. brachyphyllum Pax, but that because this name was antedated by M.

brachyphyllum Welw., they proposed a new name, M. paxianum. No description was

provided.

Rei GAISTREY:

AMARYLLIDACEAE

A New SPECIES OF CRINUM

Crinum foetidum Verdoorn, sp. nov., foliis latis et humifusis C. graminicola Verdoorn

atque C. delagoensi Verdoorn simile, sed ab ambobus foliis hebetato-turcoisinis non

nitidoviridis, perianthii segmentis acuminatis, carneis non rubicundis vel purpureis

carinatis, praecipue differt; ab aliis speciebus Africae australis plantis foetidis et

seminibus papillosis coloribus sepiaceis differt.

Plantae obtritae foetidae. Bulbus globosus, 15-18 cm diam., abrupte in collum

brevem vel longum productus. Folia 10-16, hebetato-turcoisina, subhumifusa, distincte

ciliata; exteriores circiter 70 cm longa, 12 cm lata. Pedunculus subpatens, 14-22 cm

longus, leviter compressus, 2-5 « 1-5 cm crassus. Bracteae involucrates acuminatae,

circiter 10 cm longae, base 3 cm latae. Umbella circiter 11-flora. Pedicelli 0-1-5 cm

longi. Perianthium infundibuliforme; tubus flavovirescens, circiter 10 cm longus;

lobi albidi, carneocarinati, circiter 10 cm longi, acuminati, apiculati, in dimidio superiore,

revoluti. Capsula subglobosa, rostrata. Semina papillosa-rugosa, coloribus sepiaceis.

RATE le

Type: Transvaal, Waterberg, 70 miles north of Vaalwater, Louw 3460 (PRE,

holo.).

Most parts of the plant emit a foetid odour when crushed or broken. Bulb globose,

15-18 cm diam., abruptly narrowed into a long or short neck; tunics membranous.

Leaves few, 10 to 16, spreading along the ground, a dull (matt) turquoise-green colour,

about 70 cm long, 12 cm broad, margin cartilaginous, obscurely dentate, distinctly

ciliate with hairs mostly over 1 mm long. Jnflorescence with a rather short, 14-22

cm long, peduncle which is subspreading, somewhat compressed, usually 2-5 x 1-5

cm thick. Spathe-valves acuminate from a broad base, about 10 cm long, 3 cm broad

at the base. Umbel about 11-flowered. Pedicels 0-1-5 cm long. Perianth more or

less funnel-shaped; tube green to pale yellowy-green, about 10 cm long; segments

white with a delicate pink dorsal band, about 10 cm long, outer about 2 cm broad,

inner about 2:5 cm broad, all acuminate and recurved in the upper half, apiculate

(apicule often dark rose coloured and up to | cm long on the outer segments, shorter

on the inner). Stamens declinate, filaments white; style red in upper portion. Capsule

subglobose, with the perianth base persisting like a beak. Seeds sepia coloured, rough

with raised papillose ridges.

Found in deep coarse sand in the hot, dry western and north-western Transvaal,

and northern South West Africa. Probably occurs in the northern Cape, Botswana

and Rhodesia, but not yet confirmed.

TRANSVAAL.—Rustenburg: farm Albion near Vaalpenskraal, Verdoorn 2495. Waterberg: 70 miles

north of Vaalwater, Louw 3460 (type); west of Vaalwater near Sterkstroom, Verdoorn 2497. Pot-

gietersrus; Steilloop, on Magalakwin, Verdoorn 2498.

SoutH West ArFrricA.—Okavango: 8 miles east of Runtu, De Winter 3777. Gobabis: Télken

1003.

In November, 1964, Dr. W. J. Louw of Potchefstroom University, brought material

of this Crinum to the Botanical Research Institute. He had collected it on sandy flats

in the north-western Transvaal about 70 miles north of Vaalwater. It was obvious

that it did not match any of the known South African species. The material was

complete with bulbs, leaves, flowers and fruits represented. Features noted were the

broad, matt, bluey-green leaves with rather long ciliae, the distinctly acuminate perianth

segments which were predominantly white with a delicate pink keel, and the beaked

fruits. Two years later, that is in November, 1966, a search was made for this species

in the wild. It was seen in three places in the western and north-western Transvaal.

The first group was found about 37 miles west of Thabazimbi in coarse, deep sand

(see Plate 1). Interesting points noticed, in addition to those recorded about Dr.

Louw’s specimens, were the slightly zygomorphic flowers with revolute segments and

the watery leaves which, when torn, emit a foetid odour. Next the species was seen

about 9 miles north-west of Vaalwater and lastly at Steilloop on the Magalakwin

River. At the last-mentioned site the seeds were seen. They differ from all other

South African species in that they are a sepia colour and are rough with raised papillae.

The broken or fading capsules had the same unpleasant smell as the leaves have.

Because of the ciliate leaves, the descriptions and types of two tropical species

so far unknown in South Africa were investigated, namely C. harmsii Bak. and C.

crassicaule Bak. C. harmsii was ruled out after seeing the type specimen and because

it is described as having only 3 flowers in an umbel, the leaves only 5 cm broad and

Baker puts it in the subgenus Platyaster which means that the flowers are erect and

hypocrateriform, whereas ours are trumpet- or funnel-shaped and nodding. The

photo on page 348 in Baum’s Kunene—Zambesi Expedition (1903) may illustrate C.

harmsii, the type of which was collected on that expedition, although the illustration

is published without a name.

C. crassicaule was described from a specimen which Baines painted and which

consisted merely of an inflorescence. In this connection Baines writes in his Explora-

tions in South West Africa, page 188: “‘ Monday 21st (Oct. 1861) . . . I sketched

a very beautiful umbel of white and pale purple flowers brought home last night by

Chapman, apparently a kind of amaryllis. The main stem (peduncle) was flattened,

an inch and a quarter wide and a quarter thick, the bracts had fallen down and

whithered . . . . The flower had been too long gathered to restore itself to form

when placed in water”. It was solely on this specimen, which was sent to Kew by

Baines, that Baker based the species C. crassicaule (see Amaryllidaceae, page 85).

Several weeks after painting this inflorescence, while still at Koobis, Baines painted

another Crinum, on the 13th November (see page 220 of his Explorations). Coloured

photographs of the painting done in October as well as the one done in November,

were kindly supplied to me by our Liaison Officer at Kew, Mr. E. G. H. Oliver, and,

in my opinion, the latter painting represents a species distinct from the one painted

in October, but is very definitely conspecific with the species here described, C. foetidum.

The October painting of the half-faded inflorescence, the type of C. crassicaule,

seems to represent a species with erect perianth-tubes and stamens arranged as in the

subgenus Stenaster. This interpretarion is supported by the fact that at least two

species in the Stenaster or Pachyaster groups occur in the same area.

It would seem from Dr. N. E. Brown’s notes, many years later, written on the type

specimen of C. crassicaule and on the painting done by Baines in November, that

he thought they represented the same species. On the type he has written that it was

collected at Koobis, Ngamiland, November 13, 1861, whereas it was collected and

painted in October, and on the November painting he has written “ Crinum crassicaule

Baker, Handb. Amaryll. p. 83, 1888”, which it is not. Dr. Brown evidently examined

these when he was naming the Lugards’ collection of plants from Ngamiland. In

the Kew Bulletin 1909, where the list of names is published, N. E. Brown, on page 142,

lists Mrs. Lugards’ No. 45 as Crinum crassicaule and he there describes his concept

of the species which, judging from his notes mentioned above, probably includes more

than one element. Mrs. Lugards’ painting of No. 45 shows an erect perianth-tube

and straight, erect filaments suggestive of the Stenaster or Platyaster groups. For

this reason it may be C. crassicaule, but from her painting the species is not recognizable.

In addition to the November painting by Baines, the photograph published in

the Guide to the Victoria Falls by H. Wild, page 135, illustrates a Crinum, which 1s

apparently conspecific with Crinum foetidum.

I. C. VERDOORN

PLATE 1.—Crinum foetidum. 1, in flower; 2, in fruit, from Thabazimbi District (Verdoorn 2495).

Photos: D. Edwards.

CAPPARACEAE

NOTES ON BOSCIA

Boscia is an African genus except for one species, B. arabica Pest., which occurs

in southern Arabia. Most of the species occur in drier parts, but no representatives

are found in north-western Africa. Boscia is very similar to the genus Maerua as also

pointed out by De Wolf in Kew Bull. 16: 80 (1962). The flowers of the two genera

are very similar, but in Boscia petals are usually absent, the receptacle tube is very

short and the androgynophore practically non-existent. All these characters are not

found in the genus Capparis and it is thus surprising that a few South African species

of Boscia have remained included in Capparis for such a long time. Burchell, Trav. 1

(1822) described them and Sonder in FI. Cap. 1: 60 (1860) did not recognize how

different these species were from the rest of Capparis and, instead, placed a few species

of Maerua with apetalous flowers in the genus Boscia. At the time the latter genus

was not well defined, but soon afterwards Oliver, Fl. Trop. Afr. 1:92 (1868), gave

the genus full recognition and a few South African species such as B. foetida Schinz

were correctly identified and described in the genus Boscia. Pestalozzi in Bull. Herb.

Boiss. 6, Appl. 3: 1-152 (1898) undertook the first general investigation of the genus,

particularly its anatomy. His study was, however, greatly limited by the small amoun-

of material used. Gilg & Benedict in Bot. Jahrb. (1915) revised all the African Capt

paraceae. They transferred Capparis albitrunca Burch. to the genus Boscia, but not

the very similar species C. oleoides Burch. ex DC. merely because of the presence of

petals. With recent revisions of the tropical species available, viz. Hutchinson &

Dalziel, Fl. Trop. West Afr. 1: 89 (1927); Wild in Fl. Zamb. 1: 229-35 (1960) and

Elffers et a/. in Fl. Trop. East Afr. (Capparidaceae) 50-58 (1964), and with the wide

range of South African material at my disposal, I was prompted to make the following

innovations for my treatment of the genus in the Flora of Southern Africa, Vol. 13.

B. oleoides (Burch. ex DC.) Toelken, comb. nov.

Capparis oleoides Burch. ex DC., Prodr. 1: 248 (1824); Sond. in Fl. Cap. 1: 62

(1860). Type: Cape, Bushmans River, near Rautenbach’s Drift, Burchell 4200 (K,

holo.; PRE!). C. coriacea Burch. ex DC., Prodr. 1: 248 (1824). Type: Cape moun-

tains on the south-west side of Graaff-Reinet, Burchell 2898 (K, holo.!). C. clutiaefolia

Burch. ex DC., Prodr. 1: 248 (1824); Sond. in Fl. Cap. 1: 62 (1860). Type: Cape,

near Blaauwkrans, Burchell 3881 (K, holo.!).

This species is incorrectly placed in the genus Capparis, because of the presence

of a receptacle tube, valvate sepals and sclereids in the mesophyll. A receptacle tube

with a corona on which the petals are inserted and valvate sepals are also found in the

genus Maerua, but the sclereids in the mesophyll of B. oleoides are typical of those

found in the genus Boscia. No sclereids have been found in the leaves of several

species of Maerua investigated. In fact, B. oleoides is so similar to B. albitrunca that the

two species have often been confused. However, B. oleoides can be distinguished

from the latter by its ridged branches with alternate leaves, usually terminal inflorescence

and the presence of petals. It occurs in dry vegetation in the eastern Cape as far inland

as Graaff-Reinet and the nearest locality of B. albitrunca to this is near Hope Town

or Victoria West.

B. tomentosa Toelken, sp. nov. ab speciebus omnibus Bosciae in Africa australi

tomento stellato-piloso differt.

B. polyantha sensu Suesseng., Heine & Roessler in Prodr. Fl. S.W. Afr. 47:4

(1966).

33261—5

Giess 3321

Kaokoveld

Rautanen 500

(type of B. rautanenii)

Karibib

Keet 1670

Rehoboth

Marloth 1415

Otjimbingue

Dinter 4897

Klein Karas

ics _0lU

Aroab

ee oe

Kakamas

a a [aa NRE TD Te) area TT Rec ae) a (i earn (re |

0:5 | 15 2 2°5 3 355 4 45 5 SP) 6 65

Fic. 1.—Histograms showing the variation of the width of the leaves in B. foetida subsp. foetida.

Fifty leaves of each specimen were measured accurately to 0-5 mm. The specimens were arranged

to illustrate the increase in the width of the leaves as the subspecies radiates out towards the north,

east and south of the area of distribution of the narrow-leaved form.

Arbusculae rare arbores ad 5 m altae. Rami tomentosi, glabrescentes, fere albi.

Folia non fasciculata; lamina late ovata vel elliptica, obtusa ad emarginatam in apicem,

mucronulata, in basem breve angustata vel truncatas 2-4 (-5-5) cm longa, (1-) 1-5-2-5

cm lata, rigide succulenta, tomentosa praecipue stellatopilosa; petiolus (0-4-) 0-6-1

cm longus, tomentosus. Jnflorescentia terminalis, paniculata; pedunculus 2-5 cm

longus, tomentosus; pedicellus 0-3-0-6 cm longus, tomentosus. Bractea setacea,

2-4 mm longa, tomentosa. Sepala ovata ad oblonga, 2-3 mm longa, extus tomentosa,

intra glabra. Corona crassa, succulenta, annularis, denticulata. Stamina (5) 6-8;

filamentum 3-4 mm longum, glabrum. Gynophorum 2-4 mm longum, glabrum.

Ovarium ovoideum, ovulis 12; stigma capitatum, sessile vel paene sessile. Fructus

non visus.

Type: S.W. Africa, Kaokoveld, Otjinunga, De Winter & Leistner 5749 (PRE,

holo.).

Shrubs or rarely trees up to 5 m high. Branches tomentose becoming glabrous,

almost white. Leaves alternate not fascicled; lamina broadly ovate or elliptic, obtuse

to emarginate at the apex, mucronulate, shortly tapering or truncate at the base,

2-4 (-5:5) cm long, (1—) 1-5-2-5 cm broad, stiff fleshy, tomentose with mainly stellate

hairs; petiole (0-4-) 0-6-1 cm long, tomentose. Sclereids in the mesophyll without

foot, not branched, hardly reaching the centre of the leaf, in clusters, similar on both

sides of the leaf. Inflorescence terminal, paniculate; peduncle 2-5 cm long, tomentose;

pedicel 0:3-0-6 cm long, tomentose. Bracts setaceous, 2-4 mm long, tomentose.

Sepals obovate to oblong, 2-3 cm long, tomentose outside, glabrous within. Corona

a thick ring, denticulate. Stamens (S—) 6-8; filaments 3-4 mm long, glabrous. Gyno-

phore 2-4 mm long, glabrous. Ovary ovoid, with 12 ovules; stigma capitate, sessile

or nearly so. Fruit unknown.

In dry bushveld on the north-western border of South West Africa and also in

Angola.

SoutH West ArricA.—Kaokoveld: near Otjinunga, De Winter & Leistner 5749; 5785.

Siissenguth, Heine & Roessler (1966) interpreted the species as B. polyantha Gilg

but, on investigating the type specimen, Antunes A100 (B), it was found that its branches

and leaves are pubscent with unicellular hairs, the flowers are densely clustered in an

axillary inflorescence, the flowers are not fleshy and the sepals are lanceate-elliptic.

In all these characters it differs from B. tomentosa. This latter species does not seem

to have any direct affinity with any South African species of Boscia. The stellate hairs

are multicellular and are formed by the fusion of the lower part of several adjoining

unicellular, epidermal hairs.

B. foetida Schinz in Verh. Bot. Ver. Prov. Brandenb. 29: 49 (1886). Type: S.W.

Africa, Keetmanshoop, Schinz 326 (Z, holo.!).

This complex species is distinguished from other species, except B. microphylla,

by its tomentose fruits and the sclereids in the mesophyll of the leaf with a well

developed foot. B. microphylla, which is obviously very closely related but kept as

a separate species at the moment, may have to be incorporated in this complex when

a wider range of material becomes available, particularly as it appears from the present

study that the B. foetida complex is split into a number of geographically separated

taxa. Four subspecies can be recognized mainly on the number of stamens.

Trees or shrubs always branching from the base; pedicels with spreading hairs:

Shrubs or trees with main branches ascending, up to 3 m high; peduncle

O:3= Ie ycmmlongsestamenselll lo peer eee ence cice titer icieiiiieicyerar (a) subsp. foetida

Shrublets decumbent, not higher than 30 cm; peduncle absent; stamens 11-12 (-14) (b) subsp. minima

Trees with one trunk at least 1 m high; pedicels glabrous:

Stamensplil—I5;s pedicelsmi=2 cmulongaen- aaa ee a (c) subsp. longipedicellata

Stamens 5-7 (8); pedicels 0:5-0-8 cm long.................-seeeeaee (d) subsp. rehmannianc

(a) subsp. foetida

B. foetida Schinz in Verh. Bot. Ver. Prov. Brandenb. 29: 49 (1886); Pest. in Bull.

Herb. Boiss. 6, Appl. 3: 136, t.2, fig. 1 (1898); Siisseng., Heine & Roessler in Prodr.

FI. S.W. Afr. 47: 3 (1966). B. rautanenii Schinz in Viert. Naturf. Ges. Ziirich 51: 193

(1906); Siisseng., Heine & Roessler in Prodr. Fl. S.W. Afr. 47: 4 (1966). Type: S.W.

Africa, Karibib, Rautanen 500 (Z, holo.!). B. kalachariensis sensu Dinter in Fedde

Repaaloeso2 a Os):

This subspecies is usually found in rocky outcrops in the dry southern and north-

western South West Africa and the adjoining northern Cape Province.

B. rautanenii is only a narrow-leafed form of subsp. foetida, occurring in the districts

of Swakopmund and Karibib. The histogram (see Fig. 1) illustrates the gradual increase

in width of the leaves as the subspecies radiates out towards the north, east and south

of the area of distribution of this form.

(b) subsp. minima Toe/ken, subsp. nov.

Haec subspecies ab aliis habitibus fruticulis ad 30 cm alta ramis decumbentibus;

ab subsp. foetida absentia pedunculi et sclerenchymate dissimili; ab subsp. /ongipedi-

cellata pedicello breviore, piloso; ab subsp. rehmanniana pedicello piloso et numero

staminum differt.

Type: Transvaal, Warmbad, near Makapanstad, Codd 8013 (PRE, holo.).

Shrublet not higher than 30 cm, cushion-like with decumbent branches. Leaves

with lamina oblanceate to elliptic, 0-5-1-3 cm long, 0-2-0-4 cm broad, isobilateral.

Sclereids in the mesophyll pointed, not branched at the apex, only well developed

on the adaxial side. Inflorescence racemose, usually fascicled with 2—5 flowers; peduncle

absent; pedicel 0-3-0-6 cm long, hairy. Stamens 11-12 (-14).

Found on limestone outcrops often near pans, or on clay soils near rivers in the

north-eastern Cape, western Transvaal and eastern Botswana.

Cape.—Mafeking: 50 miles west of Mafeking, Acocks 18772; near Mosita, Brueckner 529; 14

miles east of Sedilamolamo, Leistner 565.

TRANSVAAL.—Thabazimbi: 2-3 miles west of Makoppa, Theron & Marsh 253. Warmbad: near

Makapanstad, Codd 8013.

This subspecies is very similar to subsp. foetida. However, it usually grows in

areas that are temporarily swamped and even when it is found outside such marshy

habitats it does not change its decumbent habit. Although this subspecies shows a

slight overlap with the subsp. rehmanniana, the two were never found in the same area

and appear to be ecologically separated.

B. longipedicellata Gilg in Notizbl. Bot. Gart. Mus. Berl. 14: 188 (1940). Type:

Natal, Weenen, Peniston in PRE 24195 (PRE, iso.!).

This subspecies occurs in dry bushveld in central Natal. It has often been confused

with B. albitrunca, but can be distinguished by its hairy fruits and discolorous leaves.

The leaves of subsp. rehmanniana in the Lebombo Mountains in northern Natal and

Swaziland often attain similar sizes, but no intermediate stamen number has yet been

recorded.

(d) subsp. rehmanniana (Pest.) Toelken, comb. nov.

B. rehmanniana Pest. in Bull. Herb. Boiss. 6, Appl. 3:95 (1898); Burtt Davy, FI.

Transv. 1: 123 (1926); Wild in Fl. Zamb. 1: 235 (1960). Type: Transvaal, Klippan,

Rehmann 5134 (Z, lecto.!). B. microphylla Oliv., Fl. Trop. Afr. 1: 93 (1868), partly,

as to specimen Baines & Chapman. B. kalachariensis Pest. in Bull. Herb. Boiss. 6,

Appl. 3: 98 (1898). Type: Botswana, Lake Ngami, Fleck 247 (Z, holo.!). B. filipes

Gilg in Bot. Jahrb. 33: 221 (1903); Wild in Fl. Zamb. 1: 234 (1960). Type: Mozam-

bique, Lourenco Marques, Schlechter 11707 (B, holo.!; BOL!; NH!; PRE!). 2B.

seineri Gilg & Engl. in Engl., Pflanzenw. Afr. 3, 1: 242, fig. 158D—-F (1915), nomen

nudum.

Capparis albitrunca var. parvifolia Sim, For. Fl. Port. E. Afr. 2, t.3, fig. 4 (1909).

Type: Mozambique, Lourenco Marques, Sim 5157 (PRE, holo.!).

This subspecies is found in the dry bushveld of the central and northern Transvaal,

Swaziland and north-eastern Natal, extending its distribution into Mozambique,

Rhodesia and northern Botswana. The subsp. rehmanniana differs from all the other

subspecies by its fewer stamens. In the Transvaal a pattern of variation, probably

not entirely due to a difference in rainfall, can be observed from the east to the west,

varying from big leaves, 1-3 in a fascicle and with no or few sclereids to much smaller

leaves, often more than five per fascicle and numerous sclereids. Similarly, in the east

the gynophore is hairy becoming gradually glabrous towards the west. Consequently

B. filipes, as distinguished by Wild (1960), cannot be upheld.

B. microphylla Oliy., Fl. Trop. Afr. 1:93 (1868), partly, excl. specimen Baines

& Chapman; Exell & Mendonca, Consp. Fl. Ang. 1: 65 (1937); Siisseng., Heine &

Roessler in Prodr. Fl. S.W. Afr. 47: 3 (1966); emended. Type: Bumbo, Welwitsch

983 (K, lecto.!).

The species is based on two specimens: Welwitsch 983 and Baines & Chapman

s.n., which are now considered to belong to two different species. The latter specimen

has been identified as belonging to B. foetida subsp. rehmanniana. So, in order to retain

the species name in its generally accepted sense, Welwitsch 983 was chosen as the

lectotype and the description slightly emended to exclude the Baines & Chapman

specimen.

H. R. TOLKEN

NOTES ON CAPPARIS

Recent revisions such as those of De Wolf in Fl. Trop. E. Afr. (Capparidaceae)

58 (1964) and Jacobs in Blumea 12: 385 (1965) have contributed much to a clearer

delimitation of the genus Capparis. Typical characters such as the stipulate spines,

the convex receptacle, imbricate sepals and often the presence of more than two carpels

have been particularly emphasized.

In Africa comparatively few species of Capparis occur, but most of them are

widely distributed. A re-evaluation of the many taxa described from Southern Africa,

in the light of the revisions mentioned, has necessitated the following name changes :—

C. sepiaria L., Syst Nat. ed. 10: 1071 (1759); De Wolf in Fl. Trop. E. Afr. (Cap-

paridaceae) 63 (1964). Type: two specimens in LINN, viz. 664-4, “ Ind. hab. ad sepes ”

Anonymous; 664-5, India, Madras, Sandras, Koenig.

var. citrifolia (Lam.) Toelken, comb. nov.

C. citrifolia Lam., Encycl. Bot. 1: 606 (1785); Eckl. & Zeyh., Enum. 14 (1835);

Sond. in Fl. Cap. 1: 62 (1860); Wild in Fl. Zamb. 1: 237 (1960). Type: Cape, without

precise locality, in Herb. Lamarck (P, holo.; PRE, photo.!). var. longifolia

Hochst. in Flora 27: 290 (1844). Type: Cape, Uitenhage, Winterhoek, Krauss s.0

(TUB, holo.!). var. sylvatica Eckl. & Zeyh. ex Sond. in Fl. Cap. 1: 612 (1860);

]

Eckl. & Zeyh., Enum. 14 (1835), nomen nudum. Syntypes: Cape, Uitenhage, Ecklon

& Zeyher (BOL!; PRE!; SAM!); Drege (PRE!); Krauss; Gamtoos River, Thunberg

(UPS; PRE, photo.!). C. capensis Thunb., Prodr. 92 (1800); Fl. Cap. 430 (1823).

Type: Cape, Gamtoos River, Thunberg (UPS, holo.; PRE, photo.!). C. volkameriae

DC., Prodr. 1: 247 (1824); Gilg & Ben. in Bot. Jahrb. 53: 199 (1915). Type: based

on Volkameria capensis Burm.f. C. laurifolia Gilg & Ben. in Bot. Jahrb. 53: 193

(1915). Syntypes: Cape, Kaimansgat, Mund & Maire s.n. (B; PRE, photo.!); Cape,

Drege 7595 (B!); Knysna, Pappe s.n. (B!; SAM!). C. woodii Gilg & Ben. in Bot.

Jahrb. 53: 194 (1915). Type: Natal, Durban, Wood 546 (B, holo.; BOL!; SAM}).

C. sepiaria is a very widespread species being recorded from northern Australia,

East Indies, Malaysia, India and most parts of Africa particularly the eastern areas.

De Wolf (1964) recognizes three varieties in tropical east Africa, but this does not

include the typical variety which is said to be very similar to the var. subglabra. Var.

subglabra also occurs in the northern Transvaal. The second South African taxon,

var. citrifolia, which occurs mainly in Natal and the eastern Cape Province, is not

so similar to var. subglabra as De Wolf infers, but is rather like var. stuhlmannii (Gilg)

De Wolf in its stouter appearance and coriaceous leaves. Var. citrifolia differs from

var. stuhlmannii in that it produces spreading hairs (rarely absent), the margins of

the sepals are ciliate and there are up to 15 ovules per ovary. In itself var. citrifolia

is very variable and extreme forms are very different. In forests usually west of Port

Alfred the plants are glabrous, often without spines on the branches and produce

long lanceate leaves up to 8 cm long. In dry bushveld, on the other hand, densely

pubescent plants with leaves rarely longer than 4 cm and well-developed spines are

found. Although even the flowering times are often different, intermediate forms

between all these characters have been observed. From Estcourt a form is recorded

with unusually long and narrow leaves rather resembling the coppice growth of var.

citrifolia in the eastern Cape.

Volkameria capensis Burm.f. might be the oldest name for this taxon, but under

present circumstances it is regarded as a nomen dubium. The diagnosis does not give

any clue as to the identity of the plant and the type specimen cannot be found.

However, the description of C. volkameriae DC. which is based on Burman’s species,

mentions recurved stipulate spines, ovate leaves and c.30 stamens. This obviously

refers to var. citrifolia, but contradicts Burman’s diagnosis which states that the plant

is without spines. In var. citrifolia spineless specimens have often been observed

particularly in the western part of the variety’s distribution, a part which had probably

been explored before the time of Burman’s description. However, the difference of

such a conspicuous character indicates that the two authors must have been working

on different specimens. Consequently, it is considered that the identity of Volkameria

capensis Burm.f. cannot be evaluated unless the type specimen can be traced. C.

capensis Thunb., though possessing the same specific epithet as V. capensis, does not

refer to Burman’s species. This was pointed out by Dandy in Bothalia 7: 427-8 (1961).

C. fascicularis DC., Prodr. 1: 248 (1824); De Wolf in Fl. Trop. E. Afr. (Cap-

paridaceae) 65 (1964). Type: Ghana, Brass (BM, holo.; PRE, photo.!).

Two varieties are recognized in South Africa and can be distinguished as follows :—

Leaves oblong, oblong-elliptic to elliptic-lanceate, usually emarginate; inflorescence with 1-3

flowers in the axils of the leaves towards the end of branches, rarely on short lateral branches

(a) var. fascicularis

Leaves lanceate, acuminate; inflorescence axillary racemose.................-.- (b) var. zeyheri

(a) var. fascicularis. De Wolf in Fl. Trop. E. Afr. (Capparidaceae) 65 (1964).

C. transvaalensis Schinz in Vjschr. Naturf. Ges. Ziirich 57: 556 (1912); Marais

in Bothalia 8: 165 (1964). Type: Transvaal, Mahilaskop, Schlechter 4510 (Z, holo.;

BOL!). var. calvescens (Gilg & Ben.) Marais in Bothalia 8: 165 (1964). C.

schlechteri Schinz, l.c. 555 (1912). Type: Cape, Tsitsa River, Schlechter 6385 (Z,

holo.!). C. calvescens Gilg & Ben. in Bot. Jahrb. 53: 195 (1915). Type: Natal, Tugela,

Wood 8472 (B, holo.; NH!). C. rudatisii Gilg & Ben., l.c. 198 (1915); Wild in FI.

Zamb. 1: 239 (1960). Syntypes: Natal, Port Shepstone, Friedenau, Rudatis 1388

(B, holo.; PRE!); Weenen, Wood 4438 (B, holo.; BOL!; NH!). C. solanoides Gilg

& Ben., I.c. 197 (1915). Type: Natal, Little Noodsberg, Wood s.n. (B, holo.; SAM!).

C. flanaganii Gilg & Ben., I.c. 197 (1915). Type: Cape, Komga, Flanagan 809 (B,

holo.; BOL!; GRA!; PRE!; SAM!). C. marlothii Gilg & Ben., I.c. 198 (1915).

Type: Cape, Hermanus?, Marloth 2599 (B, holo.!; PRE!).

(b) var. zeyheri (Turcz.) Toelken, comb. nov.

C. zeyheri Turcz. in Bull. Soc. Natur. Mosc. 27: 324 (1854); Sond. in Fl. Cap. 1: 63

(1860); Gilg & Ben. in Bot. Jahrb. 53: 197 (1915). Type: Cape, Krakakama Forests,

Zeyhe* (BOL!; PRE!). C. volkameriae sensu Eckl. & Zeyh., Enum. 14 (1835).

This species which is widespread in Africa can be recognized by its characteristic

leaves and sessile inflorescence, i.e. several flowers in the axil of a leaf. In the leaf,

the first and second pair of secondary veins are usually much longer and more

pronounced and join the primary vein at a very acute angle. The flowers are slightly

zygomorphic with the anterior sepal usually slightly saccate and the adjoining petals

are usually broader with a pronounced villose base.

In South Africa the species occurs in a wide range of habitats and shows an

interesting series of variation with distribution. A complete range of 8-23 stamens

is found in a decreasing series from north to south. Thus the South African form

could not be ascribed to either var. elaegnoides or var. fascicularis as intepreted by

De Wolf (1964). Similarly, in northern Swaziland and around Barberton, 14-18

stamens are often produced and thus the critical difference between var. transvaalensis

and var. calvescens as distinguished by Marais (1964) falls away. In central Natal

the number of stamens is about ten decreasing gradually to eight in the vicinity of King

William’s Town. Concomitant with the change of stamen number, is a gradual decrease

in flower size, which is particularly noticeable in the size of the sepals. Specimens

from the vicinity of Bathurst and King William’s Town show a marked tendency for

the flowers to be borne on short lateral branches with usually one or two flowers at

the node, but very often without a leaf subtending this axillary, sessile inflorescence.

in var. zeyheri a delicate axillary ‘“‘raceme”’ is found. However, occasionally two

flowers per node are produced indicating that the inflorescence is a raceme-like panicle.

Actual intermediates between var. zeyheri and var. fascicularis have not been seen,

but their close contact in the area around King William’s Town and Bathurst suggests

that not even subspecific rank can be applied. Var. zeyheri differs from var. fascicularis

in its lanceate, acuminate leaves. Also, var. zeyheri occurs usually in coastal forests,

whereas var. fascicularis is found in inland forests or bushveld.

Var. zeyheri extends its distribution slightly more west than var. fascicularis and

is found just west of Port Elizabeth. Var. fascicularis, however, has never been recorded

west of this and the inscription on the holotype of C. marlothii (a synonym) in Marloth’s

hand as being collected at Hermanus must be an error; the isotype in PRE was collected

near King Willim’s Town, which seems highly feasible judging by the characteristic

inflorescence exhibited by both these specimens.

H. R. TOLKEN

A New SPECIES OF MAERUA

Maerua brevipetiolata Killick sp. nov., M. rosmarinoidei (Sond.) Gilg & en.

affinis, sed plantis constanter semiscandentibus, foliis secundis, foliolis brev ribus

latioribusque, petiolis multo brevioribus, receptaculo campanulato, petalis edactis.

disco annulari fimbrillato non lobato inaequaliter laciniato differt.

Plantae semiscandentes, ad 3 m altae. Folia (1) 3-foliolata, breviter petiolata,

glabra; lamina linearis vel anguste elliptica, 1-2-3-2 cm longa, 2-4 mm lata, foliolo

mediano lateralibus longiore, apice obtuso mucronulato, basi cuneata, margini nonnihil

revoluto, costa supra depressa subtus prominenti; petiolus 0-5-7 mm longus; petiolulus

1-1-5 mm longus. Jnflorescentia floribus terminalibus paucis racemosa; pedicelli

4-10 mm longi. Receptaculum campanulatum, 4 mm longum, 4 mm latum; discus

annularis, fimbrillatus, semicarnosus, 0-6-1 mm longus. Sepali nonnihil naviculares,

late ovati, 7 mm longi, 5 mm lati, apice leviter uncinato, margine nonnihil revoluto

ciliolato. Petala redacta, ovata, 1-4 mm longa, 0-5 mm lata, unguiculata. Andro-

phorum 3 mm longum. Stamina c. 30, candida (teste Compton 30088); filamenta

1-2 cm longa; antherae oblongae, 1-3 mm longae basifixae. Gynophorum 1-4 cm

longum; ovarium oblongum, 2 mm longum, stigmate capitato. Fructus ellipsoideo-

cylindricus, 1-8-2-5 cm longus, 0:9-1-1 cm diam., colliculatus. Semina subglobosa,

c. 3 mm diam. Fic. 2.

Type: Natal, Ingwavuma Poort, c. 500 feet, 18 July, 1960, Compton 30088 (PRE,

holo.).

Scrambler up to 3 m high. Leaves (1) 3-foliolate, shortly petiolate, glabrous;

leaflets linear or narrowly elliptic, 1-2-3-2 cm long, 2-4 mm wide, the middle leaflet

longer than the laterals, apex obtuse, mucronulate, base cuneate, margin somewhat

revolute, midrib depressed above, prominent below; petiole 0-5-7 mm long; petiolule

1-1-5 mm long. Inflorescence of few-flowered terminal racemes; pedicels 4-10 mm

long. Receptacle campanulate, 4 mm long, 4 mm wide at mouth; disc annular, fim-

brillate, with erect and some incurved fimbrillae, semi-carnose, 0-6 mm long. Sepals

broadly ovate, somewhat boat-shaped, 7 mm long, 5 mm wide, apex slightly uncinate,

margin somewhat revolute, ciliolate. Petals reduced, ovate, 1-4 mm long, 0-5 mm

wide, clawed. Androphore 3 mm long. Stamens about 30, white (teste Compton 30088);

filaments 1-2 cm long; anthers oblong; stigma capitate. Fruit ellipsoid-cylindric,

1-8-2-5 cm long, 0-9-1-1 cm diam., colliculate. Seeds subglobose, c. 3 mm diam.

This species was first collected in 1956 by Murdoch near Big Bend in Swaziland.

Several years later it was collected by Professor R. H. Compton at Ingwavuma Poort

in Northern Zululand (not in Swaziland as indicated on the label of Compton 30088).

Both these collectors found the plant in flower. In September 1968 the author paid

a special visit to Ingwavuma Poort in order to obtain fruiting material of the species

The precise locality had been given to the author by Professor Compton. The plant

was found (in fruit) growing in a mixed community of Portulacaria afra, Acacia spp.,

Combretum spp., Euclea schimperi var. daphnoides, Cladostemon kirkii, Balanites

maughamit, Maerua rosmarinoides etc. occurring on the southern side of the road

about half way between the Swaziland—Natal border gate and the picnic spot amid

fine specimens of Acacia xanthophloea on the banks of the Ingwavuma River at the

foot of Cecil Mack’s Pass.

NATAL.—Ingwavuma: Ingwavuma Poort, Compton 30088; Killick 3936.

SWAZILAND.—Lubombo: 2 miles N.E. of Big Bend, Murdoch 71.

M. brevipetiolata differs from M. rosmarinoides in the following respects: it is

always a thin-stemmed scrambler whereas M. rosmarinoides can be a tree, shrub or

sometimes a scrambler or climber; the leaves are Cerro green (Ridgeway) and arranged

in one plane instead of very dark green and pendulous, and the leaflets are shorter

and broader; the petioles are much shorter (hence the epithet brevipetiolata); the

receptacle is campanulate rather than cylindric; the petals are reduced and the disc

is annular and fimbrillate instead of lobed and unequally laciniate.

D. J. B. KILLIck

“RHONA—

COLLETT

Fic. 2.—Maerua brevipetiolata. 1, flowering twig, natural size; 2, flower, x 2; 3, portion of receptacle

showing disposition of disc, x 2; 4, disc, x 5; 5, fruit, natural size. 1-4, Compton 30088:

5, Killick 3936.

COMPOSITAE

A NEw SPECIES OF SCHISTOSTEPHIUM FROM THE NATAL DRAKENSBERG

Schistostephium radicale Killick & Claassen, sp. nov. S. griseo (Harv.) Hutch.

affine, sed plantis herbaceis parvioribus, foliis radicalibus subpectinatis, capitulis

heterogamis differt.

Herba perennis, 5—6 cm alta, rhizomate ramoso subterraneo. Caules 1 vel plures,

basi ramosi, conferti, cano-villosi. Folia alternalia, radicalia, subpectinata, petiolata;

rhachis 1-6-6 cm longus; pinnae numerosae, lobis 1-5 secundis linearibus 1-2 mm

longis; petioli 1-3-5 cm longi, basi amplexicaules. Pedunculi solitarii, elongati, fili-

formes, 12-15 cm longi, sparse foliosi. Capitula heterogama, disciformia, floribus

marginis femineis, discique bisexualibus. Jnvolucrum subhemisphaericum, 7-15 mm

diam.; bracteae 3-seriatae; extimae subulatae, 4 mm longae, basi 1 mm latae, longo-

ciliatae; interiores spathulatae, marginibus apicibusque late scariosis, 5 mm longae,

1-5 mm late, ciliatae. Receptaculum planum vel nonnihil convexum, nudum. Flores

marginis: dimidium inferiore tubulosum, 1-3 mm longum; superiore campanulatum,

aequilongum; lobi 4, triangulares 1 mm longi, glandulosi; ovarium obovatum, 3 mm

longum, compressum, anguste carinatum; stylus teretus, 1-5 mm longus, ramis oblongis

truncatis | mm longis. Pappus 0. Flores disci floribus marginis similes, sed bisexuales;

antherae lineari-oblongae, 1-8 mm longae, basi cuneatae, apice appendicibus ovatis.

Fic. 3.

Type: Natal, Underberg, Sani Pass, locally common on rock outcrops in alpine

grassland just below summit of Drakensberg, 8,900 feet, 25 January, 1966, Killick

& Vahrmeijer 3760 (PRE, holo.).

Perennial herb, 5-6 cm high with branched underground rootstock giving rise

to | or more basally branched tufted stems, grey villous. Leaves alternate, radical,

compound, subpectinate, petiolate; rhachis 1-6-6 cm long; pinnae numerous; indivi-

dual pinnae of 1-5 (from base and apex to middle) linear, secund lobes 1-2 mm long;

petioles 1-3-5 cm long, amplexicaul at base. Peduncles solitary, elongate, filiform,

12-15 cm long, sparsely and minutely leafy. Heads heterogamous, many flowered,

with marginal female florets and bisexual disc florets. Jnvolucre hemispherical, flattish,

7-15 mm diam.; bracts in 3 rows; outermost bracts subulate, 4 mm long, 1 mm wide

at base, long ciliate; inner bracts spathulate, margins and apex broadly scariose, 5 mm

long, 1-5 mm wide, ciliate. Receptacle flat to somewhat convex, nude. Marginal

florets: lower half tubular, 1-3 mm long; upper half campanulate, 1-3 mm long;

lobes 4, triangular, 1 mm long, glandular; ovary obovate, 3 mm long, compressed,

narrowly winged; style terete, 1-5 mm long, with oblong, truncate branches, 1 mm

long. Pappus 0. Disc florets: identical with marginal florets but bisexual; anthers

linear-oblong, 1-8 mm long, cuneate at base, with ovate apical appendage.

Known from only two localities in the Natal Drakensberg, namely Sani Pass

and Giant’s Castle Pass.

NATAL.—Estcourt: rare in alpine grassveld in Giant’s Castle Pass, alt. 9,000 feet, 23.1.1968, Killick

3907. Underberg: Sani Pass, Killick & Vahrmeijer 3760 (PRE, holo.).

The nearest affinity of S. radicale is S. griseum (Harv.) Hutch. These two species

can be distinguished from all other species of Schistostephium by possessing solitary

capitula borne on elongated peduncles. S. radicale differs from S. griseum in the

following respects: the plant is a low-growing radical herb instead of a virgate suffrutex ;

the leaves are subpectinate with minute pinnae not more than 2 mm long, instead of

pinnatipartite with pinnae up to 20 cm long and the capitula are heterogamous and

not homogamous. The authors are indebted to Mr. E. G. H. Oliver, the Institute’s

jaison officer at Kew, for confirming that the plant is a new species.

D. J. B. KiLLick and C. G. T. CLAASSEN

3, 1-S-lobed pinnae;

All from Killick &

2-1. X (0:

1, habit, natural size; 2, portion of leaf;

4, inner involucral bract; 5, disc floret; 6, anther; 7, style.

FiG. 3.—Schistostephium radicale.

Vahrmeijer 3760.

CRASSULACEAE

A Curious FoRM OF CRASSULA NATANS THUNB. FROM THE NATAL DRAKENSBERG

During a recent collecting expedition to the Loteni-Giant’s Castle area of the

Natal Drakensberg, the author discovered a curious aquatic Crassula (Killick 3869)

common in pans at about 7500 ft (Fig. 4). At first it was thought that the plant might

represent a new species: it could not be exactly matched in the National Herbarium,

Pretoria, or at Kew. However, on closer examination of the plant, it was decided

that it was merely a form of the extremely variable Crassula natans Thunb. It differs

from typical C. natans and the known forms of the species as distinguished by Schonland

in Ann. Bol. Herb. 2: 49 (1918) in that the stems are conspicuously swollen, fleshy

and short-noded basally and filamentous and long-noded distally. Also, the terminal

leaves are congested to form rosettes (which float on the surface of the water) and the

flowers are terminal instead of being situated in the axils of the cauline leaves.

Dissection of a rosette revealed the following: the terminal leaves which are

obovate and decussate have been congested through extreme abbreviation of the

uppermost internodes; the flowers, though appearing terminal, are axillary with 1

or 2 flowers per leaf axil.

Two specimens in the National Herbarium from the south-western Cape approach

Killick 3869 in growth form. Andreae 594 has a terminal rosette of leaves, but the

stems are very much longer and scarcely swollen and fleshy at the base. Drege s.n.

resembles Andreae 594, but the stems are distinctly swollen and fleshy at the base,

although not as markedly as in Killick 3869.

The pans in which the Drakensberg plant occurs frequently dry up during winter,

which may account for the fleshy nature of the basal part of the stems.

The petals of Killick 3869 are white, tinged with purple, the anthers are pale blue

and the squamae are dark mauve. The carpels are 1-ovulate.

D. J. B. KiILLick

CRUCIFERAE

A New ComBINATION IN SILICULARIA

In continuation of my studies on the South African Cruciferae (see Bothalia 8:

166-169, 1964; 9: 97-112, 1966) it has been found necessary to make a new com-

bination in the genus Silicularia.

Compton distinguished this genus from Cycloptychis on the 1-seeded, imperfectly

septate fruits which are not beaked. In the absence of fruits it is not always easy to

distinguish these genera from each other nor, indeed. from Brachycarpaea, Schlechteria

or some species of Heliophila.

The very apt epithet of S. sigillata Compton has to be changed since it has been

found to be conspecific with Heliophila polygaloides Schltr. based on Schlechter 8900

collected at an altitude of 5000 feet in the ‘““Koude Bokkeveld’’ (Ceres distr.) on

September 8th 1896. The new combination and its synonyms are as follows:

Silicularia polygaloides (Sch/tr.) Marais, comb. noy.

Heliophila polygaloides Schltr. in Bot. Jahrb. 27: 137 (1899), non Compton 1953.

H. nubigenoides Compton in J. S. Afr. Bot. 19: 152 (1953), nom. illeg.

Silicularia sigillata Compton, l.c. 147, fig. la.

W. Marais

Fic. 4.—Crassula natans forma. 1 and 2, habit, x 2; 3, flower, x 10 (Killick 3869).

GRAMINEAE

A NEW SPECIES OF ERAGROSTIS FROM SOUTH WEST AFRICA

Eragrostis pygmaea De Winter, sp. nov., affinis E. kingesii De Winter, sed spiculis

minoribus, caryopside orbiculari embryone caryopsidem aequanti, foliius pilis longis

mollibus obtectis facile distinguitur.

Gramen annuum, erectum vel suberectum, parvum. Cul/mi plerumque recti vel

raro geniculati, 1-nodosi, non ramosi, sparsim pilosi. Vaginae pilosae. Ligula ciliata.

Foliorum laminae expansae vel plicatae, nervis scabridis. Panicula subdense contracta,

ramis solitariis vel binis angulatis scabridis rhachidi glandulosa. Spiculae virides vel

flavae, 4-7 mm longae, 1-1-25 mm latae, 5—16-florae, glumis lemmatibusque maturitate

deciduis paleis rhachidi tortuosa persitentibus; g/wmae inaequales, inferioribus quam

superioriores brevioribus; lemmata navicularia, ovata, apice acuta, costa scabridiuscula,

nervis lateralibus glabris eglandulosis. Stamina 3; antherae 0-2 mm longae, valde,

late oblongae. Caryopsis orbicularis, opaca, pallide brunnea. Fic. 5.

Type: South West Africa, Swakopmund District, Cape Cross, + km from coast,

Giess 8706 (PRE, holo.; K; M; US; WINDHOEK).

Erect or semi-erect annual up to 7 cm high. Culms unbranched, straight or more

rarely geniculate, usually 1-noded, more rarely 2-noded, basal internodes short, glabrous,

upper ones with scattered long soft bulbous-based hairs, without glands below the

nodes. Sheaths chartaceous to almost membranous, lax, with scattered bulbous-based

hairs between the ribs, eglandular. Ligule a fringe of stiff hairs. Collar very incon-

spicuous. Leaf-blade expanded, up to 2 cm long and 3 mm wide, with long bulbous-

based hairs mainly between the nerves below and sparsely scabrid on the nerves on both

surfaces, eglandular. Panicle moderately to densely contracted, rigid, erect, 2-4 cm long

and 1-0-1-5 cm wide; rhachis ribbed, scabrid on the ribs with a few scattered bulbous-

based hairs, glands usually present below the point of junction between branches and

rhachis; branches single or 2-3 together, bearing branchlets from near the base.

Spikelets bright green to pallid, about 4-7 mm long and 1-1-25 mm wide, 5—16-flowered,

breaking up from below, glumes and lemmas deciduous at maturity leaving the zig-zag

rhachis with the pales attached to it. G/umes unequal, the lower shorter than the upper,

each much shorter than the lemma it subtends, one-nerved, boat-shaped, lower about

0:75 mm, upper 1-1-25 mm long, scabrid on the keels. Lemmas boat-shaped, broadly

ovate when flattened, 1-5 mm long and 1:0 mm wide, 3-nerved, nerves prominent,

smooth except for the keels which are scabrid. Pales slightly more than half the length

of the lemmas, 2-keeled, the keels strongly curved and scabrid. Lodicules 2, cuneate,

fleshy, 0-2 mm long. Stamens 3; anthers 0:2 mm long, very broadly oblong, purple

in colour. Ovary glabrous, styles distinct, stigmas plumose. Caryopsis opaque,

orbicular, very smooth, pale brown. Embryo nearly as long as the grain. Hilum

punctiform, basal.

This minute annual is found in shallow depressions on sandy flats in the central

and coastal Namib Desert of South West Africa. This area has an average annual

rainfall of about half an inch only and many years may pass without any rains falling

at all, in particular localities.

After rains, these plants shoot up and flower and fruit with amazing rapidity.

Development may be completed with the moisture available from only one shower.

Heavy sea mists are, however, experienced at and up to 20 miles from the coast. It

is possible that these plants derive some benefit from the mist by being moistened.

SoutH West ArricA.—Omaruru: south of the Mesemberge, Giess 9651. Swakopmund: 10 miles

east of Henties Bay on road to Usakos, De Winter & Hardy 8050; Cape Cross, + km from coast,

Giess 8706.

E. pygmaea is closely allied to, but easily distinguished from E. kingesii. The

latter is apparently confined to the deserts around Luderitz south of the high sand dunes,

which lie between Luderitz and Walvis Bay, while E. pygmaea, on the other hand, has

so far only been found north of the high sand dunes.

“RHONA

COLLET

Fic. 5.—1, Eragrostis pygmaea, spikelet, x 10 (De Winter & Hardy 8050). 2, E. kingesii, spikelet,

x 10 (Kinges 2236). a, lower glume; b, upper glume; c, lemma; d, palea; e, anthers; f, mature

caryopsis (all x 15).

The type material was collected by Mr. W. Giess, who kindly made his material

available for description when he suspected it to be new. The De Winter & Hardy

collection consists of only one sheet, because this was the only material available at

the time of collecting and is hence only represented in PRE.

B. DE WINTER

A New GENERIC RECORD FOR SOUTH AFRICA

During January 1966, while plant collecting in the Southern Drakensberg, Dr.

D. J. B. Killick and Mr. J. H. Vahrmeijer found a small aquatic grass common in

a seepage area at the top of the Sani Pass (c. 9,000 ft) very close to the Lesotho border.

The grass was identified as Catabrosa aquatica (L.) Beauv. and this was subsequently

confirmed by Dr. B. de Winter at Kew.

The gathering, Killick & Vahrmeijer 3730, represents a new generic record for

South Africa. C. aquatica is known from Europe, temperate Asia, North America

and North-west Africa, where it is found in moist places usually at fairly high altitudes.

It is a stoloniferous perennial with culms 10-40 cm tall, usually rooting from the

lower nodes. The leaf-blades are up to 10 cm long, 2-8 mm wide and with a mem-

branous ligule. The inflorescence consists of an open, ovate to oblong, panicle, 3-10

cm long with the branches spreading in somewhat distant whorls. The yellow to green

or brown spikelets are borne on short pedicels and are approximately 3 mm long with

1-3 florets.

The occurrence of this grass so far from its chiefly temperate abode poses an

interesting problem of distribution. It may be that with further plant exploration

the grass will be found on the mountains of tropical east Africa, a common migration

route for temperate plants including grass species such as Aira caryophyllea L., Koeleria

cristata (L.) Pers. and the genera Anthoxanthum and Deschampsia.

J. G. ANDERSON

LILIACEAE

HAWORTHIA KOELMANIORUM: A CORRECTION

Haworthia koelmaniorum Oberm. & Hardy

In Flowering Plants of Africa, 38: t. 1502 (1968) the epithet of this newly described

Haworthia was inadvertently given the wrong ending (viz. koelmaniora). It is herewith

amended to H. koelmaniorum, to commemorate its discovery by the Koelman family.

A. A. OBERMEYER

MALVACEAE

A New COMBINATION IN PAVONIA

Pavonia senegalensis (Cay.) Leistner, comb. nov.

Hibiscus senegalensis Cay., Diss. 3: 160 (1787). Type: Senegal, Adanson s.n. (MA,

holo.!). H. baumii Guerke in Warburg, Kunene-Sambesi Exped. 299 (1903). Type:

Angola, Cuito River, Baum 760 (B, holo.t, Z!).

Pavonia hirsuta Guill. & Perr. in Guill., Perr. & A. Rich., Fl. Senegamb. Tent. 1: 51

(1831). Type: Senegal, Safal, Leprieur s.n. (P, holo.!). var. microphylla Ulbr.

in Bot. Jahrb. 57: 117 (1920); var. 8. Guill. & Perr. in Guill., Perr. & A. Rich., Fl.

Senegamb, Dent. i: 5i (831)) ype: Senegal; PRerrottet ; No: 1) 7 (Rs holos!) 5 VB:

insignis Fenzl ex Webb, Fragm. FI. Aethiop. 42 (1854). Type: Sudan, Cordofan,

Kotschy 216 (P!). P. zawadae Ulbr. in Bot. Jahrb. 48: 371 (1912). Syntypes: S.W.

Africa, Arub, Zawada sub Dinter 1347; Omantumba, Dinter 3323; Caprivi Strip,

near Sesheke, Seiner 48, 64 (Bf).

On examining the type of Hibiscus senegalensis Cav. in Madrid it became clear

that Pavonia hirsuta Guill. & Perr. is conspecific. The area of the species extends from

the Sudan and Senegal to northern South West Africa and the northern Cape. Although

no cited material of P. zawadae was seen, it is clear, from the very detailed original

description, that this species cannot be separated from P. senegalensis.

O. A. LEISTNER

SELAGINACEAE

A New SPECIES OF SELAGO

Selago trauseldii Ki/lick, sp. nov., nulla affinitate arcte obvia.

Plantae perennes, lignosae, erectae, c. 60 cm altae, interdum basi ramosae, caulibus

pubescentibus. Folia fasciculata, conferta, elliptica, 4-6 mm longa, 1-1-5 mm lata,

puberula. Capitula subglobosa, floribus densis, 7 mm diam., in corymbum disposita.

Bracteae 7-10, anguste ovatae vel subulatae, 4-5 mm longae, 1-5 mm latae, concavae,

puberulae. Calyx tubulosus, 5 mm longus nonnihil bilabiatus; lobi inaequales, lineari-

subulati, ciliati, lobis anticis duobus | mm longis, lobo postico 3 mm longo, lobis

lateralibus duobus 2 mm longis. Corolla tubulosa, alba vel pallide purpurea, nonnihil

bilabiata; tubus cylindricus, basin versus angustatus, 4-5 mm longus; labellum

anticum 2-lobatum, lobis late ellipticis, 2 mm longis, 1-3 mm latis. Stamina 4,

didynama, superioribus 1-5 mm longis, inferioribus 1-8 mm longis; antherae 0-6 mm

longae, medifixae. Ovarium ellipsoideum vel obovoideum, 0-8 mm longum, 2-loculare;

ovula in quoque loculo 1; stylus tenuis, 3-5 mm longus, stigmate simplici. Fic. 6.

AVDM

Fic. 6.—Selago trauseldii. 1, single capitulum, = 5; 2, bract, x 7; 3, calyx, x 7; 4, corolla opened

out, < 7; 5, pistil, « 7 (Killick 1637).

Type: Natal, Bergville District, Cathedral Peak Forest Research Station, Indumeni

Valley, 6650 feet, 11 January, 1952, Killick 1637 (PRE, holo.).

Plants perennial, woody, erect, about 60 cm tall, sometimes branched at the base,

stems pubescent. Leaves fasciculate, crowded, elliptic, 4-6 mm long, 1-1-5 mm wide,

puberulous. Heads subglobose, densely flowered, 7 mm diam., arranged in corymbs.

Bracts 7-10, narrowly ovate to subulate, 4-5 mm long, 1-5 mm wide, concave,

puberulous. Calyx tubular, 5 mm long, somewhat bilabiate; lobes unequal, linear-

subulate, ciliate, 2 anticous lobes 1 mm long, posticous lobe 3 mm long, 2 lateral lobes

2mmlong. Corolla tubular, white or pale purple, somewhat bilabiate; tube cylindrical,

narrowed towards the base, 4-5 mm long; anticous lip 2-lobed, lobes usually elliptic,

1-5 mm long, 1-2 mm wide; posticous lip 3-lobed, lobes broadly elliptic, 2 mm long,

1-3 mm wide. Stamens 4, didynamous, upper 1-5 mm long, lower 1-8 mm long;

anthers 0-6 mm long, medifixed. Ovary ellipsoid-obovoid, 0-8 mm long, 2-locular;

ovules 1 in each locule; style slender, 3-5 mm long, stigma simple.

Selago trauseldii is known from only two localities in the Drakensberg: it was

first collected by the author in the Cathedral Peak area in 1952 and then 15 years later

by Mr. W. R. Trauseld in the Giant’s Castle Game Reserve further south. In both

areas it grows in Themeda triandra Grassland on the Little Berg. The plant flowers

from Janauary to March.

NatTAL.—Bergyille: Cathedral Peak Forest Research Station, Killick 1637. Estcourt: Giant's

Castle Game Reserve, Trauseld 751.

33261—6

With its rigid, erect habit and round heads arranged to form a corymbose inflore-

scence, S. trauseldii is a distinct species with no obviously close ally. The flower colour

is variable: in the Cathedral Peak specimens the flowers are pale purple, while in the

Giant’s Castle specimens they are white. The species has been named in honour of

Mr. Trauseld who, through his collectings, has added so much to our knowledge of the

flora of the Natal Drakensberg.

D. J. B. KILLick

STERCULIACEAE

A New SPECIES OF HERMANNIA

Hermannia umbratica Verdoorn, sp. nov., H. malvaefoliae praecipue foliis cordatis

accedens sed inter alia floribus minoribus geminatis, petalis glabratis differt.

Planta procumbens; rami tenues, rami, petioli, pedunculi et pedicelli stellato-

pubescentes saepe pilis brevibus vel longis pluri-cellularis apice glanddulosis immixtis.

Stipula oblonga, oblongo-lanceolata vel deltoideo-oblonga, nonnunquam lobata, 2-6

mm longa, stellato-pubescens pilis brevibus apice glandulosis immixtis. Folia ovato-

oblonga vel suborbiculata, basi cordata, margine crenata, supra sparse stellato-

pubescentia, infra grosse stellato-pubescentia et minute papillosa; petiolus 4-16 mm

longus. JInflorescentia 2-flora, folia opposita; pedunculi 10-20 mm longi; pedicelli

2-12 mm longi, ad apicem cernui; bracti plerumque 3, 1-2-5 mm longi. Calyx circa

3-5 mm longa, quasi medium quinquefidus sparse stellato-pubescens pilis brevibus

apice glandulosis immixtis. Petala lutea, anguste obovata, 5-6 mm longa, leviter ad

medium angustata, ultra medium margine inflexa glabra. Stamina filamentis cruciatis,

1-5 mm longis, lobis lateralibus apice setosis; antheris 2 mm longis, acutis sparse

ciliatis. Ovarium stellato-tomentosum. Capsula tenuiter stellato-tomentosa circiter

3-5 mm longa, lobis obtusis. Semina reniformia, laevigata, nigra hilo albido.

Type: Transvaal, Potigetersrus, Pyramid Estate, Galpin 8924 (PRE, holo.).

Procumbent plant with slender branches, the branches, petioles, peduncles and

pedicels stellate-pubescent with long, many-celled, gland-tipped hairs intermixed, often

one or other of these types of hairs predominant. Stipules oblong, oblong-lanceolate

or deltoid oblong, sometimes oblique and rarely 2-3-lobed, 2-6 mm long, stellate-

pubescent with short gland-tipped hairs intermixed. Leaves ovate-oblong to suborbicular,

cordate, margins irregularly crenate, upper surface sparsely pubescent with 1—3-rayed

stellate hairs, under surface coarsely pubescent with several-rayed stellate hairs; petiole

4-16 mm long. Inflorescence leaf-opposed, geminate; peduncle 16-20 mm long;

pedicel 2-12 mm, cernuous near the apex; bracts usually 3, 1-2-5 mm long. Calyx

about 3-5 mm long, lobed to just beyond the middle, sparsely pubescent with stellate

and short gland-tipped hairs; tube widely campanulate; lobes about 1-5 mm long,

more or less deltoid. Petals yellow, rather narrowly obovate, 5-6 mm long, lower

half with narrowly inrolled margins, glabrous. Stamens with cruciform filaments,

about 1-5 mm long, lateral arms with apical setae; anthers about 2 mm long, acute,

sparsely ciliate. Ovary stellate-tomentose; styles slender, erect. Capsule thinly stellate-

tomentose, about 3-5 mm long, enclosed in the faded calyx and corolla, carpels rounded

at the apex. Seeds reniform, smooth, black with a conspicuous whitish hilum.

Found in shade on rocky slopes in the dry wooded country of the central Transvaal.

Recorded from the Pretoria, Potgietersrus and Lydenburg districts.

TRANSVAAL.—Pretoria: Wonderboom Poort, C. A. Smith 6153. Potgietersrus: Pyramid Estate,

Galpin 8924 (type). Lydenburg: near Steelpoort, Codd & Dyer 7725.

This species resembles H. malvaefolia in the trailing habit, slender stems and

cordate leaves. It differs in several respects which are not readily detected, such as the

leaves in H. malvaefolia being more orbicular, mostly broader than long and more

regularly crenate; the flowers usually solitary and slightly larger, 8-10 mm long, and

the petals distinctly pubescent dorsally. H. malvaefolia is found on the high mountains

in the eastern escarpment at altitudes above 5500 ft in contrast with the dry wooded

country of the central Transvaal from which our species comes.

The late Mr. N. Pillans recognized this as an undescribed species and it would

have been a pleasure to name it in his honour but the epithet “ pillansii ” has already

been used in Hermannia. Collectors’ notes on all the specimens seen mentioned that

the plants grew in the shade and this suggested the name H. umbratica.

I. C. VERDOORN

New NAMES IN HERMANNIA

Hermannia antonii Verdoorn, nom. nov.

H. rehmannii (Szyszyl.) K. Schum. in Engl. Mon. Afr. Pflanz. 5: 76 (1900), nom.

illeg., non Szyszyl. (1887).

Mahernia rehmannii Szyszyl. Polypet. Thalam. Rehm. 147 (1887). Type: Transvaal,

Rehmann 6648 (K, Z).

Szyszylowicz described both Hermannia rehmannii and Mahernia rehamnnii in

1887 when working on the plants collected by Anton Rehmann in South Africa. The

syntypes of the Hermannia were collected in the Cape and the Orange Free State, while

the Mahernia came from the Transvaal highveld. Hermannia rehmannii, the Cape

species, has been found to be conspecific with H. bryoniifolia Burch. (1824). Schumann

in Engler’s Mon. Afr. Pflanz. 5: 56 recognized this and placed it in synonymy under

H. bryoniifolia. In the same publication, Schumann transferred Mahernia rehmannii

to the genus Hermannia, making the combination H. rehmannii (Szyszyl.) K. Schum.

According to present day rules of nomenclature, this is not permissible and the species

is therefore now given a new name using, for historical purposes, Rehmann’s Christian

name.

Incidentally, it may be helpful to mention, firstly, that Schumann, when making

the combination H. rehmannii, placed H. brachymalla in synonymy, but today these

are recognized as distinct species; and, secondly, under H. bryoniifolia he cites Rehmann

3249 as coming from the Transvaal which is incorrect. The Rietpoort given as the

locality could not be in the Transvaal, not only because the species does not occur

there, but because of the low collector’s number, 3249. By the time Rehmann reached

the Transvaal his numbers were all in the six thousands.

Hermannia repetenda Verdoorn, nom. nov. H. hirsuta Schrad. et Wendl., Sert.

Hannov. 10, t. 4 (1795-1798), nom. illeg., non Mill. (1768).

Miller’s Hermannia hirsuta obviously describes Hermannia althaeifolia L. and has

been placed in synonymy under that species.

The later homonym, H. hirsuta Schrad. & Wendl., here given the new name H.

repetenda (meaning regained or come upon again), was described from plants flowering

in the ‘‘ Herrenhduser Garten ’’, Hannover, the seed having come from the Cape of

Good Hope. A figure accompanies the description and from this, together with type

material preserved in the Stockholm, Leningrad and Gottingen herbaria, the species

is readily identified. It is, however very rarely found represented in herbaria today

but a recent search for the species in the wild by the author has revealed that it occurs

in the Van Rhyns Pass, the Pakhuis Pass and on the Piketberg in the western Cape

Province. Its scarcity in herbaria may be attributed to its being palatable to stock and

thus seldom found by collectors.

“RHONA-

COLLET] T

Fic. 7.—Hermannia cuneifolia var. glabrescens. 1, upper portion of a branch, x 1; 2, leaf, x 7;

3, flower, x 7; 4, petal, x 7; 5, stamens surrounding the ovary, x 7; 6, stamens, x 7; 7, ovary

and styles, x 7 (Smith 4407).

A wrong application of the name was commended by Harvey who in the Flora

Capensis (1860) mistakenly cited a figure, Jacquin, Schoenbr. t. 127, as a synonym

of H. hirsuta Schrad. & Wendl. This figure represents a closely related but specifically

distinct species which is found abundantly in herbaria (not palatable?) and which is,

today, recognized as H. aspera Wend]. Jacquin had mistakenly called it H. scabra Cav.

I. C. VERDOORN

New COMBINATIONS IN THE GENUS HERMANNIA

Hermannia burchellii (Sweet) Verdoorn, comb. nov.

Mahernia burchellii Sweet, Hort. Brit. ed. 1: 57 (1827). Type: Plate 224, Bot. Reg.

3 (1817). M. grandiflora sensu Ker Gawl. in Bot. Reg. 3: t.244 (1817), partly, as to

description, plate and Burchell 2333; Harv. in Fl. Cap. 1: 217 (1860), partly, as to

Burchell 2333. ——-var. burchellii Harv., |.c. (1860). Type: Plate 224, Bot. Reg. 3

(1817).

The specimen figured in the Botanical Register, plate 224 (1817), was taken from

a plant introduced by William Burchell from South Africa and raised in a London

nursery. Burchell is quoted as writing that it came originally from the ‘‘ vast sandy

plains northwards of the town of Litakuun”’, that is north-east of the present-day

Kuruman. In that publication the specimen is wrongly identified as being conspecific

with Hermannia grandiflora Ait., which was introduced by Masson some years earlier

and differs, among other things, in having the leaves glabrous or fairly sparsely glandular

pubescent and rather deeply lobed on the margins, instead of densely setellate-pubescent

to tomentose at least on the lower surface and shallowly crenate on the margins. The

flowers of the two species are very similar and if the genus Mahernia were maintained

they would, on account of the filaments, be placed in that genus. This accounts for

the combination Mahernia grandiflora (Ait.) Burch. ex Ker Gawl. in the Botanical

Laingsburg and Carnarvon districts whereas Burchell’s species is found only in the

Kalahari north of the Orange River which Masson did not reach. The painting in

Paterson’s “‘ Travels” facing page 60, which is also cited in the Botanical Register

as being the same species as Burchell’s plant, is Hermannia stricta (E. Mey. ex Turcz.)

Harv., found only along the lower reaches of the Orange River. It has flowers resembling

H. grandiflora and H. burchellii, but is not closely related, differing in the type of

inflorescence and the long-horned capsules.

Hermannia cuneifolia Jacg. var. glabrescens (Hary.) Verdoorn, comb. nov. Lecto-

type: Drege s.n. (K!).

H. pallens Eckl. & Zeyh. var. glabrescens Harv. in Fl. Cap. 1: 190 (1860), partly,

as to Drege specimen labelled H. multiflora and annotated by Harvey as a variety

of H. pallens, excl. Barber s.n. in Herb. Hook. (K).

When reviewing the genus Hermannia for the Flora of Southern Africa it was found

that H. pallens Eckl. & Zeyh. (1835) is synonymous with the earlier species H. cuneifolia

Jacq. (1797). It was also found that the species may be separated into two groups of

varietal rank with more or less distinct areas of distribution, overlapping only on the

borders. This necessitated a clear definition of the variety H. pallens var. glabrescens

Harv.

When describing this variety Harvey cited two specimens, one collected by Drege

and the other by Mrs. Barber. On inspecting these sheets, sent on loan from Kew,

it was found that the Drege specimen, which happens to be mentioned first under the

variety, best matched the diagnosis which reads “leaves glabrescent, very sparingly

Fic. 8.—Corchorus sulcatus. 1, rootstock and in bottom right-hand corner a small stone to illustrate

how the roots curved around and under the loose stones; 2, upper portion of a prostrate branch,

showing the persistent peduncles and pedicels from which the fruits have fallen; 3, section of the

under surface of leaf, x 7; 4, sepal x 3; 5, obovate petal, x 3; 5a, suborbicular petal from

neighbouring flower x 3; 6, androgynophore bearing stamens and ovary x 3; 7, capsule x 2;

8, seed x 2.

scaly’. The specimen collected by Mrs. Barber in the type locality of H. pallens is

obviously merely an odd specimen of the typical variety, differing in that it has a few

leaves which are glabrescent on the upper surface, but they are densely “ scaly ”

(lepidote stellate) below. The Drege specimen on which the leaves are “ sparingly

scaly ’’ and the label bears the words “ H. pallens var. subglabra”’ in Harvey’s hand-

writing is therefore here selected as the type of Harvey’s variety. It happens to match

the group presently segregated as a variety and found mainly in the transitional zone

between Karoo and grassland, which stretches roughly from the vicinity of Beaufort

West north-eastwards in a widening band through Middelburg to Aliwal North and

through Herbert district into the Orange Free State, reaching Lesotho in the east.

The locality where Drege collected the type specimen is by implication the ‘‘ Hex-

rivierbergen”’ for in Zwei Documenta that is given as the only locality for Drege’s

concept of H. multiflora. This is to the south-west of the area of distribution of the

variety as indicated by the presently available material. There may be some uncertainty

about the exact locality of the Drege specimen, but there is no doubt about its identi-

fication. It is conceivable that further specimens may be found outside the main

distribution area.

H. cuneifolia var. glabrescens differs from the typical variety in the shrublets being

generally 30-40 cm tall, more repeatedly branched, the branchlets shorter, rigid and

early glabrescent; the leaves more sparsely lepidote-stellate; the inflorescence congested

at the apices of numerous, short, lateral twigs, usually with only 3 to 5 flowers in each;

flowers smaller, just over 5 mm long (instead of 8-10 mm long); the calyx more narrowly

campanulate and slightly narrowed at the throat; and the petals glabrous or nearly

so and cuneate into the claw instead of being distinctly ciliate to densely pubescent

along the margins and abruptly narrowed into the claw.

Cape.—Ceres: Bokkeveld, Hexrivier Mts.?, Drege s.n. (K, lecto., W); Beaufort West: Nieuwveld

Mts., Esterhuysen 2748; Sunnyside, Esterhuysen 5056. Murraysburg: Van Heerden 1. Cradock:

Brynard 43; Long 770; 772; Modderfontein, Acocks 12811. Middelburg: Horn s.n.; Conway Farm,

Gilfillan in herb. Galpin 2955; 5507. Richmond: Leopards’ Vlei, Bolus 15341. Colesberg: Botha

in Bloemfontein Uniy. Herb. 7547. Aliwal North: Thode A1840. Herbert: Thornhill, Leistner 1422.

O.F.S.—Philippolis: Smith 4485; 4497. Rouxville: Ecklon & Zeyher loc. 114 in Linnaea 19.

Zastron: Maree 1. Fauresmith: Smith 413; 429a; 4344; 4373; 4407; 4430; 4541; Henrici 1815;

1864; Verdoorn 1140. Thaba Nchu: Roberts 2666. Bloemfontein: Gemmell in Bloemfontein Univ.

6440; Thode A521.

LesoTHo.—Leribe: Dieterlen 755.

TILIACEAE

A New SPECIES OF CORCHORUS

Corchorus sulcatus Verdoorn, sp. nov. C. asplenifolio Burch. affinis, sed foliis

subtus dense sericeo-villosis, a C. confuso Wild pedicellis fructuosis valde recurvis

differt.

Planta perennis, caulorhiza lignosa, ramulis prostratis sericeis vel sericeovillosis.

Folia petiolata; lamina sulcata, ovata, subrotunda, vel anguste ovato-oblonga, 1-3-2

cm longa, 4-16 mm lata, supra sericea vel sparse villosa, glabriuscula, infra dense

adpressa sericeo-villosa, margine crenato-dentata, nervis lateralibus supra valde impressi

infra prominentibus; petioli 2-10 mm longi, sericei vel villosi. Cymae 1—3-florae,

suboppositifoliae, subsessiles vel pedunculis 1:5 mm longis, villosis; pedicelli 1-4 mm

longi, villosi; bracteae anguste lineares, acuminatae, sparse pilosae. Alabastra sub-

globosa, sericea vel villosa, minute mucronata. Sepala anguste lanceolato-elliptica,

c. 6 mm longa, 1-75 mm lata, extus pilosa. Petala flava, obovata, sepalis subaequilonga,

6 mm longa, 2 mm lata (interdum petala suborbicularia, sepalis breviores, 1-75 mm

lata), basi breviter unguiculata ungue parce ciliato. Androgynophorum c. 0-75 mm

altum:; stamina numerosa, filamentis 4 mm longis, antheris 0-75 mm longis. Ovariuim

3-loculare, anguste oblongo-ellipticum, strigosum; stylus c. 4 mm longus, glabriusculus.

Capsula cylindrica subarcuata, strigosa vel villosa 15-25 mm longa, 2-2-5 mm diam.,

trivalvis, pedunculo 2 mm longo, pedicellis valde recurvatis c. 4 mm longis. Fic. 8.

Type: Transvaal, Potgietersrus District, 10 miles south of Roedtan, Codd &

Verdoorn 10376 (PRE, holo.).

Perennial with a woody rootstock, branches short, usually under 30 cm long,

prostrate, sericeus or sericeo-villous. Leaves petioled; lamina sulcate between the

lateral impressed nerves, ovate, subrotund or narrowly ovate-oblong, 1-3-2 cm long,

4-16 mm wide, upper surface sericeous or sparsely villous becoming almost glabrous,

lower surface densely and persistently appressedly sericeo-villous, margins crenate-

dentate, lateral nerves deeply impressed on the upper surface, prominent beneath;

petiole 2-10 mm long, sericeous or villous. Cymes 1—3-flowered, sub-opposite the leaves,

subsessile or with a peduncle up to 1-5 mm long, villous; pedicels 1-4 mm long,

villous; bracts narrowly linear, acuminate, sparsely pilose. Buds subglobose, sericeous

or villous, minutely mucronate. Sepals narrowly lanceolate-elliptic, c. 6 mm long,

1:75 mm broad, pilose without. Petals yellow, obovate about as long as the sepals,

6 mm long, 2 mm broad (on some flowers petals suborbicular and shorter than the

sepals, 1-75 mm broad), shortly clawed, claw very sparsely and minutely ciliate.

Androgynophore about 0:75 mm long; stamens numerous with filaments about 4 mm

long, anthers 0-75 mm long. Ovary 3-locular, narrowly oblong-elliptic densely strigose;

style about 4 mm long, more or less glabrous. Capsule cylindric, somewhat curved,

strigose or villous, 15-25 mm long, 2-2-5 mm diam., 3-valved; fruiting peduncle

2 mm long, pedicels strongly recurved, c. 4 mm long, persistent.

In the Transvaal this species was found growing scattered on an extensive limestone

outcrop about ten miles south of Roedtan on the Marble Hall road. It occurred

quite frequently in this patch and the roots were found to be curling over and under

the rounded stones common in this formation. Growing with it in the same area

was Melhania griquensis which is found commonly on limestone outcrops in Griqualand

West. It is, therefore, not surprising that a specimen from the Barkly West area

collected on limestone. was found to agree with the Roedtan specimens.

Care.—Barkly West: Gong Gong, Acocks 1445.

TRANSVAAL.—Potgietersrus: 10 miles south of Roedtan, Codd & Verdoorn 10376; Verdoorn 2511.

In nature the dark green leaves spread out from the prostrate branchlets with

face upwards showing the characteristic sulcate surface caused by the deeply impressed

lateral veins. On pressed specimens the corrugations disappear to a certain extent,

but there are usually some leaves on each specimen showing this feature. The under

surface of the leaf is densely and persistently covered with long, silky appressed hairs.

On the Cape specimen this pubescence tends to be sericeous, that is silky and appressed,

whereas in the Transvaal specimens it is sericeo-villose, the hairs being slightly curly

or wavy and not so obviously silky.

C. sulcatus approaches the variable species C. asplenifolia in the procumbent

habit and the similarly shaped capsules borne on strongly recurved pedicels. It is

distinguished from it mainly by the pubescence on the under surface of the leaves,

their shape and the usually shorter more permanently prostrate branches. In C.

asplenifolia the leaves are narrowly oblong-lanceolate to almost linear and glabrous

to hispid with bulbous-based hairs, whereas in C. sulcatus they are ovate to narrowly

ovate-oblong and, as stated above, densely, persistently sericeo-villous beneath.

Bothalia 10, 1: 83-88

New and Interesting Records of South African

Fungi, Part VI

G. C. A. van der Westhuizen* and K. T. van Warmelo*

ABSTRACT

Six species of fungi, recorded for the first time in South Africa, are described. The species are

Gelasinospora cerealis Dowding from roots of Eucalyptus saligna; Spegazzinia tessarthra (B. & C.)

Saccardo from Zea mays; Melampsora larici-populina Klebahn from Populus deltoides; Saccobolus

depauperatus (Berk. & Br.) Phill. from horse dung; Humicola stellata Bunce from grass hay and Chaeto-

mium cochliodes Palliser from garden soil.

Six species are described and discussed below. Dried down cultures or specimens on

natural substrata of all these species have been deposited in the mycological collection of

the National Herbarium (PRE), at 590 Vermeulen Street, Pretoria.

1. Gelasinospora cerealis Dowding in Can. J. Res. C. 9: 295 (1933); Cain, ibid.

28: 566 (1950); Von Arx and Miiller, Beitr. Kryptogamenfl. Schweiz 11: 293 (1954).

RiguKesy lhe os

On potato-dextrose agar colonies grow rapidly covering the plate in 3-4 days

and forming a cottony to woolly, white mycelial mat. Perithecia single or cespitose,

superficial or with base partly immersed, dark brown or black, pyriform, bare over upper

parts but with dense mycelium towards the base, 800-1100 « 600-900; neck papilliform

180-250p long and 180-250 wide at the base, ostiolate and lined with periphyses; wall

thick coriaceous composed of dark-brown, angular cells externally and subhyaline cells

in the inner layers. Asci cylindrical, hyaline, unitunicate, truncate and distinctly

perforate at the apex tapering below into a fairly short stipe, 175-280 « 22-5-37-Oy, eight

spored, or, sterile olivaceous-brown, thick-walled and pitted, 137-205 « 20-25; paraphy-

ses lacking but hyaline, irregular, elongate, thin-walled, cells and narrow, hyaline thin-

walled filaments up to 3-Ou in diameter are present among the asci. Ascospores

uniseriate, hyaline at first soon darkening to pale brown or olivaceous and developing a

thick pitted epispore which darkens to deep olivaceous black, finally opaque, broadly

ellipsoid, rounded at the ends, occasionally slightly apiculate at one end, 25—40 « 20-27:5

u mostly 32 « 25p.

Specimen examined: PRE 43070, (Mycological Herbarium) on potato-dextrose

agar, isolated from diseased roots of Eucalyptus saligna, Tzaneen, Northern Transvaal.

March 1964.

The dark-coloured ascospores with thick, pitted epispore of this fungus are charac-

teristic of the genus Gelasinospora Dowding of which four species have been described.

These have been recorded on dung, roots and stems of various plants (Von Arx and

Miiller, /.c.).

* Plant Protection Research Institute, Private Bag 134, Pretoria.

EXPLANATION OF FIGURES

Fic. 1-4.—Gelasinospora cerealis. Fig. 1, perithecia (x 15). Fig. 2, young ascus and ascospores

(x 400). Fig. 3, dark coloured, sterile ascus (< 400). Fig. 4, maturing ascospore showing pitted

epispore (x 500).

Fic. 5—6.—Spegazzinia tessarthra. Fig. 5, conidium without spines (x 500). Fig. 6, echinulate

conidium ( 500).

Fic. 7-8.—Melampsora larici-populina. Fig. 7, capitate paraphysis showing thickened wall (x 1000).

Fig. 8, uredospore (= 1000).

Fic. 9-12.—Saccobolus depauperatus. Fig. 9, details of hymenium showing immature ascospores and

paraphyses (x 520). Fig. 10, single ascus (x 520). Fig. 11, operculum after ascospore release

(x 520). Fig. 12, mature ascospore cluster after release showing mucilaginous capsule (x 520).

Fic. 13-14.—Humicola stellata—Fig. 13, detail of conidiophore and conidium (x 1500). Fig. 14,

mature conidia showing variation in shape (= 1500).

The fungus described here agrees best with Dowding’s (/.c.) description of G.

cerealis but differs somewhat in having larger perithecia than in the original description.

This is an unimportant difference since it was noticed that the perithecia formed on

diluted potato-dextrose agar were much smaller. The range of dimensions given here

for the asci and spores are wider than those in the descriptions by Dowding (J/.c.), Cain

(/.c.) and Von Arx and Miiller (/.c.). The dimensions given above cover those given by

these authors, but do not overlap with the dimensions for the closely related species,

G. calospora (Mouton) C. et M. Moreau which has smaller, more elongate spores than

G. cerealis.

The dark-coloured, thick-walled sterile asci described above were seen frequently

in cultures of the South African isolate. In the young state these could be distinguished

by the absence of differentiated contents and early thickening and pitting of their walls

while still hyaline. Later, they turn almost as dark as the spores and their numbers

increase as the cultures age. Dodge (Mycologia 26: 360-376, 1934) figured similar

dark-coloured indurate, sterile asci from cultures of Neurospora tetrasperma irradiated

with X-rays. This ascus abortion was found to be due to a lethal factor which was

carried over to new generations by ascospores containing two nuclei of opposite sex at

their origin.

The perithecia lack true paraphyses but large, irregular thin-walled cells present

among the asci, are similar to those described by Cain (/.c.) in ascocarps of G. reticulo-

spora and G. calospora.

This fungus agrees with the description of G. cerealis which had been isolated only

from the crowns of wheat and oats. Despite the difference in hosts, it appears to be

identical with Gelasinospora cerealis Dowding.

This is a new host record for this species and the first record of the occurrence of

this genus in South Africa.—G. C. A. v. d. W.

2. Spegazzinia tessarthra (B. & C.) Saccardo in Syll. Fung. 4: 758 (1886); Damon,

Bull. Torrey Bot. Cl. 80: 162 (1953).

Figures: 5, 6.

Colony on malt agar slow-growing, up to 75 mm in diameter after 5 weeks at 25°C.

Mycelium submerged, hyaline, branching, thin-walled, septate; fertile hyphae in sporo-

dochia, simple, fuscous, ascending, elongate or reduced, bearing single, terminal conidia;

sporodochia separate, scattered, up to | mm in diameter; conidia fuscous, four-celled,

the cells arranged in the form of a Maltese cross, thin-walled, dark, smooth, 12-16 x 6—

8u, or strongly echinulate, 16-24 8—14y, spines 2:0—10u.

Specimens examined: PRE 43723, on malt agar, isolated from Zea mays seed,

Lichtenburg, Transvaal, June 1966; PRE 36958 on wood of Acacia mearnsii, Atholl

Experiment Station, Natal, February 1949; PRE 10003, Fungi Malayana No. 289, on

Oryza sativa, Mt. Maquiling, Laguna Province, Philippines.

The material described here agrees well with the description by Damon (/.c.). The

South African isolates are closely similar to the Philippine material, determined by

Saccardo (/.c.), in our herbarium. Of interest here is the occurrence of one South African

specimen on wood of Acacia mearnsii which differs from Damon’s (/.c.) statement that

this fungus is common on decaying monocotyledonous plant parts.

Another species of the genus, Spegazzinia meliolae Zimm., is recorded by Doidge

(Bothalia 5, 731, 1950) on various species of Meliola and related genera in South Africa.—

GiGrAnvads We

3. Melampsora larici-populina K/ebahn, in Zeitschr. f. Pflanzen-krankh. 12: 43

(1902); Sydow, Monographia Uredinearum 3: 346 (1915); Gremmen, Tijdschr. Plziekt.

60: 245 (1954); Gaumann, Beitr. Kryptogamenfl. Schweiz 12: 132 (1959); Hennebert,

Agricultura 12; 661-670 (1964).

Figures: 7, 8.

Uredosori hypophyllous, causing pale, greenish yellow spots on the upper surfaces,

mostly in scattered groups over the entire lower surface, small, up to 1 mm in diameter, at

first covered by the raised epidermis, later surrounded by the broken epidermis, powdery,

orange-yellow to golden-yellow; uredospores oblong to oblong-obovoid, 30-42 « 15—20y,

golden yellow to orange-yellow in mass, smooth at the apex but markedly echinulate-

spinulose towards the base, the wall colourless, 2-3 thick but thickened equatorially

to 6-9 resulting in a dumb-bell shaped lumen; paraphyses pyriform capitate 44-96u

long, the upper part 15—20u in diameter with wall up to 15 thick in the uppermost

part, decreasing to 3-4 in diameter along the thin-walled stalk.

Specimens examined: PRE 43694 on Populus deltoides, Waterkloof, Pretoria, Trans-

vaal, March 1967; PRE 43691 on Populus deltoides hybrids, Pietermaritzburg, Natal,

February 1967; PRE 43692 on Populus sp. (Chilean poplar), Pietermaritzburg District,

Natal, February 1967.

This rust species had not been seen in South Africa before but the specimens cited

above were brought to the author’s attention within three weeks of one another during

the very wet late summer of 1967. It is the second rust species recorded on Populus spp.

in South Africa. The other species, Melampsora aecidioides (D.C.) Schroet. had been

recorded by Doidge (Bothalia 5: 394, 1950). M. aecidioides differs from M. Jarici-

populina by its smaller ovoid uredospores which have their walls evenly thickened

and small echinulae evenly distributed over the entire surface. Its uredosori are pale

yellow in colour.

As described here, the South African collections agree well with the descriptions by

Sydow (/.c.), Gremmen (/.c.), Gdumann (/.c.) and Hennebert (/.c.). The paraphyses of

the South African material are larger than the dimensions given by Gdumann but agree

with those of Sydow. The identity of this fungus was kindly confirmed by Dr. G. L.

Hennebert of the Catholic University of Louvain, Belgium.—G. C. A. v. d. W.

4. Saccobolus depauperatus (Berk. & Br.) Phill. in Brit. Discom. 296 (1887); Seaver,

F.J. The North American Cup Fungi (Operculates) 95 (1942).

Figures: 9, 10, 11, 12.

Apothecia scattered or gregarious, superficial, attached by short central stipe only,

minute, rarely exceeding 250 in diameter, smooth, hymenium plane or slightly convex,

usually hyaline but occasionally with dilute violet pigment in the sub-hymenial layers

or with a yellow pigment in the paraphyses; asci clavate with truncate apices, operculate,

narrowing to slender bases, hyaline, eight-spored, 55-80 15-20, interspersed with

paraphyses; spores at first loosely dispersed in the apex of the ascus, finally uniting into

an elongated spore mass, 30-40 « 12-5-14-Oy; individual spores at first hyaline, darken-

ing through deep violet to brownish violet, appearing black by reflected light, ellipsoid

with narrowed ends, smooth or occasionally pitted, 12:5-14-0 5-7-5; paraphyses

septate, slender, hyaline or occasionally coloured, reaching a diameter of 3—4w.

Specimen examined: PRE 43901 (Mycological Herbarium), dried apothecia on

horse dung, Onderstepoort Veterinary Research Institute, Oct. 1967.

Although the apothecia examined do not agree in every detail with the description

of S. depauperatus, they are nevertheless considered to be representative of this species.

The observed differences, viz. the absence of the lilac colour in the excipular cells, the

larger variation in size of the asci and the slightly larger ascospores and spore-clusters,

are not considered to be sufficient to warrant the proposal of a new species. The

variation in the colours observed is considered as expressive of strain differences and

not of specific differences. The observation by Berkeley and Broome (Ann. Mag. Nat.

Hist. III. 15. 1865) that the apothecia are yellowish when young, could have been due to

the yellow pigment in the paraphyses being more noticeable during the stages preceeding

ascospore maturity.

This species has been found on horse dung from different localities in the Transvaal

and Natal and appears to be widespread.

This is the first record of the occurrence of this genus and species in South Africa.

—K.T. v. W.

5. Humicola stellata Bunce in Trans. Brit. Mycol. Soc. 44: 372-376 (1961); Cooney,

D.G.and R. Emerson. Thermophilic fungi. W. H. Freeman and Co. London. 80 (1964).

Figures: 13, 14.

On naturally infected grass the mycelium is hyaline, smooth-walled, regularly septate,

not constricted at the septa, 1-2-1-4 in diameter; conidiophores hyaline, aseptate,

smooth-walled, arising from a hyphal cell, without basal septum, 2:0-9-4 1-4-2-7p;

conidia unicellular, hyaline, becoming dark brown at maturity, appearing black by

reflected light, may be sessile on the vegetative mycelium or borne on conidiophores,

spore wall thickening slightly with age, smooth, extremely variable in shape, ranging

from ellipsoid to tetrahedral to cuboid to stellate, older spores usually with a prominent

single refractive globule.

Specimen examined: PRE 43881 (Mycological Herbarium), colonies on dried grass

hay, Johannesburg District, June 1967.

This fungus was found on bales of dried veld hay which had been exposed to rain.

It was found at this locality only and is the first record of the occurrence of this genus

and species in South Africa.—K. T. v. W.

6. Chaetomium cochliodes Palliser in N. A. Flora 3: 61 (1910); Ames, L. M. A

monograph of the Chaetomiaceae. U.S. Army Res. and Dey. Ser. 2. 17 (1963).

On potato-malt-cellulose agar (Ames, /.c.) aerial mycelium is lacking. Perithecia

numerous, ostiolate, subglobose, greyish green to yellow-green, attached to the substrate

by dark rhizoids, seldom producing cirrhi, 300-355 « 265-312; terminal hairs numerous,

of two types: (1) thick, dark brown, paling towards the tips, slightly roughened, with few

terminal convolutions; (2) thin, olivaceous or lighter, undulate or coiled; lateral hairs

numerous, dark brown, paling towards the tips, straight, slightly roughened; asci

deliquescing before spore maturity, clavate, hyaline, eight-spored, 80-95 x 9-12y;

ascospores hyaline when immature, darkening to olive brown at maturity, irregularly

biseriate, ellipsoid, bi-apiculate, 8-0-9-5 « 6-8.

Specimen examined; PRE 42931 (Mycological Herbarium), dried culture, isolated

from garden soil, Pietermaritzburg, Natal, Sept. 1962.

This is the first record of the occurrence of this species in South Africa.—

IX Wg Wao AV

Bothalia 10, 1: 89-120

An Ordination of the Vegetation of Ntshongweni,

Natal *

J. W. Morris

ABSTRACT

The physiography, soils, climate and dense woody vegetation are briefly described for Ntshongweni

a cone-shaped hill in Natal, South Africa (29° 51’ S and 30° 43’ E)

. A primary ordination of the

woody plants, based on Bray & Curtis’s (1957) method, was carried out using 60 quadrats. Four

stand noda were delimited and another four noda within a secondary ordination of a cluster of quadrats

which could not be interpreted within the primary ordination. Edaphic and atmospheric moisture

conditions and slope aspect were proposed as the main site factors correlated with species performance.

Tables of density, local frequency and constancy for species occurring in each nodum were drawn up

SOUTHERN

AFRICA

Study area

gi SN = 30°00'S

‘ Sa

1 SS j

\ N | INDIAN

‘

v S OCEAN

i Zee 2S @ Camperdown

hee 4 s

/ '

Y \ ‘ x

1 0 S\ee

2 iN PEC es

A \ Vs, Se

\ 0, S

a” Lay, Seales

er, t y x |

q 2XShongweni Pe sK ky

} @ Richmond é Dam WD

\ XN |

} 30°30'E

; = |

i Legend Sa y |

p we Miazi_ river SS pee

e c ---- Main roads = aad . - = 30°00'S

\ E

/ g y

ao" g Towns and cities ~ Sp, 4

5 0 Kilometres 15 Y

—————— nl

31°00E

Fic. 1.—Location of Ntshongweni, the study area.

INTRODUCTION

Ntshongweni is the name of a steep-sided, cone-shaped hill situated in the Mlazi

River Valley, Natal at the intersection of co-ordinates 29° 51’ S and 30° 43’ E (Fig. 1).

The anglicized name of the hill is given to Shongweni Dam, completed in December,

1927, at the foot of the hill. Since construction, the dam and land in the vicinity,

including Ntshongweni, have been controlled by the Durban Municipality.

*From a thesis accepted for the Gearee of M. Se. ley the Department oe Boone Univ ersity of Natal,

Pietermaritzburg.

Gleason (1926) and Ramensky (cited in Whittaker, 1962), working independently

at about the same time, first proposed the Individualistic Hypothesis of vegetation struc-

ture. The Hypothesis holds that no two communities are strictly identical in floristic

composition. Instead, communities exhibit continuous variation in detailed compo-

sition and cannot be readily delimited as clearcut units. No discontinuities in compo-

sitional variation occur, except where there are discontinuities in the physical environ-

ment. Ordination studies by many ecologists, including Curtis & McIntosh (1951),

Whittaker (1956 and 1960), Bray & Curtis (1957) and Curtis (1959), have shown that

variation in composition is continuous, except where the environment is discontinuous,

and have thus fostered the Individualistic Hypothesis. On the hill, Ntshongweni,

there is a gradual change in slope aspect from north-west through north, east and south to

south-west and an altitude gradient of over 200 m. It was considered that the effect

of aspect and altitude on the vegetation of the hill was eminently suitable for study by

ordination and that the vegetation was likely to conform to a continuum in the manner

indicated by the Individualistic Hypothesis.

Ordination, the technique used in this account, is the uni- or multi-dimensional

arrangement of stands so that a statement of stand position, relative to other stands

or to the axis or axes of the model, conveys the maximum amount of information about

its composition. It is thus an ecological tool for summarizing and ordering information

about the distribution and relative abundance of plants. Rewarding use can be made of

PLATE 1.—View of Ntshongweni from the north-west.

ordination as a framework for indicating potential environmental correlations because

the complexity of environmental factors determining plant distribution may be better

measured indirectly, through studying the plants, than by direct measurement of the

environment.

PHYSIOGRAPHY AND SOILS

The hill, Ntshongweni, is bounded on its north and south sides by the deep valleys

of the Sterkspruit and Mlazi Rivers. From the surface of Shongweni Dam at an altitude

of about 300 m, the land rises, gently at first, and then steeply, to high sandstone cliffs

that almost surround a small summit plateau at over 500 m above sea level (Fig. 2).

The cliffs are composed of a band of fine-grained rock, the Orthoquartzitic Marker Band

of the Table Mountain Series (Rhodes & Leith, 1966). The Marker Band is underlain

by the sandstones, grits and shales of the Basal Zone of the Table Mountain Series

that rests unconformably on Basement Complex Granite (Dodson, 1951). The narrow

valley floors are of sandy alluvium.

The soils of the summit plateau and area east of the eastern fault (Fig. 2) are deep,

grey- or red-brown, coarse, sandy loams with low organic-matter content. On the steep

north- and south-facing slopes of Ntshongweni the soil is a shallow, greyish-brown,

gritty loam. Large and small sandstone talus boulders add heterogeneity to the granite-

derived soils. The influence of talus is most apparent immediately beneath the Marker

Band cliffs.

CLIMATE

Insolation and temperature

On account of the steep topography, the south-facing slopes of Ntshongweni are

often in shadow and only intercept direct insolation in the early morning and late after-

noon. West-facing slopes, receiving insolation in the afternoon, become hotter than

east-facing slopes, because ambient temperatures are higher in the early afternoon

than in the morning. Thus the north- and north-west-facing slopes, which receive most

direct sunlight every day, become hotter and have a higher rate of evaporation than south-

facing slopes. It was assumed that a gradient of insolation and temperature from high

to low existed from the north-facing slopes to the south-facing slopes.

During the 25 year period when temperatures were recorded (Table 1), the absolute

maximum was 43-9°C at the dam on the valley floor. During the same period, the abso-

lute minimum temperature was 2-8°C. The difference in mean monthly temperature

between summer and winter was less than 10°C. Frost was not observed by the writer

at Shongweni Dam and is apparently rare.

TABLE 1.—Temperature data for Shongweni Dam Weather Station, 1932 to 1946,

from Weather Bureau (1954a)

Mean monthly Absolute maximum | Absolute minimum

Vanuany/ae cement: coer eer eut C5) 40°6 Yow

Rebrulany sence icine 22:6 38-9 11-1

Manche iectirparcccnvocie cor roey esi: | Mya3} 39-4 11-1

Aprile nen recent 20-8 36°7 8-3

Maya es eee Cee 18-4 35°6 5-0

June.. 16:0 33-9 4-4

ITI N/a Sena Gbps eee aerated 15-5 | 32:8 2°8

PNUD US Ea crrcrsteysrterererccretoe ce ensaaks 16-5 | 36:7 3-9

September rereee een eenneaae 18-2 40-0 6-1

Octobersetacsce ey ee aon. 19-7 | 41-7 8-3

Novemiber.-2+- 52... savas Le 21-2 | 41-7 7-2

IMeceni bene ins eet ne 22-0 43-9 9-4

332617.

Rain and fog

Nearly all geographers, according to Jackson (1952), and many ecologists have in the

past explained wind direction and South Africa’s summer rainfall in terms of a low

pressure trough over the interior during summer and an anticyclonic circulation over the

interior during winter. Jackson has shown, however, that this explanation is not

consistent with known wind directions. During summer and winter an anticyclone is

centred over the eastern Transvaal so that the normal air circulations are similar in

summer and winter and the seasonal reversal of prevailing wind direction, as suggested

previously, is not possible. The two prevailing winds at Shongweni Dam are north-

east and south-west in approximately equal proportions throughout the year (Schulze,

1965). North-east winds blow during fair weather and south-west winds accompany

cold fronts and rain.

‘N

TINAT NS

aren MW. :

aie

0 metres

KOZ Basement Granite

as Table Mountain

Sandstone

Alluvium

_— a Faults

O Sample sites Contour interval: 64m

Fic. 2.—Geological map of Ntshongweni (partly after Dodson, 1951).

Accounts of the synoptic situations associated with frontal activity and thunder-

storms in Natal are described by Tyson (1964a and b and 1965). Rain is brought by

cold fronts which sweep along the east coast of South Africa throughout the year at

two- to three-weekly intervals. Precipitation from these fronts is mainly confined to the

summer months. Cold fronts are preceeded by hot, dry, gusty, north-west Berg Winds

which blow for one to three days at a time. In summer, following the passage of a

cold front, the wind becomes south-westerly, the temperature drops rapidly, a thick

cloud cover builds up and gentle, continuous rain may be expected. In winter, a

temperature drop follows a cold front and a cloud layer may develop but rain is unusual.

Rain is also brought by summer storms of high intensity and short duration, during

which 25 mm of rain, or more, may fall in two hours.

The rainy season at Shongweni Dam, during which 80 per cent of the mean annual

precipitation may fall, extends from October to April. Mean annual rainfall for the

period 1929 to 1950 was 773 mm (Weather Bureau, 1954b). The highest recorded rainfall

of 1165 mm occurred in 1943, and the lowest of 519 mm in 1949.

In summer, fog may cover the higher slopes of Ntshongweni for two or three days

at a time following the passage of a cold front. Exact frequencies of fog are not known

but it was observed about four times per month during the summer of 1966. During

the dry season, radiation fogs may form. These fogs were seen on only two occasions

at Shongweni Dam during 1966. Radiation fogs are associated with the formation of

inversions in valleys at night and are dispersed by the sun’s warmth each morning.

VEGETATION

General description of vegetation

History

The vegetation of the study area was subjected to intense disturbance prior to the

building of Shongweni Dam and the declaration of the catchment immediately above the

dam as a Nature Reserve in about 1920. Comparison of the aerial photograph taken in

1937 (Plate 2: 1) with that taken in 1959 (Plate 2: 2) shows the change in cover which

took place over 22 years in the absence of tree felling and cultivation by Bantu. Large

portions of the peninsula surrounding the Bantu Kraal and areas south-east of the summit

of Ntshongweni which lacked trees in 1937 had, by 1959, a good treecover. Throughout

the area, the cover of trees increased, the summit of Ntshongweni showing least increase

although individual bush clumps appear bigger in the later photograph.

Major vegetation types

The edges of the flat summit of Ntshongweni and of the surrounding tableland

form a sharp boundary between the physiognomically and floristically different plateau

and valley vegetation types. Differences area possible result of two major environmental

factors. Firstly, the poorly drained, more-or-less level plateau differs with regard to

soil moisture status from the steep, well-drained valley slopes. Secondly, the presence of

summer fog on the upper slopes and plateau makes this environment moister than that

of the lower valley slopes. The sandy soils of the plateau are underlain by the water-

impermeable Orthoquartzitic Marker Band. Evidence for the presence of the Marker

near the soil surface is given locally by the occurrence of Syzygium cordatum, a tree

usually found along streams, but able to grow on the plateau where drainage is impeded

and the water table is high.

Only the valley vegetation was analysed quantitatively. A brief description of the

vegetation of the plateau and valley is given below, prior to an account of the quantitative

study.

APPROXIMATE SCALE

metres

0 100 200 300 400 500

[Osea esas career Cerio Vere Seman Veneers Cor foe Rice Bees |

PLATE 2.—1, Aerial photograph of Ntshongweni taken in 1937; 2, same, 1959. Photos: Trigonome-

trical Survey Office.

1. Plateau vegetation

(a) Aristida junciformis Grassland

A grassland of perennial, tufted grasses up to 1 m high in autumn, is the principal

plateau community. Aristida junciformis, which is unpalatable to cattle except in the

early part of the growing season, is dominant in terms of cover. Other grasses include

Andropogon amplectens, Cymbopogon validus, Eragrostis capensis, E. chapelieri, E.

plana, Panicum natalense and Trichoneura grandiglumis. Spring aspect forbs are

Cassia plumosa, Eriosema salignum, Hypoxis sp., Pentanisia angustifolia, Polygala

hotentotta, Rhynchosia totta, Tephrosia macropoda and Wahlenbergia undulata. In

autumn, tall, half-woody dicotyledons, such as Pseudarthria hookeri and Leonotis

dysophylla, are prominent.

(6) Faurea saligna Woodland

The narrow-crowned tree, Faurea saligna, usually occurs on shallow, sandy soils

in almost pure stands. Individuals are up to 4-5 m high with 2-6 m between crowns.

Occasionally, Acacia nilotica and Albizia adianthifolia also occur. Many F. saligna

trees, cut off about 20 cm above ground level, have produced a multiple-stemmed coppice.

As F. saligna is prized as firewood by Bantu (pers. comm. Professor A. W. Bayer), it

was probably felled for this purpose.

Where the soil is deeper, a number of tree species form an open, deciduous, Com-

bretum molle-Acacia nilotica Woodland. Distance between the flat, spreading crowns

of the trees, which are up to 4-5 m high, varies from 1 m to about 6m. Combretum

molle and Acacia nilotica are dominant, but A. robusta, Albizia adianthifolia, Syzygium

cordatum, Apodytes dimidiata, Euphorbia ingens and Tarchonanthus trilobus are also

frequent. Shorter trees, growing to 3-5 m, include Acacia gerrardii, Sapium integ -rrimum,

Maytenus heterophylla, Dichrostachys cinerea and Vangueria infausta. "Two herbaceous

climbers of this community are Clematis oweniae and Riocreuxia torulosa. Aristida

junciformis forms a continuous understory between the trees.

2. Valley vegetation

(a) Rocky Hillside Scrub

A scrub vegetation is found on the very steep, talus-littered slopes immediately

beneath the cliffs which crown Ntshongweni. On the north-facing slopes, in the

immediate vicinity of the sandstone cliffs, Crassula portulacea, Tarchonanthus trilobus,

Canthium ciliatum, Aloe arborescens and Urera tenax are common. Where the soil is a

little deeper, trees 4-5 m high from an open, uneven canopy. Trees on the north-facing

slopes include Euphorbia tirucalli and Combretum molle as dominants, with Brachylaena

elliptica, Euphorbia ingens Hippobromus pauciflorus, Spirostachys africana, Commiphora

harveyi, Dombeya rotundifolia and Grewia occidentalis also occurring. The succulents,

Euphorbia ingens, E. tirucalli, Crassula portulacea and Aloe arborescens, are a feature of

this community. As rainfall runoff is high, the presence of these more drought-tolerant

succulent species is possibly a response to the relatively dry edaphic conditions.

On the south-facing slopes, Aloe arborescens and Euphorbia grandid’ns are common

between rocks. Both are up to 1-5 m high. Pavetta gracilifolia, which is rarely over

30 cm high and is heavily browsed, is a common shrub of this community. J/boza

riparia, Sansevieria thyrsiflora and Senecio fulgens are undergrowth herbs. An interes-

ting herbaceous climber in an early stage of the woody plant succession on the south-

facing slopes is the fern Microgramma owariensis, which appears to be restricted in

distribution to this community. Petopentia natalensis is another common, herbaceous

climber.

(6) Aloe-Dombeya Woodland

One of the most easily-distinguished communities of the study area is Aloe- Dombeya

Woodland, found on the steep, lower, north-facing slopes of Ntshongweni on mineral

soils derived from deeply weathered granite. The most conspicuous component of the

community is Aloe candelabrum which grows to 2 m high and has large, succulent leaves.

Other woody plants of the community are 2-5-4 m high and form an open woodland

with a grass understory 60 cm high. Important species are Dombeya rotundifolia,

Combretum molle, Dichrostachys cinerea, Euphorbia tirucalli, Brachylaena elliptica and

Cussonia spicata. The largest tree clumps contain, in addition, Spirostachys africana,

Commiphora harveyi, Grewia occidentalis, Canthium locuples, Acacia gerrardii and

Dalbergia oboyata. Common grasses of the field layer are Sporobolus pyramidalis and

Eragrostis curyula, while Tagetes minuta and Bidens pilosa are common weeds. The

presence of these four last-mentioned species indicates a previous history of disturbance.

PLATE 3.—Aloe-Dombeya Woodland.

The valley slopes east of Ntshongweni are occupied by Acacia spp. Woodland.

The aspects are north-east-, east- and south-east-facing, but varying angle of slope and

earlier interference by man throughout the area have produced a mosaic of plant com-

munities. Soils, usually derived from Table Mountain Sandstone, are deep, except on

steep slopes. The trees are usually 2-4 m high, but can be 6 m high occasionally. Many

are deciduous and have widespreading crowns. Trees which reach the open or closed

canopy in Acacia spp. Woodland include Acacia nilotica, A. karroo, A.gerrardii, Dichro-

stachys cinerea, Grewia occidentalis, Acacia robusta, Clerodendrum glabrum, Combretum

molle, Acacia caffra, Rhus chirindensis, Sclerocarya caffra, Strychnos decussata and

Ozoroa paniculosa. Shorter trees and shrubs include Xeromphis rudis, Brachylaena

elliptica and Sapium integerrimum. On the mesic, south-east-facing slopes Dombeya

rotundifolia, Hippobromus pauciflorus, Maytenus heterophylla, Heteropyxis natalensis,

Dombeya tiliacea, Protorhus longifolia and Erythrina lysistemon are found in addition to

the trees mentioned previously.

On parts of the south-east-facing slopes occur almost pure stands of Heteropyxis

natalensis, a short tree 2 m high, together with a few Acacia caffra trees. This woodland

may have resulted from the invasion of old fields on the south-east-facing slopes by

H. natalensis and A. caffra.

(d) Euphorbia tirucalli Succulent Scrub

A small, dense stand of Euphorbia tirucalli on the peninsula which juts out from

Ntshongweni into the dam now grows on the site of what was at one time a Bantu Kraal.

The site of the abandoned kraal can be seen in the aerial photograph taken in 1937

(Plate 2: 1) and is still visible, as a ring of dark vegetation, in the 1959 photograph

(Plate 2:2). In this area, Euphorbia tirucalli trees are usually single-stemmed, about 6 m

high, and average 1 m between boles. Acacia nilotica is found occasionally beneath

the canopy as a straggly tree 3-4 m high and Dalbergia armata occurs as a woody

climber. A few scattered herbs of Acanthaceae, including Hypoestes aristata and

Phaulopsis imbricata, occur where the shade is not too dense. In view of the spread of

E. tirucalli on the north-facing slopes of Ntshongweni and its dominance on the old

Kraal site, the absence of E. tirucalli from the Acacia spp. Woodland is remarkable.

E. tirucalli was probably planted at the Kraal as a hedge, a practice of the Bantu in the

area to this day.

(e) Spirostachys africana Woodland

On the hot, dry, north-west-facing slopes of Ntshongweni Spirostachys africana

Woodland occurs. These slopes are made up from a number of small gulleys separated

by ridges. The ridges have drier soils as a result of higher runoff and better drainage

than the gulleys, where both surface and subterranean water concentrates.

Spirostachys africana, with a thick, black, rough bark, and a diameter at breast

height of up to 30 cm, is the dominant tree. Associated trees on the ridges are Euphorbia

ingens, E. tirucalli, Hippobromus pauciflorus, Combretum molle, Grewia occidentalis

and Ptaeroxylon obliquum. Commonly occurring small trees and shrubs, which form

a discontinuous, short tree and shrub layer up to 3 m high, include Brachylaena elliptica,

Ehretia rigida and Tricalysia lanceolata. Where the canopy is broken, Aloe candelabrum

occurs. Components of the dense ground flora include Achyranthes aspera, Sansevieria

thyrsiflora, Hypoestes aristata, Hibiscus pedunculatus and Panicum deustum.

The canopy is denser, there are fewer deciduous components, and the shade cast is

much deeper in the gulleys than on the ridges. Spirostachys africana and Acalypha

sonderiana are co-dominant in gulleys and other common components include Combretum

molle, Commiphora harveyi, Euclea natalensis and Euphorbia tirucalli. The shade in

summer would appear too dense for the development of a dense ground flora as the

ground is usually almost bare.

(f) Hippobromus-Acalypha Woodland

This occurs on the south-facing slopes of Ntshongweni where insolation is lower and,

as a result, temperatures and evaporation rates are lower than elsewhere in the study

area. Soils are derived from granite with varying degrees of admixture from the sand-

stone above. Under the sandstone cliffs, slopes are steep and soils shallow, but at

lower altitudes soils become deeper and slopes more gentle. The south-facing slopes,

like the north-west-facing slopes, are made up of edaphically drier ridges with greater

runoff and subsurface drainage than the intervening gulleys.

Canopy trees found on the ridges include Hippobromus pauciflorus, Ziziphus mucro-

nata, Heteropyxis natalensis, Dombeya rotundifolia, Rhus chirindensis, Euphorbia ingens,

Grewia occidentalis, Combretum molle, Protorhus longifolia and Xylotheca natalensis.

A discontinuous layer of dark-green leaved shrubs, up to 2 m tall, includes Euclea

natalensis, Azima tetracantha and Maytenus undata below the canopy. A 60 cm high,

lush, ground flora of grasses and dicotyledonous herbs, includes Panicum deustum and

Sansevieria thyrsiflora.

In the gulleys dense shade is cast by the thick, closed canopy of dark green leaves.

Common trees include Acalypha sonderiana, Dombeya tiliacea, Grewia occidentalis,

Baphia racemosa and Euclea natalensis. A large number of other species contribute

occasionally to the canopy stratum. Tricalysia lanceolata and Dovyalis rhamnoides

are common shrubs, while Popowia caffra occurs frequently as both a shrub and a

climber. Common components of the herb layer are Achyranthes aspera, Argyrolobium

tomentosum, Crocosmia aurea, Cyathula cylindrica, Hypoestes aristata, Oplismenus

hirtellus, Panicum deustum, Phaulopsis imbricata, Plectranthus purpuratus, Priva cordi-

folia and Sansevieria thyrsiflora.

On the boundary between gulley and ridge, a dense, impenetrable tangle of woody

lianes often develops. The lianes, including Dalbergia obovata, D. armata, Scutia

myrtina, Acacia ataxacantha and Capparis tomentosa with stems up to 7 cm in diameter,

form a dense tangled growth from ground level to a height of about 4m. Emergent

trees 4-5-6 m high, including Grewia occidentalis, Clerodendrum glabrum and Dombeya

rotundifolia, occur occasionally. Erythroxylum emarginatum and Popowia caffra

are evergreen shrubs or scramblers up to 2 m high that occur in the tangle. The ground

flora is poor.

Quantitative analysis of valley vegetation

Methods

1. Sampling

The valley vegetation on the sides of Ntshongweni and on the peninsula which

juts into Shongweni Dam were sampled for the quantitative study. Density values for

trees over 1 m high were recorded from 60 square sample areas, which are termed

stands, of side 30-5 m (4 acre). Stands were located by restricted randomisation,

with the advantage of sampling the area efficiently and allowing statistically-valid

comparisons of the samples to be made. Aspect, angle of slope, rock type and soil

depth were the site factors recorded for each stand.

Species-area curves were plotted (Fig. 3) to determine a satisfactory sample size.

The smallest size used was 2:3 sq m (25 sq ft) and increasing sizes were obtained by

doubling the area to a maximum of 595-4 sq m (6,400 sq ft). The method is unsatisfac-

tory according to Greig-Smith (1964) as the different sizes are not independent and, there-

fore, the number of species may be exaggerated. A more satisfactory method is the use of

separate, randomly placed samples of each size (Greig-Smith, 1964). The latter method

was not used at Ntshongweni, however, because the scale of pattern is such that samples

located at random would almost certainly land in more than one community and thus

overestimate the number of species per community. Although a marked change in the

rate of increase of species number with increasing sample size occurs at the 74:4 sq m

(800 sq ft) size the number of species continues to increase appreciably to the maximum

sample area. It was thus decided to use samples larger than 595-4sqm. The minimal

area would not be reached before the sample was either too large for a density count to be

made in a reasonable time, or more than one community would be included in the sample

as a result of the pattern of gulleys and ridges. The 595-4 sq m sample size was eventual-

ly chosen, therefore, as being the largest stand consistent with the scale of pattern of the

vegetation and that could be analysed within a reasonable amount of time.

20 B

species

Number

Area (sq. m)

Fic. 3.—Species-area curves for three localities.

The square stands were divided into four strips [30-5 m by 7-6 m (100 ft by 25 ft)]

and each strip into four squares of side 7-6 m (25 ft) for ease of sampling, so that local

frequency data were available from the 16 sub-samples comprising each stand, and so

that samples with areas from 7-6 m? to 30:5 m? were available for analysis.

2. Ordination

The ordination was based on that developed by Bray & Curtis (1957) and described in

detail by Beals (1960). Initially, coefficients of similarity (C) between every possible

pair of stands were calculated. Expressed as a percentage for two stands, M and N,

the coefficient is given by:

Cun) = aap x 100

where a is the sum of the density values of all species in stand M, b is the sum of the

values in stand N, and w is the sum of the lower values recorded for species common to

stands M and N. The coefficient may have a value of 100 if the stands are identical in

species composition and density, a value of zero if the stands have no species in common

or an intermediate value dependent upon the degree of similarity of the stands.

100

The number of similarity coefficients in the matrix of 60 stands sampled at Shongweni

Dam was 1770. Two hundred and eighty of these coefficients had values under 10 per

cent, 756 under 20 per cent, and 1153 under 30 per cent. Applying Beals’s criteria of

end stand selection to these very low similarity coefficients resulted in tight clustering of

stands in the centres of the first and second axes. Inspection of the coefficients of

similarity between pairs of stands situated in the tight clusters showed that many were

very dissimilar from each other and should not have been placed as close together as

they were. The level of distortion was thus high and the axes inefficiently utilized.

49 26+ °55 59-32

“8

Key to all figures showing quartile

distributions within ordinations.

absent: - Ist quartile: O

2nd quartile: M 3rd quartile: O

‘ Ath quartile: @

Fic. 4.—a, Primary ordination where numbers enclosed by solid lines refer to stands mapped in Fig. 5

Axes are in percentage dissimilarity. A = Acalypha sonderiana Nodum; B = Aloe candelabrum

Nodum; C = Euphorbia ingens Nodum: and D = Euphorbia tirucalli Nodum: b, distribution of

Table Mountain Sandstone (T) and Basement Complex Granite (G) within the ordination; c,

quartile distributions of total density per stand within the ordination; d, quartile distributions

of number of species per stand within the ordination.

101

The problem was not overcome by using three dimensions. These considerations lead

to the modification of Beals’s criteria for end stand selection as described below (see

also Morris, 1967).

The sum of the similarities of each stand with every other one was found and the

standard deviation of each mean similarity calculated. Stands with low sums of simi-

larity but high standard deviations were chosen for end stands. Choosing stands with

low sums of similarity ensured that the axis was relatively long while a high standard

deviation meant that the stands were spread along the axis and not clustered in one section

of the axis.

Results

1. Primary ordination

(a) Assessment for noda

Four groups of stands with similar floristic composition, termed noda, and named

after the species with the highest mean density in the stands making up the group,

are delimited within the primary ordination, shown in Fig. 4a and mapped in Fig. 5.

The stands clustered in the centre of the ordination, between the delimited noda, are

later shown to be heterogeneous and to form more than one nodum. A clear discon-

tinuity exists between the stands of the Euphorbia tirucalli Nodum and all other stands

(Fig. 4a). The other three noda grade into the central cluster and their delimitation

was done after considering the positions of the stands, in the field and within the

ordination, and of species distributions within the ordination. Other, equally justifiable

delimitations are possible. The justification for the delimitation of the Euphorbia

ingens Nodum, which appears inseparable from the stands of the central cluster, is

given later. It is described in this section because of its close proximity in the field

to the Euphorbia tirucalli Nodum. Noda are delimited to aid discussion and do not

indicate that the Individualistic Hypothesis does not hold.

(6) Relationships of noda to site factors

Stands of the Acalypha sonderiana Nodum are located on the upper slopes of

Ntshongweni facing north-west, north and east (Fig. 5). On the north-west-facing

slope the Nodum extends down from the cliff surrounding Ntshongweni, and on the

east-facing slope a tongue, terminated by stands 24 and 26, extends down from the

cliff towards the Sterkspruit River. Investigations in the field and of the aerial photo-

graphs showed that these tongues follow gulley courses. From Fig. 4b, most stands

of the community are seen to occur on Basement Granite. Most stands are located

under the Table Mountain Sandstone cliffs on very steep slopes with shallow soils.

The four stands making up the Aloe candelabrum Nodum are found along the

Sterkspruit River on the lower north-facing slopes of the hill. The shallow soils are

derived from Granite (Fig. 4b). The convex ground form, in contrast to the concave

form where stands of the Acalypha sonderiana Nodum are found at this altitude, suggests

less favourable soil moisture conditions in the Aloe candelabrum Nodum.

The Euphorbia ingens Nodum is located on the peninsula which extends into

Shongweni Dam, east of Ntshongweni, where the deep sandy soils are derived from

Table Mountain Sandstone (Fig. 4b). The central portion of the peninsula, on which

all four stands are located, is level.

Two stands, constituting the Euphorbia tirucalli Nodum, were found on the site

of the old Bantu Kraal. Being so far removed within the ordination from all others,

these stands have floristic compositions which are very different from all the other

stands. Overall site conditions appear no different from stands of the surrounding

area occupied by the Euphorbia ingens Nodum.

102

Instead of actual density values, quartile values are plotted within the ordination

as species behaviour is then easier to comprehend. The total number of non-zero

density values, arranged in order of magnitude, is divided into four groups (quartiles)

containing, as near as possible, equal numbers of stands.

Shongweni

Dam

o 6feet i999

200

o metres

B Acalypha _sonderiana Nodum Contour interval : 61m

O Aloe _candelabrum “

O Euphorbia ingens

Fic. 5.—Map of Ntshongweni showing noda delimited in the primary ordination For clarity, stands

delimited in the secondary ordination are omitted.

With few exceptions, high total densities (Fig. 4c) are found in the Acalypha

sonderiana Nodum and in the Euphorbia tirucalli Nodum, whereas low total densities

are found in the Aloe candelabrum Nodum and in the Euphorbia ingens Nodum. Stands

of the central cluster have both high and low total densities. All four noda are notable

for the low number of species per stand (Fig. 4d), while most stands of the central

cluster have many species per stand. Stands of the Acalypha sonderiana and Euphorbia

tirucalli Noda thus have high densities of a few species while the stands of the Aloe

candelabrum and Euphorbia ingens Noda have low densities of few species.

103

Only selected species are plotted within the primary and secondary ordinations.

Some unplotted species appear scattered at random within the ordinations while some

show, less clearly, the patterns shown by species which are plotted.

Stands containing high densities of Acalypha sonderiana (Fig. 6a) are almost

entirely restricted to the Acalypha sonderiana Nodum, while E. tirucalli (Fig. 6b) occurs

in stands of the Euphorbia tirucalli and Acalypha sonderiana Noda but not in the

Euphorbia ingens Nodum, which occupies the rest of the peninsula. Mean densities,

mean local frequencies and constancy for species occurring in stands of the Acalypha

sonderiana Nodum are given in Table 2. Mean density is per stand of 930-25 sq m

(4 acre) and local frequency is derived from the 16 sub-samples of which each stand

consists. An idea of the variability about the mean values may be obtained from the

standard deviations of each mean, given also in Table 2. Constancy is the frequency

of occurrence of the species in the stands of the nodum, expressed as a percentage of

the total number of stands in the nodum. The floristic composition of an “‘ average ”’

stand within the area occupied by the nodum is given by the table. The high densities

of A. sonderiana and E. tirucalli in the Acalypha sonderiana Nodum suggest that

site conditions are very suitable for their growth.

TABLE 2.—Mean density and mean local frequency, with standard deviations (s.d.),

and constancy for species occurring in more than two of the 13 stands

of the Acalypha sonderiana Nodum with a mean density greater than unity

; Density Frequency

Species Constancy

mean s.d. | mean s.d.

PA\calyphatsondenianasseneee ener: 151-6 87-2 12-8 5)73) pos}

SBuphonbialtinucalliteeee meee eee: 75:4 60-8 11-8 5-0 100-0

Spirostachystatricanase eee een: 25-4 18-4 10:3 6:0 84-6

Hippobromus pauciflorus................ 15-5 M23) 3-8 4-3 84-6

(Commo efi tO. 5 .00000090000000000000 ISje3} S97 U5 5-6 92-3

Grewiaroccidentaliseepererememareieritere Oe] 9-4 so) 3-6 92-3

Canthiumpmundianumeereee eee eee eee 8-5 20°6 DIS 3-6 69-2

Dombeyantiliaccamenerncrrmrerier irre 8-5 Teo 5:1 AD 84-6

“Brachylacnalellipticasepeerreemceeecereaas 7) Weu 5-0 3}05) 92-3

Bhretiamcidaeeeeercrecreici creer 59 4-9 Asi5 3-4 84-6

Wrerastemaxapacrtn cece resoan oceans OC we Se8) 12:9 2:4 6-4 46-2

Bucleamatalensissepepaceoaeeenerie ei 4-7 4-2 3-8 3-4 84-6

FO MINS HUCNSoccoccccx0v0db00009000K00 4-0 4:6 2:9 3-2 69-2

Conimiphorayhanveyienpeeee ee eee eee 3°8 2°27) 2-8 2:0 100-0

BUphOLbiayingensseeeeeeee ree: 2:7 Dey Wp) 28} 76:9

Schreberayalatatearmemcriac csc 1-9 3-9 (55) Soil 46-2

Ruttenlickiasverrnucosase-eenn eae ese oe: 1:8 3-1 los) 2:6 38-5

INOS GANGS Hy, ocosoovvn0500000000000 17) 2-9 0-9 i105) 46:2

Ochnatarboreaseree ee eo oes oe oe ee iLo7/ Deil 1-5 oS} 84-6

Cussoniayspicataspepete ie dockets eat 1-4 eS ilo 12 61-5

Xcaciavrobustas eee ee erect enone: ilo iLes) 1-0 1-4 61-5

Erythroxylum emarginatum.............- oil 03} 0-9 1:8 38-

Motalrofes Okothemspecicseeemeeeneeeeiae 14-7

Motalimeanidensity-peeceeeeeeeeocoeeie: | 369-0 |

* Plotted within the primary ordination

104

Fic. 6.— Quartile distributions of a, Acalypha sonderiana; b, Euphorbia tirucalli; c, Aloe candelabrum;

d, Brachylsena elliptica; e, Combretum molle; f, Dombeya rotundifolia densities within the primary

ordination. See Fig. 4 for key to symbols.

105

Stands on the upper slopes of Ntshongweni are shown by the ordination to be

similar to stands in gulleys near the valley bottom. The floristic similarity between

these altitudinally-distinct localities can possibly be explained by the compensatory

effect of soil moisture and “‘ atmospheric’’ moisture regimes. Run-off is rapid and

drainage good on the steep upper slopes while gentler slopes and thus better moisture

retaining capacities are found at lower altitudes. This gradient from dry to moist is

compensated for by an atmospheric moisture gradient from moist to dry. The air is

often moister and precipitation possibly higher on the upper slopes, through the presence

of fog, than at the base of the hill.

Species which have distributions centred on stands of the Aloe candelabrum Nodum

include Aloe candelabrum, Brachylaena elliptica, Combretum molle and Dombeya rotun-

difolia (Fig. 6c-f). No species is restricted entirely to the four stands making up this

community but all extend into the central cluster and two extend into the Acalypha

sonderiana Nodum as well. Reference to Table 3 shows that in addition to the four

species mentioned above, Spirostachys africana has a high mean density value in these

stands. The stands of the Aloe candelabrum Nodum were located in Aloe-Dombeya

Woodland, described in the general account of the vegetation. Distributions within the

ordination of species commonly occurring in the Nodum suggest that its boundaries

are not as clearly defined as was indicated in the descriptive account.

TABLE 3.—Mean density and mean local frequency, with standard deviations (s.d.),

and constancy for species occurring in more than one of the four stands

of the Aloe candelabrum Nodum with a mean density greater than unity

| Density Frequency

Species Constancy

mean s.d. | mean s.d.

eer ee | Oi ne OMOEA ER oe

*Aloe candelabrum.................-.---. | 74:5 54-7 | 133 3-4 | 100-0

“Brachylacnavellipticasrer reer eee eae eee Os 20:7 15-0 0-8 100-0

=(Choyneloyeyquian TOI, 5 60c000660000000000000 44-0 14 6nn | lis) 0-6 100-0

Spirostachystatnicanameeerereeie eee DI5) 15-3 9-3 2-4 100-0

*Dombeya rotundifolia................... 13-8 8-8 8:5 4°8 100-0

Hippobromus pauciflorus................ 9-5 10-5 3-8 2:4 100-0

WECOAMNS WOW, oc600050000000000000000 5/95) 5:3 3-8 3}97 100-0

Dichrostachysicinereasereeeeeree eee aoe 4-8 4°8 3-8 3-4 100-0

Buphorbiasingenseemrememeecer eae rr 4-8 2:9 4-0 2-6 100-0

ACaciaycafina saris mcke -ccrsic octet stareneaieees 4-3 Deo) | Bors} 221 100-0

Grewiaioccidentalistee-re cee: 4-0 4-7 3-3 3397) 100-0

Commiphorayhanveyiesseeeeeeeee oe: 3-8 sj05) 3-5 3-0 100-0

Pehretiamipidaserresasercnmcmien concer 3-8 5-6 2-8 3-6 75-0

Dombeyaitiiaccasseeeeerereeree ese: 3-5 4-7 | ileg} 5 50-0

INhusypenthenrieneamaemeeeecee rea 3-0 1:8 3-0 1-8 100-0

@ussoniagspicatalee eee rere errr | ZY O29 | Be 0-9 100-0

Maytenuspheterophyllaeanseees ees | 1-8 neil 1S o7/ | 50-0

Acaciasniloticaseer epee cena: | 3 LoS} 1-0 2 | 50-0

MotaliofMisothemspecieseeree ee eerie 8-0

mhotalgmeanidensityapeeeeae tee eerie 274-5

* Plotted within the primary ordination

106

Species with highest mean densities in the Euphorbia ingens Nodum (Table 4) are

Euphorbia ingens, Dichrostachys cinerea and Acacia nilotica. The area covered by this

Nodum was subjected to intense disturbance in the relatively recent past, a possible

reason for the low total density and lack of species with high densities. Acacia nilotica

and Dichrostachys cinerea are known to be secondary species and some other species

occurring in this Nodum may also be secondary.

TABLE 4.—Mean density and mean local frequency, with standard deviations (s.d.),

and constancy for species occurring in more than one of the four stands

of the Euphorbia ingens Nodum with a mean density greater than unity

Density | Frequency

Species | | Constancy

mean s.d. | mean Svale |

EUphonbian cn steeer eee eaee ere errr | 24-3 8-9 10-0 Do | 100-0

Dichrostachys cinerea.................... DPS) 9:4 10-3 Sos) | 100-0

ACRGE), MCWCG.s 0000000000000 ccc oDD000N0E 11-8 8-9 5:8 4-6 | 100-0

Canthiumiciliatumeeaeeeeee enone: | Sets} 9-4 5:8 4:9 | 100-0

Grewiaroccidentaliss-eeeeeene ae enero: 8-3 5-1 4-8 Aoil | 100-0

ERO VES UCU. ccoconaccaacosccoccaccs 7-0 Gell | Ges oy | 100-0

/NGEYSE) HOU Bc oc oopcc0 000 c0cevcooeo0000 6:5 S95 | 38 3:9 75:0

Clerodendrum glabrum.................. 6:3 1:3 | 4:8 1:3 100-0

Rhus; pentherion. accuse weiner 4-5 2-4 | 3-8 2-1 100-0

iNcdcia perrardiin a, ee eee 3-8 4-1 | 2-5 2-4 | 75:0

Sclerocanyaycatita eee naeeEe errr | Bes Ded | Bes 0-6 | 100-0

ACoyonoyrawion WON cccoc00000000000000000 } 38 Alo | DoS 3:0 | 100-0

Ganthiumsventosumpeaeaeeeaeeeeen eee 3-0 DoD M5) 1-3 100-0

EucleamatalensiSeepenceeneenicniceinrcee | 2-8 3-1 | 2-0 A), | 75-0

Maytenus heterophylla................... D5 3) | ies 2-4 | 50:0

: |

Bhiretiasnigidar so c\hise sec acnnis sees shee eeee 1-8 2-1 | 1:8 Dei | 50:0

+ Brachylacnavellipticaseeeereecereee nrc 1-5 0:6 1-3 0-5 | 100-0

Canthium mundianum................... 1:5 le? | i103} 1:3 75:0

Comminhoralianycyieeeee eee eee | eS ey, ER 11 75-0

Total of 20 other species.................. | 10-2

Motalemeankdensityareer eect | 136-0 |

* Plotted within the primary ordination

The most common species in the two stands of the Euphorbia tirucalli Nodum

is E. tirucalli (Table 5). Its distribution, shown within the ordination in Fig. 6b, is

limited to the stands of this Nodum and the Acalypha sonderiana Nodum. Species

occurring in both stands but not mentioned in the general account of the vegetation

are Dichrostachys cinerea, Acacia robusta, Euphorbia ingens, Canthium ciliatum, Clero-

dendrum glabrum and Protorhus longifolia. With the exception of P. longifolia, which

has a low density anyway, these species are also components of the Euphorbia ingens

Nodum, which surrounds the small Euphorbia tircucalli Nodum, so that the stands

apparently bridge the ecotone between the two noda. The marked difference between

the two noda is, however, still apparent by comparing Tables 4 and 5.

107

TABLE 5.—Mean density and mean local frequency, with standard deviations (s.d.),

and constancy for species occurring in at least one of the two stands of

the Euphorbia tirucalli Nodum with a mean density greater than unity

| |

Density Frequency |

Species | | Constancy

| mean s.d. | mean s.d. |

-Buphorbiaytinucalliseae eee cece | 285-0 65-1 16-0 0-0 | 100-0

Dichrostachys cinerea.................... W145 16-3 6-0 4:2 | 100-0

INEOA TONSA asconccccaccv0cc0e0000009 || Iles 201 Wes) Asi || 100-0

Euphorbiavingensseeeeeeee eee eee 6°5 7:8 3-0 2°8 100-0

(Acaciaimiloticaseeeee ee eeeeeeee cern: Se) 723 |) Bess 3305) | 50-0

Canthium\ciliatum. 3... 2 5.3. 35) Oo || Zo 0-0 100-0

Clerodendrum glabrum.................. 2°5 2-1 1-5 0:7 100-0

/AGROA, HITE ooos00000000000000000006 2-0 2:8 1:5 2-1 50:0

Azimaytetnacantha=peeeeeeeerne neers 2-0 2:8 195) Dll 50-0

Aibiziaadianthifoliaeereeeeeeeeeeeeee |) 2-0) 1452-01 251 |) 100-0

wiremaronientallSheeeeeeeerer eee are 2-0 0:0 2-0 0-0 100-0

Dombeyaitiliaceaseraa-aee eee eee ones 1-5 “1 1-0 1-4 50-0

IIDO GA TAA Gsoa0c0000000000000000000 1-5 2:1 1-5 Aoi | 50-0

Grewiaroccidentaliste-eeeeeeoe ene ener | 1:5 2-1 1-5 Det | 50-0

Protorhusilongifoliaseeeee eee eee eee | 1:5, 0-7 1:5 0:7 | 100-0

ZIZIPHUS MUCLONataseEee eee eee eee: | opalee) Ao 1-5 ail | 50-0

Total of 12 other species.................. | 4-5

Motalgmeanidensityaeneee eee cece | 348-0

* Plotted within the primary ordination

2. Secondary ordination

(a) Assessment for noda

A secondary ordination (Fig. 7a and b) with three axes, X, Y and Z, accounted

for the variability between the stands constituting the central cluster of the primary

ordination (Fig. 3a). Because three axes were required for the secondary ordination,

it is not surprising that a meaningful arrangement of stands of the central cluster could

not be obtained on the two-dimensional, primary ordination.

Five groups of stands, referred to as noda and named after the species with highest

mean densities in the stands of the nodum, are delimited within the secondary ordination.

The delimitation was made after inspection of stand position in the field (Fig. 8) and

within the ordination, and after inspection of species distributions within the ordination

in the same way as for the primary ordination.

As intermediate stands exist between neighbouring noda there are very few real

discontinuities within the ordination and other, equally justifiable. delimitations are

possible. The secondary ordination also supports the Individualistic Hypothesis of

Gleason (1926).

33261—8

108

Fic. 7.—a, X-Y axes of the secondary

ordination; b, Y-Z axes of the secondary

ordination. Numbers refer to stands mapped

in Fig. 8 and axes are in percentage dis-

similarity. A=Dombeya tiliacea Nodum;

B=Hippobromus-Heteropyxis Nodum; C=

Hippobromus-Spirostachys Nodum; D=

Spirostachys africana Nodum; and E=

Euphorbia ingens Nodum (mapped in Fig.

5); c, distribution of Table Mountain Sand-

stone (IT) and Basement Complex Granite

(G) within the X-Y ordination; d, quartile

distributions of total density per stand

within the X—Y ordination; e, quartile distri-

butions of number of species per stand

within the X-Y ordination. See Fig. 4 for

key to symbols.

109

The Euphorbia ingens Nodum is discussed under the primary ordination section.

Within the X-Y ordination (Fig. 7a), it is centrally placed and would appear closely

related to the four surrounding noda. On the Y-Z ordination (Fig. 7b), however,

it is shown to be clearly separated from the other noda so that its separation on the

primary ordination is justified. The Euphorbia ingens Nodum is most closely related

to stands 21 and 22 of the Spirostachys africana Nodum and stands 13, 15 and 19 of

the Hippobromus-Heteropyxis Nodum, according to Fig. 7b. The above-mentioned

stands lie west and south-west of the peninsula which is covered by the Euphorbia

ingens Nodum. Inspection in the field showed that, in addition to being close to the

peninsula geographically, these five stands were similar, physiognomically and floris-

tically, to the Euphorbia ingens Nodum. Being further from the Bantu Kraal site

and, therefore, further from a centre of past disturbance, possibly accounts for stands

13, 15, 19, 21 and 22 being, in some ways, intermediate in floristic composition between

the Euphorbia ingens Nodum and the Hippobromus-Heteropyxis and Spirostachys

africana Noda.

Ntshongweni

Dombeya __tiliacea Nodum Contour interval: 61m

Hippobromus — Heteropyxis

Hippobromus —Spirostachys

Spirostachys africana

#oao ®@

Fic. 8.—Map of Ntshongweni showing noda delimited in the secondary ordination

110

The two-dimensional ordinations of Fig. 7a and 7b are two projections of a three-

dimensional model of the actual forty-dimensional structure so that some distortion

of relative stand positions is to be expected. On the Y—Z axes, the Dombeya tiliacea

Nodum and the Hippobromus-Spirostachys Nodum are superimposed. The stands of

these two communities are, in fact, not intermingled as they appear to be, but lie

one behind the other, as shown by the X-Y axes. In the same way, the stands of the

Euphorbia ingens Nodum on the X-Y axes should not be thought of as lying in the same

plane as the other four noda but as either above or below the plane of the other

four noda.

Bearing in mind that the Euphorbia ingens Nodum is distinct from the other four

noda, only the X—Y axes need be used for further discussion of the secondary ordination.

The Y—Z axes contain less information than do the X—-Y axes, as two of the noda

which can be recognized in a three-dimensional model overlap on the Y-Z axes. A

further reason for using the X and Y axes in preference to the Y and Z axes is that

the former are almost orthogoral while the latter are non-orthogonal.

(6) Relationships of noda to site factors

Stands 2, 8, 12 and 16 of the Dombeya tiliacea Nodum occur under the Ortho-

quartzitic cliffs of Ntshongweni where slopes are very steep and soils are derived from

both Table Mountain Sandstone and Granite (Fig. 7c). The other stands of this

Nodum occur on Granite at lower altitudes, but a deep alluvium overlies the Granite

where stands | and 3 occur on the edge of the Mlazi River floodplain. Stand 42 of

the Nodum occurs on the north-facing side of Ntshongweni on the minor south aspect

which can be seen in the aerial photographs (Plate 2). The south-aspect influence is

sufficiently strong to give this stand a floristic composition similar to that of stands

on the main south-facing slopes.

Surface and subterranean drainage is good on the steep, upper slopes but at lower

altitudes where slopes are more gentle, movement of subterranean water is slower so

that soil-moisture conditions are more hydric at lower than at higher altitudes. Con-

versely, atmospheric moisture conditions are more hydric at the higher altitudes because

summer fogs, when present, usually clothe only the crest of Ntshongweni. The com-

pensatory effect of soil and atmospheric moisture regimes discussed above is a possible

explanation for the similarity, shown by the ordination, between stands from the upper

and lower south-facing slopes. Low insolation is a characteristic of all the south-

facing slopes.

The Hippobromus-Heteropyxis Nodum also occurs on the south-facing slopes of

Ntshongweni. Some stands are located between the upper and lower groups of stands

of the Dombeya tiliacea Nodum and the others occur south-west of the peninsula.

Stand 38 occurs next to stand 42 of the Dombeya tiliacea Nodum. Most stands occur

on Granite. Field inspection showed that the stands of this community occurred on

ridges with good surface and subterranean drainage. Stands of the Hippobromus-

Heteropyxis Nodum are thus probably more mesic, either edaphically or atmospherically,

than stands of the Acalypha sonderiana and Aloe candelabrum Noda which were

delimi.ed within the primary ordination.

All but one of the stands of the Spirostachys africana Nodum occur on Table

Mountain Sandstone, along the lower reaches of the Sterkspruit River. Stands are

near a perennial, subterranean water supply and on gently sloping ground so that

the moisture content of the soil is good. As most of the stands are east-facing insolation

is of intermediate intensity.

111

Measurements of actual soil-moisture content, atmospheric moisture content and

quantity of insolation would have to be made to confirm the interactions suggested

in Fig. 9. Being on the south-facing slopes, insolation is relatively low in the Dombeya

tiliacea and Hippobromus-Heteropyxis Noda but soil moisture content is probably

higher in stands of the Dombeya tiliacea Nodum than in those of the Hippobromus-

Heteropyxis Nodum. Soil moisture content is probably also high in stands of the

Spirostachys africana Nodum. Conversely, soil moisture content is low on the steep,

north-facing slopes occupied by the Hippobromus-Spirostachys Nodum. Relatively high

insolation is experienced by stands of the Hippobromus-Spirostachys and Spirostachys

africana Noda as they occur on north-, north-west- and east-facing slopes.

Two gradients from hydric to zeric may be thus recognized (Fig. 9). The Dombeya

tiliacea Nodum which is considered to be most hydric (low insolation and high soil

moisture content) grades into the Hippobromus-Spirostachys Nodum which is con-

\ x

~ ee 53

NN “ <

\ AS

°° .*

S e

° NN

WN < \

Nene N

eed

Fic. 9.—Summary of suggested interactions of insolation and soil moisture on the X-Y axes of the

secondary ordination. Only positions of end stands are given. A = Dombeya tiliacea Nodum:

B = Hippobromus-Heteropyxis Nodum; C = Hippobromus-Spirostachys Nodum; and D = Spiros-

tachys africana Nodum.

sidered most xeric (high insolation and low soil moisture content) through the Hippo-

bromus-Heteropyxis Nodum (low insolation and low soil moisture content) and the

Spirostachys africana Nodum (high insolation and high soil-moisture content).

(d) Species behaviour

Stands of the Dombeya tiliacea and Hippobromus-Heteropyxis Noda generally

have low total densities while high total densities are found in most stands of the

Hippobromus-Spirostachys and Spirostachys africana Noda (Fig. 7d). Stands of all

four noda have many species per stand, the lowest number of species per stand being

found in stands of the Dombeya tiliacea Nodum (Fig. 7e). The Dombeya tiliacea

Nodum is, therefore, characterized by relatively few species and low total densities.

Many species occur in the stands of the other noda, stands of the Hippobromus-

Heteropyxis Nodum having low total densities and stands of the Hippobromus-Spiros-

tachys and Spirostachys africana Noda having high total densities. Very few species

are restricted in distribution to the stands of any one nodum. As a rule, a species

reaches it highest density in the stands of one nodum, but also occurs in the noda on

either side.

TABLE 6.—Mean density and mean local frequency, with standard deviations (s.d.),

and constancy for species occurring in more than two of the ten stands of

the Dombeya tiliacea Nodum with a mean density greater than two

Density | Frequency |

Species | Constancy

| mean s.d. | mean _ s.d.

a eicaen ae earte S ee ew om a =

= Dombeyartiliaccaseer er eeeeeneeeeennierntn 32:5 35-3 9-3 4-9 90:0

Grewia occidentalis...................... 19*5 16:8 9-0 4:7 100-0

*Tinicalysiaylanceolataseeee renee 15-7 21-2 4-6 Sjo5} 80-0

Acalypha sonderiana.................... | 15:2 24-9 4-8 6:2 60-0

Euphorbiayingenshe- eee eee 14:4 31:9 25) 2:9 80:0

+Eihnretiahigidaeereeereee corer nee cin 11-3 15:0 5:3 3-7 90-0

Bucleanatalens|Seeeeeeeeeecernecernrnrnne 11-1 9-3 | 6:5 4-5 90-0

*Hippobromus pauciflorus................ 8-1 10:8 | 3:9 4-0 80-0

*Brachylaenavellipticassreaeeeeeeeeeninenee 4:9 6-1 2°8 3-0 | 60-0

*Erythroxylum emarginatum.............. 3:8 3:8 2-8 2°4 | 70:0

PAZimantethacanthasneeererert erie 3:4 3:4 2:8 AG | 70-0

Ptaeroxylon obliquum................... 3:4 7:8 | 1-7 32) || 50-0

Rhus chirindensis....................... 3309) 3-5 2:4 2:0 | 90-0

Canthium mundianum................... 2:9 4:2 lei 1-6 50-0

*Dovyalis rhamnoides.................... 2:8 4-1 1-6 Aoi 50:0

Xylotheca kraussiana.................+0. | De Ze) I ie as | 40-0

INGESA, HOU onoccn000dcdcouvsc0eun0000 2:4 3:5 1-6 1:6 | 70-0

Maytenus heterophylla................. 2:4 2:9 ile7/ 1:8 70-0

Rhus penthert. nee et ee eee 2:3 A\08} | ilo 2-0 | 50:0

ZAZiphuspmuctonatasse serene 7128} 2:6 2-0 Do? 60:0

*(Protorhus longifolia).................... (1-8) (3:7) | (0-9) (1-6) | (50-0)

Total of 34 other species.................. 33-8

MOotalpmeanidensityaeeereeeieierteaen insite 197-9

* Plotted within the secondary ordination

Fic. 10.—Quartile distributions of a, Azima tetracantha; b, Dombeya tiliacea: c, Dovyalis rkamnoides 3

d, Erythroxylum emarginatum; e, Protorhus loigifolia; f, Tricalysia lanceclata densities within

the X-Y ordination. See Fig. 4 for key to symbols.

114

Species which have distributions centred on the stands of the Hippobromus-

Heteropyxis Nodum, for example, Dombeya rotundifolia, Heteropyxis natalensis and

Hippobromus pauciflorus (Fig. 11 a-c) are widespread throughout the area covered by

stands of the secondary ordination.

Species which have distributions centred on the stands of the Dombeya tiliacea

Nodum include Azima tetracantha, Dombeya tiliacea, Dovyalis rhamnoides, Erythrox-

ylum emarginatum, Protorhus longifolia and Tricalysia lanceolata (Fig. 10). Of these,

only Azima tetracantha is restricted, almost entirely, to this Nodum. Because the stands

of the Euphorbia ingens Nodum are in a different plane from those of the other com-

munities, crescent-shaped distribution patterns are shown by many species within this

ordination (for example: Protorhus longifolia and Tricalysia lanceolata in Fig. 10).

TABLE 7.—Mean density and mean local frequency, with standard deviations (s.d.).

and constancy for species occurring in more than two of the eleven stands

of the Hippobromus-Heteropyxis Nodum with a mean density greater

than two

| Density | Frequency |

Species | Constancy

| mean s.d. mean ec |

*Hippobromus pauciflorus................ 60:6 46-6 10-7 Soy | 100-8

*Heteropyxis natalensis................... 38:0 44-0 7:4 6:8 | 81-8

(Ciel OCAGISNAWS.coccc0500c0c005000000 || ANC) 18-5 11-2 5-0 100-0

Combretum molle....................... ey) 9:8 5-9 4-2 81-9

“Dombeyartiliaccasepereneeeneeereern 9-4 13-0 3-9 4-6 54:5

Dichrostachys cinerea.................... 8:9 13-7 3-5 Abo | 54°5

#A Cacia Cafital ert yack ccek oie ee 8-0 14-7 3-6 So 45-5

ShecOmanialcapensiS Hanne eer erent 7:3 14:0 | 2:8 4-6 36:4

*Brachylaena elliptica..................... 6:9 9:2 | 4:4 3-9 | 90-9

Acaciayniloticaserasceo oie aoe eee 6:6 8-0 4:1 4-1 81-8

*Dombeya rotundifolia................... 6°5 6:1 4°5 AB | 90-9

=Bhretiains(daseeeeeeesee eee eee 6:0 6:0 3°6 os) | 90-9

Bucleamnatalensiseeereeenenneerrcnecnenee 5:6 5:8 3-8 3-6 81:8

Maytenus heterophylla................... 5)°2 44 | 3-7 2-6 | 100-0

Euphorbiainsensseeseeeeeeee eer 4:5 40 | 3:6 De 90-9

Putterlickia verrucosa.................... 3:6 4-6 De 2-9 | (Qa

Ziziphus mucronata..................... 3-1 3-1 2:6 2-4 100-0

Glalisenayanisataeeeeeeeeere ea nrne 2°8 3-3 1:8 Don, | 63-6

hagaraicapensiSeeee een enone nner Dod) 4:6 1:8 2:9 54-5

D-CROMIO VG HUGS, cooc000cv0000s000000006 Dod 2:3 Dra De || 63-6

Canthium ciliatum...................... 2.4 GG | tel Do 23}

Commiphora harveyi.................... 2.4 1:9 | 2-3 lo” | 72-7

Canthiumiventosumteeen eee ease eeeeeeer D2 2:4 1-6 1.9 63-6

IMESACTUS WHCEN. ccocon0dcv0cv000000000 2.2 3-4 ilo7/ Dos | 45-5

*Tricalysia lanceolata..................... 2.1 4-4) 1:5 Dal | 21)

Wray quite) JVEIS) 6 00000000000003000000 oil Je | ie) Soil | 45-5

Total of 29 other species.................. W307

Tove! ERIN COSI, 5 cc0coba00000000000008 262.7 |

* Plotted within the secondary ordination

Fic. 11.—Quartile distributions of a, Dombeya rotundifolia; b, Heteropyxis natalensis; c, Hippobromus

pauciflorus; d, Acacia caffra; e, Acacia gerrardii; f, Brachylaena elliptica densities within the

X-Y ordination. See Fig. 4 for key to symbols.

116

Apart from species which are plotted within the ordination in Fig. 10, Grewia occiden-

talis, Acalypha sonderiana, Euphorbia ingens and Euclea natalensis have high mean

densities in stands of the Dombeya tiliacea Nodum (Table 6). Grewia occidentalis

is not pjotted, because it is ubiquitous through the study area, being absent from only

two of the 60 stands. Protorhus longifolia is included in Table 6, even though it does

not qualify for inclusion, as it has a clear pattern of distribution within the ordination

and isknown to be a species appearing early in the succession to forest. Its distribution

within the study area on the south-facing slopes indicates the area which, in the absence

of disturbance, has the potential of being covered by forest.

Apart from the species whose distributions within the ordination are given, Grewia

occidentalis, Combretum molle and Dombeya tiliacea have high mean densities in the

Hippobromus-Heteropyxis Nodum (Table 7).

TABLE 8.—Mean density and mean local frequency, with standard deviations (s.d.),

and constancy for species occurring in more than two of the eight stands

of the Hippobromus-Spirostachys Nodum with a mean density greater

than two

= — = — = —— ——S = = — = — — — a

Density Frequency |

Species | | Constancy

| mean s.d. | mean s.d. |

= ae eee Vana te a. (ae, Pel

*Hippobromus pauciflorus................ | 70-1 36:7 | 12-6 Ae 100-0

*Spirostachystainicanaseeee eee | 40:5 Aleit jf Wel 5j03} 8-57

Combretumimollesnee seen eee | 23-1 13°8 10-0 4-1 100-0

| |

GROW, OCICSUAMBs,cc0ccn0c5coos0vcccc || AleD 19 10-6 2-9) | 100-0

~Brachylacnavellipticatereeeee meee nn ieLOnS lilo 8-5 3-9 100-0

ZAloeicandelabrumkene eee eee eee | 12:6 7:4 6:4 3-9 87-5

ERO VG TWANG, ooococcoscascccds00000e | ilileS e4h | 7/85) 2°8 100-0

SIMCHA HECA.cocsoonvcvocccnacng000acce 10:9 8-4 6:3 3-3 100-0

*Acaciaicafitac cn Seer eras mean anor | 9-8 DoS) 5°33 4-4 75-0

Acalypha sonderiana.................... 8-8 So7 | WO 3305) 37-5

*Dombeya rotundifolia................... 8-6 8:4 | 4:9 4-3 | 87-5

Maytenus heterophylla................... 7:5 70 | 5:5 3-4 | 75:0

Canthium mundianum................... | 7:4 7-1 | 4-0 3-7 | 75:0

*Heteropyxis natalensis................... 7-0 5:8 | 2:4 4:8 62-5

IRIS JOANN, 66 cascconsandcccdnGnobace 6°6 5:0 | 5-4 33 87-5

Dichrostachys cinerea.................... | 4:9 Goch || shed 3-8 75:0

Euphorbia ingens).......0.)..0000s0000- | 4-9 3:0 3:8 Doil | 100-0

Bucleamnatalensisneee reenter eee | dog} 1-9 3-9 1-9 87-5

Miaytenusnind ata annie aeerenenent 4-1 4-7 355 o7/ 62-5

Putterlickia verrucosa.................... 305) 376 |. 2:3 1:8 75-0

Acacia nilotica.. Sil 2-9 | 2-5 Bes | 75-0

| |

ACACIA CCLIAT CER EEE Rea GE nee ereieen 2-9 Soil || ech 23} 62:5

Clausena anisata..................s see, A Ao | le Gel 37-5

ZL VMNNUS HOUOROOAMA cooonecc0vn,00000000 Ded) 2:8 | 2-1 De3} 62:5

FEKANE CHOSIISs ococcgon0000000000000000 2:3 1-5 2-0 1:2 87-5

Total of 27 other species.................. 19-2

MOfalsmeanidensityareeeeee Cece 317:3

ce Plotted within ‘the secondary ordination

ILIL7/

Fic. 12.—Quartile distributions of a, Aloe candelabrum; b, Spirostachys africana; c, Xeromphis rudis,;

d, Canthium locuples; e, Ehretia rigida; f, Tecomaria capensis densities within the X-Y ordination

See Fig. 4 for key to symbols.

118

Species with relatively high densities and high constancy values in the stands of

the Dombeya tiliacea Nodum, but are absent from or have low densities in stands of

the Hippobromus-Heteropyxis Nodum, are Dombeya tiliacea and Fricalysia lanceolata.

Acacia nilotica, Combretum molle, Dombeya rotundifolia, Heteropyxis natalensis and

Hippobromus pauciflorus are absent from, or rare in, the Dombeya tiliacea Nodum,

but have relatively high densities in the Hippobromus-Heteropyxis Nodum.

Acacia caffra, A. gerrardii and Brachylaena elliptica have distributions centred on

the stands of the Hippobromus-Spirostachys Nodum (Fig. 11 d-f). Acacia gerrardii

is restricted, almost entirely, to this Nodum, while Brachylaena elliptica has a very

wide distribution from the stands of the Hippobromus-Heteropyxis Nodum to the

Spirostachys africana Nodum.

TABLE 9.—Mean density and mean local frequency, with standard deviations (s.d.),

and constancy for species occurring in more than two of the eight stands

of the Spirostachys africana Nodum with a mean density greater than two

2 |

Density | Frequency

Species Constancy

mean s.d. | mean s.d. |

= — — — _—— — — — — ee ———— me

*Spirostachys africana.................... 43-1 27-4 12:4 4-6 | 100-0

*Brachylaena elliptica..................... 19-1 16:2 8-9 5-9 87:5

DGS WANG, 5 5000000000000000000000 17:3 10-6 9-0 4-4 | 100-0

Combretumpmollesaeeeeeeeee ne eenenne 14-5 8-2 7:8 D3 | 100-0

*Hippobromus pauciflorus................ 14:0 15-6 5-1 4:6 87-5

Clausenavanisa tannin nen eren nae | 13-0 13332) 4-5 3°5 | 87-5

Grewiaroccidentalisee ere enter nenr | 12-9 7:3 75 29) || 100-0

“Aloercandelabrum-ereee eee eee ee een: 10:5 14-7 5°33 dog | 87-5

Canthium mundianum................... 9-1 9:4 ) 4:5 me) 100-0

Buphorbiasingensseeseeaene eer erenner 8-1 5)03} 4-6 225) 100-0

Ochnayarborcaee eee eee eee enenore 7-3 84 | 4-0 4-0 | 75-0

Ecleabnatalensiseeeeenennneeeeennrnenn 6-3 4-2 4:4 2:8 87-5

xXvlothecaykraussianadeeee eee aera | 6:3 15-0 2:1 4:2 | 37-5

ACA Nt NVTTION NOCIONI. 00000000000000n0000 6:1 U3 D9) 3-6 | 62:5

PMecOMmantagcapens |S eee eee Eee nate 6-1 Te3 4:0 AG | 62:5

Dichrostachys cinerea.................... SY) 5-8 3°5 32 | 75.0

Acaciamiloticarennnssetaa noe se eee } SS) 5-1 4:6 4-4 62.5

“Ehretiasnigidannny tee eae nme | 55) 3-0 4:6 225) 100.0

Acalypha sonderiana.................... 5:3 8-7 25) DoT | 50-0

May tenusiundataaeeerecenrecnnrnnenrnnn: 4:9 Yeu 3:6 4-3 | 75:0

Maytenus heterophylla................... 4-6 S°7/ 29 3307) | 62-5

Putterlickia verrucosa.................... 4-4 4-1 3:3 2:6 87-5

Municakdiscololtyeeereeteeeneneernentr 4-1 10-9 1-1 2°5 37/25)

Rihtistpentherisneaskeren eee eonenan 3:9 3-6 2-3 2-1 | 100-0

Canthium ventosum..................... 3-5 3-7 2°6 1:9 | 87-5

*Heteropyxis natalensis................... 3-0 3-9 1:9 Don | 62:5

C@anthitinc latte | dow 5-0 2-1 B05 | 75-0

*Dombeya rotundifolia................... 2:7 2-6 2:5 Des | 75:0

Total of 27 other species....... Rc uyaene 28:6

MoO taleme dec enis ty Ameer nner anne 278-3

* Plotted within the secondary ordination

119

Apart from the species after which the Hippobromus-Spirostachys Nodum is named

and species plotted within the ordination, Grewia occidentalis, Combretum molle and

Dichrostachys cinerea have high mean densities (Table 8).

Aloe candelabrum, Spirostachys africana and Xeromphis rudis (Fig. 12 a—c) occur

in both the Hippobromus-Spirostachys and the Spirostachys africana Noda. Canthium

locuples (Fig. 12d) is restricted, almost entirely, to the Spirostachys africana Nodum

while Ehretia rigida and Tecomaria capensis (Fig. 12e and 12f) occur in the stands of

the Spirostachys africana Nodum and on the south-facing slopes of Ntshongweni.

Apart from species plotted within the ordination, Combretum molle, Clausena anisata

and Grewia occidentalis have high mean densities in stands of the Spirostachys africana

Nodum (Table 9).

GENERAL CONCLUSIONS

Ordination was successful in aiding the writer’s understanding of the vegetation

and more knowledge was gained about the vegetation than if a descriptive survey

alone had been carried out. Too much variation was inherent in the vegetation for

it to be illustrated in two or three dimensions as used by Whittaker (1960) and by

Bray & Curtis (1957). Two ordinations, using a total of five dimensions and a modified

method of end stand selection, were required. Thus, used with discretion, ordination

was a valuable tool in the study of this subtropical vegetation.

Correlations between species behaviour, illustrated by a stand ordination, and

certain site factors were proposed. Actual measurements of factors, like quantity of

insolation and soil moisture content, necessary to substantiate the proposed correlations,

were precluded by a lack of time. Such measurements would have to be included in

a more thorough investigation.

Evidence in support of the Individualistic Hypothesis is given, particularly by the

secondary ordination. Although it was possible to delimit noda, it was shown that

site factor and floristic gradients exist between noda.

ACKNOWLEDGEMENTS

It is a pleasure to acknowledge the interest shown and encouragement given to

me by Professor A. W. Bayer during the preparation of my thesis. For advice, I am

also grateful to Dr. D. B. Woods who supervised the initial stages of the work. I am

also indebted to Professor C. H. Bornman and the staff of the Botany Department,

University of Natal, Pietermaritzburg, especially to Drs. K. D. Gordon-Gray and

O. M. Hilliard for assistance and advice; to the Secretary, Department of Agricultural

Technical Services for seconding me to the University of Natal to complete my study;

to the Chief and staff of the Botanical Research Institute, Pretoria, in particular, Drs.

D. Edwards and J. H. Ross, Messrs E. J. Moll and R. G. Strey, for encouragement

and help; to Dr. J. O. Grunow and Messrs. C. J. Ward, P. D. Tyson, R. Tumner,

C. J. Vernon and G. L. Webb for fruitful discussion and advice, I wish to thank Mrs.

G. J. Coetzee and Miss H. L. Tomlinson for typing the manuscript and Messrs B. B.

Mkhize and S. S. Nxumalo for assistance in the field.

REFERENCES

Beats, E., 1960. Forest bird communities in the Apostle Islands of Wisconsin. Wilson Bull. 72,

156-181.

Bray, J. R. & Curtis, J. T., 1957. An ordination of the upland forest communities of southern

Wisconsin. Ecol. Monogr. 27: 325-349. ;

Curtis, J. T., 1959. The vegetation of Wisconsin: an ordination of plant communities. Madison:

University of Wisconsin Press.

120

Curtis, J. T. & McIntosu, R. P., 1951. An upland forest continuum in the prairie-forest border

region of Wisconsin. Ecology 32: 476-496.

Dopson Re G., 1951. Geology of the Shongweni District, Natal. M.Sc. thesis, University of Natal,

Durban.

GLEASON, H. A., 1926. The individualistic concept of the plant association. Bull. Torrey Bot.

Club 53: 7-26.

GreEIG-SmiTH, P., 1964. Quantitative plant ecology. 2nd Ed. London: Butterworths.

Jackson, S. P., 1952. Atmospheric circulation over South Africa. S. Afr. Geogr. J. 34: 48-60.

Morris, J. W., 1967. Descriptive and quantitative plant ecology of Ntshongweni, Natal. M.Sc. thesis,

University of Natal, Pietermaritzburg.

Ruopes, R. C. & Leitu, M. J., 1966. Lithostratigraphic zones in the Table Mountain Series of Natal.

Trans. Geol. Soc. S. Afr. 71 (in press).

SCHULZE, B. R., 1965. Climate of South Africa. Part 8. General Survey. Pretoria: Government

Printer.

Tyson, P. D., 1964a. Berg Winds of South Africa. Weather, Lond. 19: 7-11.

Tyson, P. D., 1964b. A summer storm over Pietermaritzburg. J. Geog., Stellenbosch 2: 23-29.

Tyson, P. D., 1965. Berg Wind over Durban. Weather, Lond. 20; 115-116.

WEATHER BurREAU, 1954a. Climate of South Africa. Part 1. Climate Statistics. Pretoria: Government

Printer.

WEATHER BurREAU, 1954b. Climate of South Africa. Part 2. Rainfall Statistics. Pretoria: Government

Printer.

WHITTAKER, R. H., 1956. Vegetation of the Great Smoky Mountains. Ecol. Monogr. 26: 1-80.

WHITTAKER, R. H., 1960. Vegetation of the Siskiyou Mountains, Oregon and California. Ecol.

Monogr. 30: 279-338.

WHITTAKER, R. H., 1962. Classification of natural communities. Bot. Rey. 28: 1-239.

Bothalia 10, 1: 121-128

An Investigation of the Plant Ecology of the

Hawaan Forest, Natal, using an

Ordination Technique

E. J. Moll

ABSTRACT

An account of the forest ecology using a slightly modified Wisconsin ordination technique (Bray

& Curtis, 1957) is given. The river-facing slope is considered preclimax, and the sea-facing slope

subclimax, to the climax forest on the flat land. In addition, a secondary element, resulting either

from recent disturbance or, possibly, from recently drier environmental conditions, is shown to be

present in the climax forest.

INTRODUCTION

In recent ecological literature there has been a swing away from traditional descrip-

tive ecological accounts towards quantitative methods of analysis. One of these methods,

the modified Wisconsin ordination technique developed by Bray & Curtis (1957), was

investigated with the aim of testing its usefulness when applied to South African

vegetation, particularly forest. This technique can be used to compare quantitatively

floristic data for different stands or examples of vegetation, and assumes that the

floristic composition of a site is an expression of the environment of that site.

Vegetation samples are compared one with another, to obtain an Index of Similarity

2w ; ;

TG me 5* 100, where w is the sum of the lesser scores for each species common

to both stands, and a and b are the sum of the scores for each sample.

The sample plots are then arranged along axes, in this study two axes X and Y,

so that the linear distance between plots is an indication of relative similarity or

dissimilarity between plots. Because the method of endplot selection used by Bray &

Curtis (1957) did not make efficient use of factor space, endplots with high mean

dissimilarities and high standard deviations of the mean were used (Morris, 1967;

1969).

METHODS

Data, suitable for ordination, were collected during an investigation of the Hawaan

Forest (Moll, 1968a). This Forest is situated close to the sea on the North Coast

highway, approximately 10 miles (16-1 km) north of Durban at the intersection of

co-ordinates 29° 42’ south and 31° 06’ east. The Forest occurs on a flat area which

slopes eastwards down to the sea in one direction, and steeply northwards to the Mhlanga

River in the other direction (Moll, 1968b). The area analysed consists of some 100

acres (40-5 ha) of Coast Forest on sandy dune soils. Density data for trees and shrubs,

i.e. the number of stems per plot, were collected from fifty 33 ft (10-1 m) square plots

constituting a total sample of approximately 1-3 per cent of the Hawaan Forest.

Woody plants with a diameter at breast height (d.b.h.) of at least three inches (7:6 cm)

or at least 15 ft (4-6 m) high, were considered trees. All smaller woody plants were

122

considered shrubs. The fifty plots were located by restricted randomization, a grid

being superimposed on a large scale map of the Forest, and three plots were located

within each grid square by pacing out two random co-ordinates.

RESULTS

Two axes, X and Y, were sufficient to account for most of the floristic variation

between samples of the trees and shrubs. The two-dimensional scatter diagrams,

where points represent sample plots, are shown in Figs. 1-3. Isolines have been

employed to draw attention to high and low values. Correlation coefficients between

200 randomly selected interpoint distances and their corresponding calculated dissimi-

larity values were found to be significant, with a value of 0-61 on the tree ordinations

(Figs. 1 and 2), and a value of 0-50 on the shrub ordination (Fig. 3). This significant

correlation indicates that most variation within the data is accounted for by the X

and Y axes used.

Fic. 1.—Tree ordination showing the distribution of: a, plot aspect (F = flat, S = sea, R = river);

b, angle of slope (F = flat, G = gentle, S = steep); c, canopy tree density (1 = 2-4, 2 = 5-6,

3 = 7-10, 4 = 11-18); d, estimated mean tree height in feet (1 = 20-23, 2 = 24-26, 3 = 27-29,

4 = 30 and more); e, estimated d.b.h. in inches (1 = less than 4, 2 = 4-5-9, 3 = 6-8, 4= 8:1

and more); f, Cola natalensis (. = absent, 1 1,2 = 2,3 = 3,4 = 5-6); g, Xylotheca kraussiana

(. = absent, 1 1,2 = 2,3 =3, 4 =4-5); h, Cavacoa aurea (. = absent, 1 = 1, 2 = 2, 3 =3,

4 — 5); i, Celtis africana (. = absent, 1 = 1, 2 = 2).

123

Figs. la and 1b show the distribution of two environmental factors, aspect and

angle of slope, which were estimated for each sample plot while in the field. The Y

axis of the ordination is mainly responsible for separating the flat plots, the gentle

sea-facing plots and the steep river-facing plots. There are, apparent however, certain

anomalies which should be explained. The sea-facing plot and the two river-facing

plots at the upper extreme of the Y axis are undisturbed forest. The remaining river-

facing plot and three sea-facing plots, which are apparently similar to the flat plots,

do in fact have high coefficients of dissimilarity, but a third axis would be necessary

to expose these differences. Such discrepancies are understandable when one considers

that a multidimensional system has been expressed in two-dimensions.

Fig. 1c shows that the tree densities are related to both the X and Y axes. The

exact relationship between aspect and density is not clear, but it is apparent that the

majority of the most dense plots are river-facing and sea-facing. Some flat plots do,

however, have a high density.

Fic. 2.—Tree ordination showing the distribution of: a, Drypetes natalensis (. = absent, | = 1,

2 = 2); b, Teclea gerrardii (. = absent, 1 = 1, 2 = 2-3, 3 = 5); c, Dovyalis sp. nov. (. = absent,

1=1, 2=2):; d, Deinbollia oblongifolia (. = absent, 1 = 1, 2 = 3); e, Mimusops obovata

(. = absent, 1 = 1,2 = 2); f, Strychnos decussata(. = absent, 1 = 1); g, shrub density (1 = 6-18,

2 = 24-37, 3 = 41-69, 4 = 70-88); h, Uvaria caffra (. = absent, 1 =1, 2= 24, 3 = 6-9,

4 = 10-30); i, Peddiea africana (. = absent, 1 1,2=2,3=3,4=412).

33261—9

124

Estimated canopy height has been plotted in Fig. Id, the tallest trees occurring

in the flat plots. Fig. le shows the mean estimated d.b.h., the largest trees being in

the flat plots.

Figs. 1f and li, and Figs. 2a—2f, show the distribution of the ten tree species with the

highest densities on the ordination. One species, Strychnos innocua, is not illustrated

here as it shows no relationship to the major environmental situations considered here

and apparently has a random distribution. Cola natalensis (Fig. 1f) occurs mainly

in those flat and sea-facing plots which have the highest tree density. Xylotheca

kraussiana (Fig. 1g) occurs almost exclusively in the steep river-facing plots where the

trees are generally small and with a high density, although two sea-facing plots and

one flat plot do each contain only a single specimen. Cavacoa aurea (Fig. 1h) occurs

in flat, sea- and river-facing plots which are gently sloping, though two steep-river-

facing plots do each contain a single tree. The plots containing C. aurea vary in density,

but contain the largest forest trees. Celtis africana (Fig. li) has a similar distribution

on the ordination to Cola natalensis, but is confined mainly to the most dense plots.

Drypetes natalensis (Fig. 2a) is generally limited to the sea- and river-facing plots,

where the trees are predominantly small and fairly dense. Teclea gerrardii (Fig. 2b)

is found mainly in the sea-facing plots with moderate density and medium-sized trees.

Dovyalis sp. nov. (Fig. 2c) has a fairly random distribution on the ordination and is

similar to Cola natalensis. Deinbollia oblongifolia (Fig. 2d) is limited entirely to river-

facing plots and to two sea-facing plots where the trees are small and dense. Mimusops

obovata (Fig. 2e) occurs mainly in the sea-facing plots, but also occurs in two of the

river-facing plots. Strychnos decussata (Fig. 2f), with the exception of one river-facing

plot, occurs exclusively in flat plots of low density and big trees.

Fic. 3.—Shrub ordination showing the distribution of: a, plot aspect (F = flat, S = sea, R = river);

b, angle of slope (F = flat, G = gentle, S =steep); c, shrub density (1 = 6-18, 2 = 24-37

3 = 41-69, 4 = 70-88; d, Notobuxus natalensis (. = absent, 1 = 1-5, 2 = 12-26, 3 = 34-52);

e, Baphia racemosa (. = absent, 1 = 1-4, 2 = 5-10, 3 = 11-30, 4 = 61); f, Carissa bispinosa

(. = absent, 1 = 1-2, 2 = 3-7).

125

Fig. 2g shows the shrub density data plotted on to the canopy tree ordination.

No clear trends are discernible, but this is understandable because the shrub species

are not as dependent on the external environment as are the tree species. It is apparent,

however, that most river plots have a comparatively low shrub density.

Figs. 2h and 21 show the distribution of two selected shrub species, Uvaria caffra

and Peddiea africana, on the tree ordination. Neither shows any marked relationship

to the tree species performance. These two shrub species are given as examples of the

unclear relationship between shrub and tree layers.

FLAT

LOW TO MEDIUM TREE DENSITY

TALLEST TREES OF MAX. D.B.H.

MEDIUM SHRUB DENSITY

Cavacoa aurea

Strychnos decussata

RECENT

DISTURBANCE

FLAT

MEDIUM TREE DENSITY

TALL TREES, HIGH D.B.H.

MAX. SHRUB DENSITY

Cola natalensis

Celtis africana

Dovyalis sp.

SEA

GENTLE SLOPE

aah HIGH TREE DENSITY

MEDIUM SIZED TREES

LOW SHRUB DENSITY

Teclea gerrardii

Mimusops obovata

Drypetes natalensis

RIVER

STEEP SLOPE

HIGH TREE DENSITY

SMALLEST TREES

MIN. SHRUB DENSITY

Xylotheca kraussiana aN

Deinbollia oblongifolia PR E S U B

Drypetes notolensis LI CLIM f X

Fic. 4.—Diagrammatic summary, from the tree ordination, of suggested ecological gradients.

126

Figs. 3a and 3b show the distribution of aspect and angle of slope on the shrub

ordination. The relationships between these two environmental factors are not as clear

on the shrub ordination as they were on the tree ordination. A diagonal relationship

to the two axes separates most of the gently-sloping to steeply-sloping, sea and river

plots from the flat plots.

Fig. 3c shows the shrub density on the ordination. The relationships are not

clear, but the flat plots tend to have the highest densities of shrubs.

Figs. 3d to 3f show the distribution of selected shrub species on the ordination.

Only three species are shown. These are representative of the unclear shrub patterns

on the ordination that appear to be unrelated to any of the major environmental data

considered here. Notobuxus natalensis (Fig. 3d) is the one exception and shows a clear

pattern, occurring in the flat plots where it contributes markedly to the high shrub

density.

The results of the tree ordination are briefly summarized in Fig. 4, and suggested

successional relationships are also given.

DISCUSSION

By considering the limited environmental data that were collected in the field

together with the tree density data and species behaviour, and with some experience

of species performance in other parts of Natal, it is postulated that the forest on the

flat sites represents the highest stage of development and can be considered climax.

It is on the flat land that soils are deepest and where maximum penetration of rain

water occurs. The rather varied tree density in some of the flat plots was at first a

little perplexing but, when canopy tree species were plotted on to the ordination, it

was apparent that either recent disturbance, by way of the occasional removal of a

tree, or by a natural tree-fall, or recently drier environmental conditions, accounted

for the much higher density in these plots. The shrub density data also indicated that

there had been some recent disturbance, as many of the flat plots had a dense shrub layer

indicative of a disturbed canopy. In fact, plots at the upper end of the Y axis may

be considered as the best climax forest. The forest on the sea-slope may be considered

a subclimax, limited by salt-spray, and the forest on the north-facing river slope, which

is a xerocline, may be considered as preclimax.

Considering the canopy trees, it is postulated that Cavacoa aurea and Strychnos

decussata (Fig. 1h and 2f) are true climax species in Hawaan.

Xylotheca kraussiana and Deinbollia oblongifolia (Fig. 1g and 2d) are heliophytes

which occur almost exclusively in river plots. An occasional tree occurs on the sea-

facing slope or on flat land, probably in an old gap.

Drypetes natalensis (Fig. 2a) occurs in an equal number of sea- and river-facing

plots, and in two flat plots. It appears that this species prefers a fair amount of light,

as the two flat plots where it is found are towards the seral end of the Y axis.

Mimusops obovata (Fig. 2e), a species that favours xeric sites, occurs mainly in the

sea-facing plots, but is also present on the river slope. It occurs at the upper end of

the Y axis and is probably indicative of subclimax forest.

Teclea gerrardii (Fig. 2b) has a similar behaviour to M. obovata, but is almost

completely confined to sea-facing plots.

Cola natalensis has the highest density and is the most frequent species in Hawaan.

Its distribution on the ordination (Fig. 1f) is across the successional trend postulated

(see Fig. 4), being found mainly in flat plots, but it also occurs in sea- and river-facing

127

plots. From field observations it was apparent that most C. natalensis trees were

fairly small and I suggest that the frequency of C. natalensis is due to recent disturbance.

This suggestion is partly supported by the shrub density data where the plots with the

highest densities of shrubs are related to the distribution of C. natalensis.

The last two species, Celtis africana (Fig. li) and Dovyalis sp. (Fig. 2g), have a

similar distribution to Cola natalensis although less well defined. Celtis africana is

a fairly fast-growing tree which can tolerate both open canopy conditions and drier

environmental conditions, as can also Dovyalis sp.

It is obvious that some relationship exists between the river xerocline and the sea-

slope, the former probably bearing a seral relationship to the latter. This relationship

is shown by the distribution of Xylotheca kraussiana (Fig. 1g), Deinbollia oblongifolia

(Fig. 2d) and Mimusops obovata (Fig. 2e) on the ordination.

The distribution of shrub data on the canopy and the shrub ordinations was not

clear and, at the present level of investigation, the shrubs are not as ecologically

meaningful as the trees. A possible reason for this is that the shrubs, being partly

protected by the trees, are not as dependent on the external environment. In addition,

the shrub layer is more sensitive to local disturbance, such as the occasional removal

of trees and natural treefalls. A combination of these factors has masked the shrub

behaviour. Only the distribution of Notobuxus natalensis (Fig. 3d) was clear, this

species being found exclusively in flat plots under the best canopy.

CONCLUSIONS

The climax forest species growing on flat sites are the tallest trees with maximum

d.b.h. Most important climax species are Cavacoa aurea and Strychnos decussata.

The forest of the gently-sloping sea slope is subclimax, and the most important sub-

climax species are Teclea gerrardii, Mimusops obovata and Drypetes natalensis, the

last species occurring on both the river and sea slopes (Fig. 4). The forest of the steep

river slopes is at a preclimax to that on the flat sites, typical species being Xylotheca

kraussiana and Deinbollia oblongifolia.

The modified Wisconsin ordination technique has given a good indication of the

behaviour of the plant species considered. The manner in which this has been achieved

is by simple, pictorial illustrations of the relationships of the plant species to certain

environmental factors. The results of a previous investigation, based on density and

frequency values, of the forest ecology (Moll, 1968a), were similar, but less obvious,

in spite of the presence of the three distinct topographic sites in the Hawaan Forest.

It is possible that if the size of the sample plot was manipulated to get the optimum

size of sample, the results may have been even more informative.

Certain more refined statements concerning the plant ecology have been possible

with the ordination technique than were possible with the previous study by Moll

(1968a). A good example of this is that the ordination showed that Cola natalensis,

Celtis africana and Dovyalis sp., although common in climax forest, are not true climax

species, but can occur where there is comparatively recent disturbance. Such a

conclusion was not easily drawn from a previous study of the Forest.

ACKNOWLEDGEMENTS

In particular I wish to thank Dr. D. F. Woods of the Botany Department, University

of Natal, Pietermaritzburg, who introduced me to ordination, and to Mr. J. W. Morris

for much valuable advice. I would like to acknowledge the use of the facilities of the

Botany Department, University of Natal, Pietermaritzburg. Finally, I would like to

express my very sincere thanks to Dr. D. Edwards for his many valuable criticisms

and comments.

128

REFERENCES

Bray, J. R. & Curtis, J. T., 1957. An ordination of the upland forest communities of southern

Wisconsin. Ecol. Monogr., 27: 325-349.

MoLL, E. J., 1968a. An account of the plant ecology of the Hawaan Forest, Natal. J. S. Afr. Bot.,

34: 61-76.

MOLL, E. J., 1968b. A plant ecological survey of the Three Rivers Region, Natal. Unpublished report

to the Natal Town and Regional Planning Commission.

Morris, J. W., 1967... Descriptive and quantitative plant ecology of Ntshongweni, Natal. Unpublished

M.Sc. thesis, University of Natal, Pietermaritzburg.

Morris, J. W., 1969. An ordination of the vegetation of Ntshongweni, Natal. Bothalia 10: 89-120.

129

Book Review

Fics (Ficus Spr.) of HONG KONG by DENNIs S. HILL. Hong Kong: Hong Kong University Press. 1967.

Pp. viii + 128, 178 figures, 4 tables, 65 black and white plates. Price HK $60.

It has been known for a long time that a unique symbiotic relationship exists between chalcid

wasps belonging to the family Agaonidae and plants of the genus Ficus: the agaonid wasps can only

develop in the gall flowers of the figs and the insects are the sole means of pollination for the fig flowers

and are hence responsible for maturation of the fruit.

Dr. Hill, an entomologist, has made a careful study of the association between the figs and fig-wasps

of Hong Kong. The investigation, which lasted three years, involved making continuous ecological

observations on 200 trees representing 27 species of Ficus, 14 of which produced fertile fruit. In all,

Dr. Hill collected over 25,000 figs from which he obtained 65 species of chalcid wasp plus numerous

other insects. Dr. Hill was fortunate in being able to enlist the aid of Professor E. J. H. Corner of

Cambridge, expert on the Asian and Australasian species of Ficus, in identifying the Hong Kong species.

Following his Introduction and Methods in Chapters 1 and 2 respectively, Dr. Hill discusses in

Chapter 3 the characteristics, fig anatomy and pollination of the genus Ficus, and provides a systematic

list of the Hong Kong figs as well as a key to species.

Chapter 4, which comprises the bulk (90 pages) of the book, contains a description of the mor-

phology and ecology of the Hong Kong Ficus species. In Chapter 5 a provisional synoptic catalogue

of world genera of fig-wasps is provided.

In Chapter 6 Dr. Hill tabulates the different chalcid species that were collected from the 14 infected

species of Ficus and comes to the extremely interesting conclusion that, under natural conditions, the

Agaonidae of Hong Kong are completely host specific. In the Sycophaginae, however, host specificity

does not appear to be the rule.

Chapter 7 is devoted to a discussion of the high degree of adaptive radiation shown by the Hong

Kong Ficus species, their dispersal and the origin, evolution and migration of Ficus.

Chapter 8 comprises a useful bibliography, which is intended to contain all the more important

contributions towards a study of Ficus species (except F. carica) and fig-wasps.

In Chapter 9, which is an appendix, the previous names for the Hong Kong species of Ficus, and

the Ficus species recorded from China, Taiwan, Ryuku Islands and Japan are tabulated. Chapter 10

comprises an index to the species of Ficus and genera of fig-wasps.

There are 178 excellent figures and 65 black and white photographs of varying quality.

The thought comes to mind that if host specificity in the Agaonidae is a general rule, then a study

of the South African agaonids occurring in our fig species might assist in deciding whether some of the

alleged differences between closely related species such as Ficus burkei (Miq.) Miq., F. natalensis (Miq.)

Hochst., F. petersii Warb. and F. craterostoma Warb. ex Mildbr. & Burret are valid. Although the

revision of the South African figs is in the capable hands of Dr. De Wolf of Georgia, U.S.A., such

a study would serve as an invaluable check. It is to be hoped that some entomologist in this country

will undertake an investigation of the figs and fig-wasps of South Africa. Dr. Hill’s study would serve

as an admirable model.

D. J. B. KILvick

}

"99

7 iia -

‘iy

ki 7 a "

Prare 1 — Illtyd Buller Pole Evans.

Bothalia 10, 2: 131 —135

Illtyd Buller Pole Evans (1879 - 1968)

M. D. Gunn

Illtyd Buller Pole Evans C.M.G., M.A. (Cantab.), D.Sc. (Wales), LL.D.

(Rand), F.L.S., pioneer in botanical research in South Africa, died in his nine-

tieth year on 16th October, 1968, at Umtali, Rhodesia. Born at Llanmaes near

Cardiff, Wales, on 3rd September, 1879, he was the son of an Anglican clergy-

man, Rev. Daniel Evans, M.A., whose wife Caroline Jane Pole came from

a very old West Country family.

After attending Cowbridge Grammar School he entered the University College

of South Wales and Monmouthshire, graduating with the degree of B.Sc. in 1903.

He proceeded to Cambridge (Selwyn College) as a research student in botany,

specializing in mycology and plant pathology under Prof. H. Marshall Ward F.R.S.,

gaining his research degree in 1905. In July of the same year he was appointed

to the post of Mycologist and Plant Pathologist in the newly constituted Transvaal

Department of Agriculture in Pretoria, where he joined J. Burtt Davy who, two

years earlier, had been appointed as Chief of the Division of Botany.

It was a case of starting from zero and, in spite of severe difficulties imposed

by lack of office accommodation and laboratory facilities, a programme of research

was soon under way in a make-shift greenhouse near his cffice, and at Skinners

Court Experiment Station near Pretoria, on rusts in cereals and other plant

diseases. The results were published in a steady flow of scientific papers and

popular articles dealing with mycology and plant disease problems in the Transvaal,

and he also found time to give advice to neighbouring territories.

In 1912 his services received recognition from the Union Government with

the creation of the Division of Mycology and Plant Pathology under his charge.

The following year improved accommodation for the Division became available

at Vrede Huis, situated in eight acres of ground at the foot of Meintjies Kop, on

which the Union Buildings were in process of being constructed.

The year 1913 saw further changes when, with the retirement of Burrt Davy,

the Division of Botany was amalgamated with the Division of Mycology and Plant

Pathology, and the Transvaal Colonial Herbarium was transferred to Vrede Huis.

In the same year, under an agreement with Medley Wood, the Division took over

control of the Natal Herbarium at Durban, with sufficient ground for carrying

out experiments on tropical and subtropical crops. At this time his staff consisted

of three mycologists: Dr. Ethel M. Doidge, appointed in 1908; Dr. P. A. van der

Bijl, appointed in 1911, who became Mycologist in Charge of the Natal Herbarium

and later Professor of Plant Pathology at Stellenbosch University; and Miss A. M.

Bottomley, appointed in 1913. In addition there was Miss S. M. Stent, appointed

in 1904 in Burtt Davy’s Division, who was in charge of the phanerorgamic her-

barium and economic section of the Division.

132

In the early days of his career in South Africa, Pole Evans became keenly

interested in the rich flora of the country. At first he paid special attention to

Aloes, which he brought together in the grounds of the Division, resulting in

probably the largest collection of South African species then in existence. In

1915 and 1917 he published descriptions of many new species of Aloe in the

Transactions of the Royal Society of South Africa. Cycads (Encephalartos) also

claimed his attention and specimens which he collected still stand in the grounds

of the Division and at the Union Buildings.

An example of the strong and decisive action which he was always prepared

to take occurred in 1916 when Citrus Canker was reported in the Transvaal.

The disease spread rapidly and drastic action, involving the destruction of infected

nurseries and commercial orchards, was called for. After a lengthy campaign,

involving large sums of money spent in compensation, the Division succeeded in

completely eradicating the disease which otherwise would have had a crippling

effect on the citrus industry.

Another problem referred to Pole Evans concerned the serious wastage

experienced in shipments of citrus fruits during 1919. In 1920 he published

his findings in his ““Report on Cold Storage Conditions for Export Fruit at Cape

Town,” while a second report was issued in 1921 in conjunction with three

members of his staff entitled ‘“‘Further Investigation into the Cause of Wastage

of Citrus Fruits from South Africa.” In 1925 a Low Temperature Laboratory

at Cape Town was created as a result of his persistently stressing the need for

research staff and equipment to investigate the engineering and biological problems

involved in the precooling, transport and storage of fruit. From its inception to

the time of his retirement, the Laboratory enjoyed his personal interest and

support.

For some years he had advocated a comprehensive botanical survey of

South Africa. The objectives and advantages of such a survey were eventually

brought directly to the notice of the then Prime Minister, General Botha, with the

result that the Minister for Agriculture, in July 1918, approved of an Advisory

Committee for Botanical Survey with the Chief of the Division of Botany and

Plant Pathology as Director of the Survey. The Advisory Committee included

the following prominent botanists, in an honorary capacity, tc co-ordinate botanical

research in the various parts of the country: Mrs. L. Bolus and Dr. R. Marloth

of Cape Town, Prof. S. Schonland of Grahamstown, Prof. J. W. Bews of Pieter-

maritzburg, and Prof. G. Potts of Bloemfontein. Also nominated to serve on

the Committee were Sir Arnold Theiler, Director of Veterinary Research, and Mr.

C. E. Legat, Conservator of Forests. One of the most important aims of the

Survey was to publish local floras, memoirs and handbooks dealing with the

vegetation. The first memoir was published in 1919 and the series still continues

to this day.

Pole Evans, who travelled widely throughout South Africa recording and

photographing the major types of vegetation, published « preliminary account

cf his observations in 1917 in the official Yearbook of the Union of South Africa,

in an article entitled “The Plant Geography of South Africa,” with an accompany-

ing map in colour. Later, as President of the South African Association for the

Advancement of Science in 1920, he enlarged on the subject in his presidential]

address: ‘““The Veld, its Resources and Dangers,” which was published in the

South African Journal of Science 17 : 1 — 34 (1920). In this paper he classified

the country into 19 botanical regions and gave a brief ecological characterisation

of each region.

133

In 1920 the first part of ‘““The Flowering Plants of South Africa,” a serial

magazine based on hand-coloured illustrations, was issued under his editorship.

The Government sanctioned the publication on condition that the necessary

funds for the cost of publication were met by private subscription. With confidence

in the succees of the venture, Pole Evans raised enough money through personal

contact with liberal donors in South Africa and overseas to continue the work for

many years. In 1946, some years after his retirement, the title was changed to

“The Flowering Plants of Africa” and, from 1948, the Gcvernment has assumed

full responsibility for its publication both in English and Afrikaans.

Another periodical, intended primarily as a medium for the publication of

botanical papers and monographs emanating from the National Herbarium, was

initiated by Pole Evans in 1921. He named the official organ of the Division

Bothalia to commemorate General Louis Botha, first Prime Minister of the Union

of South Africa and Minister of Agriculture until 1913, to whose policy and

influence much of the rapid agricultural development in South Africa after Union

was due.

The field of research under the direction of Pole Evans was enlarged in 1927

to include the Divisions of Entomology and Horticulture and, with the inclusion of

a Field Husbandry section in 1929, the title of the organisation was changed to the

Division of Plant Industry. In addition to its extensive research programme,

this Division was responsible for regulations dealing with plant imports and

quarantine, locust control, veld conservation, nursery inspection and the transport

of perishable products.

His interest in the natural vegetation led to the establishment of the Dongola

Botanical Reserve in the dry bushveld of the northern Transvaal. Here some

of the finest examples of baobabs (Adansonia digitata) were to be seen. Unfor-

tunately the Reserve was abandoned after his retirement. Ir 1926 the Veld Reserve

at Fauresmith in the Orange Free State was initiated for the scientific study of

indigenous grasses and bushes of the Karoo region. With the establishment of

an up to date laboratory at this centre and the appointment of Dr. Marguerite

Henrici, a series of valuable publications on the physiology and nutritional value

of these plants was made possible. Further research centres were developed for

the introduction and study of indigenous grasses at Prinshof and Rietondale in

Pretoria, and for subtropical horticultural crops at Nelspruit in the eastern Transvaal.

He had for many years supported and taken a prominent part in associations

devoted to science. In 1905 he joined the S.A. Ornithologist’s Union, later to be

amalgamated with the Transvaal Biological Society, of which he was a foundation

member and President in 1911. The title of the Society was later changed to the

S.A. Biological Society and he continued as a council member, receiving the Scott

Memorial Medal in 1919, awarded by the Society for his research contributions.

In 1907 he was elected a Fellow of the Linnaean Society (London) and became

a member of the S.A. Philosophical Society, later to become the Royal Society

of South Africa, and of this Society also he was a Fellow. He was a strong

supporter of the South African Association for the Advancement of Science,

becoming President of Section C in 1916 and, as mentioned previously, of the

Association as a whole in 1920. In 1922 the Association awarded him the South

African Medal and Grant for outstanding scientific achievements. The C.M.G.

was conferred upon him in 1921 and, in 1933, the University of the Witwatersrand

awarded him an honorary LL.D. degree.

On the occasion of the visit of the British Association for the Advancement

of Science to South Africa in 1929, a handbook entitled ‘‘Science in South Africa”

134

was published in which the chapter on “Vegetation of South Africa’ with an

accompanying map was contributed by Pole Evans. This was the fore-runner of

the well-known vegetation map on the scale of 1 : 3,000,000. published as Botanical

Survey Memoir No. 15 (1936).

He was a member of the Editorial Board of the Empire Journal of Experi-

mental Agriculture. In Vol. 1 (1933) of the Journal he wrote on “Agricultural

Possibilities of Some of the African Grasses” and later, in Vol. 18 (1950), on ‘‘The

Possibilities of Beef Production in Southern Africa.” In 1935 he attended the

Imperial Botanical Conference held in London, where he read a paper entitled

“Pasture Research in the Union of South Africa,’ which was published in 1936.

In 1934 the House of Assembly expressed concern at the serious deterioration

of the natural vegetation cover and the threat to the country’s water resources

caused by indiscriminate veld burning on the mountains. A programme submitted

by Pole Evans was adopted as a basis for immediate action and this included the

formation of a Pasture Research and Veld Management Section within the Division

of Plant Industry. A series of Pasture Research Stations was established in

representative vegetation regions in the Transvaal, Natal and the Cape Province

in order to formulate sound principles of veld management.

This was the final official project he was to launch before his retirement in

September 1939. In July 1939, in the preface to the bulletin entitled ‘Pasture

Research in South Africa, Progress Report No. 2,” he wrote that: “Grass is the

foundation of man’s existence in our land as in all others. It is surprising therefore

that there should be any who are slow to recognize this and some even loth to

admit it.” “It is my obvious duty again to draw your attention to the fact that large

areas of the country which formerly were rich and flourishing pastoral grounds

are now wholly depleted of their grazing and are rapidly becoming desert wastes.

Nothing but the establishment of well-equipped pasture research stations in these

areas can bring any permanent relief and restore health tc the land and wealth

to the people.” ‘To a small body of men and women under Dr. J. W. Rowland’s

able direction, credit must be given for lifting the veil that has so long obscured

the dangerous trend of South African agriculture. Few thought that pasture

research could be of much benefit to the Country as a whole and many begrudged

the little that was spent on it in comparison with other State services. Yet. it

has remained for pasture research to point out our folly in the past and to indicate

safer and sounder methods of approach for the future.”

South Africa owes a debt of gratitude to Pole Evans for his farsightedness

and drive in building up a large and active body of research workers who were

inspired by his own dedication and energy. One of his most important contributions

was in stressing the value of the indigenous plant cover, and of grasses in particular,

in conserving soil. His search for grasses suitable for grazing and conservation

purposes took him beyond the borders of South Africa. Several visits were paid

to Botswana and his travels in this territory are recounted in Botanical Survey

Memoir No. 21 (1948). In 1938, at the request of the Kenya Government for nis

advice with regard to soil erosion and pasture problems, he undertook his most

extensive expedition, travelling a distance of 13,000 miles through tropical Africa,

and collecting over 700 living grasses, a large number of seed samples and nearly

2,000 botanical specimens. This expedition is described with numerous illustrations

in Botanical Survey Memoir No. 22 (1948).

In 1922 he married Miss Mary R. H. Thomson B.A. (Cape) M.Sc. (Lond.),

who had joined his staff as Mycologist in 1919, and who fully shared his wide

botanical interests. Mainly due to her untiring devotion, he was able to accomplish

135

his many undertakings. After his marriage he made his home at Irene, some miles

south of Pretoria and near Doornkloof, the home farm of his friend General

Smuts. A keen horticulturalist, he introduced many indigenous plants into his

garden. In the early 1950’s he settled near Umtali in Rhodesia and continued

to collect interesting indigenous plants in the surrounding area for several years

until confined to his home through indifferent health.

Though sadly hampered by infirmity during his last few years and no longer

able to walk, he retained his interest in natural history and enjoyed particularly the

bird life in the beautiful garden he had created. He is survived by his widow

and two children: Dr. Jean Pole Evans and Mr. Reginald J. Pole Evans M.B.E.

Many plants which he collected proved to be new tc science and several

commemorate his name, for example, Aloe pole-evansii Christian, Gladiolus pole-

evansii Verdoorn and a fungus, Puccinia pole-evansii Doidge. In view of his inter-

est in grasses, it is fitting that a grass he discovered in the mountains of Lesotho

should be given the generic name Polevansia De Winter in his honour.

EG ty

» & al >

> S18

Bothalia 10, 2: 137 — 328

Cultural Characters and Carpophore Construction

of Some Poroid Hymenomycetes*

G. C. A. van der Westhuizent

ABSTRACT

The cultural characters and construction of the carpophores of 24 species of poroid

Hymenomycetes were studied. The microstructures formed in culture and oxidase reactions

of the cultures were compared with the microstructures present, the construction and type

of decay of the carpophores from which they were made. The type of interfertility of seven

species was determined.

Intercollection pairings of haploid mycelia derived from single basidiospores and the

technique of dikaryotizing a large haploid mycelium growing in culture by pairing it with

a small dikaryotic mycelium, were used to confirm the identity of different collections of

eight different species.

The literature on the classification, structure and anatomy of the carpophores and pure

culture studies of Hymenomycetes, was reviewed.

It was found that the 24 species were distributed among nine of the groups proposed

by Nobles (1958) on the basis of their cultural characters. The structures formed in culture

were also found to be present in the carpophores so that the carpophores could also be

assigned to the some groups as their cultures. The carpophores did not indicate the same

relationships as the cultures however. Differences in the micromorphological characters

of hyphae and in the types of hyphae present in carpophores of species in the same group

were found. Differences in construction of the carpophores were noticed in species with

similar types of hyphae. Micromorphological characters of hyphae and the microstructures

as well as the construction of the carpophores are constant for each species. Differences

and similarity of micromorphological characters and construction of carpophores of different

species are not adequately conveyed by the concept of hyphal systems.

All seven species tested displayed the tetrapolar type of interfertility. Six of these

are associated with white rots. In the intercollection pairings, dikaryotization and clamp-

formation of the haploid test mycelium could not be achieved with Polyporus dichrous and

Polyporus pubescens.

* Thesis submitted to the University of Pretoria in partial fulfilment of the requirements

for the degree of Doctor of Science. ;

+ Plant Protection Research Institute, Department of Agricultural Technical Services, Pretoria.

» KY 1© GROUES OF GoseES GUUCHE! ge com om com amp cm om om om ame

. Descriptions of species

eon D

138

CONTENTS

Acknowledgements

AMtTOUUCHON: aus aac ee ug eT eae ee Me

Review of Literature .... .

Classification of the poriod Hymenomycetes - Dope eee eran Senn MIB i Kae ean tn |

Structure and anatomy of the carpophore in taxonomy

Pure culture studies

liMfaterral’s andiormethod Sit ccx, cvs ences cers cece eters Oe as ere

15) TLIC) G Fal) keane ee Metre rear 2.) |e mE Sr RTA eae eater Genin dubia. cai

TW (Jeo Vo y0 Kite nets CRIES Werder ee eg eeeons Riva ae NTS tory, peasy ean A oN le Stee lakers eon Gas

5.1 Group 7 .... pe eae ct ee ern CTA Nebite eetd VEST coke: cite) .caeh vane. “Woe

Poly porus GUUSTUS. <2 act ees teeta) a ane ee

5-2) Grouplo ao eke

Polyporus dichrous

Dos}, GOW) WB sates como core: spas bjeeduey saci abbey iccrceeh rece ch acral acme ere rn a ce ee

NEC ZILESIES CDLQTIG ie. cs youet) og Coa Re Ee

TGC ZILCS RAD EG Verac ls MEN seenel wecssyarcat ts urea reco Erna) CeRte (lero ce aa ager Re oe Me

SrA GROW P Sy leer anne

OMNES PUA COVA cre a Nee wcrerre cet te ettoy GN Bete rege Ree en ere Eee eee

SSiniGroupy25 ee) week Beth suzy pace eates Use Sree ae eee ee

Daedaleaquercina® sc.. scicodema: cia ie Rc Pee

TVAMELCS SINOESEA™ fees cee, ects, sate ey NO, Re ee

Prametes *FOSCOlAs eu) vee Al We ke ae eee Pee ane, mn ent ae

FIOMES COGN AOKI ca. Rea, Gee tee ee een ee ee

5.6 Group 32,. PM ere eee cea

Polystictus subiculoides-

Se EGrOuUpi tds! 2 (So) eee SR SR ee id Re SEL Sey been ee eae nc nee

Polyporus versicOlor Wetecn acs eee ee eee ene en ee

Trametes suaveolens

enzitesobetulina: = 0 secrire, fend ene, teen Pann) Oi meena ne ames Eee ene ae

POLY Pops PUBESCONS. rin cs, ask tach a Ps! eat, count eer crea) Ce ere

Trametes, méyenii 0...

PCN ites DAHSOUl> cain. hess: shel pee Gee Weer te oa ged ig eee

Polyporus occidentalis 0.0. coe

Trametes cingulata :

OLY /POLUSAVINOSUSI tC mnt ee Re a A I Ae, tes Re done Tera

L538) (Group sla hae ee eR A eT ee See, ete eee

as Daedalea confragosa

TV AMMVELES \COFTURGLA! es (een sek tees). Gass ie ee

WL ONGQONG, CEHUIS ack ce eek Bees, cots) tee ee ee

HinQMetes: GCWDUNGCIGIG. fern in ae ee EO ee

SIO Groups (53) «acd pa te eae ey SS mee

Polyporus sacer

DISCUSSION Eee

Summary

Tables

139

ACKNOWLEDGEMENTS

This work was made possible by a bursary granted by the South African

Public Service Commission and the Department of Agricultural Technical Services

which enabled me to conduct research at the Mycology Section, Plant Research

Institute, Research Branch, Canada Department of Agriculture, Ottawa, with kind

permission of Dr. R. A. Ludwig, former Director of the Plant Research Institute

and Dr. J. Walton Groves, former Head of the Mycology Section. To Dr. Mildred

K. Nobles, Mycologist to the Plant Research Institute, under whose guidance this

work was done, I am particularly grateful for her interest, helpful discussion and

friendly advice in scientific and personal matters

To the Curators of herbaria mentioned in the texts for the loan of specimens,

the staff of the Mycology Section in Ottawa for their assistance and friendship, to

my promoters, Prof. H. P. van der Schijff and Dr. C. J. Rabie, of the University

of Pretoria, for their guidance and to my colleague, Mr. K. T. van Warmelo, for

assistance with preparation of the manuscript, I extend my sincere thanks.

1. INTRODUCTION

The increasing use of timber and the afforestation of new regions of South

Africa necessitates the detection and control of factors that cause losses of trees

and timber. The recognition and identification of fungi which cause diseases of

living trees and decay of timber, are important in this respect.

Identification of decay fungi, most of which are placed in the Hymenomycetes,

poses many problems however. The fruit-bodies which are required for identifi-

cation of a causal organism may be lacking or they may be abnormal or in a state

in which they can not be identified with certainty. In many cases the wood-

decaying fungus may be isolated in pure culture but this is no assurance that the

fungus will be identified unless it is one of the relatively small number of species

of which the characters in pure culture have been described. The characters of

cultures of many wood-rotting Hymenomycetes can not be related to the existing

descriptions of their carpophores. There is as yet no correlation between generic

characters of fruit-bodies and cultural characters. Identification of these wood-

rotting fungi from culture must go directly to the species. Studies relating cultural

characters to characters cf the carpophores from which cultures were made, thus

need to be undertaken.

Most of the known species of Hymenomycetes in South Africa have been

described by a number of older workers, listed by Doidge (1950), around the turn

of the century. Later, Van der Bijl (1922 a, b, c; 1924; 1925; 1926) described

a number of poroid Hymenomycetes but most of these descriptions are based

on gross morphological characters. Not many of these descriptions include details

of spore characters. Since morphological characters are known to be variable, it is

often difficult to identify specimens with the aid of such descriptions. More

precise descriptions are thus required.

The South African fungal flora includes a wide range of species, some of

which are known from the cool temperate regions of the Northern Hemisphere,

while others are truly subtropical or tropical species. The relationships between

these taxa are, in most cases, unknown. Most of the zeneric type species have

been described from the cool temperate regions where a number of systems of

classification of the poroid Hymenomycetes were also developed. It is thus

necessary to compare the taxa from the warmer regions with these generic types

in order to determine their generic affinities as well as their relationships with

other species in the same genera. Only in this way can a natural system of

140

classification of these fungi be developed. This need was also emphasized by

Lowe (1963 a) recently.

Attempts to recognize relationships between species of poroid Hymenomycetes

are of little value unless anatomical characters are taken into consideration. This

view had been emphasized by the work of Corner (1932 a. b, 1947, 1948, 1950,

1953), Cunningham (1946, 1947, 1948 a-h, 1949 a, b, 1950 a, b, 1954, 1963),

Pinto-Lopes (1952), Nobles (1956, 1958 b, 1965, 1967), Bondartzeva (1961, 1963)

and Teixeira (1958, 1960, 1962 a,b). These workers have shown that the micro-

scopic characters of the hymenial structures and hyphae that make up the carpo-

phore, are more constant and reliable than the largely morphological criteria that

have been used hitherto. Only Cunningham (1946, 1947, 1948 a-h, 1949 a,b,

1950 a, 1963) and Pinto-Lopes (1952) have proposed systems of classification of

the Polyporaceae, based on their anatomical characters. These systems were

however not generally accepted by students of this group. Nobles (1958 b)

grouped cultures of 225 species of Polyporaceae on the basis of their biochemical

activities, hyphal modifications and spore shape into 36 groups which she suggested

to be natural taxa of generic or higher rank. This concept was favourably received

by Bondartzeva (1961) and others as a new approach to the problem of polypore

taxonomy, with great promise for the development of a natural system of classifi-

cation of these fungi. Nobles’ thesis however, is based on cultural characters where

many hyphal modifications are known which have never been described from

carpophores. It is thus necessary to undertake correlative studies of the cultures

and carpophores in order to determine whether the structures found in culture

are also present in the carpophores. Correlative studies which reveal the presence

of identical vegetative structures in both carpophores and cultures should thus

indicate relationships between carpophores similar to those indicated by cultural

characters.

In some of Nobles’ (1958 b) groups, however, species are included which differ

widely in habit and morphology of their fruit-bodies although the cultures showed

them to be similar in regard to hyphal morphology, spore shape and biochemical

activity. Corner (1953) showed that certain species which may show superficial

resemblances, may differ widely in the types of hyphae present as well as the

arrangement of the hyphae or in the construction of the carpophores. It thus

appears that together with hyphal morphology, the construction of the fruit-body

must also be taken into account when considering relationships between different

species of polypores. Although it has been known since the publication of

Corner’s (1932 a, b) classical papers that different types of hyphae are present

in carpophores of poroid Hymenomycetes, relatively few serious attempts have

been made to study these characters and apply these concepts to the solution

of taxonomic problems. Studies in which cultural characters are correlated with

the characters of hyphae and other microstructures and the construction of the

carpophores of as many species of poroid Hymenomycetes as possible, should

therefore provide the information which should make a natural system of classifi-

cation of these fungi possible.

The present study was undertaken as a basis for future taxonomic work on

the South African species of poroid Hymenomycetes. No taxonomic study had

been undertaken on the South African polypores since the time of Van der Bijl

(1922 a, b, c, 1924, 1925, 1926) and it has become necessary to apply modern

techniques and concepts to the study of these fungi.

For this purpose a number of species were collected at random to obtain

species with diverse characters and affinities. The cultural characters and carpo-

phore characters of these species were studied to determine: (i) how Nobles’

141

(1958 b) concepts may be applied to them; (ii) whether the structures formed in

their cultures are also present in their carpophores; (iii) whether the phylogenetic

relationships indicated by their cultures also exist between the carpophores; (iv) to

compare the cultural and carpophore characters of these species with these

characters of other species, especially generic type species, in order to obtain

possible indications of their phylogenetic affinities. In this way it was hoped to

provide accurate descriptions of a number of common species to serve as a sound

basis for future taxonomic studies of these fungi.

It has been shown by Davidson, Campbell & Blaisdell (1938), Overholts

(1953), Nobles (1958 b) and Bondartzeva (1961) that the oxidase reaction of

cultures of polypores, the type of decay caused by these fungi and their host

preferences, are valuable characters for the identification of species and may

also be of considerable taxonomic importance. The oxidase reactions of the fungi

included in this study, were thus determined and their type of decay and host

range recorded for these reasons.

Work by Vandendries (1922, 1923, 1924 and 1933), Mounce & Macrae

(1936, 1937, 1938), Nobles (1943, 1967) and Boidin & des Pomeys (1961), among

others, has shown that pairings between mycelia grown from single basidiospores

may yield valuable information on the identity of morphologically similar fungi

even when collected in different parts of the world. Nobles (1958 b) also advanced

the thesis that species of the Polyporaceae causing brown rots, have the bipolar

type of interfertility whilst species with simple clamp connections which cause

white rots, have the tetrapolar type of interfertility. Where possible, single

basidiospore cultures were thus prepared from the different South African species

in order to determine whether this also applied to the South African species. It

was also attempted to pair single basidiospore mycelia from South African

collections with cultures of Canadian origin in order to confirm the identity of

different collections whenever possible.

As this study was intended as an exploration of the usefulness of the modern

techniques in the taxonomy of South African polypores, no attempt was made to

determine the full synonymy of the different species. Instead, only generic

synonyms are cited and the species are described under their basinyms or their

more generally used binomials. No new combinations or genera were made or

created and possible phylogenetic relationships are merely indicated in the

descriptions.

2. REVIEW OF LITERATURE

CLASSIFICATION OF THE POROID HYMENOMYCETES

In his early classification of the pore fungi, Fries (1821) recognized five genera.

viz. Daedalea, Polyporus, Merulius, Boletus and Fistulina of which the last three

are not considered to be genera of the Polyporaceae by most workers today. The

generic distinctions were based on hymenial configuration but this soon proved

to be inadequate. In later works, Fries (1828, 1838, 1874) added new genera

or accepted genera proposed by other workers until, in 1874, he recognized eight

genera of Polyporaceae and seven others in which the hymenium is borne in

tube-like structures. Hymenial configuration and gross morphological characters

such as the nature of the surface of the carpophore, were the basis of these

generic concepts. Other workers soon added more genera in recognition of the

desirability of splitting up Fries’ unwieldy and heterogeneous groups into smaller

142

more natural ones. Among the first workers in this respect were S. F. Gray

(1821) who added nine genera and Quelét (1886) who listed the polypores of

France under 15 genera, ten of which were new. Karsten (1880, 1881, 1889) split

Fries’ genera Polyporus and PDaedalea, into 26 differeat genera most of which

are considered to be acceptable genera today. Murrill (1907 b, 1908), recognized

74 genera of Polyporaceae, which he divided into 4 sub-families on the basis

of habit and hymenial configuration of the carpophore. The basis for segrega-

tion of the genera, was partly gross morphological characters and to a lesser

extent, anatomical characters, mainly spore characters. Lloyd (1898, 1905, 1909,

1913, 1916, 1920, 1922) in his Mycological Writings on the other hand recognized

only 12 genera in the Polyporaceae. The basis for these was mainly gross

morphological characters. Anatomical characters were considered to be of value

at the species level only.

Van der Bijl (1922 a, b, c, 1924, 1925, 1926) published descriptions of the

South African species of polypores during this period. Morphological characters

were the basis for this work in which anatomical details are often lacking. The

generic concepts in this work were those of Fries (1838, 1874) and Lloyd

(1898-1925). Only 8 genera were recognized and distinguished on the basis of

hymenial configuration and pore shape, method of attachment of the carpophore

and consistency of the pileus. The resupinate forms included in the genus Poria,

were omitted. More recent works on Polyporaceae by Overholts (1953), Lowe

(1946, 1947, 1948, 1957, 1958, 1963 b, 1966), Lowe & Gilbertson (1961 a, b) and

Gilbertson (1961) are still based on the generic concepts of Fries (1821, 1838,

1874) although many micromorphological characters are included in their descrip-

tions of species.

The trend to delimit genera on the basis of micromorphological characters

was initiated by Patouillard (1887) who recognized 39 genera delimited on the basis

of microscopic and morphological characters (Patouillard, 1900). These concepts

were later applied and extended by other workers, namely Carlton Rea (1922),

Bourdot & Galzin (1928), Donk (1933), Pilat (1936), Imazeki (1943), Bondartzev

& Singer (1941), and Bondartzev (1953) to the polypore floras of their respective

countries. Among these workers a tendency towards recognizing an increasing

number of genera in the polypores is clearly evident ranging from 10 genera in

the British Isles (Rea, 1922) to 61 genera in the European part of the U.S.S.R.

(Bondartzev, 1953).

Systems of classification of the Polyporaceae in which the micromorphology

of the hyphae or the concept of hyphal systems as advanced by Corner (1932 a, b)

were used to characterize genera, were proposed by Cunningham (1946, 1947,

1948 a-h, 1949 a, b, 1950 a) and Kotlaba & Pouzar (1957). Cunningham (1946,

1947, 1948 a-h, 1949 a, b, 1950 a) applied these concepts of hyphal systems to his

studies of the Polyporaceae of New Zealand in which he recognized only 12 genera.

Pinto-Lopes (1952) regarded Corner’s (1932 a, b) and Cunningham’s (1947,

1948 a-h, 1949 a, b, 1950 a) concepts and application of hyphal systems to be

of little value. He instead regarded the type of septation and thickening of the

walls of hyphae as the only criteria of taxonomic value. On this basis he proposed

a system of classification of the Polyporaceae in which he recognized 22 genera

divided among seven sub-families.

Kotlaba & Pouzar (1957) proposed a system of classification of polypores

of Czechoslovakia based on hyphal systems in which they recognized 48 genera,

seven of which were new. Most of their concepts were based on the work of Cun-

ningham (1946, 1947, 1948 a-h, 1949 a, b, 1950 a) and Teston (1953 a, b).

143

Bondartzeva (1961) critically reviewed the more recent systems of classification

of the polypores. She rejected the systems of Cunningham (1947, 1948 a-h, 1949

a, b, 1950 a) and Pinto-Lopes (1953) as artificial and instead regarded that of

Kotlaba & Pouzar (1957) as more natural but incomplete. In her opinion the

system of Bondartsev & Singer (1941), which was later adopted with certain

modifications by Bondartzev (1953), is the most natural one since it considers

structure or texture in relation to anatomy and morphology of the fungi.

It is evident that the basis for classification of the poroid Hymenomycetes

have undergone profound changes since Fries (1821) published his system. It

is also evident that there is an almost total lack of agreement on generic concepts

in these fungi. These different systems of classification and generic concepts

had been reviewed by Cooke (1959) who listed about 300 genera which had been

proposed for the Polyporaceae. He considered about 100 of these to be valid

and usable.

Donk (1960, 1962) discussed the origin, usage and status of the generic names

proposed for polypores and agreed with Cooke (1959) in many respects. Later,

in his conspectus of the families of the Aphyllophorales, Donk (1964) recognized

the impossibility of including in the family Polyporaceae, genera of which the

characters had not been clearly defined. Consequently the Polyporaceae were not

discussed in full in that work.

Despite this confusion two main trends are noticeable in a survey of these

works, viz.: (i) a tendency toward the recognition of a larger number of genera

of poroid Hymenomycetes and (ii) a change in the relative importance of the

taxonomic criteria away from the macroscopic morphological characters of the

older taxonomists towards the micromorphological and anatomical characters

considered to be more important by the modern workers. This must mean that

there is a growing conviction among mycologists that the micromorphological

characters are more constant and reliable in taxonomic studies than the macroscopic

characters and therefore more capable of indicating phylogentic relationships. This

in turn would allow the grouping of species and consequently more precise generic

deliminations.

It must be emphasized however that most of the systems of classification

and generic concepts proposed so far, were based on species found in the cool

temperate regions of the Northern Hemisphere. Very few tropical or sub-tropical

species, of which many occur in South Africa, have been described or included

in these classifications. This shortcoming was recognized by Lowe (1963 a) when

he stated: “I have been and will for some time be preoccupied with species

concepts. Until these are cleared up for a large proportion of the polypores

particularly the tropical species, I cannot consider, without crippling misgivings,

the larger aspects of generic separations.” The accurate description of micromor-

phological characters of the hyphae and other microstructures in fruit-bodies is

thus of the greatest importance in the taxonomy of these fungi.

STRUCTURE AND ANATOMY OF THE CARPOPHORE IN TAXONOMY

The first important work on the taxonomic value of micro-structures and

anatomy was published by Patouillard (1887) who included details of hyphal

morphology, structure of the surface of the pileus and characters of basidia,

spores and cystidia, in his generic descriptions. Later he used these characters

to delimit genera of the Hymenomycetes (Patouillard, 1900)

144

Ames (1913) studied structure of the fruit-bodies in relation to generic concepts

of 130 North American species. She considered anatomical features such as

consistency and hyphal arrangement of the trama, modifications of the surface

of the pileus, the relation of the hymenophore to the pileus and spore characters,

to be of great taxonomic value. Characters of the cystidia were regarded as too

variable to be of any value above the species level while spore colour was

regarded as the most important spore character and preferable to context colour

as a criterion for generic delimitation. Ames concluded that the character of the

flesh or the consistency of the fruit-body indicate the broader relationships within

the Polyporaceae most clearly but that in the recognition of genera a complex

of characters rather than separate characters must be considered. On the basis

of her studies, which did not include the resupinate species, she recognized 16

genera among the temperate North American species.

The great advance in the study of micromorphology of basidiomycete carpo-

phores was made possible by the work of Corner (1932 a, b) who introduced

the concept of hyphal systems. He demonstrated that the fruit-body of Polyporus

xanthopus (Corner, 1932 a) is constructed of three types of hyphae which differ

morphologically in respect of type of septation, wall-thickness, morphology,

ontogeny and function. The generative hyphae are hvaline, thin-walled and

nodose-septate and are the basic hyphae from which ell other hyphae as well

as the basidia are produced. ‘The skeletal hyphae are thick-walled, unbranched

and aseptate and form the main structural elements of the pileus. They arise from

lateral branches of the generative hyphae. The binding hyphae are thick-walled

and aseptate and also arise from generative hyphae but are of limited growth and

have many short, tortuous branches which bind the other hyphae into the tough,

leathery tissues of the pileus. These three types of hyphae thus constitute a

fruit-body with a trimitic hyphal system. The fruit-body of Fomes laevigatus

(Corner, 1932 b) on the other hand consists only of thin-walled, simple-septate

generative hyphae and thick-walled, aseptate, unbranched, skeletal hyphae. Because

the binding system is lacking, the fruit-body of Fomes laevigatus has a dimitic

hyphal system. The term ‘“‘monomitic hyphal system’ was proposed to describe

the construction of fruit-bodies in which only generative hyphae are present. Later,

Corner (1953) showed that certain species have fruit-bodies with dimitic hyphal

systems which consist of generative and binding hyphae. Asterodon spp. and

Asterostromella spp. (Corner, 1948) were shown to have other specialised structures

in their dimitic fruit-bodies, which are similar to structures found in the fruit-bodies

of Aleurodiscus spp., Hymenochaete spp. and Fomes spp. But Corner regarded

the elaborate fruit-body of Polystictus xanthopus, with trimitic hyphal system, as

more highly evolved than the dimitic or monomitic types (1932 b). He showed

that differences in colour and texture of fruit-bodies are determined by the

characters of the crust and hyphal systems while the microscopic structure of the

upper surface determines whether it will be smooth, mat, velutinate, tomentose,

laccate and so forth (Corner, 1932 a, b; 1953).

Corner (1950) used the concepts of hyphal systems as well as other microscopic

characters to delimit genera in a monographic treatment of the clavarioid fungi

and later of the cantharelloid fungi (Corner, 1966).

Corner thus showed that thin-walled, septate hyphae are present together

with other hyphae, modified in various ways, in the tissue of fruit-bodies of

Hymenomycetes. He further demonstrated the interrelationships of these different

kinds of hyphae and showed that the consistency of the tissues and the nature

of the upper surface depend on the nature of the hyphae present in the fruit-

body. He also indicated the phylogenetic significance of the different kinds of

hyphae and their possible use in the classification of these fungi.

145

Humphrey & Leus (1931) made anatomical studies of the upper surfaces

of the pilei of Ganoderma spp. They found that there were anatomical differences

in the surfaces of these species and that these anatomical characters were constant

for each species.

K. Lohwag (1940) made anatomical and morphological studies of the upper

surfaces of a number of European fungi. Many of these were type species of

genera which had been proposed by various European workers at different times.

Lohwag distinguished five main types of covering of the pilei, viz.:

1. The derm which consist of hyphae which run more or less perpendicular to

the surface. Of this, there are three kinds, (a) the Aymeniderm in which

the elements are tightly packed, resembling a hymenium; (b) the trichoderm

consisting of hair-like elements, either separate or bundled together, and

closely joined and, (c) the palisadoderm similar to the hymeniderm but con-

sisting of slender, loose elements.

2. The hymenophoral cover, which consists of a sterile hymenophore.

3. The cutis, which consists of elements arranged parallel to the surface giving

the smooth glabrous appearance.

4. The cortex which consists of a denser matting of more or less modified hyphae

of the context.

5. The crust, which consists of a hard and sharply distinguished layer on the

surface without regard to its structural origin.

Later, H. Lohwag (1941) adopted K. Lohwag’s terminology in his studies

on the anatomy of the Asco- and Basidiomycetes and introduced the term paraderm

to describe the surface covering which consists of a pseudo-parenchymatic structure

built up of isodiametric cells.

Furtado (1965 a), in his studies of the relation of the microstructures to the

taxonomy of the Ganodermoideae, reformulated the older concepts of the types

of structure of the pilear cover in more exact terms. He also introduced the terms

“‘lacca-like substance” and ‘‘laccate appearance” to replace “‘lacca”’ and “‘laccate”’,

as previously used by many authors, because the chemical nature of this substance

is unknown. He regarded the ‘cortex’? and ‘“‘derm” as the two major categories

of structures found in the Ganodermoideae. The term ‘“‘cortex’’ he applied

to “‘a structure lacking any distinctive layer’ but characterized by a “continuous

and progressive condensation of the context hyphae towards the periphery.” The

““derm”’ was defined as ‘“‘all types of structures in which the hyphae are anti-clineal

to the pilear surface.” Five types of derm, viz., the Hymeniderm, Palisadoderm,

Trichoderm, Paraderm and Indeterminate derm, were recognized. The first four

terms agreed with those of K. Lohwag (1940) and H. Lohwag (1941) while the

fifth was proposed for ‘the type of derm found in a structure formed of incrusted

and intermingled hyphae in which the original arrangement cannot be traced

precisely.” These studies clearly demonstrated the varied nature of the upper

surface and the morphological differences that exist between the hyphae that

comprise the various types of covering of the pileus. The proposed terms, however.

do not indicate which type of nyphae, if more than one type of hypha is present

in the pileus, undergo the modifications to produce a specific kind of upper

surface.

After publication of Corner’s concepts of hyphal systems (1932 a, b) and

Lohwag’s (1940) work on the nature of the upper surface of fruit-bodies several

workers realized the potential usefulness of these concepts in Hymenomycete

taxonomy and called for urgent application of Corner’s methods to studies of

146

Hymenomycetes. Corner (1954) a) stated that only by a study of hyphal characters

could a natural system of classification of the Hymenomycetes be worked out.

Both Corner (1954) a) and Wakefield (1948) mentioned the existence of series

of Hymenomycetes related by structure but differing in hymenial configuration

and characters. They stressed the importance of separating species and genera

by considering the sum of all the characters present in the carpophore. Kotlaba

(1964) restated the views that the microstructures of carpophores and spore

characters are the only constant and reliable characters in Hymenomycete taxonomy.

He stated: “The importance of these characters lie in their particular combinations.

Furthermore, the same character may have different taxonomic value in different

groups and cannot be generalized.”” He thought that genera of higher fungi should

be delimited on the basis of a complex of characters while species may be delimited

on single characters only.

Cunningham (1946, 1947, 1948 a, b, c, d, e, f. g, h, 1949 b, 1950 a, 1954)

was the first worker to apply Corner’s concept of hyphal svstems to the classification

of polypores. In his studies of the Polyporaceae of New Zealand, in which he

recognized 17 genera, he regarded the types of hyphal systems present in the

fruit-bodies, together with the absence or presence of clamps on the generative

hyphae, the colour of the hyphae and the type of basidia produced, as important

at the generic level. These criteria, however, were not used consistently. While

some genera were characterized by the absence of clamp connections, e.g. Fomes

Kickx, Fomitopsis Karsten, Coltricia Mich. ex S. F. Gray and /nonotus Karst.,

species both with and without clamp connections were Jef! in Poria and Merulius

Hall. ex Fr. Similarly, he included in some genera, e.g. Lenzites Fr., Trametes

Fr., Coriolus Quél. and Daedalea Fr., only species with trimitic hyphal systems

while species with monomitic, dimitic, and trimitic hyphal systems were included

in Poria. This inconsistency is perhaps not surprising since a study of Cunningham’s

(1946, 1954) definitions of the different types of hyphae, do not reveal distinct

and constant differences between skeletal and binding hyphae. It appears that

Cunningham himself was not too clear about differences between these types of

hyphae.

Pinto-Lopes (1952) also proposed a system of classification of the Polyporaceae

based mostly on the characters of the hyphae which comprise the fruit-body.

In a bio-taxonomic study of polypores, he concluded that the characters of the

hyphae were fixed and genetically constant under many different conditions of

growth in the carpophores as well as in culture on artificial media. He distinguished

between three types of hyphae, viz. primary, secondary and tertiary. Hyphae

produced by germinating basidiospores are primary hyphae but they become

secondary hyphae, which may form clamp connections, after fusion with other

genetically compatible primary hyphae. Differentiated hyphae, which are

characterized by thickening of the wall, are the tertiary hyphae. According to

him the carpophores of all species consist of secondary and tertiary hyphae of

which the micromorphological characters are constant. In order to prove the

constant characters of the tramal hyphae, he also investigated those characters in

artificial culture and found that: (i) secondary and tertiary mycelium were always

present in cultures although it was difficult to distinguish the tertiary mycelium

in some cases; (2) species with clamp connections on the secondary hyphae of

the carpophore also have them on the secondary hyphae in culture; (3) with some

exceptions, the colour of the mycelium in culture was the same as that of the

hyphae of the carpophore and was produced in the same way. He concluded that

the same types of hyphae are present in both carpophores and cultures. Pinto-Lopes

concluded that each species has a plan of anatomical organization which is always

constant in all carpophores and that carpophores of the same species always have

147

the same structure in spite of differences in the appearance of the upper surface

of the carpophores. Hyphal characters have great taxonomic value while characters

such as surface of the carpophore and consistency of the context are of lesser

value. Macroscopic characters such as carpophore shape, pore shape, tube length

and microscopic characters such as spore shape and spore colour he regarded

as having no taxonomic value. He agreed with Ames (1913) that certain micro-

scopic structures such as cystidia are useful aids for the recognition of species

but are too variable to be of taxonomic importance above this level. Pinto-Lopes

regarded the characters of the secondary hyphae as of prime value and _ the

characters of tertiary hyphae of secondary value but the characters of the

secondary and tertiary hyphae taken together are of prime value and permit the

division of the family into sub-families. On this basis he distinguished eight main

groups in the Polyporaceae with the following hyphal characters:—

A. Secondary hyphae with clamp connections:

(1) tertiary hyphae hyaline, with clamp connections and walls not thickened

or slightly thickened;

(2) tertiary hyphae hyaline, with clamp connections, walls thickened;

(3) tertiary hyphae hyaline without clamp connections, walls more or less

thickened, and

(4) tertiary hyphae yellow or brown, without clamp connections and walls

more or less thickened.

B. Secondary hyphae without clamp connections:

(1) tertiary hyphae hyaline, septate and walls slightly thickened;

(2) tertiary hyphae yellow, septate, and walls slightly thickened:

(3) tertiary hyphae hyaline, aseptate, walls much thickened, and

(4) tertiary hyphae yellow or brown, aseptate and walls much thickened.

These distinctions formed the basis for his system of classification of the

Polyporaceae consisting of 22 genera distributed among 7 sub-families.

The work was severely criticized by Corner (1954 b) for the author’s views

on hyphal modifications and his disregard for characters other than those of the

hyphae. His lack of close adherence to the International Code of Botanical

Nomenclature also drew criticism (Cooke, 1959; Teixiera, 1962 b) and the work

did not meet with acceptance among mycologists. Nevertheless, Pinto-Lopes’ work

is of considerable value because of the accurate and reliable observations on the

hyphal characters of the species described and his confirmation of the existence of

different types of hyphae in the fruit-bodies of a large number of species of

Polyporaceae. Despite its shortcomings, it cannot be disregarded by any student

of micromorphology and taxonomy in the Polyporaceae.

Both Corner and Pinto-Lopes thus focussed the attention of taxonomists

sharply on the varied nature of the microscopic characters of the hyphae in the

carpophores and have indicated their value in the taxonomy of this group. Both

workers have indicated the important differences betweei the undifferentiated and

differentiated hyphae in fruit-bodies of Hymenomycetes.

The use of hyphal characters and Corner’s (1932 a, b) concepts of hyphal

systems have been applied to taxonomic studies of polypores by a number of

different workers. Teston (1953 a) stated that Corner’s (1932 a) and Cunningham’s

(1946) definitions of hyphal systems do not distinguish clearly between skeletal

and binding hyphae. In her study of the hyphal systems of 100 species of

Polyporaceae from the Bourdot herbarium in the Museum of Natural History

148

in Paris, Teston (1953 a, b) often found it difficult to place particular hyphae

in one of the three systems and to decide whether a particular species was mono-

mitic, dimitic or trimitic. She reported that thickened walls and lack of septa

distinguish the skeletal systems from the generative system but that intermediate

stages, i.e. hyphae with thickened walls and clamp connections can also be present.

In less complex species, such intermediate hyphae, (mediate hyphal system, Corner,

1932 a) would be numerous and function as a pseudo-skeletal system. If species

of each genus were arranged in order of decreasing importance of the pseudo-

skeletal system, an almost continuous series is obtained in which it is difficult

to separate clearly the monomitic species from the dimitic. In species where

clamp connections are lacking on the generative hyphae, the distinction is even

more difficult and can be based only on wall thickness which varies progressively.

With regard to the binding hyphae, Teston (1953 a,b) believes that they cannot

be defined absolutely, but only by comparison with the skeletal hyphae. Although

distinctions exist in form and diameter and staining reactions, the binding and

skeletal systems are related through a system of intermediates. Teston agreed with

Cunningham (1946) and Pinto-Lopes (1952) that absence or presence of clamp

connections is of great importance in relation to structure. Species which do

not possess clamp connections on the generative hyphae do not attain the same

complexity of structure which is found in species with clamp connections on their

generative hyphae. She concluded that species can be arranged in order of

increasing complexity, from monomitic species through all intermediates to trimitic

species, within each genus. But the characters of differentiated hyphae cannot

serve as a basis for taxonomy because it will lead to the fragmentation of genera.

It is a badly defined character which may vary with size and age of the fruit-body.

Only rarely does it permit the recognition and definition of natural groups. It can

be used to advantage to arrange species in such groups.

Teston’s (1953 a, b) observations thus indicate that the hyphal modifications

are not as clearcut and fixed as Corner (1932 a, b, 1954), Cunningham (1946) and

Pinto-Lopes (1952) believed. Teston’s conclusions differ from these author’s views

that hyphal characters are of primary importance in the taxonomy of the Poly-

poraceae although she does admit that they may permit the recognition and

definition of natural groups.

Hansen (1958) in her study of the anatomy of the Danish species of

Ganoderma, confirmed some of Teston’s (1953 a, b) observations and conclusions.

Hansen found that the skeletal hyphae differed from Corner’s (1932 a, b, 1953)

definition in that they often have one or more branches near their distal ends.

Their main stems are thick-walled and aseptate and arise at clamp-connections on

thin-walled, generative hyphae. The lateral branches act as ties although the main

stems are arranged longitudinally. In the dissepiments, these lateral branches of

the skeletal hyphae take over the binding function completely. Binding hyphae

are present in the context only and are of the bovista type. Hansen concluded

that the differences in the skeletal systems of the species examined are of a

qualitative as well as quantitative nature and not constant enough for use in the

delimitation of species.

The American workers, Lowe and Overholts, on the other hand, largely

ignored the concepts of hyphal systems in their work on American polypores.

Overholts (1953) in his account of the Polyporaceae of the North-eastern United

States and Canada, included details of spores, basidia, cystidia and other micro-

structures and hyphae in his descriptions of species. He often used the term

“hyphal complexes” in descriptions of species which possess binding hyphae in

Corner’s (1932 a) terminology. No attempt was made to use these characters for

generic delimitation or classification of the species described. Similar work was

149

published by Lowe (1946, 1947, 1948, 1956, 1957, 1958, 1961, 1963 b, 1966)

who included details of characters of the spores, hyphae and other microstructures

of the large number of species described by him. These characters were used for

diagnostic purposes at the species level only while the generic concepts were

Friesian.

Banerjee & Debi (1956) attempted to relate micromorphological and structural

differences with morphological differences in the fruit-bodies of different collections

of Polystictus xanthopus. They could distinguish three morphologically different

types of fruit-body, viz.:

1. thin fruit-bodies with long. narrow, excentric stipes and minute, regular pores;

2. small, thick, fruit-bodies with thick, excentric stipes and regular but larger

pore mouths, and

3. thin, sessile fruit-bodies with hydnoid to irpicoid pores.

Forms intermediate between all three types, were found. The basidia and

spores of fruit-bodies of the first two types were similar in size and shape. The

basidia of the third type of fruit-body were larger than those of the other two

types and so were the spores, which also differed in shape. Fruit-bodies of the

first two types consisted of generative hyphae with clamp connections and mediate

hyphae, skeletal hyphae and binding hyphae. In fruit-bodies of the third type.

no binding hyphae were present but only much branched generative hyphae which

resembled binding hyphae. The authors described binding hyphae as much-

branched, thick-walled, and without clamp connections. They concluded that two

varieties of Polystictus xanthopus exist since ‘‘the separation into three types by

macroscopic characters alone cannot be substantiated in al! cases by microscopic

characters and in other details.’”’ The larger pore, basidium and spore dimensions

as well as the absence of binding hyphae, distinguish the third type of fruit-body

from the other two. These workers thus did not regard these differences to be

sufficiently important and constant to justify recognition of separate species.

Teixeira (1956, 1958, 1960, 1962 a, b) on the other hand firmly believes that

the microstructures and hyphal morphology of the carpophores of polypores are

the only characters of taxonomic value. Teixeira & Rogers (1955) noticed that

Aporpium caryae, which has a poroid hymenial surface, also has cruciate-septate

basidia. They transferred this species to the Tremellales which are characterized

by such basidia. Teixeira (1956) published details of his methods of studying

the construction of the carpophore. Essentially, these consist of carefully teasing

apart thick sections, from different parts of the carpophore, cut parallel to the

direction of growth of the hyphae and dissecting out individual hyphae with the

aid of fine needles under 50x magnification of the dissecting microscope. The

morphology of the hyphae and other microscopic structures and their inter-

relationships, are then studied under the oil immersion lens. This method is

essentially similar to that described by Corner (1932 a, b, 1953). Recently Fidalgo

(1967), published a sophisticated method of obtaining intact hyphae for microscopic

examination from carpophores by means of ultrasonic vibrations.

Teixeira (1958) applied his method to study the microstructure of Laricifomes

Officinalis. He showed that this fungus, which is the type species of the genus

Laricifomes Kotlaba & Pouzar, differs in a number of structural details from

Fomes fomentarius the type species of Fomes Kickx, (Donk, 1960). He also

demonstrated (Teixeira, 1960) that the generative hyphae of a number of common

North American species have clamp connections at the septa although this fact

is not mentioned in a number of important reference works in this field. Clamp

connections were absent from the generative hyphae of two of these species.

150

Teixeira (1926a) afterwards applied his method to the study of the microstructure

of the basidiocarps of species of the genus Fomes Kickx. In this work he amended

the generic description to include species of which the surface is covered by a

definite crust over a chestnut-coloured context consisting of thin-walled, hyaline

generative hyphae with clamp connections at the septa and thick-walled, differen-

tiated hyphae without clamp connections, the skeletal and binding hyphae. The

ends of the skeletal hyphae at the upper suface are agglutinated into the tough,

hard crust. Hairs over the crust are produced by terminal proliferation of these

skeletal hyphae. Although the skeletal hyphae are aseptate, generative hyphae with

septa and clamp connections often produce branched structures which resemble

the binding hyphae. Together with the type species Fomes fomentarius, two other

species were recognized. Species were distinguished on the characters of the

crust and the size of the pores. On the basis of this work, Teixeira concluded that

the microstructures of the fruit-bodies, such as basidia, spores and other hymenial

structures, as well as the nature of the generative hyphae and the specialised

branches which they produce, are genetically constant in character and therefore

more reliable in taxonomy than the morphological features which are still used

to delimit genera in the Polyporaceae. These conclusions were restated and

supported by numerous examples from earlier literature in a review of the taxonomy

of the Polyporaceae published later (Teixeira, 1962 b).

O. Fidalgo & M. E. P. K. Fidalgo also used Teixeira’s (1956) methods to

study the hyphal systems and taxonomy of a number of genera and species of

polypores. By these methods, M. E. P. K. Fidalgo (1958) demonstrated that

Lenzites cinnamomea Fr. differs from Gloeophyllum sepiarium (Wulf. ex Fr.)

Karst. in the characters of the skeletal and binding hyphae. She later proposed

the genus Phaeodaedalea (M. E. P. K. Fidalgo, 1961) similar to Gloeophyllum

Karst. and Hexagona Fr. in its trimitic hyphal system, but differing from these

enera in having brown, globose spores. She also showed that Trametes odoratus

Fr. is characterized by a dimitic hyphal system (M. E. P. K. Fidalgo 1962). The

genus Osmoporus Sing., of which Trametes odoratus is the type is thus distinct

from Gloeophyllum Karst. of which the type species, Gloeophyllum sepiarium

(Wulf. ex Fr.) Karst., has a trimitic hyphal system.

O. Fidalgo (1958) concluded that Ptychogaster rubescens is the chlamydos-

porous form of Polyporus guttulatus Peck because of similarities between hyphae

from cultures of Polyporus guttulatus and those from the fruit-bodies of Ptychogaster

rubescens. Later, (O. Fidalgo 1962 a, b) he showed that Bornetina corium Mangin

& Viala is the imperfect state of Diacanthodes novoguinsensis (P. Henn.) O. Fid.

and that both are characterized by a monomitic hyphal system with clamp

connections on the hyphae and a tendency to be dimitic.

In collaborative studies, Fidalgo & Fidalgo (1962) described the hyphal

systems, the morphology and construction of the sporocarp of five species of

polypores. They also proposed the new genus, Pseudofistulina, (Fidalgo & Fidalgo,

1963) for which extensive descriptions of the micromorphology of the hyphae and

other structures, as well as descriptions of the construction of various parts of the

fruit-body, were presented. This genus differs from Fistulina Bull. ex. Fr., as

typified by Fistulina hepatica Huds. ex Fr., by the absence of clamp connections

on the hyphae, the presence of a derm composed of acanthophyses and hyaline,

thin-walled spores instead of yellow, thick-walled spores. Both genera have mono-

mitic hyphal systems.

The presence or absence of clamp-connections on the hyphae of Hymeno-

mycetes and their importance in taxonomy has been the subject of much discussion

by various workers. Not much attention was given to these structures by Bourdot

151

& Galzin (1928), Donk (1933), Overholts (1953), and Lowe (1948, 1956, 1957,

1958, 1963 b) who often reported them to be absent from hyphae of species

where they are now known to be present (Pinto-Lopes & Farinha, 1950). For

taxonomic purposes, the presence or absence of clamp connections are regarded

to be of value at the species level only by some workers such as Hesler & Smith

(1963) and Smith (1966) while others, notably Pinto-Lopes (1952) Singer (1962)

and Teixeira (1962 a, b) believe them to be of value at a higher level. The former

view is supported by Smith’s (1966) observations that clamp connections regularly

occur only in a small number of species of some genera in the Gasteromycetes.

Pantidou (1961), and Pantidou & Groves (1966) found that clamp connections

were present in mycelium of species of Boletaceae grown in cultures but which do

not have clamps on the hyphae of their fruit-bodies while in other species, clamp

connections and simple septa were present in the same mycelium. Smith (1966)

suggested that since the numbers of species with clamps vary in different groups,

a quantitative study of the proportion of clamped septa in both clamped and

clampless species is needed as well as mating compatibility studies between single

spores of clamped and clampless species. Until such studies had been undertaken

he regards the absence or presence of clamp connections on the hyphae of the

basidiocarp as of significance at the specific level only. On the other hand the

views of Pinto-Lopes (1952), Singer (1962) and Teixeira (1962 b) are supported

by the work of Cunningham (1946, 1947, 1948 a-h, 1949 a, b, 1950 b), Teixeira

(1958, 1960, 1962 a, b), O. Fidalgo (1958, 1963), M. E. P. K. Fidalgo (1961, 1962),

Fidalgo & Fidalgo ((1962, 1963, 1966, 1967) and Furtado (1964, 1965 a, b, 1966,

1967) who found clamp connections to be constantly absent or present on the

generative hyphae of different species of Polyporaceae, and who characterized

genera on this basis. The position was well summarized by Singer (1962) who

stated: “If the presence or absence of clamp connections is used as a character

in taxonomy it is essential to make sure that the specimen studied is not merely

a parthenogenetic form of a normally bipolar or tetrapolar species. If this possibility

is excluded we have further to deal only with species with normal sexuality that

have lost their ability to form clamp connections and homothallic forms, species

or genera that find themselves in the same condition. Under these circumstances

the presence or absence of clamp connections must be accepted as a valuable

character.” In Donk’s (1964) recent treatment of the Aphyllophorales, it is evident

that some groups contain species with clamp connections only while others contain

species with simple septa only. Donk is of the opinion that “‘the value of clamps

as a taxonomic feature differs from group to group and may even appear erratic

within rather small taxa of lower rank such as species.” Furtado (1966) pointed

out that “‘in the clamped species studied experimentally, clamp connections are

formed only in one specific heterokaryon, the dikaryon.” He proposed that since

clamp formation is controlled genetically, it is necessary te study the cytogenetic

condition of the hyphae whenever the pattern of septation is decisive for definition

of any taxa or the proposal of any hypothesis.

The concept of hyphal systems and the use of hyphal characters in taxonomic

studies have also been applied to non-poroid Hymenomycetes. Ragab (1953)

reported that most species of the Hydnaceae have monomitic hyphal systems while

some are dimitic. In some genera monomitic as well as dimitic species are

found. He included a key to 14 genera in the Hydnaceae, based on hyphal

characters as well as morphology.

Cunningham (1963) applied his concepts of hyphal systems in studies of the

Thelephoraceae of New Zealand. The hyphal systems were however used as a

taxonomic character only at the generic or even sub-generic level. Cunningham

(1963) attached greater importance to the microscopic structures of the hymenial

152

layer and differences in habit and hymenial contiguration. Cunningham stated:

“There is not the marked differentiation in hyphal systems in the Thelephoraceae

that is so noticeable a feature of the Polyporaceae; far the greater number of genera

possess species with both monomitic and dimitic systems, hyaline and brown

hyphae and are with or without clamp connections. A few genera however do

show some differentiation.”

Talbot (1951, 1954 b, 1958 a, b) used micromorphological characters to delimit

genera and species in his studies of the South African resupinate Hymenomycetes.

In his descriptions, full details of hyphal characters and microscopic structures

were included but the concept of hyphal systems is not always evident. He regarded

the nature and absence or preseace of microscopic structures in the hymenium as

of greater taxonomic importance. In this respect he agreed with Cunningham

(1963) and other workers.

Reid (1959, 1962, 1963, 1965) also placed much emphasis on micromorpho-

logical characters of hyphae and microscopic structures in his taxonomic studies

which include mainly the lower Hymenomycetes.

Welden (1960), who revised the American species of Cymatoderma Jungh.

on the basis of anatomical studies, could distinguish four main types of hyphae

in the context of these fungi, viz.:

(1) long sub-solid to solid hyphae rarely branched or clamped, arising gradually

or abruptly from thin-walled hyphae,

(2) similar but narrower and more tortuous hyphae,

(3) solid to sub-solid, narrow, short or long branching hyphae, and

(4) thin-walled, relatively wide branching and clamped hyphae.

Intermediates between these may also be found. Welden (1960) found that

he could not apply ‘Corner’s (1932 a, b) terminology to the hyphae and hyphal

structure despite the fact that Reid (1959) had divided the genus into dimitic

and trimitic sections. Welden considered some of the branching hyphae to perform

skeletal as well as binding functions while some of the thick-walled hyphae with

clamp-connections served generative as well as skeletal functions. He concluded:

“IT do not wish to negate the useful terms ‘dimitic’ and ‘trimitic,’ but a strict

interpretation of Corner’s terms does not appear applicable to the American

species of Cymatoderma unless all the species are considered trimitic’”. He used

macroscopic morphological characters as well as microscopic characters of the

hyphae and hymenial and sub-hymenial structures to delimit species.

Lentz (1960) made extensive use of micromorphological and hyphal characters

to characterize type species of Stereum Pers. ex. S. F. Gray and allied genera

after he realized that his earlier descriptions (Lentz, 1955) of the genus Stereum

in the upper Mississippi Valley was based largely oi macroscopic and gross

microscopic characteristics.

Slysh (1960) used characters of the microscopic structures and hyphae of the

fruit-bodies of Peniophora Cooke to describe the species found in New York State

and surrounding regions. He reported that only one species, Peniophora greschikii

is composed of two distinct types of hyphae, i.e. dimitic. All the others are mono-

mitic but there are differences in the arrangement of the kyphae in different species.

He further reported that some species have simple-septate hyphae, while they may be

nouose-septate in others. Hyphal septation was used to distinguish between species of

two out of the total of eight sections into which he divided this genus.

153

Maas Geesteranus (1962, 1964) found hyphal characters and anatomical

structure to be of fundamental importance for the delimitation of genera in the

Hydnaceae, and described diff2rences in the morphology and arrangement of

hyphae of a number of species whose carpophores consist of generative hyphae

only. He stated: ‘“The necessity and importance of the anatomical structure for

the correct understanding of a genus becomes at once apparent when one considers

Steccherinum as it was conceived by Banker and extended by subsequent authors

Pens Also the apparent difficulty experienced by some authors sharply to delimit

Hydnellum and Sarcodon disappears as soon as the hyphal system in both genera

is taken into account.”’ On the basis of this study of the generic types he concluded:

“It is more than likely that, with the spines as the sole character in common, the

connection of many of the hydnaceous genera, will have to be sought not within

the “Hydnaceae’ but, irrespective of hymenial configuration, with groups now

widely separted.”’

Bondartzeva (1963) discussed the use of the anatomical criteria for the

taxonomy of the Aphyllophorales. She accepted Corner’s (1932 a, b) concepts

but concluded that the type of hyphal system is mainly of generic value in the

taxonomy of this group. In cases, however, where adaptation to conditions of

an external medium could have produced changes in the hyphal system, while

other features make it evident that the species belong to the same genus, the type

of the hyphal system may be either of supra-generic or infra-generic importance.

She further regarded the type of hyphal system as an element of adaptive evolution.

This view is supported by: (i) the relationship between ihe consistency of the con-

text of the sporocarps and consequently their anatomical structure and the way of

life of the species, and (ii) the limited number of types of hyphal systems and the

occurrence of identical systems in species which differ widely in respect of other

characters but are similar in ecological relationships. The anatomical structure,

in her opinion, is an important indication of the life forms but not of the basic

line ot evolution. The anatomical structure is thus an element of partial and

not general evolution and hyphal systems of the fruit-bodies cannot be considered

as one of the basic features in the development of a phylogenetic system. In this

respect, Bondartzeva believes that the problem of convergence, which occurs widely

in the Hymenomycetes, requires elucidation since it offers the key to the under-

standing of the complex phylogeny of this group of fungi.

Smith (1966) in his discussion of the hyphal structure of the basidiocarp stated

that the mitic system (Corner’s hyphal systems, 1932 a, b) is not very suitable

for application to the Agaricales and thought it “ineffectual to set up special terms

for generalized situations when they can be properly evaluated only by careful

attention to detail.” He thought the the terms such as “generative hyphae” fail

to express adequately what is actually seen under the microscope in individual

species. He considered descriptions of hyphal modifications, cell shape and

microstructures, especially cystidia, to be important in taxonomic studies and agreed

with Bondartzeva (1963) by stating that most hyphal modifications in the basidio-

carps appear to be adjustments to meet the problem of moisture loss. The diversity

in form and content of end cells of hyphae is due to the different ways which

different species have evolved to meet this problem.

Donk (1964) used the concept of hyphal systems together with macro- and

micromorphological characters to characterize the families of the Aphyllophorales.

He regarded these hyphal systems and micromorphological characters as of great

importance in taxonomy. His descriptions of families were preceded by concise

reviews of the anatomical features.

154

The morphology of microscopic structures and hyphal modification was

reviewed by Talbot (1954 a) and Lentz (1954). Both authors discussed in detail

the morphology of the structures found in the fruit-bodies and hymenia of the

Hymenomycetes, their origin and probable functions. Many examples were

cited and misconceptions in the terminology corrected.

The application of hyphal characters and characters of microstructures by

different workers to the taxonomic problems of the Hymenomycetes, has thus

met with varying degrees of success. The concept of hyphal systems has been

rejected by some and enthusiastically adopted by others, but all the workers

mentioned here show in some way that hyphal morphology and modifications

cannot be ignored in Hymenomycete taxonomy.

Pure CuLTuRE STUDIES

Hyphal morphology and modifications have also been shown to be of basic

importance in the recognition of Hymenomycetes in pure cultures. Numerous

workers who studied decay of timber have demonstrated that pure culture studies

are important diagnostic tools in such work. Certain workers (e.g. Boidin, 1964)

also regard pure culture studies to be of great value in taxonomic studies of

Hymenomycetes.

Among the pioneers in this field were Long & Harsch (1918) who studied

a large number of species in culture and introduced the terms still used to describe

the texture of the mat. Fritz (1923) described 18 species, destructive to balsam

fir, in detail and indicated the characteristics which should be used in the

identification of decay fungi in pure culture. She indicated the importance of

microscopic characters. Baxter (1924-1945) made extensive use of cultural

characters in his taxonomic studies of resupinate polypores but described micro-

morphological characters of the cultures only on rare occasions. Baxter used 2

per cent malt agar at temperatures of 25°C, 30°C and 35°C. Humphrey & Siggers

(1933) studied the temperature relations of decay fungi and found that they

could be grouped into three groups on the basis of their growth rates at different

temperatures.

Bavendamm (1928) observed that fungi which cause white rot, were capable

of darkening the colour of media containg gallic acid or tannic acid or other related

compounds by oxidation. Species which cause brown rot did not cause darkening

of the medium. This observation was later confirmed by Davidson, Campbell &

Blaisdell (1938) who referred to this phenomenon as the “oxidase reaction”. In

a study involving 210 species of decay fungi, these authors found that these fungi

could be divided into eight groups on the basis of their reactions when grown on

malt agar media containing gallic acid or tannic acid. Of the fungi tested, 166

produced diffusion zones and 36 did not produce diffusion zones on either medium

while seven fungi gave inconsistent results. Of the 36 fungi that were negative

for extra-cellular oxidase, 30 were associated with brown rots while 151 species

out of the 166 that were positive, were associated with white rots. These reactions

proved to be useful diagnostic characters in the identification of cultures of decay

fungi and were later incorporated by Davidson, Campbell & Vaughn (1942) in

their descriptions of cultures of fungi causing decay of living oak in the Eastern

United States. In this work, extensive use was made of microscopic characters

of the hyphae and other specialized structures in the mat as well as macroscopic

appearance of the cultures. These workers also introduced a key in which various

characters of the different cultures were expressed by alphabetical symbols. This

allowed the incorporation of new species into the key with a minimum of disruption.

155

Refshauge & Proctor (1936), studied Australian wood-decaying fungi in

culture. They found that most of the fungi associated with white rot of timber were

also capable of decolourizing certain dyes which had been added to the media.

This was also found to be due to oxidation.

Jorgensen & Vejlby (1953) described a method for the preparation of an

extract from red cabbage leaves to determine the presence of polyphenol oxidase

enzymes in cultures of wood-rotting fungi. Etheridge (1957) suggested the use

of meal of white spruce as an indicator medium for the presence of oxidase enzymes

in white-rot fungi. More extensive studies of the occurrence and function of

oxidase enzymes in wood-rotting fungi were carried out by Lyr (1955, 1963) and

Luthardt & Lyr (1965).

The oxidation of phenolic compounds by wood-rotting fungi was reviewed

by Kaarik (1965) in her extensive study of the oxidation of 20 different phenolic

compounds by a large number of decay fungi in pure culture. By applying drops

of alcoholic solutions of the phenolic compounds to growing cultures of the fungi

she found that four main types of reactions occurred, viz.: (i) production of

tyrosinase only; (ii) production of laccase only; (iii) production of both laccase

and tyrosinase, and (iv) production of neither laccase nor tyrosinase. On the basis

of their reactions the mycelia could be divided into four groups which in turn could

be sub-divided according to the intensity of the reactions and specific reactions

to specific compounds.

Campbell (1938), in a study of 32 species of Fomes in culture, included

extensive microscopic details of hyphae and structures formed in culture. The

presence or absence of brown diffusion zones around mycelia grown on media

containing gallic acid and tannic acid, was used as a diagnostic character in this

work also.

Cartwright (1929, 1931), studied decay fungi in pure culture in England.

In collaboration with Findlay he later described the cultural characters of many

fungi causing decay of soft wood and hard wood trees and timber (Cartwright

& Findlay, 1946). Their descriptions, which were devised primarily for the

recognition of fungi from decayed timber, include details of microscopic structures,

hyphae, physiological data and details of the decay.

A number of workers described wood-rotting fungi in culture from specialized

habitats or a single host. Snell (1922) described fungi causing decay of building

timber and in cotton mills. Walek-Czernecka (1933) described cultures of fungi

from decayed railway sleepers in Poland. Davidson, Lombard & Hirt (1947)

described fungi causing decay in wooden boats, while large-brown-spored house-rot

fungi in the United States were described by Davidson & Lombard (1953). Earlier,

Davidson & Campbell (1943) reported on nine species of decay fungi from black

cherry. Robak (1942) described six species of decay fungi from pine in Norway.

A lack of similarity of methods adopted was apparent among these workers.

This made it very difficult if not impossible to compare results reported by

different authors. An attempt to overcome these difficulties was made by Nobles

(1948) with the publication of descriptions of the cultural characters of 126 species

of wood-rotting fungi from Canada. For this purpose the topography and colour

changes of the mat as well as other macroscopic characters, the reactions on gallic

acid and tannic acid media and micromorphological details of hyphae and other

structures were combined in the descriptions of the various species. These

characters were also reflected in a key for the identification of species, in which

a number of characters, both macroscopic and microscopic, were represented by

different numerals arranged in 11 vertical columns. This key, like that devised

156

by Davidson ef al. (1942), is capable of continued expansion by the incorporation

of new species. Later Nobles (1958 a) devised a rapid test for the presence of

extra-cellular oxidase enzymes in cultures of decay fungi. By the application

of a drop of an alcoholic solution of gum guaiac directly to a culture, the presence

of extra-cellular oxidase enzymes is indicated by rapid blueing of the gum guaiac

solution. No colour change occurs when the solution is applied to cultures of

species that cause brown rots. Parallel tests on 33 species with the gum guaiac

solution and the standard Bavendamm (1928) metiod, in which cultures are grown

on malt agar containing gallic acid and tannic acid, gave identical results for nearly

90 per cent of the species. Inconsistent reactions were obtained by both methods

from 19 species.

Nobles’ methods were adopted by Van der Westhuizen (1958, 1959) for

descriptions of South African wood-rotting fungi in culture. Very similar methods

were used by Da Costa, Matters & Tamblyn (1952) for their descriptions of

Australian wood-rotting basidiomycetes in culture.

Studies of pure cultures have been used by various workers to distinguish

between species with morphologically similar carpophores. In many cases inter-

collection pairing of mycelia each derived from a single basidiospore, a technique

pioneered by Bensaude (1918), Vandendries (1922, 1923) and Kniep (1928), has

been employed to enhance and confirm the results obtained from pure culture

studies. The formation of hyphae with clamp connections when clampless mycelia

each derived from a single basidiospore, are grown together in pairs, is regarded

as positive proof of conspecificity of the mycelia and consequently of the spores

and carpophores from which they were obtained. By using this method Mounce &

Macrae (1936) showed that Lenzites sepiaria, Lenzites trabea and Trametes

americana, which have very similar carpophores, are indeed three distinct species,

while Lenzites thermophila is conspecific with Lenzites trabea because mycelia

from single basidiospores from carpophores assigned to these two species, were

interfertile. Later Mounce & Macrae (1937) found that no clamp connections

formed when monosporous mycelia of Fomes roseus and Fomes subroseus were

paired thus confirming the validity of the two species, which differ further in respect

of spore shape and cther minor characters of the carpophores. Nobles (1943)

showed by this method that different cultures isolated from decay in pines were

identical to those derived from a carpophore of Poria microspora and distinct

from those of Trametes serialis which had been considered to be the cause of the

decay. More recently, studies of cultural characters, including in most cases

tests for interfertility, were used by various workers to solve taxonomic problems

in Hymenomycetes. Among these were Bose (1952) and McKay (1959) with

studies of Polyporus cinnabarinus and Polyporus sanguineus, McKeen (1952) with

studies of three species of Peniophora, Nobles, Macrae & Tomlin (1957), various

species of polypores; Harmsen, Bakshi & Choudnury (1958) with two species of

Merulius; Sarkar (1959) with six species of Coriolellus; Davidson, Lentz & McKay

(1960) with Stereum spp. causing pecky cypress; Denyer (1960) with two species of

Flammula; Harmsen (1960) with Merulius spp.; Weresub & Gibson (1960) with

Stereum pini; Aoshima, Lentz & McKay (1961) with Stereum taxodii; Boidin &

Des Pomeys (1961) with various resupinate lomobasidiomycetes; Lombard,

Davidson & Lowe (1961) with Fomes ulmarius and Poria ambigua; Nobles & Frew

(1962) with Pycnoporus,; McKay (1962) with Polyporus palustris and other brown

rot species: Lombard & Gilbertson (1965, 1966) with various Poria spp.; Macrae

& Aoshima (1966) and Macrae (1967) with Hirschioporus spp.; Sen & Sehgal (1967)

with twelve Indian polypores and McKay (1967) with Polyporus meliae and two

similar species.

Sy)

In some of these studies use was made of the ‘Buller phenomenon’’ to

determine the identity of many of the cultures studied. Buller (1931) showed that

a large monokaryotic mycelium of Coprinus lagopus was dikaryotized rapidly by

a small inoculum of dikaryotic mycelium of the same species placed at its periphery.

Kawamura (1941) showed that a haploid mycelium of Polystictus sanguineus could

be dikaryotized by mating with a theoretically incompatible dikaryotic mycelium,

produced by mating two haploid mycelia, neither of which was compatible

with the test haploid mycelium. Terra (1953) summarized the literature on the

“Buller phenomenon” and reported similar dikaryotization of a large hap-

loid mycelium by small dikaryotic mycelia with Schizophyllum commune, Leu-

coporus brumalis, Cytidia flocculenta, and Panus stipticus. In these cases

the haploid and dikaryotic mycelia were obtained from carpophores of the same

species but collected in different regions. Boidin & Des Pomeys (1961) used this

method in a study of certain species of resupinate Homobasidiomycetes. Nobles

& Frew (1962) though that the “‘Buller phenomenon’ could be a valuable tool

for the confirmation of identificaticn of cultures if it could be repeated with most

species of the Hymenomycetes. In their study of the genus Pycnoporus Karst.

these authors examined 103 cultures from many parts of the world. The identity

of 57 isolates was confirmed by means of mating t2sts between single spore cultures

while 46 isolates were identified by means of the “Buller phenomenon’’. Three

types of reactions were observed, viz.: (i) a positive reaction in which the haploid

test colony, originally composed of hyphae with simple septa, was converted to

the dikaryotic condition, as shown by the presence of hyphae with clamp connections

around the periphery; (ii) a negative reaction in which the haploid test fungus

continued to grow in the haploid condition whilst the dikaryotic inoculum grew

to produce a sector distinguishable by difference in colour or texture or growth

rate, and (iii) a negative reaction in which the dikaryotic inoculum failed to grow;

the test fungus grew around it, without dikaryotization taking place. By means

of these tests, the isolates of Pycnoporus Karst. were divided into three groups

of interfertile isolates. These three groups also exhibited similarities in cultural

characters and carpophore morphology and anatomy within each group. The

authors were thus able to distinguish between the three species of orange-red

polypores which had originally been described under the specific epithets of

Polyporus cinnabarinus Jacq. ex Fr., Polyporus coccineus Fr. and Polyporus

sanguineus L. ex Fr. Van der Westhuizen (1963) used the “Buller phenomenon”

to confirm the conspecificity of eight collections of Cerrena unicolor (Bull. ex Fr.)

Murr., a white rot fungus which has the bipolar type of interfertility. Studies with

pure cultures are thus valuable tools in taxonomic studies of the Hymenomycetes.

After extensive studies with pure cultures, Nobles (1958 b) presented a guide to

the taxonomy and phylogeny of the Polyporaceae on the basis of their micromor-

phological characters and oxidase reactions in culture. In a study of 212 species,

evidence was provided that the family is composed of two main groups: a

primitive group consisting of species that produce no extra-cellular oxidase and,

if heterothallic, show the bipolar type of infertility; and a more advanced group

made up of species that produce extiacellular oxidase and if heterothallic, show

the tetrapolar type of infertility in those species whose hyphae are regularly

nodose-septate and a third, intermediate and minor group of species in which the

advancing hyphae are simple-septate but the older hyphae become nodose-septate,

and which also display the bipolar type of infertility. Within each group, species

were arranged in smaller groups on the basis of their hyphal characters in culture

and basidiospore characters. In this way a total of 36 groups of species resulted.

Some of these groups appear to represent natural taxa while others appear to

require further sub-divisions suggesting supra-generic grouping, but whether

homogenous or not, the groups can be arranged in the order of increasing complexity

158

of their hyphal structures to form a sequence suggesting their phylogenetic

development. Nobles warned however that these groups can be accepted as

representing natural taxa only if the cultural characters used in the segregation

and arrangement of the species have recognizable counterparts in the carpohores

of these species. The evaluation of the taxonomic significance of the groups thus

requires correlated studies of the micromorphological characters of carpophores and

cultures of the species in each group. Later Nobles (1965) again presented these

views in a diagnostic key with brief descriptions of 149 species of wood inhabiting

Hymenomycetes devised primarily for their recognition in culture.

Some important facts are evident from Nobles’ work, viz.: (1) there is little

correlation between cultural characters and morphology, on the one hand and

morphology and hymenial configuration of the carpophores, the traditional bases

for generic delimitation, on the other; (2) hyphal characters such as septation

and modification, in combination with spore characters, are regarded as of supra-

generic or generic importance; (3) despite statements to the contrary by Whitehouse

(1949) and Raper (1953, 1954) strong evidence is presented in favour of the view

that the type of infertility is constant for each genus of the Polyporaceae; (4) the

absence or presence of extra-cellular oxidase in culture is of prime phylogenetic

importance; and (5) divisions between groups are based on a complex of

characters. In a critical review of the most recent systems of classification of the

Polyporaceae, Bondartzeva (1961) commented favourably on Nobles’ views which

she regarded as a sound approach to the development of a phylogenetic system

of classification. Lowe (1963 a) did not agree with Nobles’ views. He believed

that the Hymenomycetes with short-lived fructifications composed of thin-walled.

nodose-septate hyphae, are the more advanced species. He saw a parallel in the

flowering plants, where fast-growing herbaceous annuals with resistant seeds are

regarded as more advanced than woody perennials.

Studies in which structures found in culture were correlated with those of

the carpophores as suggested by Nobles (1958 b), have since been undertaken

by various workers. Sarkar (1959) studied six species of the genus Coriolellus

Murr. which were all found to be similar in cultural characters and carpophore

morphology. These six species which cause brown rots, do not produce extra-

cellular oxidase enzymes in culture. Five of them were shown to have the bipolar

type of infertility while one, Coriolellus malicola, was found to be homothallic.

Sarkar showed tha ttaxonomically important characters found in cultures were

also present in the corresponding carpophores from which the cultures were made.

Nobles & Frew (1962) in an exhaustive study of the orange-red polypore genus,

Pycnoporus Karst., presented cultural, morphological and genetical evidence in

support of the recognition of three species in this genus. All three species cause

white rots of both hardwood and coniferous trees. The authors showed that

hyphae, similar to those found in culture, are also present in the carpophores.

Basidia and basidiospores too, were alike in both the cultures and carpophores

but iodia which were present in most cultures were not found in the carpophores.

Van der Westhuizen (1963) showed that the thick-walled as well as thin-walled

hyphae of Cerrena unicolor (Bull. ex Fr.) Murr. have clamp connections at the

septa in both the cultures and fruit-bodies. This fungus differed in this respect

from the type species of the different genera to which it had been referred.

Farinha (1964) described the characters of the hyphae from carpophores and

cultures of 30 species of polypores according to the methods of Pinto-Lopes (1952).

She concluded that the microscopal characters of the secondary hyphae of the

carpophores and those from cultures of the same species are identical. This

IS)

also applied to some kinds of tertiary hyphae, but other forms of growth were

apparently lacking in culture. She thought that a large number of species should

be studied in detail before a general terminology for the different types of hyphae

could be devised. Such a terminology should take the different forms of growth

of the hyphae, throughout the life cycle, into consideration and should be applicable

to hyphae from the carpophores as well as from cultures on artificial media.

Lombard & Gilbertson (1965) described the cultural characters and carpophores

of 14 species of Poria with negative or weak oxidase reactions. These species

displayed different hyphal characters but in most species the hyphae present in

the cultures were also present in the carpophores. Only in those species in which

nodose-septate hyphae with irregularly thickened walls were present in the

cultures, were those hyphae not found in the carpophores as well.

It thus appears that most of the structures formed in cultures, may also be

found in the carpophores. A statement to this effect had been made by Pinto-

Lopes (1952) but was not supported by any evidence at that time. It must

however be determined whether this is true for all species of Hymenomycetes and

whether some of the hyphal modifications described in cultures, such as cuticular

cells and hyphae with interlocking projections, are also present in the carpophores

of the relevant species and, finally, whether the relationships indicated by cultural

characters can also be demonstrated to exist between the fruit-bodies.

3. MATERIALS AND MEHODS

MATERIALS

Culture media:

1.25 per cent Difco malt agar;

1.25 per cent Difco malt agar + 0.5 per cent tannic acid;

1.25 per cent Difco malt agar + 0.5 per cent gallic acid.

Extra-cellular oxidase enzyme test solution:

0.5 gm gum guaiac in 30 cc of 96 per cent alcohol, (Nobles, 1958 a).

Mounting media for microscope preparations:

5 per cent aqueous KOH solution (Talbot, 1951);

1 per cent aqueous phloxine solution (Talbot, 1951);

Lactophenol (Smith, 1960).

Specimens for study:

Dried fruit-bodies from the following herbaria were examined: National

Herbarium, Mycological Collection, Pretoria (PRE); P. A. van der Bijl Herbarium,

University of Stellenbosch (STE); Canada Department of Agriculture, Mycological

Herbarium, Ottawa, Ont. (DAOM); New York Botanical Gardens, New York,

(NY); Royal Botanic Gardens, Kew, England, (K); Farlow Herbarium of Crypto-

gamic Botany, Cambridge, Mass., U.S.A. (FH).

Abbreviations of these herbaria are according to Lanjouw & Stafleu (1964).

Fresh, living fruit-bodies were collected at random in different localities, as

indicated in the descriptions of different species, in Canada and South Africa.

Cultures were available for study from the collections in the Mycology Section,

Department of Agricultural Technical Services, Pretoria and the Mycology Section,

Plant Research Institute, Canada Department of Agriculture, Ottawa.

160

MeTHODS

Cultural characters were studied by growing the different fungi on 1.25 per

cent Difco malt agar plates. Agar was poured to a depth of about 4 mm into 90

mm petri dishes for this purpose and the plates were inoculated at the side. Four

to six plates were made of each isolate. The cultures were incubated at room

temperature, 72° — 76°F, in the dark for six weeks. Cultures were examined

macroscopically and microscopically at weekly intervals and details of texture,

colour and topography were recorded. Details of micromorphological characters

of structures from different parts of the mycelial mat, mounted in a mixture of

equal parts of KOH solution and phloxine solution (Talbot, 1951) or lactophenol

and examined with the aid of the high dry and oil immersion lenses, were recorded

and illustrated by means of camera lucida drawings and photomigrographs according

to methods described by Nobles (1948) and Van der Westhuizen (1958).

The production of extra-cellular oxidase enzymes by the growing mycelia were

detected either by the application of a drop of alcoholic gum guiacum solution

to the growing mycelium, (Nobles, 1958 a) or by growing the fungi on plates of

malt agar containing 0.5 per cent gallic acid and 0.5 per cent tannic acid for

seven days (Davidson, Campbell & Blaisdell, 1938). The appearance within a few

minutes of a blue colour in the gum guaiacum solution and dark coloured zones

in the malt-gallic acid and malt-tannic acid media, presented positive proof of the

production of extra-cellular oxidase enzymes by the fungus under test.

Cultures of single basidiospores were obtained by suspending freshly collected

fruit-bodies in damp chambers. Under these conditions basidiospores were shed

on sterile glass slides placed under them. The spores were then suspended in

sterile water. Small quantities of this suspension, (about | ml) were poured

over malt agar plates which were incubated for about 24 hours. Afterwards,

single germinating spores were picked off under a dissecting microscope by means of

a sterile inoculating needle and transferred to malt agar slants. The absence

of clamp connections on the mycelia which developed from these spores, was

regarded as an indication that they originated from single spores.

The type of interfertility of individual species was determined by mating pairs

of inocula from single basidiospore cultures on malt agar slants. For this purpose

16 cultures, each grown from a single basidiospore obtained from one carpophore,

were used. Inocula from these single spore cultures were placed in pairs, about

15 mm apart on malt agar slants and incubated until the mycelia developing from

them met and mingled on the slant. Inocula from each set of 16 single basidiospore

cultures were mated in this way in all possible combinations. After incubation

the resultant mycelia on these slants were examined microscopically for the

presence of clamp connections which would indicate compatible mating types in the

parent single basidiospore mycelia. The results were plotted in a pairing table,

as illustrated by Macrae (1967) among others, but are presented here in the

abbreviated form used by Yen (1950), Nobles, Macrae & Tomlin (1957) and Nobles

& Frew (1962).

Pairings between different isolates of species for which single basidiospore

cultures were available, were made by mating three or four single spore mycelia

from each collection in all possible combinations in the way described above.

The formation of clamp connections on the resultant mycelia was regarded as

positive proof of the conspecificity of the collections from which the single spore

cultures were obtained.

When single spore cultures were not available for all collections examined for

each species, attempts were made to determine conspecificity of the collections by

161

dikaryotizing a large monokaryotic mycelium with a small dikaryotic mycelium

placed at the periphery of the growing monokaryotic colony, according to the

“Buller phenomenon” as described by Buller (1931), Terra (1953), Boidin & Des

Pomeys (1961) and Nobles & Frew (1962). The appearance of clamp connections

on the peripheral hyphae of the monokaryotic mycelium within four to seven days

after contact between the growing mycelia, was regarded as positive proof of genetic

compatibility of the two mycelia and hence conspecificity of the two collections

involved.

To study the anatomy and hyphal morphology of the carpophores, thick radial-

longitudinal sections were cut from them and examined as described by Teixeira

(1956). Small portions about 5 x 2 x 1 mm were removed from different parts

of these sections at the margin, the upper surface, upper and lower context, tubes

and stipes when present, and soaked in a mixture of equal parts of KOH and

phloxine. With the aid of sharpened sewing needles, these portions were then

gently teased out and dissected by transmitted light under 25 x magnification

of a dissecting microscope to obtain undamaged hyphae and other micro-structures

for examination. Excess material was removed from the slide and the dissected

parts were covered with a coverslip and examined in the KOH-phloxine mixture

with the aid of the oil immersion lens. Both fresh and dried fruit-bodies were

examined in this way.

Radial-longitudinal sections of fruit-bodies were also cut with the aid of

a freezing microtome at a thickness of 15u and mounted in lactophenol or the

KOH-phloxine mixture in order to study the relationships and orientation of hyphae

in the carpophores.

The hyphal characters of both cultures and carpophores were described in

accordance with Nobles’ (1948, 1958 b, 1965) terminology, but terms proposed

by Corner (1932 a), Teixeira (1962 b) and Donk (1964) were also used.

All colours in quotation marks are according to Ridgway (1912).

The description of each species given below, was compiled from examination

of all the specimens cited for that species. ‘Collections of which the cultures were

also examined, are indicated by an asterisk (*) placed before the herbarium

number. All the sets of drawings were similarly compiled from drawings of

structures from the different collections.

4. KEY TO GROUPS OF SPECIES STUDIED

In the following descriptions of species studied, the species are placed in

groups based mainly on their cultural characters, as proposed and numbered by

Nobles (1958 b). A key to these groups is given below:

1. Extra-cellular oxidase reaction in culture negative 2

1. Extra-cellular oxidase reaction in culture positive 5

2. Only thin-walled, nodose-septate hyphae formed in culture 3

2. Thin-walled, nodose-septate hyphae and fibre hyphae formed in culture 4

2. Thin-walled, nodose-septate hyphae, fibre hyphae and nodose-septate

hyphae with irregularly thickened walls formed in cultures ... ... Group 25

3. Basidiospores subglobose to ovoid Be cal Group il

3. Basidiospores allantoid z f Group 9)

4. Mycelial mat in culture white .... Group 18

4. Mycelial mat in culture brownish Group 13

162

5. Mycelial mat consisting of thin-walled, nodose-septate hyphae only Group 32

5. Mycelial mat cS of thin-walled, nodose-septate hyphae and

fibre hyphae .. eh tah he crac R ait Me. Meee ened ase .Group 45

5. Other special structures | present | in ‘ihe mat "besides thine walled!

nodose-septate hyphae and fibre hyphae re

6. Cuticular cells also present .... Ae gee .. Group 51

6. Hyphae with irregular or interlocking projections present as welll . ae Group 53

Se DESCRIPMONSTORSSEE CIES

5.1 Group 7

Cultures of species in this group form white mycelial mats which do not

produce extra-cellular oxidase enzymes; their thin-walled hyphae have simple

clamps at the septa, do not form thick-walled, aseptate fibre hyphae and rarely

form chlamydospores. Their basidiospores are globose, ovoid or ellipsoidal.

Their interfertility is of the bipolar type with multiple allelomorphs for heterothallism

at one locus only.

Fic. 1— Polyporus adustus. (a) Carpophores of DOAM 53500; (b) culture of PRE 42039

at six weeks.

Polyporus adustus Willd. ex. Fr., Syst. Myc. 1, 363, 1821;

Boletus adustus Willd., Fl. Berol., p. 392, 1787:

Bjerkandera adusta (Willd. ex Fr.) Karst., Medd. Soc. Fauna Fl. Fenn. 5,

38, 1879:

Leptoporus adustus (Willd. ex Fr.) Quél., Ench. Fung., p. 177, 1886;

Gloeoporus adustus (Willd. ex Fr.) Pil., Atl Champ. Eur., III, (1), 157, 1936.

163

Cultural characters

Growth is fairly rapid, the plates being covered in two to three weeks. The

advancing zone is even, with hyphae raised and extending to the limit of growth.

Mat at first white, loose, thick, cottony to woolly over large areas but soon forming

a zone of pale “‘cream-buff’ colour about half-way across the plate where the

mycelium appears more compact and tends to agglutinate in strands forming a

vague network over the surface of the agar. Mat remains white, raised, but after

four weeks collapsing in the pockets of the network and becoming more appressed

felty, assuming a dirty white colour. The network of agglutinated hyphae increases

slowly in extent and darkens in colour to “‘pale ochraceous buff,’ while irregular,

rounded, lumps of compact, felty mycelium ranging in colour from “‘tilleul buff”

to “avellaneous” or “‘Dresden brown” may form along the sides of the dish,

becoming poroid later. At six weeks the mat may be appressed, thin, felty,

pellicular for some distance around the inoculum, raised in the darker areas, but

collapsed elsewhere with small patches of thin, white, felty, mycelium similar to

that around the inoculum, appearing on the agar surface. Odour faint to fairly

strong, unpleasant mushroomy. Reverse is bleached. Oxidase reaction with gum

guaiac is variable being either negative or weakly positive to strongly positive in one

case. On gallic acid and tannic «cid media, no growth o1 a trace of growth may

take place but no diffusion zones are produced. Reactions observed from different

isolates are tabulated in Taste I.

Advancing mycelium: hyphae hyaline, thin-walled, branching at or near the septa,

nodose-septate with deeply staining contents and often with short, repeatedly

branched lateral branches submerged in the agar, 2.5 — 4.0u in diameter, (Fig. 2a).

Aerial mycelium: (a) hyphae hyaline, thin-walled, nodose-septate, branching at the

septa as in the advancing zone, 2.0 — 4.5 u (Fig. 2 a); (b) nodose-septate hyphae

as in (a) but with walls somewhat thickened and tending to be refractive or

occasionally solid, 3.0 — 5.2u in diameter. In coloured areas similar hyphae are

numerous but with the contents dark-coloured, ofte1 with simple septa present as

well or, in parts, without contents (Fig. 2b); (c) narrow, hyaline hyphae, repeatedly

branched and tapering to slender tips 0.4 — 0.6u in diameter arise from thin-

or thick-walled nodose-septate hyphae and become tightly interwoven to form

the tough, pellicular patches on the agar in some isolates (Fig. 2c); (d) oidia hyaline,

elongate, ovoid, cylindrical or irregular, thin-walled 3.0 — 9.0u x 2.0 — 3.0u

very numerous in some isolates (Fig. 2d); (e) chlamydospores intercalary and

terminal, ellipsoid, ovoid, thick-walled, with staining contents 7.0 — 13.0 x 4.5 —

9.0u wide, found in two isolates (Fig. 2e).

Fructifications: composed of nodose-septate hyphae with slightly thickened walls

as in (b), tightly interwoven and with dark coloured contents, 2.5 — 3.5u in

diameter; basidia arising on these hyphae, small hyaline, ovoid to narrowly

clavate, with 4 slender, straight sterigmata, 11.0 — 15.0 x 4.0 — 5.5, sterigmata

2.0 — 2.5u; spores hyaline, smooth ovoid or ellipsoid, thin-walled, with a small

apiculus 3.0 — 3.6 x 2.4 — 3u (Fig. 2f).

Submerged mycelium: hyphae hyaline, nodose-septate, walls slightly thickened, often

with a number of small lateral branches arising close together (Fig. 2g).

Carpophore characters.

Carpophore annual, lignicolous, grouped or occasionally compound, sessile.

effused-reflexed to almost entirely resupinate, occasionally dimidiate: pileus

applanate, flabellate or spathulate often imbricate and laterally connate, soft, some-

what spongy when fresh drying to hard, rigid, 1.5 x 3 — 10 x 0.1 — 0.8 cm; surface

Fic.

FICURE2:

2.— Polyporus adustus. a - f. Hyphae and structures from cultures: (a) thin-

walled, nodose-septate hyphae from advancing zone; (b) thick-walled, nodose-

septate hyphae; (c) narrow, branching hyphae; (d) oidia; (e) chlamydospores;

(f) basidia and spores; (g) submerged hypha with lateral branches.

h-p. Hyphae and structures from carpophores: (h) thin-walled, nodose-septate

hypha with dark-coloured contents; (k) basidia; (m) basidiospores; (n) thick-

walled, nodose-septate hyphae from context; (p) nodose-septate context hyphae

with lateral, branched, binding processes.

165

velutinate to somewhat villose, occasionally glabrescent, smooth or concentrically

sulcate, pale creamy white when fresh often with a faint blueish tint, drying to

smoky grey, cinereous or pale tan; margin acute, thin, entire, occasionally lobate.

concolorous or slightly darker and drying to black, sterile below for 1 — 2 mm;

pore surface grey to greyish black or darker where bruised. drying to dark brownish

grey; pores angular, minute 4 — 7 per mm; dissepiments even or somewhat

irpicoid, tubes concolorous, in one layer, up to 2 mm deep; context white or nearly

so, often ochraceous 1 — 6 mm thick, azonate or occasionally with a dark zone

and with a brown or blackish line above the tubes and extending into the

dissepiments.

Hyphal characters: hyphae nodose-septate, more or less straight or flexuous to

tortuous, branching at or near the septa, the branches more or less straight or

flexuous to very tortuous, walls hyaline or faintly coloured, thin, or variously

thickened with contents hyaline or dark brown, 1.8 — 5.0u diameter (Fig. 2h).

Hymenium: basidia short, broad clavate to narrow ovoid 11.0 — 15.0 x 4.0 — 5.5u

with 4 slender, straight sterigmata 2.0 — 2.5u long and borne in clusters on the

short, thin-walled, terminal and sub-terminal cells of the tortuous hyphae of the

dissepiments (Fig. 2k); basidiospores ellipsoidal to ovoid, hyaline, smooth, thin-

walled, with a small apiculus, 3.0 — 3.6 x 2.4 — 3.0u (Fig. 2m).

Construction. At the margin, the pileus consists of branching, thin-walled, nodose-

septate hyphae intertwined and extending to the extreme edge. Contents of some

hyphae are dark-brown imparting the dark edge to some pilei (Fig. 2h). The upper

context consists of hyaline, nodose-septate hyphae with walls only slightly thicker

than in the margin or with walls much thickened, clamp connections conspicuous,

branching at or near the septa 2.8 — 4.75u in diameter (Fig. 2n). These adhere

together in strands running parallel to or inclined upward from the direction of

growth of the pileus. Individual hyphae bend away from these strands to join

others so that an open lattice-like structure is formed. Hyphae from the context

project upward to a common level forming the upper surface, where they are

agglutinated by a faintly brownish, laquer-like substance into small tufts or, the

ends free, hyaline, with dark contents or empty, thin-walled, 2.0 — 4.7u in

diameter. The lower context consists of nodose-septate hyphae, as in upper

context, with slightly thicker, occasionally almost solid. hyaline walls and dark

brown contents, turning downward toward the trama and dissepiments, intertwined

and tortuous and narrowing to 1.8 — 3.0u and walls somewhat thinner. From

these, lateral branches arise with numerous septa and conspicuous clamp con-

nections, very tortuous, profusely branched, narrow 1.5 — 3.0u, tightly binding

other hyphae and branches into a smooth, dense tissue (Fig. 2p). The edges of

the dissepiments consist of slightly interwoven parallel. hyaline, nodose-septate

hyphae 1.8 — 3.0u in diameter.

Decay and hosts

Polyporus adustus causes a white rot of various species of hard woods (Cart-

wright & Findlay, 1946; Nobles, 1948).

Specimens examined

* Indicates culture studied as well.

Herb. DAOM: 9209. on Fagus grandifolia, Chelsea, Que., Oct. 1937; 10157, on Thuja

occidentalis, Oakville, Ont.. Jan. 1941; *17571, on Populus tremuloides, Brule, Alta., Oct.

1947: *17575, on Ostrya virginiana stump, Chelsea, Que., Oct. 1947; 17889, on Populus

166

tremuloides, Candle Lake, Sask., Oct. 1959; *22576, on Populus sp., Sheep Island, Lake

Opinicon, Ont.. Sept. 1950; *53500, on hardwood, Pack Forest, Warrensburg, N.Y. Oct. 1959.

Herb. PRE: 15607, on Podocarpus elongatus, Harold Forest, Natal, Sept. 1915; 24859,

Kirstenbosch, C.P., June 1929; 30383, on Fagus sp., U.S.D.A., Washington, F. P. 52045;

31420, Stellenbosch, C.P., July 1919; 31462, Stellenbosch, C.P., May 1921; 35949, Oxshott,

England, Oct. 1946; *42039, on Populus sp. log, Stellenbosch, Aug. 1959; *42328, on hardwood

log, Dorset, Ont., Sept. 1962 (DAOM 94016); *42329, on hardwood log, Dorset, Ont., Sept.

1962 (DAOM 94010); *42332, on hardwood log, Dorset, Ont., Sept. 1962 (DAOM 73987);

*42350, on dead deciduous tree, Dorset, Ont., Sept. 1962 (DAOM 94014); *42365 on Acer

sp. log, Packenham, Ont., Aug. 1962 (DAOM 94007).

Interfertility studies

When tested for the production of extra-cellular oxidase in culture, the isolate

from South Africa showed a strong positive reaction while some of the cultures

of Canadian origin showed no reaction. Nobles (1958 a) reported a similar

variable reaction for Polyporous fumosus which Overholts (1953) described as

closely similar to and at times indistinguishable from Polyporus adustus. Nobles

(1958 b) further reported Polyporus fumosus as having the tetrapolar type of

interfertility while Polyporus adustus has the bipolar type of interfertility. It was

therefore necessary to determine whether the South African isolate was interfertile

with other isolates and therefore conspecific with Canadian collections of Polyporus

adustus. For this purpose four cultures obtained from single spores from the

South African isolate PRE 42039, were paired with similar cultures isolated from

Canadian collections of Polyporus adustus. These single spore cultures were

paired in all possible combinations on malt agar slopes. The results are presented

in TaBLE 2 according to the method of Nobles, Macrae & Tomlin (1957), Macrae

(1967) and others.

From the results, it is clear that all the Canadian isolates of Polyporus adustus

used, were interfertile. The South African isolate only had a low degree of

interfertility with the Canadian isolates of Polyporus adustus. In three out of

a total of 48 pairings numerous clamps were formed. This South African collection

differed from the Canadian collection only in respect of the more intense reaction

of its culture when tested for extra-cellular oxidase, but was closely similar in all

other respects. It can thus be concluded that this South African isolate is

conspecific with the Canadian isolates of Polyporus adustus but belongs to a

different geographical race. The existence of geographical races had been reported

by Mounce & Macrae (1938) for Fomes pinicola, a species with bipolar interfertility.

Discussion

All hyphae formed in cultures of P. adustus are nodose-septate with hyaline

walls. Hyphae from various parts of the culture differ only in the thickness of the

wall, overall diameter and manner of branching while differentiated hyphae are

lacking. Differences in cultural characters between different isolates exist only

in the absence or presence of accessory spcres and the intensity of the reaction

for extra-cellular oxidase when tested with gum guaiac solution.

The narrow, repeatedly branched hyphae in the pellicular patches of the mat

of some of the isolates have not been reported before. Chlamydospores were

reported from cultures by Cartwright (1931) and oidia by Nobles (1948). Chlamy-

dospores were present in some of the cultures only but oidia were present in all

cultures although their numbers varied considerably in the different isolates.

This description of the cultural characters agrees fairly closely with that of

Nobles (1948) and earlier descriptions by Bose (1930), Cartwright (1931), Cartwright

& Findlay (1946) and Davidson, Campbell & Blaisdell (1938).

167

Irregularities in the reaction of this fungus when tested for extra-cellular

oxidase in culture, have been reported by Davidson et al. (1938) and Nobles

(1958 a) who found that the reaction may vary from negative to weakly-positive

on gallic acid and tannic acid media. Davidson ef al. (1938) stated that ‘‘these

fungi may require from 7 to 14 days to form brown diffusion zones but the reactions

are always positive.” Nobles (1958 a) recorded a strong positive reaction with

her gum guaiac test. The results given in Table 1, however, show that no positive

reactions were obtained on gallic and tannic acid media even after 14 days and

that results with the gum guaiac test also varied from negative to positive. Yet all

workers reported that Polyporus adustus causes a white rot. Lyr (1955) reported

that Polyporus adustus were among the 13 species out of 103 wood-rotting fungi

tested by him which formed peroxidase in small amounts. Most of those species

which formed peroxidase were also able to utilize lignin by oxidation and cause

white rots. He thought that the peroxidase could act like laccase, the extra-cellular

oxidase enzyme present in most white-rot, lignin-destroying Hymenomycetes, in

this respect. It thus seems probable that the weak and erratic reactions observed

when Polyporus adustus was tested for extra-cellular oxidase may be due to small

and varying amounts of peroxidase enzyme formed by different isolates of this

species. Differences in the intensity of the reaction wher different isolates of 173

species of wood-rotting Hymenomycetes were tested for oxidase enzymes by the

application of drops of various compounds, were reported by Kaarik (1965).

The above description of the fruit-body of Polyporus adustus agrees with

that of Overholts (1953) who described the hyphae in the carpophore as “‘hyaline,

considerably branched with cross-walls and clamps, 3 — 6u in diameter.’ Teston

(1953 b) and Cunningham (1948 b) reported that the carpophores are monomitic

and consist of thin-walled, generative hyphae with numerous clamp connections.

Pinto-Lopes (1952) and Farinha (1964) stated that the secondary hyphae are

hyaline and nodose-septate, while the tertiary hyphae are nodose-septate and with

walls never much thickened. Donk (1933), reported that fruit-bodies of this

fungus consist of thin-walled, nodose-septate hyphae, more or less parallel in

arrangement and forming a loose tissue in the upper part of the trama but closely

packed and parallel in the lower part from where hyphae turn downward into the

trama of the tubes. Bourdot & Galzin (1928) similarly reported thin-walled

nodose-septate hyphae loosely interwoven in the upper context and very compact

and closely packed and parallel in arrangement in the lower context. Ames (1913)

also reported a difference in the consistency between the upper and lower layers

of the context of this fungus but did not describe the morphology of the hyphae.

All these descriptions thus agree that the fruit-bodies consist of nodose-septate

hyhpae only and have a monomitic hyphal system while most reports by earlier

workers agree that the hyphae are arranged more or less in parallel. These

descriptions thus imply that the kyphae are all similar in appearance and that

the fruit-bodies are simple in construction. From the description furnished above,

and illustrations of hyphae, (Fig. 2h, n, p) it is evident that marked morphological

differences exist between hyphae from the upper and lower context. Hyphae from

the upper context are straight and sparingly branched with the septa far apart

while hyphae from the lower context are tortuous, frequently branched and have

septa fairly close together, and often lie across the direction of growth. Hyphae

similar to those of the upper context are also present in the trama. By their

tortuous branching in the lower context, all hyphae are tightly bound into the

dense, tough tissue so characteristic of the lower context and trama of Polyporus

adustus. Differences in the morphology of these hyphae are thus related to

differences in the consistency and texture between the upper and lower parts of

the carpophore on the one hand and function of the hyphae on the other hand.

168

Many of the lateral branches of hyphae binding the tissues in the lower context,

resemble the binding hyphae described by Corner (1953) in the tissues of Polyporus

sulphureus, in that they are lateral processes which differ from the parent hyphae

solely in the method and extent of branching and in function. They do not

resemble the binding hyphae described by Cunningham (1946) which are

differentiated, branching, thick-walled, aseptate structures. It is thus clear that

there is a diversification of function among the thin-walled, nodose-septate hyphae

that make up the fruit-body of Polyporus adustus. The fruit-bodies are thus more

complex in construction than was evident from previous descriptions. This complex

construction may prove to be valuable in taxonomic considerations of this and

other monomitic species.

A comparison of the structures formed in culture with those found in the

carpophore, shows that in both, all hyphae are nodose-septate, although differences

in the thickness of the walls are evident. The thick-walled hyphae found in the

carpophore are however of the same type as those of the culture. The narrow,

branching hyphae found in the felty, pellicular patches of the mat in culture

appear to be homologous to the much branched, tortuous hyphae found in the

lower context and trama of the carpophore. Basidia and basidiospores formed

in culture are similar in all respects to those of the carpophores. With the

exception of chlamydospores and oidia, structures formed in cultures of Polyporus

adustus are thus present in the carpophores as well.

Polyporus adustus is the type species of the genus Bjerkandera Karst. (Cooke.

1959; Donk, 1960), but has also been referred to other genera by various workers.

It differs from Polyporus squamosus Huds. ex Fr. and Polyporus tuberaster Jacq.

ex Fr. the lectotypes of the genus Polyporus Mich. ex Fr. according to Murrill

(1907 a) and Donk (1960), respectively. Carpophores of Polyporus adustus do

not possess the characteristic thick-walled hyphae with tapering ends which are

present in the fruit-bodies of these two species (Corner, 1953; Overholts 1953).

Cultures of Polyporus adustus lack the dark, skin-like areas consisting of thick-

walled hyphae with interlocking projections which characterize cultures of Polyporus

squamosus and Polyporus tuberaster (Nobles 1948, 1958 b). Quélet (1886) placed

Polyporus adustus in his genus Leptoporus and was followed in this by Bourdot

& Galzin (1928) but there is uncertainty about the type species of this genus.

Polyporus mollis Pers. ex Fr. the lectotype selected by Donk (1960), differs from

Polyporus adustus by having the carpophore composed of “thick-walled, hyaline,

sparingly branched hyphae with a few inconspicuous cross walls but no clamp

connections” (Overholts, 1953). The identity of Polyporus epileucus Fr., the type

species selected by Murrill (1903), is confused with that of Polyporus spumeus Fr.

(Bourdot & Galzin, 1928). Murrill (1903), regarded the genus Leptoporus (Quél.),

as synonymous with Bjerkandera Karst. and was followed in this by Donk (1933),

Bondartzev & Singer (1941), and Bondartzev (1953), but a study of descriptions

of hyphae by Overholts (1953), Teston (1953 b), and Pinto-Lopes (1952), of species

placed in the genus Leptoporus by Bourdot & Galzin (1928), revealed many

differences in hyphal characters between these species. It thus seems advisable

to retain Polyporus adustus in the genus Bjerkandezra Karst. of which it is the type

species. The problems concerning its relationship with species of the genus

Leptoporus Quél. can be solved only after the uncertainty about the type species

of that genus had been clarified and its hyphal characters had been carefully

studied.

Polyporus adustus cannot be regarded as congeneric with Polyporus conchoides

Mont., the type of the genus Glozoporus Mont. because it lacks the hymenium

which is continuous over the edges of the pores, a characteristic of the genus

Gloeoporus (Overholts, 1953; Hansen, 1956; Donk, 1960)

169

5.2 Group 9

The mycelial mats of cultures of species in this group remain white and do

not produce extra-cellular oxidase enzymes; the’r thin-willed hyphae have simple

clamps at the septa, do not form thick-walled, aseptate fibre hyphae and rarely

form chiamydospores. Their basidiospores are cylindrical or allantoid. Their

interfertility is of the bipolar type.

Fic. 3.— Polyporus dichrous. (a) Carpophores of PRE 42093 (top) and PRE 42384 (bot-

tom); (b) culture of PRE 42384 at 6 weeks; (c) radial-longitudinal section through

carpophore showing nodose-septate hyphae and hyphal strands of context, x S00 phase

contrast; (d) delicate, thin-walled, nodose-septate hyphae of tramal layer, x 1000 phase

contrast.

Polypsrus dichrous Fr., Syst. Myc 1, 364, 1821;

Leptoporus dichrous (Fr.) Quél., Fl. Myc., p388, 188:

Gloeoporus dichrous (Fr.) Bres., Ann. Mycol. 14, 230, 1916.

Cultural characters

The mycelium grows fast to moderately fast covering the plate in two to three

weeks. Advancing zone even or slightly bayed, appressed or submerged and difficult

to see. Mat hyaline to whitish, mostly submerged, with small, scattered areas of

170

finely farinaceous to short floccose mycelium developing after four weeks. Reverse

remaining unchanged or developing a characteristic greenish-yellow colour after

two to four weeks. Odour strong, unpleasant after two or three weeks then

diminishing somewhat.

No diffusion zones formed on gallic acid and tannic acid media but trace

of growth on the latter after seven days. Oxidase reaction with gum guaiac

solution, negative.

Advancing mycelium: hyphae hyaline, branching, nodcse-septate, thin-walled,

2.0 — 4.5u in diameter (Fig. 4a).

Aerial and submerged mycelium: (a) hyphae as in the advancing zone; (b) narrow,

hyaline, unbranched, thin-walled hyphae with conspicuous clamp connections

and 1.5u in diameter (Fig. 4b); (c) nodose-septate hyphae occasionally with irregular

swellings up to 7.0u in diameter, between the septa (Fig. 4c).

Carpophore characters

Carpophore annual, lignicolous, sessile, effused-reflexed or often imbricate;

pileus 1.5 — 7.0 x 3.0 — 10.0 x 0.1 — 0.5 cm mostly pure white to faintly yellowish

and coriaceous when fresh, drying creamy to ochraceous, rigid and brittle; surface

velvety-villose to glabrous, azonate, mat when dry; margin entire, acute, concolorous,

white, sterile below; pore surface waxy, flesh-coloured to reddish-purple, poroid;

pores rounded, or somewhat angular, entire, thin-walled, 5-8/mm; tubes shallow

less than 1 mm and hymenial surface separable from pileus as a thin elastic layer;

context white, soft fibrous, thin, 1-4 mm.

Hyphal characters: hyphae branching, hyaline, nodose-septate, thin-walled or thick-

walled 1.5—8.0u in diameter (Fig. 4d. e).

Hymenium: basidia cylindrical to narrowly clavate 12.0 — 15.0 x 2.0 — 4.0u with

four short, straight sterigmata 1.5 — 2.0u, the basidia packed into a tight palisade

continuous over the dissepiments (Fig. 4f); basidiospores allantoid, hyaline, smooth,

thin-walled 3.0 — 4.0 x 0.7 — 1.0u (Fig. 4 g).

Construction. At the margin of the fruit-body the hyphae are narrow, 2.2 —5.0u

in diameter, thin-walled and slightly interwoven parallel to the direction of growth

of the pileus. Behind the margin in the upper context the hyphae become thick-

walled (sclerified generative hyphae; Donk, 1964), and the branches more divergent

and up to 8.0u in diameter. Here the hyphae tend to form intertwined strands

from which individual branches diverge to join adjacent strands thus forming a

loose, lattice-like structure (Fig. 3c). Towards the upper surface the strands

disappear and the hyphae are divergent, loosely interwoven, their ends thin-walled,

free or agglutinated into irregular tufts which form the upper surface. Below

this tissue, the loosely arranged, thick-walled hyphae pass rather abruptly into a

dense layer, 40 — 80u thick, of branching, thin-walled. tortuous, nodose-septate

hyphae tightly interwoven into a pseudo-parenchymatous tissue. From this layer

narrow hyphae 1.5 — 3.0u in diameter with very thin, delicate walls turn downwards,

branch occasionally and run parallel to or lightly intertwined with one another

towards the hymenium where they branch profusely to form a narrow, sub-hymenial

layer on which the basidia are borne. On the hyphal walls masses of amorphous

granules of gelatinous material are deposited so forming the characteristic gelatinous

tramal layer 50 — 300u thick in fruit-bodies of this species (Fig. 3 d).

The small cylindrical basidia are borne on the pseudo-parenchymatous

sub-hymenial layer 5.0 — 15.0u thick formed by the numerous terminal branches of

the thin-walled hyphae of the tramal layer. The hymenium is continuous over

the edges of the pores.

171

) os

FIGURE 4. _ tou, 96) 6000

Fic. 4.—Polyporus dichrous. a - c. Hyphae from cultures: (a) hypha from

advancing zone; (b) narrow, hyaline, unbranched, thin-walled hypha with

conspicuous clamp connections; (c) hypha with irregular swellings.

d-g. Hyphae and structures from carpophores: (d) thin-walled, nodose-septate

hyphae; (e) thick-walled nodose-septate hyphae; (f) basidia; (g) basidiospores.

Decay and hosts

Polyporus dichrous causes a white rot of angiosperm wood.

Specimens examined

Herb. DAOM: *8118, on P. contorta var. latifolia, Lumby B.C.; *11609; *22281.

Herb, PRE: 15617, Gingindhlovu, Natal, June 1915; 22285, Mycotheca Boreali Africana No.

349, R. Maire; 23479, Mont-aux-Sources, Natal, July 1928; 27739, on dead wood, Donnybrook,

Natal, Jan. 1935; 28924, dead wood, Donnybrook, Natal, Febr. 1935; 30511, on dead wood,

Potchefstroom, Transvaal, March 1939; 31602, on dry wood of Rhus viminalis, Aug. 1915;

31814, on dead wood, Nottingham Road, Natal Aug. 1917; 40205, Town Bush, Pietermaritz-

burg, June 1948; 41486, Ex Herb. Wm. Bridge Cooke No. 30137; 42093, on dead log,

Potgietersrust Dist., March 1960; *42384, on decaying log, Gatineau Park, Que., Sept. 1961;

*42436, on decaying log, Barberton, Tvl., May 1960.

Interfertility studies.

Eighteen cultures, each grown from a single basidiospore obtained from a

fresh carpophore of PRE 42384, were paired in all possible combinations on malt

agar tubes to determine the type of interfertility. It was found that Polyporus

dichrous has the tetrapolar type of interfertility with allelomorphs for heterothallism

at two loci. The results, showing the distribution of mating types among the single

spore mycelia, are presented in TABLE 3.

To test the conspecificity of collections of which cultures were available,

by means of the ‘Buller Phenomenon,” two mycelia from single spores of

Opposite mating types of PRE 42384-8 and PRE 42384-10 were used. The

method described by Nobles & Frew (1962) in their studies of species of the

genus Pycnoporus Karst., was used. Seven days after placing the dikaryotic

mycelium on plates on which the haploid mycelia were growing, the latter were

examined for the presence of clamp connections at the periphery. Results were

negative. After five days more, the plates were again examined. No clamp

connections had formed on any of the haploid mycelia, which had been inoculated

with small dikaryotic mycelia from ail the cultures studied of Polyporus dichrous.

It thus appears that the “Buller phenomenon” cannot be effectively used for

confirming the identity of cultures of Polyporus dichrous.

Discussion

This description of the cultural characters of Polyporus dichrous agrees closely

with those of Davidson et al. (1938) and Nobles (1948, 1965). The featureless

mycelial mat, the greenish-yellow colour imparted to the agar, the usual lack of

extra-cellular oxidase and the thin-walled, nodose-septate hyphae serve to dis-

tinguish this species in culture.

Cultures of Polyporus dichrous do not give positive reactions when tested

for extra-cellular oxidase on gallic and tannic acid media but may give a weak

positive reaction with gum guaiac solution (Nobles, 1965). Davidson et al (1938), and

Nobles (1948) reported that this species causes a white rot. Overholts (1953)

stated that the decay associated with it is “‘usually white, but careful dissection

often shows a small amount of definitely brown rot in the vicinity of the sporo-

phores”. He thought that the fungus probably causes a brown, carbonizing rot,

ecause of the negative reaction for extra-cellular oxidase. Petersen (1961),

reported that this fungus causes a brown rot of deciduous fruit trees in the U.S.A.

Kaarik (1965) found no reaction when two isolates of Polyporus dichrous were

tested for oxidative ability of 20 different phenolic compounds. Kirk & Kelman

173

(1965), found that although cultures of Polyporus dichrous gave negative reactions

on 9 different phenolic compounds, the fungus caused a white rot of sweet gum

test blocks. Extracts made of these blocks contained a phenol-oxidase active

against catechol and guaiacol but not 1-napthol. Extracts of cultures of this fungus

contained a weak catechol oxidizing agent inactive against guaiacol or |-napthol.

These authors concluded that ‘inability of certain wood decay fungi to oxidize

phenols in agar cannot be assumed to indicate inability to utilize lignin’. These

observations could explain the erratic results of tests for extra-cellular oxidase

by Polyporus dichrous and may indicate the presence of oxidation enzymes

different from those of many other species of polypores which cause white rot.

The hyphae formed in the cultures were always thin-walled and nodose-septate

and resembled those in the margin and gelatinous tramal layer of the fruit-body.

Thick-walled, nodose-septate hyphae like those in the context of the fruit-bodies,

were seen once in an old culture. These thick-walled, nodose-septate hyphae of

the carpophores, are “‘sclerified generative hyphae’? (Donk, 1964) and are formed

by internal thickening of the walls of thin-walled, nodose-septate hyphae. The

hyphae formed in cultures of Polyporus dichrous are thus present in the fruit-bodies

as well.

The gelatinous nature of the trama of the pores of Polyporus dichrous had

been mentioned by many earlier workers (Van der Bijl, 1922 a; Bourdot & Galzin,

1928; Overholts, 1953). The construction of this layer was first described by

Hansen (1956) as consisting of hyphae with strongly gelatinized walls. From the

present writer’s observations it is clear that the gelatinous tramal layer consists

of narrow, branched, nodose-septate hyphae with very thin walls which are placed

more or less vertically in a loose palisade-like tissue of which the interstitial spaces

are filled by a hyaline, amorphous substance (Fig. 3d). In freshly mounted, radial-

longitudinal sections of the fruit-body in KOH-phloxine, this substance could be

seen as an unstained mass extruded from the section, in the otherwise bright pink

colour of the mounting medium. In thin sections mounted in sterile distilled

water, this substance could also be seen as a slightly darker mass, oozing out of the

tissues, when observed by means of a phase contrast microscope. The thin-walled,

tramal hyphae are clearly visible in extremely thin sections of about Su or less in

thickness. This construction would explain Van der Bijl’s (1922 a) observation

that the tubes are ‘“‘separable from the context as a thin elastic layer when

moistened.”

The hyphae of the fruit-body are all nodose-septate so that this fungus has

a monomitic hyphal system as reported by Cunningham (1948 b) and Hansen

(1956), but from the above description of the fruit-body, which agrees in most

respects with that of Hansen (1956) it is evident that the thick-walled, nodose-

septate hyphae of the context support the layer of thin-walled, interwoven hyphae

from which the thin-walled, tramal hyphae and hymenium are suspended. These

differences in hyphal morphology which is associated with differences of function,

result in a much more complex construction of the fruit-body of Polyporus dichrous

than that of fruit-bodies of species of Thelephoraceae which also possess monomitic

hyphal systems (Talbot, 1951, 1954 b, 1958 b). Similar differences in morphology

and function of hyphae were also found in the carpophores of some other species

of poroid Hymenomycetes with monomitic hyphal systems such as Polyporus adus-

tus (see previous section) and Cerrena unicolor (Van der Westhuizen, 1963).

The gelatinous tramal layer and hymenium which is fertile over the edges of

the pores are important morphological characters of the carpophores of Polyporus

dichrous. These characters are not found in any of the other species in Group 9

(Nobles, 1958 b) and are absent from most species of poroid Hymenomycetes. These

174

characters are found in species of the genera Merulius Hall. ex Fr., and Gloeoporus

Montagne. The fruit-body of Merulius tremellosus Schrad. ex Fr., the type of

the genus Merulius, also has a gelatinous layer under the hymenium and appears

to resemble the fruit-body of Polyporus dichrous in construction (Burt, 1917).

Cultures of Merulius tremellosus, however, were placed by Nobles in her Group 54

(1958 b) together with other species characterized by the presence of simple-septate

hyphae in the advancing zone and nodose-septate hyphae elsewhere, a positive

reaction for extra-cellular oxidase and the bipolar type of inter-fertility. Another

species, Merulius ambiguus Berk. which also has a gelatinous layer under the

hymenium, (Burt, 1917), differs in cultural characters from Merulius tremellosus

by forming mycelial mats consisting of thin-walled, simple-septate hyphae,

occasionally with multiple clamp connections and a negative reaction when tested

for extra-cellular oxidase. Merulius ambiguus was placed by Nobles (1958 b)

in Group 28, together with other species with similar cultural characters. Some

other species, viz., Merulius lacrimans Wulf. ex Fr., the type of the genus Serpula

Pers. ex S. F. Gray, (Cooke, 1959) Poria incrassata Berk. & Curt., the type of

Meruliporia Murrill (Cooke, 1959), and Poria taxicola (Pers.) Bres. (Hansen, 1956)

all have hymenia which are continuous over the edges of the dissepiments but they

lack the gelatinous tramal layer in their fruit-bodies. (Burt, 1917).

No descriptions of the cultural characters of Gloeoporus conchoides Mont.,

the type of the genus Gloeoporus Mont., are available but the carpophores of

Polyporus dichrous and Gloeoporus conchoides are so similar in morphology and

anatomy that the two species can often be distinguished from each other only by

the absence of clamps on the hyphae of the latter species according to Overholts

(1953) and Bakshi & Singh (1961). Although this may be regarded as a difference

of generic importance between the two species, some genera of Hymenomycetes

are known to include species with as well as without clamps on their septate hyphae.

Because these two species have so many other characters in common, it seems

advisable to include Polyporus dichrous in the yenus Gloeoporus Montagne.

The possession of an hymenium which ‘s fertile over the edges of the

dissepiments, excludes Polyporus dichrous as well as other species of Gloeoporus

Mont. from the Polyporaceae. Donk (1964) advanced arguments for the inclusion

of this genus together with the genus Merulius Fr. and other genera with similar

characters in the family Corticiaceae Herter. The observations recorded here

support this proposal.

5.3 Group 13

The mycelial mats of cultures of species in this group mostly develop brown,

coloured areas due to the presence of browa pigment in the hyphal walls or

contents. No extra-cellular oxidase enzymes are produced. The thin-walled

hyphae have simple clamp connections at the septa and brown, aseptate fibre hyphae

are formed in most of their cultures. Their basidiospores are brown and ovoid or

ellipsoid-cylindric in shape. Interfertility is of the bipolar type.

175

Fic. 5.—Lenzites saepiaria. (a) Carpophores of DOAM 22745; (b) culture of DAOM

22443 at six weeks; (c) nodose-septate hypha with lateral outgrowth with spiny pro-

jections in culture, x 1000 phase contrast.

Lenzites saepiaria (Wulf. ex Fr.) Fr. in Epicr. Syst. Myc. p. 407, 1838;

Daedalea saepiaria Wulf. ex Fr., Syst. Myc. 1, p. 332, 1821;

Gloeophyllum saepiarium (Wulf. ex Fr.) Karst., Finl. Hattso. II, p. 80, 1879.

Cultural characters

Growth moderately fast to slow, the mycelium covering the plate in three to

Seven weeks. The advancing zone is even; mat thin, appressed, becoming somewhat

downy-farinaceous, or downy in younger parts with thin, sub-felty or irregular

pellicular areas in the older parts of the mat, some of which may become slightly

warted; mat white at first but turning “‘pale ‘uff’ changing to “warm buff” or

“antimony yellow” or “honey yellow” to “tawny olive” or “‘snuff brown’? to

“umber brown” in patches. Occasionally mounds or ridges of raised, floccose,

woolly mycelium appear, white at first, then darkening to ‘‘antique brown’’ or

“snuff brown” and often with irregular patches of thin, collapsed mycelium between

these mounds. Reverse unchanged or bleaching slowly; odour none or faint,

178

somewhat spicy. No diffusion zones on tannic acid and gallic acid agar and slight

growth only on the latter.

Advancing mycelium: hyphae hyaline, thin-walled, nodose..e>tate, branching at

the septa or from the clamp connections, 2.2 — 5.2u in dianseter (Fig. 6 a).

Aerial mycelium: (a) hyphae as in the advancing zone 2.0 — 3.5u in diameter;

(b) narrow, hyaline, nodose-septate hyphae, repeatedly branched, thin-walled at first,

becoming thick-walled to solid, 1.0 — 2.5u diameter, and with swollen projections

or lateral outgrowths with numerous short, spiny projecticns or lateral outgrowths

(Fig. Sc; 6b); (c) nodose-septate hyphae with thickened, brown walls and narrow

lumina with staining contents, branching near the septa or from the clamp connec-

tions, 2.2 — 4.2u in diameter (Fig. 6c); (d) fibre hyphae sub-hyaline to pale straw

yellow, occasionally branched, walls thickened to sub-solid or solid, aseptate 2.2 —

3.64 in diameter (Fig. 6d); (e) oidia long cylindrical, hyaline, thin-walled 6.0 —

13.0 x 3.0 — 4.2u (Fig. 6e); (f) chlamydospores rare, cvoid to subglobose, terminal,

6.0 — 8.0 x 6.0 — 12.0u (Fig. 6f). Submerged mycelium: (a) hyphae as in the

advancing zone; (b) nodose-septate hyphae with walls thickened or solid, branched,

numerous 1.2 — 3.64 in diameter (Fig. 6b).

Carpophore characters

Carpophore lignicolous, solitary or grouped, annual or reviving; sessile, usually

dimidiate occasionally effused-reflexed; pileus applanate to somewhat convex often

laterally connate, occasionally imbricate, coriaceous when fresh, drying to rigid,

1.0 — 10.0 cm x 1.0—8 cm x 0.2—1.0 cm; upper surface hirsute-tomentose

to fibrillose-tomentose or compactly tomentose at maturity, bright yellow-reddish

brown to dark ferruginous; margin acute, thin or thick, entire, pale cream coloured

to almost orange; pore surface yellowish brown to rusty brown, usually lamellate,

eccasionally daedaloid or poroid, lamellae often dentate, 0.5 — 1.0 mm apart and

2.0 — 5.0 mm broad; context up to 3 mm thick, umber to chestnut brown, darkening

in KOH.

Hyphal characters: (1) nodose-septate hyphae hyaline or pale yellow, thin-walled,

occasionally collapsed, branching frequently near the septa or from the clamp

connections, 2.0 —3.0u in diameter (Fig. 6g); (2) nodose-septate hyphae with

thick, yellow-brown walls, and lumina narrow or solid, 2.5 —4.0u in diameter,

with occasional thin-walled branches (Fig. 6h); (3) fibre hyphae long, unbranched

or occasionally branched, walls thickened, pale straw yellow to yellow brown with

lumina narrow or occluded but widening towards the apex, aseptate or with one or

two simple septa near the tip, 2.5 —4.0u in diameter (Fig. 6k); (4) fibre hyphae

pale brown, subsolid to solid, repeatedly branched, the branches short or long,

flexuous, aseptate 1.5 — 3.0u in diameter (Fig. 6 m).

Hymenium: basidia long-clavate, hyaline 20 — 32 x 4.0 —5.0u with four slender

sterigmata 3.9 —4.8u (Fig. 6n); basidiospores cylindrical, obliquely apiculate,

hyaline, smooth, thin-walled, 7.0—10.0 x 2.0—4.0u (Fig. 6p); cystidioles

fusiform resembling immature basidia but with slightly thicker walls 22.0 — 33.0 x

3.0 —5.0u, arising on basidial hyphae and protruding 5.0—7.0u above the

basidia (Fig. 6 q).

Construction. At the margin the pileus consists of long, unbranched fibre hyphae,

aseptate, their walls partly thickened and sub-hyaline. The fibre hyphae lie

parallel to the direction of growth of the pileus, are somewhate intertwined with

their ends projecting outward to form the margin. Interwoven with the fibre

hyphae and away from the extreme margin are numerous, hyaline, thin-walled,

nodose-septate hyphae 2.2 — 3.5u in diameter branching frequently at the septa or

from the clamp connections. Behind the margin in the upper context, the fibre

FIGURE 6.

Fic. 6.— Lenzites sepiaria. a-f. Hyphae and structures from cultures: (a) nodose

septate from advancing zone; (b) nodose-septate hyphae with lateral out-

growths; (c) thick-walled, brown nodose-septate hyphae; (d) fibre hyphae:

(e) oidia; (f) chlamydospores.

g-q. Hyphae and structures from carpophores: (g) nodose-septate hyphae;

(h) thick-walled, nodose-septate hyphae; (k) fibre hyphae, unbranched; (m)

fibre hyphae with short, flexuous branches; (n) basidium; (p) basidiospores;

(q) cystidiole.

178

hyphae have thickened, pale, straw-yellow walls and narrow lumina. They are

turned slightly upward towards the upper surface, are somewhat intertwined,

2.5 — 4.0u in diameter with their tips free and projecting at a common level to

form the pubescent, pale-coloured upper surface of the growing margin. Interwoven

with these fibre hyphae are nodose-septate hyphae with thin, pale yellow walls,

often collapsed, with deeply staining contents and branching freely or forming

H-connections. In the dark-coloured part of the context the fibre hyphae are

mostly sub-solid to solid, unbranched, 2.5 — 4.0u in diameter, their walls dark

yellow brown, the hyphae somewhat intertwined and turning upwards towards

the upper surface of the pileus. At the upper surface these hyphal ends are

agglutinated into tufts by a brownish, amorphous substance and bent over towards

the margin to form the fibrillose-tomentose upper surface of the mature, dark-

coloured part of the pileus. Below this fibrillose-tomentose layer, the fibre hyphae

are agglutinated into a trichocutis by a thin layer of pale brownish substance.

Below this layer and in the older parts of the context the nodose-septate hyphae

are interwoven with the fibre hyphae, have thick, yellow-brown walls with the

lumina narrow or occluded and the hyphae sub-solid or solid, 2.5 — 4.0u in

diameter (Fig. 6 h). Also present in the older parts of the context are fibre hyphae

with pale brown, thickened walls, aseptate and branching repeatedly, the branches

1.5 — 5.3u in diameter, short or long. tortuous, tightly interwoven with the other

hyphae and binding them into a tough tissue (Fig. 6 m). Below the upper context

the fibre hyphae turn downwards towards the trama of the pores. The fibre

hyphae are similar to those of the upper context but remain somewhat narrower.

2.2 — 3.0u in diameter and are more tightly intertwined and consequently more

flexuous. Fibre hyphae with short, numerous branches with thickened walls are

very numerous, binding all hyphae into a tough tissu2. The nodose-septate hyphae

in this part of the pileus mostly have thickened, pale brown walls and branch

freely from the clamp connections or at the septa. In the trama and towards

the hymenial surface these hyphae become more numerous, thin-walled, the walls

sub-hyaline, and with deeply staining contents, branching freely and interwoven

with the other tramal hyphae, finally protruding at the hymenial surfaces where

they bear the basidia and cystidioles on short terminal branches.

Decay and hosts

Lenzites sepiaria causes a brown, carbonizing 1ot of dead coniferous wood or

occasionally of hardwood.

Specimens examined

Herb. DAOM: *17240, on Betula papyrifera, Goose Bay, Lab., July, 1947; *17246. on

Pinus mariana, Goose Bay, Lab., July. 1947; *22276, on Populus tremuloides log, Quesnel,

B.C., Aug. 1949; *22442, on Picea stump, Algonquin Park, Ont., Sept. 1950; *22443, on

Tsuga or Pinus log, Dorset, Ont., Oct. 1950; *22745, Goose Bay, Labrador, Aug., 1949;

22761, on Picea mariana log, Goose Bay, Lab., May 1950; 30059, on Picea glauca log, Riding

Mt. Nat. Park, July 1950; 31986, base of Picea sp. North Bay, Ont., Sept. 1955.

Herb. PRE: 41887 on Picea canadensis, ex Herb. J. R. Weir, Aug. 1917; 42141 on Pinus sp.

log, Mac-Mac Falls, Transvaal, July, 1961, *42381 on coniferous log, Corkery Road, Ont.,

Aug. 1962.

Discussion

This fungus appears to be quite variable in culture. Snell (1922) reported

that the fungus is recognizable in culture by its scant aerial mycelium and powdery

appearance due to abundant oidia. Robak (1942) and Cartwright & Findlay (1946)

confirmed this and reported patches of thicker, orange-brown, velvety mycelium

on which flat, antler-like, basidia-bearing processes develop. Nobles (1948)

described the farinaceous appearance and numerous oidia of the cultures but did

179

not mention fibre hyphae. Falck (1909) described two kinds of hyphae from

cultures, viz.: (1) conducting hyphae in which the cross walls disappear at the

clamp-connections and (2) hyphae with very thick walls and narrow lumina which

give mechanical strength. These latter correspond to the fibre hyphae described

above. Fibre hyphae were not present in all cultures studied and chlamydospores

were present in cultures of DAOM 22442 and PRE 42381 only. All cultures

however had abundant oidia while the thick-walled, ncdose-septate hyphae were

present in varying numbers in all the cultures. These hyphae may be identical

to Falck’s (1909) conducting hyphae. The variability of the cultural characters

of this fungus makes its identification in culture rather difficult on occasion.

In this species, most of the types of hyphae formed in culture are also found in

the carpophores. The peculiar lateral outgrowths of short branches and the

roughened processes produced on the fibre hyphae and _ thick-walled, nodose-

septate hyphae in culture, were not found in any of the carpophores.

Bourdot & Galzin (1928) described the hyphae from carpophores of Lenzites

sepiaria as brownish and thick-walled. Cunningham (1948 h), considered this species

to fall in the genus Daedalea Pers. ex Fr., as defined by him and thereby implied

the presence of thin-walled, generative hyphae with clamp connections at the septa,

and thick-walled, brown, aseptate skeletal and binding hyphae in the carpophores.

Pinto-Lopes (1952) reported hyaline, thin-walled, nodose-septate, secondary hyphae

and brownish, narrow, aseptate, subsolid or solid tertiary hyphae in the carpophores.

Overholts (1953) stated: “hyphae mostly pale chestnut, rarely branched, thick-

walled, with no crosswalls or clamps, 3 —— 5u in diameter, a few hyphae paler

or nearly hyaline, with clamps.” Teston (1953 b) confirmed the existence of

nodose-septate generative hyphae, aseptate, thick-walled, skeletal hyphae and narrow,

sinuose, branched, binding hyphae in the carpophores of this species. The

description given above thus agrees with those of earlier workers. The binding

hyphae of this species are thus true binding hyphae as defined by Corner (1932 a,

1953). The brown, thick-walled, nodose-septate hyphae (‘‘sclerified generative

hyphae,” Donk, 1964) which are fairly numerous in the older part of the context

have not been mentioned before. The carpophores thus have the types of hyphae

and construction of species with a trimitic hyphal system as defined by Corner

(1932 a), Cunningham (1946, 1954) and Teixeira (1962 b).

Lenzites sepiaria was chosen as the type species of the genus Gloeophyllum

by Karsten, (Donk, 1960) but Fries (1821) placed it in Daedalea Pers. ex Fries

before transferring it to Lenzites Fries a few years later (1838). Cunningham

(1948 h) considered this species to be congeneric with Daedalea quercina L. ex Fr.

the type of the genus Daedalea Pers. ex Fr. (Cooke, 1959; Donk, 1960). Although

Cunningham (1948 h) defined the genus Daedalea as having a trimitic hyphal

system, it was found in the course of this work that Daedalea quercina does not

have “‘binding hyphae”’ in the carpophores (c.f descriptions Group 25). Furthermore,

cultures of Daedalea quercina do not form brown colours; the hyphae are all

hyaline and the cultures are characterized by the presence of nodose-septate

hyphae with irregularly thickened walls which are absent from cultures of Lenzites

sepiaria. Because of these differences. Lenzites sepiaria cannot be placed in the

genus Daedalea Pers. ex Fries.

Although Lenzites betulina (L. ex Fr.) Fries, the type species of the genus

Lenzites Fries, has the trimitic type of hyphal system, (Cunningham, 1948 h:

O. Fidalgo, 1957) it differs from Lenzites sepiaria by having hyaline hyphae in both

the cultures and carpophores. Its cultures produce extra-cellular oxidase and the

fungus causes a white rot. It also has the tetrapolar type of interfertility (Nobles.

Macrae & Tomlin, 1957) while Lenzites sepiaria has the bipolar type (Mounce &

180

Macrae, 1936). It thus appears that Lenzites sepiaria is not well classified with

species of Lenzites either and is best placed in the genus Gloeophyllum Karst,

of which it is the type species (Donk, 1960).

Fic. 7.— Lenzites trabea. (a) Carpophores of DAOM 22444; (b) culture of PRE 42457

at six weeks; (c) antler-like fructifications in culture: (d) wide and narrow nodose-

septate hyphae from culture, x 1000 phase contrast.

Lenzites trabea Pers. ex Fries, Epicr. Syst. Myc. p. 407, 1838;

Daedalea trabea Pers. ex Fr., Syst. Myc. I, 335, 1821;

Trametes trabea (Pers. ex Fr.) Bresadola, Hym. Hung. Kmet. 27, 1897:

Gloeophyllum trabeum (Pers. ex Fr.) Murrill, Bull. Torr. Bot. Cl. 32, 370, 1905;

Coriolopsis trabea (Pers. ex Fr.) Bond. & Sing., Ann. Mycol. 39, 62, 1941;

Phaeocoriolellus trabeus (Pers. ex Fr.) Kotlaba and Pouzar, Ceska mykologie

9 (3) 152 — 170, 1957.

Cultural characters

Growth moderately fast, plates covered in 3-4 weeks. mycelial mat reaching

a radius of about 30 mm after 1 week. Margin even, mycelium raised or in some

cultures appressed, thin, downy, usually loose cottony to woolly at first, almost

white to “‘light buff.” Mat becomes progressively more woolly with age, darkening

to “‘cream buff” or “light ochraceous buff” to “pale orange-yellow” with patches

181

of “‘orange buff” later turning to “apricot buff’ or “‘ochraceous orange” and fans

of mycelium growing over the lid of the Petri dish from the sides of the cultures.

Fruiting bodies develop from the second week onward as dark ‘apricot buff”

patches of compact, felty mycelium from which slender rounded spines, “apricot

buff”? in colour grow up, later expanding into flat, lamellate structures forming

loose, irregularly poroid fructifications. At six weeks, mat mostly raised, woolly,

radially and concentrically sulcate and “maize yellow” with darker fruiting or

incipient fruiting areas, or, with large, irregular areas of thin, downy mycelium, or,

sodden, appressed patches among areas of raised, woolly, “‘pale ochraceous buff”

or “‘maize yellow” mycelium. Reverse bleaching slowiy at first, then darkening

in patches to “cadmium yellow,” “deep chrome,” ‘‘capucine orange”’ or ‘xanthine

orange’. Odour faint or strong, suggesting garlic. No diffusion zones are formed

on gallic acid and tannic acid media; no growth on the latter medium but up to

20 mm diameter colonies on gallic acid media. No reaction occurs when gum guaiac

solution is applied to the mat.

Advancing mycelium: hyphae hyaline, thin-walled, nodose-septate, branching from

the clamp connections 3.0 — 4.5u in diameter, (Fig. 8 a).

Aerial mycelium: (a) hyphae as in the advancing zone; (b) fibre hyphae rare, long,

narrow, hyaline at first but turning yellowish gradually, unbranched or occasionally

branched, the branches long, walls thickened, refractive, lumina narrow or occluded,

aseptate, 2.0 — 3.0u in diameter, (Fig. 8 b); (c) nodose-septate, narrow, much

branched hyphae, thin-walled at first but becoming solid later, 0.7 — 1.5u arising

from normal thin-walled hyphae (Fig 8g); (d) oidia numerous or rare, cylindrical,

thin-walled, hyaline 2.0 — 3.0u :a diameter and length variable (Fig. 8 h).

Fructifications: (a) thin-walled nodose-septate hyphae as in the advancing zone;

(b) fibre hyphae numerous, long, unbranched, tortuous, yellowish-brown, walls

thickened, lumina narrow or occluded, widening near the origins and tips, aseptate,

2.2 — 3.5u (Fig. 8 b); (c) nodose-septate hyphae with pale-yellowish conspicuous

walls, much branched and forming H-connections, 2.2 — 3.5 in diameter, inter-

twined with other hyphae in the fructification (Fig. 8 c): (d) basidia hyaline, long-

clavate to almost cylindrical 19.0 — 27.0 x 6.0 — 7.5u, with 4 long, slender.

straight sterigmata 4.5 — 6.0u, borne in clusters of short branches from thin-walled,

nodose-septate hyphae (Fig. 8 4d); (e) basidiospores hyaline, cylindrical, obliquely

apiculate, smooth, thin-walled 7.2 — 9.6 — (12) x 3.0 — 3.7u (Fig. 8 e):; (f)

cystidioles hyaline, fusiform, often with peculiar branching, apical processes,

18.0 — 32.0 x 4.0 — 6.0u arising from the basidial fascicles (Fig. 8 f).

Submerged mycelium: (a) hyphae hyaline, narrow, branched, with conspicuous

clamp connections at the septa, thin-walled at first but walls thickening later,

1.2 — 2.0u in diameter (Fig. 7d, 8k); (b) nodose-septate hyphae as in the advancing

zone; (c) chlamydospores hyaline, ovoid to ellipsoid, thick-walled with deeply

staining contents 10.0 — 23.0 x 6.0 — 10.0u, intercalary or terminal (Fig. 8 m).

Carpophore characters

Carpophore annual or occasionally reviving, lignicolcus, solitary or compound,

sessile or effused-reflexed, applanate or dimidiate somewhat convex, often laterally

connate, occasionally imbricate; coriaceous when fresh but harder and more rigid

on drying, 1.0 — 10 x 1 — 4 x 0.2 — 0.8 cm; upper surface tomentose, velvety but

becoming glabrous and rugose, azonate or concentrically sulcate, grayish-brown to

cinnamon-brown or umber-brown; margin obtuse, thin, concolorous with pileus or

somewhat paler, entire or somewhat lobed; pore surface concolorous or slightly

paler than upper surface, lenzitoid to poroid or occasionally lamellate; tubes 1 — 4

mm deep, edges thin, entire, 2 — 3/mm:; context fibrous, umber-brown, | — 4 mm

thick.

182

Vic. 8.—Lenzites trabea. a-m. Hyphae and structures from cultures: (a) hyphae

from advancing zone; (b) unbranched, fibre hyphae; (c) thick-walled, nodose-

septate hyphae with H-connection; (d) basidia; (e) basidiospores; (f) branched

cystidioles; (g) thick-walled, sub-solid, nodose-septate hyphae; (h) oidia; (k)

narrow, hyaline, nodose-septate, submerged hyphae; (m) chlamydospores.

n-t. Hyphae and structures from carpophores: (n) thin-walled, nodose-septate

hyphae; (p) thick-walled, nodose-septate hyphae with tortuous branches; (q>

unbranched fibre hyphae; (s) basidia; (t) basidiospores.

183

Hyphal characters. Carpophores consist of: (1) nodose-septate hyphae, thin-walled,

hyaline, branching, often from the clamp connections, 2.4 — 4.0u in diameter (Fig.

8 n); (2) nodose-septate hyphae with thickened sub-hyaline to brownish walls,

narrowed or occluded lumina, branching and forming H-connections and forming

short, tortuous, thin-walled or thick-walled to solid, nodose-septate, lateral branches

2.5 — 4.5u in diameter (Fig. 8 p); (3) fibre hyphae long, straight or slightly flexuous,

unbranched, pale brown, walls thickened, lumina narrow or occluded and usually

widening towards the tip, aseptate, or with one or two simple septa towards the tip,

2.5—5.0u in diameter (Fig. 8 q).

Hymenium: basidia long clavate, hyaline, 19.0 — 27.0 x 6.0 — 7.5u, with four

slender, straight sterigmata 4.5 — 6.0u long (Fig. 8 s); basidiospores cylindrical,

ends rounded, obliquely apiculate, hyaline, at first, later pale yellowish-brown,

smooth ,thin-walled, 8.0 — 9.6 x 3.0 — 4.0u (Fig. 8 t).

Construction. At the margin the carpophore consists of long, unbranched fibre

hyphae with pale brown thickened walls and prominent aseptate lumina, subsolid

in the middle portion, 2.4 — 3.6u in diameter and arranged more or less parallel

but somewhat intertwined in the direction of growth. Also intertwined with the

fibre hyphae are long, hyaline, thin-walled, nodose-septate hyphae mostly branching

from the clamp connections, 2.4 — 3.6u in diameter, the branches interwoven with

the other hyphae. In the context behind the margin the hyphae turn upward

towards the upper surface. The fibre hyphae are lightly intertwined and similar

to those of the margin but their lumina are narrower, often occluded and one or

two simple septa may be present near their tips. Their ends are free and project

upward to a common level to form the finely tomentose upper surface of the young

parts of the pileus. Intertwined with these fibre hyphae, up to a position slightly

below their tips, are numerous hyaline, nodose-septate hyphae similar to those in

the margin, branching freely and forming frequent H-connections. In the older

parts of the context the fibre hyphae are slightly darker in colour and their walls

thickened for longer distances. Here the nodose-septate hyphae have thickened,

pale brown walls with lumina narrowed and sections often solid and forming

short, very tortuous, nodose-sepiate branches, tightly interwoven with the fibre

hyphae (Fig. 8 p). In some specimens a greyish, glabrous cuticle may be present

over this part of the carpophore. This may be up to 50u thick and consists of

hyaline or sub-hyaline hyphae with thick, gelatinous walls swelling in KOH,

arising from the nodose-septate hyphae of the context. and lying prostrate and

intertwined in all directions on the upper surface where they are agglutinated

into a thin, smooth trichocutis (Lohwag, 1940). In the lower context the hyphae

turn downwards into the trama Fibre hyphae resemble those of the upper context

but they are narrower, mostly with very narrow lumina, more flexuous and tightly

intertwined to form a denser tissue. In these parts, the nodose-septate hphae are

thin-walled, hyaline, branching freely and interwoven with the fibre hyphae. Many

branches of these nodose-septate hyphae are tightly intertwined with the fibre

hyphae, become thick-walled or solid and bind the hyphal elements together into

a tough tisue (Fig. 8 p). Many of these nodose-septate hyphae remain thin-walled

and ramify though the hyphal elements in the tramal tissue to form numerous,

short branches bearing clusters of basidia at the hymenial surfaces, each basidium

subtended by a basal clamp connection (Fig. 8 s).

Decay and hosts

Lenzites trabea causes a brown rot of hardwood timbers but is also found

on coniferous wood.

184

Specimens examined

Herb. DAOM: *F3823, on hardwood log, Ottawa, Ont., May 1934; F3838, on coniferous

wood, St. Andrews, N.B., Aug. 1933; F2893, on Acer sp., Durham, Ont., Aug. 1937; F6482,

on Alnus incana, Montgomery Lake, Ont., Sept. 1935; F8893, on Acer sp. log, near Durham,

Ont., Aug. 1937; F9073, Mt. Mitutoge, Japan, Oct. 1933; *9507, on Tsuga canadensis, Toronto,

Ont., 1939; *22444, on deciduous log, Waterbury Centre, Vt., Oct. 1950; *22630, ex Div.

For. Prod. South Melbourne, Aust., 1939; *30929, on Pinus sylvestris, Suomsujorri Lieksa,

Finland, 1951; *72285, Portage du Fort, Que., June 1961.

Herb, PRE: 31310; 31341, decayed wood, Stellenbosch, C.P., Nov. 1926; 31379, old posts,

Stellenbosch, C.P., Aug. 1919; 31426, decaying logs, Stellenbosch, C.P., Sept. 1919; 36818,

on Pinus sp. stumps, Harrismith, O.F.S., March 1948; 40222, on Cupressus sp. logs, Pretoria,

Tvl., March 1952; 41679, Johannesburg, Tvl., 1954; *42457, on decaying hardwood log,

Honeydew, Tvl., Jan. 1961.

Interfertility studies

Lenzites trabea has the bipolar type of interfertility with allelomorhps for

heterothallism at one locus only (Mounce & Macrae, 1936). To determine whether

the South African collection was interfertile and therefore conspecific with a

Canadian strain, four cultures, each obtained from a single basidiospore of PRE

42457, were mated on malt agar slants in all possible combinations with each of

four single spore cultures of the Canadian isolate DAOM 72285. The results

are given in TABLE 4.

The formation of clamp connections on all the paired mycelia is regarded

as positive proof that the South African and Canadian isolates are interfertile and

therefore conspecific.

Discussion

This description of the cultural characters of Lenzites trabea agrees well with

those of Snell (1922), Cartwright (1931), Cartwright & Findlay (1946) and Nobles

(1948). Cultures of this species are readily recognized by their distinctive colour

and texture, early fruiting. the brown fibre hyphae which are rare in the mycelial

mat but abundant in the fruiting structures, and the very narrow hyphae with

large conspicuous clamp connections which are found at the surface of the agar

(Fig. 7 d).

From the descriptions it is evident that not all structures formed in culture

are also present in the carpophores. The narrow hyphae with prominent clamp

connections as well as the chlamydospores formed in the agar were not found in

the carpophores. It is possible that these structures may be found in the wood

decayed by the fungus but no specimens of decayed wood were available for

examination. Chlamydospores have been found in wood decayed by species which

form them in culture (Cartwright & Findlay, 1946). The nodose-septate hyphae

and fibre hyphae formed in cultures are similar to those in the carpophores of

Lenzites trabea.

From the above description it is evident that the carpophores of Lenzites

trabea consist of two types of hyphae only, viz. unbranched, thick-walled hyphae

mostly aseptate or occasionally with one or two simple septa (skeletal hyphae,

Corner, 1932 a) and branching, nodose-septate hyphae either with thin, hyaline

walls or with the walls coloured and thickened (sclerified generative hyphae, Donk,

1964). This observation agrees with the descriptions of the hyphae of this species

by Pinto-Lopes (1952), Teston (1953 b) and Overholts (1953). According to the

definitions of Corner (1932 a, 1953), Cunningham (1946, 1954) and Teixeira

(1962 b) hyphae with clamp connections must be regarded as generative hyphae.

The carpophore of Lenzites trabea thus has a dimitic hyphal system. But thick-

walled elements of the nodose-septate hyphal systein serve to bind the tissues of

185

the carpophores. This binding hyphal system, which admittedly is not well

developed, thus consists of sclerified generative hyphae. The carpophores of

Lenzites trabea differ in construction and hyphal characters from those of Lenzites

sepiaria, a species with trimitic hyphal system (Cunningham, 1948 h; Teston,

1953 b) in which the binding system consists of “branched, aseptate, thick-walled

hyphae of limited growth” that arise as differentiated terminal cells of lateral

branches of nodose-septate generative hyphae (Corner, 1932 a). This difference

is of fundamental importance in the anatomy of the carpophores of these two

species, which therefore cannot be regarded as being congeneric.

Pinto-Lopes (1952), Corner (1954 a), Bondartseva (1961), Teixeira (1962 b) and

Donk (1964) regard the absence or presence of different types of hyphae in carpo-

phores as important at the generic or higher level. Lenzites trabea thus cannot

be regarded as being congeric with Lenzites sepiaria, the type species of the genus

Gloeophyllum Karsten (Cooke, 1959; Donk, 1960). For this same reason, Lenzites

trabea cannot be congeric with any one of the thrze species Polyporus occidentalis,

the type species of the genus Coriolopsis Murr., or Lenzites betulina, the type

species of the genus Lenzites Fr., or Trametes sauveolens, the type species of the

genus Trametes Fr. Lenzites trabea had been transferred to each of these three

genera by various authors (loc. cit.) of which all have fruit bodies with trimitic

hyphal systems (see descriptions Group 45).

Fries (1821) and Cunningham (1948 h) regarded Lenzites trabea as congeneric

with Daedalea quercina Fr., the type species of the genus Daedalea Fr. But the

nodose-septate hyphae with irregularly thickened walls which are present in cultures

and carpophores of that species (Group 25) are not present in the cultures and

fruit-bodies of Lenzites trabea and the two species can therefore not be regarded

as congeneric.

Kotlaba & Pouzar (1957) created the new genus Pheeocoriolellus with Lenzites

trabea as the type and only species. This genus is characterized by a dimitic

hyphal system and was based on Bondartsev’s (1953) and Teston’s (1953 b)

descriptions.

M. E. P. K. Fidalgo (1962) reported that hyphal analysis of Trametes odorata

(Wulf. ex Fr.) Fries, the type species of the genus Osmoporus Sing., revealed this

fungus to be dimitic, with the generative hyphae branched, hyaline, nodose-septate,

characteristically thin-walled but often thick-walled or solid and skeletal hyphae

unbranched, yellowish-brown to brown, aseptate, long, fibre-like. The nodose-

septate hyphae were thin-walled in the growing region but often thick-walled and

brownish in the older parts of the context and above the dissepiments. Fidalgo’s

descriptions of the hyphae, which confirmed the reports of Pinto-Lopes (1952)

and Teston (1953 b), agree very closely with the description of the hyphae of

Lenzites trabea as given above. These two species are thus very similar in hyphal

characters, construction of the carpophore and other morphological features. These

two species are thus congeneric and if the genus Osmoporus Sing. is accepted

as valid, Lenzites trabea should be transferred to it.

Resumé.

From these descriptions it is evident that Lenzites sepiaria (Wulf. ex Fr.) Fr.

and Lenzites trabea Pers. ex Fr. possess many characters which are common to

both species in their cultures and their carpophores. The absence of some types

of hyphae, which are present in the carpophores of Lenzites sepiaria, from carpo-

phores of Lenzites trabea, indicates that the relationship between these species,

which is suggested by these common characters, must exist at a supra-generic level.

186

5.4 Group 18

The mycelial mats of cultures of species in this group usually remain white

or may develop pale pinkish or vinaceous tints. They do not produce extra-cellular

oxidase. Their thin-walled hyphae have simple clamp connections at the septa

and remain thin-walled but septate, thick-walled fibre hyphae are also formed in

large numbers. Their basidiospores are globose, ovoid or ellipsoidal in shape. The

interfertility of species of which this character is known, is of the bipolar type.

Fic. 9.—Fomes pinicola. (a) Carpophores of DAOM 22755; (b) culture at six weeks;

KOH; (d) inter-calary chlamydospore, thin-walled nodose-septate hyphae and fibre

hyphae from carpophore.

Fomes pinicola (Sw. ex Fr.) Cooke, Grey. 14, 17, 1885;

Polyporus pinicola Sw. ex Fr., Syst. Myc. 1, 372, 1821;

Fomitopsis pinicola (Sw. ex Fr.) Karst., Bidr. Kanned. Finl. Nat. Folk, 48,

306, 1889.

Ungulina marginata (Fr.) Pat., Essai taxon. Hymen., p. 103, 1900.

187

Cultural characters

Growth is moderately fast to slow, the plates being covered in four to six

weeks. Advancing zone even at first, thin, downy, appressed, later somewhat

ragged but the mat becoming more dense towards the inoculum; white, raised,

cottony-woolly, concentric zones develop or rounded lumps or nodules which slowly

increase in size, appear on the zones of denser mycelium and may grow together

as compact, uneven masses of tough, chamois mycelium near the sides of the

plate; fruiting areas originating as shallow pores of which the sides grow upward

to form tubes, resulting in pored areas with the oldest tubes in the centre, develop

on the irregular lumps after five to six weeks. The cultures remain white or pale

“cream color” over the fruiting areas. The reverse remains unchanged. No

odour or a faint, fragrant odour is emitted. No colour change occurs when alcoholic

gum guaiac is applied to the culture. No diffusion zones are produced on gallic

acid and tannic acid media but some mycelial growth occurs on both media.

Advancing mycelium: hyphae branching at or near the septa, hyaline, thin-walled,

nodose-septate, 1.0 — 4.0u in diameter (Fig. 10 a).

Aerial mycelium: (a) nodose-septate hyphae as in the advancing zone; (b) fibre

hyphae hyaline, long, unbranched or occasionally branched, arising from nodose-

septate hyphae, thin-walled near their origins and tips but walls thickened, lumina

narrow or occluded, aseptate or with one septum neer the tip, 1.0 — 4.0u in

diameter (Fig. 10 b); (c) chlamydospores, inter-calary or terminal, ovoid to sub-

globose with thickened walls 6.0 — 18.0 x 6.0 — 9.Qu. mostly in young parts

of mycelium (Fig. 10 c).

Fructification: (a) hyphae as in the advancing zone; (b) fibre hyphae as in the

aerial mycelium but more freely branched, the branches long and tortuous, a

number of branches often arising close together from a main branch and close to

its origin, 1.0 — 4.0u in diameter (Fig. 10 d); (c) basidia long clavate, 18.0 — 25.0

x 5.5 — 7.2u with four sterigmata 2.4 — 3.0, and arising from short branches

of thin-walled, nodose-septate hyphae in the dissepiments (Fig. 10 e); (d) basidio-

spores hyaline, ovoid, or ellipsoidal with an oblique apiculum, thin-walled, smooth

4.8 — 6.0 x 3.3 — 4.0u (Fig. 10 f); (e) hymenial cystidia hyaline, long cylindrical

often tapering towards the tips and projecting somewhat beyond the basidia,

24.0 — 30.0 x 2.4 — 3.0u, arising from the basidial fascicles (Fig. 10 g).

Carpophore characters

Carpophore perennial, lignicolous, solitary, sessile, dimidiate, convex to

ungulate, hard corky to woody up to 15 x 20 x 10 cm; upper surface covered with

a resinous layer, smooth, sticky and reddish brown at first, later dark gray to black,

smooth or sulcate with age, hard; margin obtuse, thick, rounded, entire to lobate

or undulate, pallid or lighter coloured than other areas: pore surface white to

yellowish or pale buff where bruised; pores, rounded, 3 — 5 per mm, mouths

entire, dissepiments thick, even; tubes concolourous, distinctly stratified, 3 — 7

mm long each season; context pale creamy to yellowish or pale brown, concentrically

zoned, reddish brown in KOH, up to 2.0 cm thick.

Hyphal characters: carpohores consist of (a) hyaline, branching, thin-walled, nodose-

septate hyphae, with deeply staining contents, or empty. 2.2 — 4.0u in diameter,

(Fig. 10 h); (b) fibre hyphae long, unbranched, hyaline to sub-hyaline with walls

thick and refractive, lumina narrow, widening gradually towards the ends, aseptate

or with one or two simple septa near the distal end, 3.0 — 9.0u in diameter (Fig.

10 k); (c) chlamydospores hyaline, ovoid to ellipsoid, thick-walled, intercalary in

nodose-septate hyphae, 7 — 12 x 10 — 24 (Fig. 10 m).

Fic.

FIGURE IO.

10.—Fomes pinicola. a - f. Structures from culture: (a) nodose-septate

hyphae from advancing zone; (b) fibre hyphae; (c) chlamydospores; (d) fibre

hyphae with long, tapering branches; (e) basidia; (f) basidiospores; (g) hymenial

cystidium.

h-q. Structures from carpophores: (h) thin-walled, nodose-septate hyphae;

(k) fibre hyphae; (m) chlamydospores; (n) basidia; (p) basidiospores; (q) hyme-

nial cystidium.

189

Hymenium: basidia hyaline, long clavate 18.0 — 25.0 x 5.5 — 7.2u with four

slender, straight sterigmata 2.4 — 3.0u (Fig. 10 n); basidiospores hyaline, ellip-

soidal to ovoid, smooth, thin-walled, 4.8 — 6.0 x 3.3 — 4.0u (Fig. 10 p); hymenial

cystidia long, hyaline, tapering terminally, 24 — 30 x 2.4 — 3.0u, projecting

10 — 12u beyond the basidia (Fig. 10 q).

Construction: At the margin the carpophore consists of long, unbranched, thick-

walled, fibre hyphae with narrowed lumina, arranged parallel to the direction of

growth of the carpophore, somewhat intertwined and closely packed. Behind the

margin in the context are numerous hyaline, thin-walled, nodose-septate hyphae,

branching repeatedly and intertwined with the fibre hyphae, which arise from

them. The context consists of long fibre hyphae, unbranched or occasionally

branched, hyaline or sub-hyaline, sub-solid to solid, aseptate or occasionally with

one or two simple septa at the distal end, 3.0 — 9.0u in diameter and with globules

of lacquer-like material adhering to them in the zones of growth of the pileus.

These hyphae are mostly closely packed and slightly intertwined among themselves

and with branching, hyaline, thin-walled, nodose-septate hyphae, some with deeply

staining contents, others collapsed and empty, or, with intercalary thick-walled,

ovoid to ellipsoid chlamydospores 7 — 12 x 10 — 24u in the middle and lower

context. At the upper surface the ends of fibre hyphae are bent over and become

interwoven with one another and numerous branching, thin-walled, nodose-septate

hyphae with deeply-staining contents, to form a sub-surface cortex (Lohwag, 1940).

From these hyphae, a resin-like material is exuded which permeates the upper

layer of the cortex and covers the surface of the pileus in a layer up to 1500u thick

(Fig. 9 c). From the lower context the fibre hyphae bend downward, remaining

unbranched and intertwined towards the dissepiments. In the dissepiments the

fibre hyphae are unbranched, tortuous, sub-solid, aseptate, narrower than in the

context and tightly intertwined among themselves. Hyaline, nodose-septate hyphae

are intertwined and interwoven with the fibre hyphae, branching freely and

becoming increasingly numerous towards the hymenial surfaces where short branches

project to bear fascicles of clavate basidia and long, tapering cystidia.

Decay and hosts

Fomes pinicola causes a brown rot of dead, standing or fallen coniferous and

occasionally angiosperm timber, destroying both heartwood and sapwood.

Specimens examined

Herb. DAOM: *F3249, on Populus balsamifera, Victoria Beach, Man. May 1933; *F6895,

on Pinus yezoensis, Hokkaido, Japan; *F6925, on Picea glauca, Hot Springs, Alaska, Aug.

1936; *F7120, on fallen Picea exelsa, Poland, Sept. 1936; *7121, on living Prunus avium,

Poland, Sept. 1936; *8567, on Picea sitchensis, Kodiak Is., Alaska, Aug. 1938; *8568, on

Tsuga heterophylla, Mt. Arrowsmith; *9937, on dead A. saccharum, Preston, Ont., Oct. 1959;

*10787. on Pyrus malus stump, Kentville, N.S., Feb. 1942; 17924. on Picea glauca, Hudson

Bay, Sask., Sept. 1947; 17926, on Populus tremuloides, Hudson Bay, Sask., Sept. 1947;

22358, ex Herb. J. Pinto-Lopes, Lisbon; 22711, on Abies balsamea log, Labrador, Aug. 1949:

22746, on Picea sp. logs, Moosehide Mtns., Yukon, Aug. 1949; 30023, on Betula papyrifera,

Doré Lake, Sask., July 1948; 30064, on Pinus contorta slash, Spines Mill, Alta., Sept. 1950;

30152, on Betula alba, Uppsala, Sweden, Aug. 1952; 30157, on Betula alba, Uppsala, Sweden,

Sept. 1952.

Discussion

The cultural characters as described above agree well with those described

by Mounce (1929), Campbell (1938), Davidson, Campbell & Vaughn (1942),

Cartwright & Findlay (1946) and Nobles (1948, 1965). There is a decided lack of

distinguishing features in the gross appearance and hyphal characters of the mat

formed in culture but the chlamydospores which occur in the newer parts of the

mat but not in the older parts, may be a useful diagnostic feature.

190

From the above description it is clear that only thin-walled, branching nodose-

septate hyphae and fibre hyphae, typically unbranched, are present in the

carpophores. Lowe (1957) stated of Fomes pinicola: “‘context hyphae rarely

branched, thick-walled, non-septate, 5 — 8u in diameter with a small amount of

thin-walled, clamped hyphae, 2.4u in diameter; tramal hyphae similar except mostly

3 — 5u in diameter”. Overholts (1953) reported the thick-walled, aseptate hyphae

but also found “hyphal complexes composed of hyaline hyphae 3 — 4u in

diameter, present in considerable numbers”. Overholts usually reported such

hyphal complexes in species in which branched binding hyphae are present

(e.g. Lenzites betulina). Overholts did not describe the hyphae in these complexes.

Teston (1953 b) figured narrow, thick-walled branched hyphae from the tubes

of Ungulina marginatus (—Fomes pinicola, Bourdot & Galzin, 1928; Lowe,

1957) but stated that a few branched hyphae are found in the dissepiments

of the carpophores of this species. No anatomical details of these hyphae were

given however. It appears that Overholts’ ‘“‘hyphal complexes’ and Teston’s

“branching hyphae’ may correspond to the branched, nodose-septate hyphae,

some of which have thickened walls, which are very numerous in the dissepiments

of the specimens examined of this species. Structures resembling binding hyphae

as described and defined by Coraer (1932 a, 1953), Cunningham (1946, 1954),

and Teixeira (1962 a, b) were never seen in any of the specimens examined. The

above description thus agrees well with earlier reports on the hyphal characters

of this species.

From the descriptions it is clear that structures found in the cultures, are

also found in the carpophores. While the fibre hyphae of the carpophores are

mostly unbranched, branched fibre hyphae are found in the fruiting areas formed

in culture. The number of branches is however the only difference between these

structures which are otherwise identical in all other respects. The chlamydospores

found in some of the carpophores examined, are of interest since their occurrence in

carpophores of Hymenomycetes have been reported only rarely, even in species

in which they are abundantly produced in cultures. This is an indication that the

absence or presence of chlamydospores in cultures, is at best of diagnostic value

in the identification of cultures only.

Fomes pinicola has the hyphal characters, which are typical of species with

the dimitic hyphal system as described by Corner (1932 a, 1953) and Cunningham

(1946, 1954). Farinha (1964) reached a similar conclusion after a study of cultures

and carpophores of this species. The large, perennial carpophores are however

remarkably simple in construction. The fibre hyphae are straight and are more

or less parallel to one another in the carpophore. Even in the dissepiments are

they seldom very tortuous, suggesting a rather loose association of hyphae. There

further appears to be little binding of the tissues by the nodose-septate hyphae as

observed in Lenzites trabea and special binding hyphae or processes are lacking.

This simplicity of construction is emphasized by the fact that sections of the

carpophores, even from the dissepiments, are easily teased out with needles or even

squashed when placed in KOH for microscopic examination.

The simple construction and dimitic hyphal system of fruit-bodies of Fomes

pinicola, are in sharp contrast to the complex carpophores of Fomes fomentarius

(L. ex Fr.) Kickx the type of the genus Fomes (Fr.) Kickx (Donk, 1960) as described

by Teixeira (1962 a). According to Teixeira (1962 a), the fruit-bodies of Fomes

fomentarius are characterized by the dark brown context, consisting of nodose-

septate generative hyphae, dark-brown, thick-walled skeletal hyphae and much

branched binding hyphae and are covered by an indurate layer of agglutinated

ends of skeletal hyphae. Because of this great difference in hyphal composition

191

and construction of the carpophores of Fomes fomentarius and Fomes pinicola,

these two species cannot be regarded as congeric.

Although Fomes fomentarius (L. ex Fr.) Kickx is the type species of the genus

Fomes (Fries) Kickx, Donk (1960) presented strong evidence in favour of his

view that this species is also the type of the genus Ungulina Pat. and that the latter

genus is an isonym of the genus Fomes (Fr.) Kickx. If this view is accepted, it

will mean that Fomes pinicola cannot be included in either of the genera Fomes

(Fr.) Kickx or Ungulina Pat. but may be best placed in the genus Fomitopsis

Karsten of which it is the type species (Cooke, 1959; Donk, 1960).

Fomes pinicola has not been recorded from South Africa. The species

however occupies an important taxonomic position as the type of the genus Fomitop-

sis Karst., to which a number of species, including some found in South Africa, had

been referred. It furthermore appeared to be a good example of a species with

dimitic hyphal system, of which the construction had not been described, to include

in this study in order to have a sound basis for future comparison with possibly

related species.

5.5 Group 25

Cultures of species in this group form white, pale yellow or rose coloured

mycelial mats which do not produce extra-cellular oxidase enzymes. Their thin-

walled hyphae have simple clamp connections at the septa. Thick-walled, aseptate

fibre hyphae are also formed. Characteristic hyphae with numerous clamp con-

nections and their walls irregularly thickened, and the lumina much reduced but

staining deeply in phloxine, are present to a greater or lesser extent. Their

basidiospores are cylindrical o: allantoid. Their interfertility, where known, is of

the bipolar type.

Daedalea quercina L. ex Fr., Syst. Myc. 1, 333, 1821.

Lenzites quercina (L. ex Fr.) Quélet, Ench., 153, 1886;

Trametes quercina (L. ex Fr.) Pilat, Atl. Champ. Eur. 3, 329, 1936.

Cultural characters

Growth is slow, the mat reaching a radius of about 30 mm in two weeks and

covering the plate only after six weeks. The advancing margin is even with the

hyphae raised to the limit of growth. The mat is white, at first woolly with fine

strands radiating from the inoculum. From 3 — 4 weeks the older mycelium

tends to become appressed and patches of denser felty mycelium with an overgrowth

of sparse, erect, cottony mycelium begin to form on the agar in the older parts

of the mat end along the sides of the dish. Over these patches, fruiting areas may

appear as granules which later develop into coarse, warty or spine-like columns

of dense white or creamy mycelium which may become connected to each other

by ridges or strands of similar dense mycelium. At six weeks the cultures have a

thin, radiating, woolly mat with vague, radiating and concentric depressions with

the dense mycelium of the fruiting areas mostly over the older parts and against

the sides of the dish. Rhizomorphic strands may run from the inoculum to the

fruiting areas. The reverse remains unchanged. At 4 — 6 weeks a pleasant, fruity

odour is given off by the culture. No reaction is evident when tested for extra-

cellular oxidase enzymes with gum guaiac solution.

Fic. 11.— Daedalea quercina. (a) Carpophore of PRE 31394, upper surface and (b)

hymenial surface; (c) nodose-septate hyphae with irregularly thickened walls from

carpophore, x 1000; (d) culture of PRE 42366 at 6 weeks.

Advancing mycelium: hyphae hyaline, thin-walled, nodose-septate, branching at or

near the septa or from clamp connections 2 — 6u in diameter (Fig. 12 a).

Aerial mycelium: (a) nodose-septate hyphae as is the advancing zone; (b) nodose-

septate hyphae with the walls irregularly thickened and with deeply staining contents

in the lumina, either very narrow 1.5 — 2.0u in diameter, branching and solid in

parts or wide, up to 6u in diameter and mostly unbranched and tending to break at

the clamp connections (Fig. 12 b). The latter type is abundant in the felty mycelium

on the surface of the agar; (c) fibre hyphae hyaline, unbranched or occasionally

with branches, the walls thick, refractive and lumina narrow or almost occluded,

aseptate and widening only at the ends, 1.5 — 4.0u in diameter (Fig. 12 c). They

arise from thin-walled, nodose-septate hyphae and nodose-septate hyphae with

irregularly thickened walls.

Fructifications: (a) nodose-septate, thin-walled hyphae; (b) nodose-septate hyphae

with irregularly thickened walls, and (c) fibre hyphae as described above. Basidia

hyaline, long-clavate 22 — 36 x 4.5 — 6u, with 4 straight sterigmata 3.6 — 4.5u

long (Fig. 12 d), borne in clusters on repeatedly branched thin-walled, nodose-

septate hyphae (Fig. 12 d); basidiospores, short cylindrical, hyaline, smooth, thin-

Fic.

FIGURE 12.

12.— Daedalea quercina. a-f. Structures from culture: (a) hypha from

advancing zone; (b) nodose-septate hyphae with irregularly thickened walls:

g-q. Structures from carpophores: (g) thin-walled, nodose-septate hyphae;

(h) nodose-septate hyphae with irregularly thickened walls; (k) fibre hyphae;

(m) basidia; (n) basidiospores; (p) tramal cystidia; (q) skeletal cystidia.

194

walled, 4.5 — 6.5 x 2.4 — 3u (Fig. 12 e). Ends of fibre hyphae project into the

hymenium and slightly beyond it from the underlying tissues, appearing as hyaline

cystidia, 3 — 3.6 in diameter.

Submerged mycelium: (a) hyaline, thin-walled, nodose-septate hyphae as in the

advancing zone; (b) nodose-septate hyphae with irregularly thickened walls as in

the aerial mycelium; (c) chlamydospores intercalary and terminal, hyaline,

ellipsoidal, thick-walled 6 — 20 x 4 — 8u (Fig. 12 f).

Carpophore characters

Carpophores perennial, lignicolous, solitary or grouped, sessile, dimidiate;

pileus applanate, occasionally connate, imbricate; hard cerky or rigid up to 15 x 20

x 8 cm; upper surface at first finely tomentose, soon glabrous, uneven to zonate,

somewhat furrowed, finally somewhat incrusted with age, at first whitish, later

umbrinous to black; margin obtuse, entire, pallid; pore surface creamy white to

pale amber or avellaneous, poroid at first but soon labyrinthiform and about 1 mm

wide, dissepiments even, 0.75 — 1.5 mm wide; tubes up to 3 cm deep, somewhat

decurrent, concolourous with the pore surface; context whitish to pale brown,

0.2 — 1.5 cm thick, corky, zonate, with concentric zones of darker and lighter

colour, smooth, fibrous texture and darkening with KOH.

Hyphal characters: (i) thin-walled, nodose-septate hyphae hyaline, branched, with

frequent H-connections, 1.8 3.0u in diameter (Fig. 12 g); (11) nodose-septate

hyphae with irregularly thickened, refractive walls hyaline, branched, forming

occasional H-connections and contents staining deeply in phloxine, 1.8 3.6u in

diameter, rare (Fig. 12 h); (iii) fibre hyphae long, more or less straight, or flexuous,

unbranched, or, with one or two branches often with pointed ends sub-hyaline

to pale brown, the walls thickened towards the middle with extremities thin-walled

lumina prominent, staining, aseptate, or with one or two simple septa near the

tips, widest at the ends, or, walls much thickened, with lumina narrowed or

occluded and reduced to a thin interrupted line, expanding only towards the ends,

2.5 — 5.5u in diameter (Fig. 12 k).

Hymenium: basidia long clavate 20.0 — 30.0 x 4.5 — 6.5u bearing four short,

slender, straight sterigmata 2.5 — 3.0 (Fig. 12 m); basidiospores hyaline, long

ellipsoid to cylindrical and flattened on one side, thin-walled, smooth 4.8 — 6.5 x

2.4 — 3.0u (Fig. 12 n); tramal cystidia hyaline, ob-clavate, thick-walled to sub-solid

30.0 — 50.0 x 4.0 6.0u or longer, arising in the hymenium or in the trama

below (Fig. 12 p).

Construction. At the margin the fibre hyphae are mostly long, straight and

unbranched with prominent lumina, their ends often thin-walled and collapsed.

These hyphae are arranged more or less parallel to and loosely intertwined with

each other and with the numerous branched, thin-walled, nodose-septate hyphae

from which they arise. The older parts of the context are similar but the fibre

hyphae have thicker walls, becoming sub-solid and few thin-walled, nodose-septate

hyphae are present. In the upper part of the context the fibre hyphae bend upwards

gradually, are mostly branched, sub-solid, up to 5.5 in diameter more or less

straight and loosely intertwined, and arranged with their apices closely packed and

imbricate at a common level, to form the finely pubescent upper surface. Thin-

walled, nodose-septate, branching, hyaline hyphae mostly 2.5u in diameter are

intertwined with the fibre hyphae just below the level of the upper surface. From

the lower part of the margin and context, fibre hyphae turn downward towards

the dissepiments. “These fibre hyphae are narrower, mostly 2.2 — 4.0u in

diameter with more prominent lumina, often very tortuous and more frequently

branched, the branches long and flexuous and becoming tightly interwoven in all

195

directions with each other and with the numerous nodose-septate hyphae into a

denser and more compact tissue than the upper context. Some fibre hyphae remain

more or less straight. Intertwined with the fibre hyphae are nodose-septate hyphae,

mostly thin-walled but occasionally with their walls irregularly thickened and

refractive, branching repeatedly and becoming increasingly numerous towards the

surface of the dissepiments where they branch freely to form the numerous, short,

closely packed, tightly intertwined, nodose-septate branches which bear the basidia.

The ends of fibre hyphae from the trama of the dissepiments may protrude

through this layer and beyond the basidia as tramal cystidia. Short fibre hyphae,

50.0 — 90.0u in length, arising from thin-walled, nodose-septate hyphae in the

trama, or, as lateral branches of fibre hyphae and mostly narrow, 1.8 — 2.2u,

near their origin and for part of their length, but then widening suddenly into

spear-shaped ends (Fig. 12 p), may also project as skeletal-cystidia into the

hymenium (Fig. 12q). Thick-walled cystidia arise from the basidial fascicles on

the same level as the basidia.

Decay and hosts

Daedalea quercina causes a brown rot of hardwood timber and trees where

it lives saprophytically on dead parts (Cartwright & Findlay, 1946).

Specimens examined

Herb. DAOM: *F676, Quercus sp., Ottawa, Ont., Sept. 1926; *F2278, on red oak, Morton,

Ont., June 1932; F6848, on hardwood stump. Mt. Burnet, Que., Nov. 1935; F6888, onQuercus

sp., Chelsea, Que., Nov. 1935; F10198, on Quercus sp., Ile Perrot, Que., Aug. 1941; 17933, on

Quercus robur, Bavaria, Sept. 1946; 22351, on Eucalyptus sp., Portugal, ex Herb J. Pinto-

Lopes; 52788, on Quercus borealis, Wickham, New Brunswick; 53418, on Quercus sp.,

Gatineau, Que., Nov. 1950; 72046, Fungi Scandinaviae, Ellensvide, Sweden; *72510, Gatineau

Park, Que., Oct. 1961.

Herb, PRE: 1480, Kirstenbosch, C.P., June 1921; 15552, on Quercus stump, ex Herb. L. O.

Overholts; 22846, Falkenberg, Germany, Leg. Plogel; 24207, on Quercus sp. ex Hollos,

Hungarian Fungi; 31394, Stellenbosch; 36573, on stump, Falcourt, Sussex, England, Aug. 1947;

34551, on Quercus sp., Cape Town; 41570, on Eucalyptus stump, ex Herb. J. Pinto-Lopes;

*42366, on decayed hardwood, Pakenham, Ont., Aug. 1962.

Herb. STE: 1397, Kirstenbosch, C.P.; 1674, Kirstenbosch, C.P., July 1924; 2521, Kirstenbosch,

C.P., June 1928; 2742, East London, C.P., Sept. 1932.

Discussion

In the cultures, fibre hyphae are present mainly in the tough, felty, fertile

areas. On the other hand, the nodose-septate hyphae with irregularly thickened

walls, which are so characteristic of cultures of this group, are not very numerous

in the fertile parts of the culture. Instead, these hyphae make up a large proportion

of the soft aerial mycelium where they may develop into solid hyphae with solid

clamp connections.

This description of the cultural characters agrees well with those of Humphrey

& Siggers (1933), Cartwright & Findlay (1946) and Nobles (1948) but the thin-walled

swollen, globose cells, either single or in chains, usually “common in fragile,

cinnamon-buff mycelium from the upper part of the culture in test tube cultures

6 — 8 weeks old’’, reported by Davidson et al. (1942) were not seen.

From the descriptions it is evident that three kinds of hyphae are present

in carpophores of Daedalea quercina, viz.: nodose-septate hyphae with thin walls,

nodose-septate hyphae with irregularly thickened walls and fibre hyphae which

are mostly unbranched. The carpophores also possess extra-hymenial structures

but are on the whole rather simple in construction despite their usual large size.

Thin-walled, nodose-septate hyphae have been reported in the carpophores

of Daedalea quercina by Cunningham (1948 h), Pinto-Lopes & Farinha (1950).

196

Pinto-Lopes (1952), Overholts (1953), Teston (1953 b) and Teixeira (1960). The

nodose-septate hyphae with irregularly thickened walls have been reported from

cultures only by Nobles (1948, 1958 b), but these hyphae, which are so abundant

in the cultures have not been reported form the carpophores before. They are

present in the carpophores in small numbers only and were found only after

prolonged and careful searching. It was noticed in the cultures that these hyphae

were most numerous in areas away from the fibre hyphae and fructifications. It

is therefore possible that these hyphae may be more abundant in the decayed wood,

under the carpophores. In support of this view, it may be added here that large

numbers of such hyphae were seen on the surface and in the vessels of a specimen

of wood decayed by an unknown fungus which was recently examined by the

author. This fungus displayed all the hyphal characters of cultures of species

in group 25 (unpublished data).

Cystidia were not reported from the carpophores by Overholts (1953) but

Bourdot & Galzin (1928) and Talbot (1954 a) figured fusiform, thick-walled cystidia

which they regarded as hyphae projecting into the hymenium from the underlying

tissues. Many of these projecting hyphae were seen in some specimens and some

of these pseudo-cystidia (Lentz, 1954) or tramal cystidia (Donk, 1964) were rather

characteristic in form (Fig. 12 p). They resemble normal fibre haphae in all

characters except in their length and appear to be stunted fibre hyphae. Some

of these structures were seen to be borne in the same position as the basidia on

the same nodose-septate hyphae. These are regarded as skeleto-cystidia and have

thick, refractive walls and narrow, aseptate lumina like fibre hyphae but lack

their length. They fit the description of skeleto-cystidia given by Donk (1964, p.

234) very well.

The fibre hyphae from the carpophores are slightly darker in colour and

generally larger in diameter than those from the cultures but are in other respects

closely similar. As described here, they agree well with the descriptions by

Cunningham (1948 h), Pinto-Lopes (1952), Overholts (1953) and Teston (1953 b)

who stated that the fibre hyphae are more or less straight and unbranched or

rarely branched.

From the above descriptions it is evident that the structures formed in culture

are also present in the carpophores from which they were made. No chlamy-

dospores were seen in the carpophores but they have been reported in hyphae

present in the decayed wood (Cartwright & Findlay, 1946).

Cunningham (1948 h), in his characterization of the genus Daedalea Pers. ex

Fr. stated that Daedalea quecina, the type species, has a trimitic hyphal system

with the “binding hyphae aseptate, commonly of the bovista type”. Teston (1953 a)

and Kotlaba & Pouzar (1957) agreed. Teston (1953 a) 1eported that the binding

hyphae were narrower more tortuous and branched more frequently than the

skeletal hyphae. Teston’s figures (1953 b, P14: 8) show hyphae which resemble

the branches of fibre hyphae as illustrated here in Fig. 12 k. These branches

perform a binding function but they are morphologically similar to and continuous

with the straight fibre hyphae. These branches contribute to the binding system,

as described by Corner (1932 a) in the case of some skeletal hyphae in the fruit-body

of Polystictus xanthopus but they differ morphologically from the true binding

hyphae from the fruit-body of Polystictus xanthopus (Corner 1932 a) or those of

Fomes fomentarius (Teixeira, 1962 a) in their limited branching and unlimited

growth. These branches can thus not be regarded as true binding hyphae. Nor

were binding hyphae of the bovista type as described by Cunningham (1948 h), found

in the tissues of the specimens examined. Donk (1964) furthermore accepted

Corner’s (1953) view that Cunningham’s (1946) ‘binding hyphae of the bovista type”

197

are branched skeletals. This agrees with the observations described above. For

these reasons the fruit-bodies of Daedalea quercina must be regarded as having a

dimitic hyphal system in the sense of Corner (1932 a, b). Cunningham (1946) and

Teixeira (1962 b). This conclusion contradicts the reports by Cunningham (1948 h)

and Teston (1953 b).

Fidalgo (1957) discussed the nomenclatural status of Daedalea Pers. ex. Fr. and

related genera. He concluded that the only distinction between the genera Daedalea

Pers. ex Fr., Lenzites Fr. and Trametes Fr. is in the hymenial configuration, a

character which is so variable in species of this group, as to be without significance.

Daedalea Pers. ex Fr., Trametes Fr. and Lenzites Fr. are thus synonyms in his

view; but comparison between the descriptions of Daedalea quercina on the one

hand and Trametes suaveolens (L. ex Fr.) and Lenzites betulina (L. ex Fr.) Fr.,

the type species of Trametes Fr. and Lenzites Fr. respectively (Donk, 1960) on the

other hand, shows that clear and significant distinctions in hyphal characters and

construction exist between these species, viz.: hyphae with many, short, tortuous,

thick-walled, aseptate branches, arising from thin-walled, nodose-septate hyphae

and binding the tissues together (Fig. 23 d; 26 g) are abundant in the carpophores of

Trametes suaveolens and Lenzites betulina but entirely absent from those of

Daedalea quercina. The nodose-septate hyphae with irregularly thickened walls

which are found in the carpophores, and, more abundantly, in the cultures of

Daedalea quercina are entirely absent from the carpophores and cultures of

Trametes suaveolens and Lenzites betulina. Furthermore, Daedalea quercina causes

a brown rot (Overholts, 1953) and its cultures do not produce extra-cellular oxidase

while cultures of Trametes sauveolens and Lenzites betulina, which cause white rots

(Overholts, 1953), produce extra-cellular oxidase (see Group 45). These differences

in hyphal composition, construction of the carpophores and biochemical activity

of these two species indicate important phylogenetic differences which must

necessarily outweigh all taxonomic considerations based on superficial similarities

such as form, texture and colour of the carpophores. For these reasons the genus

Daedalea Pers. ex Fr. is not congeric with the genera Trametes Fr. and Lenzites

Fr. but constitutes a distinct and well-marked generic entity. Daedalea quercina

thus is the type species of the genus Daedalea Pers. ex Fr. which is characterized

by carpophores having a dimitic hyphal system consisting of nodose-septate

generative hyphae with thin walls, nodose-septate hyphae with irregularly thickened

walls, and sub-solid to solid, aseptate, skeletal hyphae unbranched or occasionally

branched.

These results must influence the taxonomic positions of a large number of

species of polypores because the genus Daedalea Pers. ex Fr. is one of the oldest

genera accepted by Fries (1821). Together with Daedalea quercina, Nobles (1958 b)

included a number of other species with similar cultural characters in Group 25.

Among these were seven species of the genus Coriolellus Murrill, including the type

Coriolellus sepium (Berk.) Murr. Sarkar (1959) in a study of six of these species,

showed that the structures formed in their cultures were also present in their

carpophores so that these six species formed a homogenous group which she

placed in the genus Coriollelus Murr. From her descriptions it is evident that

the hyphal characters of these species of Coriolzllus are very similar to those of

Daedalea quercina as described above. Other carpophore characters such as spore

shape, carpophore texture, upper surface and attachment as well as the type of

decay and host range also agree in many respects. The main differences are the

absence of the daedaloid hymenial surface of Daedalea quercina and the presence

of nodose-septate hyphae with uniformly thickened walls in its carpophores. Such

hyphae were interpreted as “early stages of fibre hyphae” by Sarkar (1959) and

were found in all the species of Coriolellus Murr. described by her. It thus appears

198

that the genus Coriolellus Murr. has so many characlers in common with the type

species of the genus Daedalea Pers. ex Fr. that these six species described by Sarkar

(1959) should be transferred to the genus Daedalea Pers. ex Fr. Certain workers,

however, consider the absence or presence of certain types of hyphae in fruit-bodies

to be of importance at the genus level (Bondartseva, '961: Teixeira, 1962b; Fidalgo

& Fidalgo, 1963, 1966), and these hyphae described as ‘early stages of fibre

hyphae” by Sarkar (1959) are absent from the carpophores of Daedalea quercina.

Although the hyphal characters of too small a number of species have been studied

with sufficient care and accuracy to properly evaluate the significance of a difference

of this nature, it appears that the transfer of these six species of Coriolellus to the

genus Daedalea Pers. ex Fr. by Aoshima (1967), is acceptable.

Fic. 13.— Trametes moesta. (a) Carpophores of PRE 42241 upper and hymenial surfaces;

(b) culture of PRE 42241 at 6 weeks; (c) fructification in culture, x 4.

Trametes moesta Kalchbrenner, Fungi Macowaniana, Grevillea 10, 56, 1881.

Cultural characters

Growth is slow the mat reaching a radius of up to 25 mm in two weeks and

up to 50 mm in six weeks. The margin is bayed and somewhat ragged with the

hyphae raised to the limit of growth, cottony at first, becoming woolly and finely

farinaceous, white, with small patches of felty mycelium developing on the surface

of the agar, the patches increasing in size but becoming granular towards the

inoculum and coalescing to form a dense, felty, mat around the inoculum and

against the sides of the dish. Fertile areas appear as small depressions or irregular

199

ridges on these parts, after 5-6 weeks. At six weeks faint, radiating grooves

may be seen in the younger more woolly mycelium with small, elongated tufts

of dense mycelium over this part. Mat white, with occasionally ‘“‘pale ochraceous

buff” tinges on the felty patches. The reverse is bleached and a faint, slightly

fragrant odour is given off. The oxidase reaction is negative when tested with

gum guaiac solution.

Advancing mycelium: hyphae hyaline, nodose-septate, clamp connections simple,

branching mostly near the septa, 2.2 — 5.5u in diameter, the contents staining

deeply in phloxine (Fig. 14 a).

Aerial mycelium: (a) thin-walled, nodose-septate hyphae as in the advancing zone;

(b) nodose-septate hyphae with irregularly thickened walls and contents staining

deeply in phloxine, branching or unbranched 1.5 — 5.0u in diameter (Fig. 14 b);

sub-solid with slight beadlike swellings with prominent lumina in those parts and

the ends usually thin-walled, 1.5 — 5.2u near the origin and widening gradually

to 4 — 5u at the widest part; others narrow, 0.7 —— 1.0: in diameter for some

distance then widening suddenly to 4 — 5u with the lumina more prominent and

continuous or in a series of ellipsoidal spaces with deeply staining contents

(Fig. 14 c).

Fructification: (a) thin-walled, nodose-septate hyphae as above; (b) fibre hyphae

as in aerial mycelium but usually narrower, 2.5 — 3.5u: (c) basidia long clavate

20 — 27 x 5.5 — 7.5u with 4 straight sterigmata, 3 — 4u long (Fig. 14 d); (d)

basidioles 1.2 — 4u wide, often with narrow branches, thin-walled, hyaline,

arising from the basidial fasicles (Fig. 14 e); (e) basidiospores hyaline, long

ellipsoidal to cylindrical and flattened on one side, with a marked apiculum, smooth,

thin-walled, 6.0 — 8.5 x 3.0 — 4.2u (Fig. 14 f).

Submerged mycelium: a thin-walled, nodose-septate hyphae as in the advancing

zone; (b) chlamydospores abundant, hyaline, subglobose to ellipsoidal, intercalary

or terminal, thick-walled, borne on thin-walled, nodose-septate hyphae 4.5 x 6.0 —

8.0u (Fig. 14 g).

Carpcphore characters

Carpophores perennial, lignicolous, solitary or grouped, sessile, dimidiate;

pileus applanate to thick convex, single, laterally connate or imbricate, rigid, hard,

corky, up to 12 x 7 x 3 cm; surface at first finely tomentose to sub-glabrous,

smooth or somewhat rugulose and slightly rimose in age, at first creamy white,

darkening to “‘pinkish buff” later to ‘“‘avellaneous” in mature specimens or blackish

in oldest parts; margin obtuse, thick, entire, creamy white when fresh, darkening to

“pale pinkish buff’ or ‘pinkish buff’; pore surface creamy white when fresh

drying to somewhat dirty white or umber in older parts, poroid to daedaloid; pores

elongate, angular, 0.5 — 1.0 mm wide; dissepiments dentate, thin; tubes whitish

0.5 — 12 mm deep, becoming stuffed with white hyphae; context “wood brown”,

zonate, fibrous. | — 10 mm thick, darkening with KOH.

Hyphal characters: (i) nodose-septate hyphae hyaline, branching, forming H-con-

nections, thin-walled, contents staining in phloxine, 1.2 — 3.0u in diameter (Fig.

14 h); (ii) nodose-septate hyphae with irregularly thickened walls, lumina irregularly

narrowed and deeply staining contents, branching ana forming H-connections.

2.4 — 4.5u in diameter, rare (Fig. 14 k); (iii) fibre hyphae long, sub-hyaline to pale

brownish, straight or flexuous, unbranched or with one to three branches, narrow

near ‘he origin but widening towards the middle, walls thickened, refractive, lumina

narrow or obliterated often visible as interrupted lines, prominent towards the

FIGURE (4.

. 14— Trametes moesta. a - g. Structures from cultures: (a) thin-walled, nodose-

septate hyphae from advancing zone; (b) nodose-septate hyphae with irregu-

larly thickened walls; (c) fibre hyphae; (d) basidia; (e) basidioles; (f) basidio-

spores; (g) chlamydospores.

h-q. Structures from carpophores: (h) thin-walled, nodose-septate hyphae;

(k) nodose-septate hyphae with irreguarly thickened walls; (m) fibre hyphae;

(n) basidia; (p) basidiospores; (q) tramal cystidium.

201

extremities, aseptate or with occasional simple septa near the tips, 2.5 — 6.0p

in diameter (Fig. 14 m).

Hymenium: basidia long clavate, 22.0 — 30.0u long, bearing four, straight, slender

sterigmata 2.5 — 3.0u (Fig. 14 n); basidiospores hyaline, cylindrical, thin-walled,

smooth 7.2 — 9.8 x 3.3 — 4.2u (Fig. 14 p); tramal cystidia sub-hyaline, tapering

towards the tips, thick-walled, with narrow lumina, 3.0 — 6.0u in diameter

projecting up to 40u beyond sub-hymenium, arising from trama as widened

terminal portion, 50 — 80u long, of short, narrow, fibre hyphae 1.5 — 3.0u in

diameter (Fig. 14 q).

Construction. At the margin the tissues consist of long, unbranched, hyaline fibre

hyphae usually with prominent lumina and thin-walled extremities, often collapsed

3.0 — 4.0u in diameter. These fibre hyphae are arranged more or less parallel

to the direction of growth of the carpophore and loosely intertwined with each

other and the numerous narrow, branching, thin-walled, nodose-septate hyphae

from which they arise. In the older parts of the context the construction

is similar but the fibre hyphae are darker in colour, their walls are thicker, the

lumina are often reduced to narrow interrupted lines and the hyphae are of greater

diameter, up to 6.0u. Few nodose-septate hyphae are present. In the upper context

the fibre hyphae are mostly sub-solid, more or less parallel to and intertwined

with each other and with their ends closely packed at a common level to form

the finely pubescent upper surface. Near the upper surface thin-walled, nodose-

septate hyphae are fairly numerous, intertwined with the fibre hyphae and branching

freely between them and across their direction of growth. In the older tissues

thin-walled, nodose-septate hyphae are rather rare. In the lower context some

fibre hyphae turn downwards to form the trama of the dissepiments. The fibre

hyphae are somewhat narrower, 2.4 — 4.5u, with wider lumina than in the upper

context and more flexuous, often with a somewhat beaded appearance and many

with one to three long branches, of similar appearance. These hyphae and their

branches are tightly intertwined with each other as ‘well as thin-walled, branching,

nodose-septate hyphae and occasional nodose-septate hyphae with irregularly

thickened walls to form the dense tissues of the lower context and trama. The

nodose-septate, thin-walled hyphae become very numerous by repeated branching

towards the hymenial surfaces where abundant, short, narrow branches produce

the basidia. From the tramal tissues short fibre hyphae with narrow lower portions

which suddenly increase in diameter towards the upper part (Fig. 14 q) and

50 — 90u in length, project into the hymenium as tramal cystidia. Some branches

of fibre hyphae may also project into the hymenium.

Decay and hosts

This species causes a brown rot of stumps of hardwood trees.

Specimens examined

Herb. PRE: 11288, coll. A. Roberts, May 1915; 34391, on indigenous hardwood, Hluhluwe

Game Res., Oct. 1935; *42241, on Acacia mearnsii stump, Kaapse Hoop, Tvl., Feb. 1961;

*42242, on Acacia mearnsii stump, Kaapse Hoop, Tvl., Feb. 1961; *42442, on Acacia

mearnsii stump, Kaapse Hoop, Tvl., Feb. 1961.

Herb. STE: 538, as Daedalea moesta Kalchbrenner.

Interfertility studies

In order to test the possibility of conspecificity between this species and

Daedalea quercina, which is very similar to Trametes moesta, four cultures made

from single spores obtained from a fructification formed in culture by Trametes

moesta PRE 42241, were paired on agar slopes in all possible combinations with

202

four cultures made from single spores of Daedalea quercina DAOM 2278. Four

days after the mycelia had met on the slopes, the cultures were examined for the

presence of clamp connections.

No clamp connections were found in any of the cultures thus indicating that

Daedalea quercina and Trametes moesta are two different species.

Discussion

This species was described by Kalchbrenner (loc. cit.) from a collection

by Tyson which could not be located for examination The specific epithet

of the specimens examined in this study is based on Van der Bijl’s description

(1922 a), the collection PRE 11288 cited by him, and collection No. 538, Daedalea

moesta Kalch. in the P. A. van der Bijl Herbarium, University of Stellenbosch.

The other specimens cited above agree very well with Van der Bijl’s description

and specimens.

This species is not well-known in South Africa and the collections cited above

are the only records of its occurrence. The three collections from Kaapse Hoop,

are probably part of the same population of this fungus in that region since they

were made from different hosts in a fairly small area. It is probable that the

species may be much more widely distributed than these records would indicate.

The cultural characters of Trametes moesta had not been described before

In culture the fungus forms nodose-septate hyphae, some with irregularly thickened

walls, and fibre hyphae, whilst no extra-cellular oxidase is produced. This species

thus displays all the characteristics of species included in Group 25 by Nobles

(1958 b).

In cultural characters Trametes moesta resembles Daedalea quercina very

strongly but differs from it in a slower growth rate, even margin, generally smoother

topography of the mat and the formation of granular, fertile areas from which

ridges more delicate than those in cultures of Daedalea quercina, arise. Cultures

of Trametes moesta differ from those of species of the genus Coriolellus Murr., as

described by Sarkar (1959) in the absence of the refractive projections from their

nodose-septate hyphae with irregularly thickened walls. In most respects their

cultures appear to be strikingly similar however, but the differences mentioned here

may serve to distinguish cultures of Trametes moesta from those of other species

if considered together with host and locality records if available.

Van der Bijl (1922 a) described the hyphae in carpophores of Trametes moesta

as “‘simple, 2 — 4u in diameter’. This agrees to some extent with the description

given above as the bulk of the hyphae in the carpophore are thick-walled, aseptate,

fibre hyphae mostly unbranched or occasionally branched. Nodose-septate, thin-

walled hyphae are abundant in the growing margin, tramal tissues and, to a lesser

extent, near the upper surface of the carpophore. Nodose-septate hyphae with

irregularly thickened walls were found in small numbers in the lower context and

tramal tissues. Since hyphae with clamp connections at their septa must be

regarded as generative hyphae, according to Corner (1953), Teixeira (1962 b),

and Donk (1964), only two kinds of hyphae, generative and skeletal hyphae are

present in the carpophores of Trametes moesta. This species thus has a dimitic

hyphal system (Corner, 1932 a, b; Cunningham, 1946, 1954).

The structures that were found in the cultures of Trametes moesta were also

present in the carpophores. The fibre hyphae were somewhat larger in diameter

in the carpophores than in the cultures and were pale brownish rather than

hyaline or sub-hyaline as in the cultures. The thin-walled, nodose-septate hyphae

were abundant in the margin and dissepiments of the carpophores but the nodose-

septate hyphae with irregularly thickened walls, were seen only rarely and occurred

203

mostly in mounts made from the olde1 parts of the lowe1 context just above the

dissepiments.

The peculiarly branched structures seen in the hymenia of some fructifications

formed in culture, were not seen in the carpophores. It was noticed at the time

that the cultures in which these occurred, showed signs cf dessication. Since these

structures were formed on the basidial hyphae in the hymenium, they were regarded

as deformed basidia fermed under dry conditions as described by Bose (1943) in

carpophores of Polyporus sanguineus and Ganoderma lucidum.

The carpophores of Trametes moesta are strikingly similar to those of Daedalea

quercina in hyphal composition, construction and morphology. The types of

hyphae found in Daedalea querciza were also found in carpophores of Trametes

moesta. In both species the fibre hyphae, which make up the bulk of the

carpophores are arranged in parallel and slightly intertwimed and unbranched in

the upper context. They are more frequently branched and tightly interwoven

in the lower context and dissepiments. In both species short fibre hyphae or

branches of fibre hyphae project as tramal cystidia from the tramal tissues of the

carpophores. Both species cause brown rots in hardwood stumps or logs, but the

non-appearance of clamp connections when single spore mycelia of these two species

were mated, indicate that they are not conspecific.

_ Although carpophores of these two species are so strikingly similar in mor-

phology and anatomy, small but consistent differences are present, viz.: the upper

surface of carpophores of Trametes moesta have pale reddish-brown colours not

common in Daedalea quercina; the pore surface of Trametes moesta is mostly

poroid with the angular pores much smaller and dissepiments more delicate than

those of Daedalea quercina. The basidiospores of Trametes moesta are longer

and more markedly cylindrical than those of Daedalea quercina. Differences of

this nature do not outweigh the great similarity in cultural characters and carpo-

phore anatomy and are regarded by most workers as of interspecific value only.

Therefore, these two species must be congeneric and Trametes moesta Kalch.

should be transferred to the genus Daedalea Pers. ex Fr. as typified by Daedalea

quercina L. ex Fr.

Trametes roseola Patouillard & Hariot, in Journal de Botanique 14, 239, 1900.

Cultural characters

Growth is slow, the mat reaching a radius of 27 mm in two weeks and covering

the plate in 5 weeks. The margin is even, with mycelium appressed for about 1 mm,

then raised, white, cottony behind the margin but becoming woolly towards the

inoculum. After 3 weeks small, rounded, lumps of dense mycelium form along a

narrow zone and on the sides of the plate, later covering the younger part of

the mat in distinct zones of pebbly mycelium alternating with zones of smooth, felty

mycelium. After six weeks the plates are covered, the mats raised, woolly, with

a deep, wide, concentric groove in the newest growth and with successive narrow,

concentric coarsely farinaceous to pebbly zones, often with “light ochraceous buff

colours and traversed by shallow radial grooves. About halfway across the mat

and towards the inoculum, the texture is cottony-woolly. Mat is white at first,

but turns a very pale “‘seafoam yellow” colour. Lumps of compact mycelium

“light ochraceous buff” or ‘‘cinnamon buff” or “dresden brown” in colour appear

on the surface after 3 — 4 weeks and later may develop minute pores over the

surface. The reverse is unchanged at first but bleaching after 4 — 5 weeks. A

faint, fragrant odour is given off. The oxidase reaction is negative when tested

with gum guaiac solution.

SHLLU GUL 40 HNEIOASERATAEEHUMOOAE UALR. ath

Fic. 15.— Trametes roseola. (a) Carpophore of type specimen, upper surface

and (b) hymenial surface; (c) culture of PRE 42443 at 6 weeks.

Advancing mycelium: hyphae hyaline, simple or branching near the septa, nodose-

septate with deeply staining contents 2.2 — 4.5u in diameter (Fig. 16 a).

Aerial mycelium: (a) hyphae as in the advancing zone; (b) nodose-septate hyphae

with irregularly thickened walls, often solid and refractive in parts, branched or

unbranched 2.5 — 4.5 in diameter (Fig. 16 b); (c) fibre hyphae hyaline at first,

later subhyaline, unbranched or very occasionally branched, walls thick and

refractive, lumina narrow, widening near the tips, aseptate or occasionally with

one or two simple septa, 2.5 — 3.5u, arising from thin-walled, nodose-septate

hyphae and of fairly slow growth (Fig. 16 c).

Submerged mycelium: (a) hyphae as in the advancing zone; (b) nodose-septate

hyphae with irregularly thickened walls as in aerial mycelium; (c) chlamydospores

terminal or intercalary, ovoid or ellipsoid with thick, refractive walls 18.0 — 30.0 x

9:0 — 12.0u (Fig. 16 d).

Carpophore characters

Carpophore annual or perennial, lignicolous, solitary. sessile, pileus conchate

to somewhat spathulate, soft corky, drying to corky, up to 5.5 x 6.0 x 1.2 cm;

surface finely pubescent, smooth, mat, azonate, ““pale ochraceous buff” to “‘light

pinkish cinnamon” or becoming fuscous in parts in older specimens; margin obtuse,

entire, thick and rounded, concolorous with upper surface; pore surface “‘seashell

pink” darkening to “‘vinaceous cinnamon” or “buff brown” and cracking on

drying; pores rounded or slightly angular, 5 — 8 mm, dissepiments even; tubes

pale yellowish, 0.5 — 1.0 mm deep, stratified in some specimens, decurrent at point

of attachment; context pale ‘‘ochraceous buff” to “‘light pinkish cinnamon” floccose

or somewhat fibrous, with occasional concentric, darker zones, and darkening in

KOH,

205

eae a UO

FIGURE 16.

Fic. 16.— Trametes roseola. a - d. Structures from cultures: (a) thin-walled nodose-

septate hyphae from advancing zone; (b) nodose-septate hyphae with irregularly

thickened walls; (c) fibre hyphae; (d) chlamydospore.

e-k. Structures from carpophores: (e) thin-walled, nodose-septate hyphae;

(f) nodose-septate hyphae with irregularly thickened walls; (g) fibre hyphae:

(h) basidia; (k) basidiospores.

206

Hyphal characters: (i) nodose-septate hyphae hyaline, thin-walled, with deeply

staining contents and branching near the septa 2.0 — 3.0u in diameter (Fig. 16 e);

(ii) nodose-septate hyphae with walls irregularly thickened and refractive, hyaline,

branching near the septa, forming H-connections, 3.0 — 6 Ou in diameter (Fig. 16 f);

(iii) fibre hyphae sub-hyaline to pale brownish, straight or flexuous, mostly

unbranched, occasionally branched, thick-walled the lumina prominent or narrow,

seldom occluded, aseptate or occasionally with one or two simple septa near the

thin-walled tips, narrow and thin-walled towards the origin, arising from thin-walled,

nodose-septate hyphae, 2.2 — 7.0u in diameter (Fig. 16 g).

Hymenium: basidia hyaline, clavate 8.0 — 15.0 x 4.5 — 6.0u with short, straight

sterigmata 1.5 — 2.2u (Fig. 16 h); basidiospores hyaline. long ellipsoidal to cylin-

drical, smooth, thin-walled 4.8 — 6.0 x 2.2 — 2.8u (Fig. 16 k).

Construction. The margin consists of long, straight, unbranched fibre hyphae,

walls subhyaline to pale brownish, arranged more or less parallel to or somewhat

intertwined with each other, their tips projected forward to form the margin of the

pileus. Branching, narrow, hyaline, thin-walled, nodose-septate hyphae, from which

the fibre hyphae arise, are intertwined and interwoven with them just behind the

margin. Towards the upper context the fibre hyphae are darker in colour, with

thicker walls and are of larger diameter than in the margin but also parallel to

each other and turning upwards towards the upper surface where their thin-walled

ends are closely packed at a common level to form the finely pubescent to sub-

glabrous upper surface. Just below the upper surface, narrow, thin-walled,

branching, nodose-septate hyphae, intertwined with the fibre hyphae, are fairly

numerous. In the older tissues few nodose-septate hyphae are present. The lower

context consists of long fibre hyphae, straight or flexuous, 3.0 — 4.0u in diameter,

occasionally branched, walls thickened, but lumina prominent, more or less parallel

to one another and slightly intertwined with one another and with small numbers

of thin-walled, branching, nodose-septate hyphae from which they arise and occa-

sional nodose-septate hyphae with irregularly thickened walls, 3.0 — 6.0 in

diameter. From the lower context the hyphae turn downwards into the dissepiments

where the fibre hyphae are narrower and become very flexuous and more tightly

interwoven than higher up in the context. Thin-walled, nodose-septate hyphae

with deeply staining walls are numerous, branching frequently and anastomosing

parallel to and across the direction of growth of the fibre hyphae and turning

outwards toward the pore surfaces with increased branching to form the tightly

interwoven, short, nodose-septate branches of the sub-hymenium bearing the basidia.

Also in the dissepiments are occasionally portions of nodose-septate hyphae with

irregularly thickened walls, continuous with the thin-walled, nodose-septate hyphae.

Decay and hosts

Trametes roseola causes a brown rot of broad-leaved trees. Hopkins (1939,

1943) reported this species as the cause of stem rot of living trees in Rhodesia.

Specimens examined

Herb. PRE: 26709, on Eucalyptus ficifolia, Pietermartizburg, 1934; 28560, on Acacia mearnsii,

Impolweni Natal, Sept. 1934; 30191, on Acacia mearnsii, Melmoth, March, 1935; 30634, on

living Prunus persica, Rhodesia, June 1937; 30204, on Acacia mearnsii wood, Melmoth,

Natal 1935; 24116, on dead wood, Albert Falls, Natal; 39016, on dead wood, ex Herb. Timber

Res. Lab., Johannesburg; *42443, on decayed hardwood log, Bushbuckridge, Tvl., Feb. 1961.

Herb. Patouillard in FH: Sheet no. 2853, 3 collections in packets marked: “Caraban

(Casamance) Leg. Chevalier’; “313, sur branche mort, Reserve foresti¢ér de Compong Chnang,

Cambodge, Juillet 1921, M. Perclot;” “Madagascar, M. Decary 1920.”

The specimen from Caraban on Patouillard’s sheet no. 2835, agrees with the

collection data mentioned in the original description (Patouillard & Hariot, 1900)

and must therefore be designated the type specimen (Fig. 15 a, b).

207

Discussion

The presence of aseptate fibre hyphae and nodose-septate hyphae, some with

irregularly thickened walls, and the absence of extra-ceilular oxidase enzymes in

its cultures, place Trametes roseola in Group 25 (Nobles, 1958 b). Its cultural

characters agree well with those of other species in this group but the pale yellow-

ish-green colour of the mycelial mat, the colour of the poroid, felty lumps in the

cultures and the slow-growing fibre hyphae are unique in cultures of Trametes

roseola and serve to distinguish this species from others in Group 25. This species

had not been described in culture before.

The South African collections of this fungus agree very well in morphological

and hyphal characters with Patouillard’s collections, one of which is designated

as the type. All the carpophores were rather small, with minute pores barely

visible to the naked eye, fairly thick dissepiments and with a soft, ‘“‘trametoid”

feel and appearance. To the type specimen is attached a small piece of hardwood,

from which it grew, showing a characteristic brown rot. No spores could be found

on the type specimens but young basidia were very numerous. There is no doubt

that the South African specimens are conspecific with Patouillard’s collections.

From the description it is evident that the carpophores of Trametes roseola

are simple in construction. They consist of three types of hyphae only, viz.

thin-walled, nodose-septate hyphae, nodose-septate hyphae with irregularly thickened

walls and fibre hyphae. The fibre hyphae are mostly unbranched and lightly

intertwined. The nodose-septate hyphae are branched but not tightly interwoven

with the fibre hyphae. There is an almost total absence of a binding function in

the hyphal elements so that the fruit-bodies feel soft and somewhat fragile.

From the descriptions, it is evident that the vegetative structures formed in

culture are also present in the carpophores. As in other species of this group,

the nodose-septate hyphae with irregularly thickened walls were very numerous and

prominent in the cultures but rare in the carpophores. In one carpophore, PRE

42443, however, these hyphae were fairly abundant in a narrow zone between

two layers of tubes and towards the middle of the carpophore where they were

visible as a faintly greenish patch in the otherwise apricot-coloured context tissue.

In gross morphological features and texture of its carpophores, Trametes

roseola resembles Trametes suaveolens (L. ex Fr.) Fr., the type of Trametes Fr.,

very closely. Comparison of their hyphal characters, however, reveals that this

resemblance is entirely superficial since the short, much branched fibre hyphae

(or binding hyphae), present in carpophores of Trametes suaveolens, are absent

from the carpophores of Trametes roseola. Nodose-septate hyphae with irregularly

thickened walls are present in carpophores and cultures of Trametes roseola but

not in those of Trametes suaveolens.

The similarities in cultural characters and hyphal characters of their carpo-

phores indicate close affinities between Trametes roseola, Daedalea quercina and

species of the genus Coriolellus Murr. described by Sarkar (1959). The small

fruit-bodies of these Coriolellus species however, contain hyphae described by Sarkar

as “immature fibre hyphae”’ or ‘incompletely differentiated fibre hyphae.” These

hyphae, from her figures, are thick-walled, nodose-septate or “‘sclerified generative

hyphae”’ (Donk, 1964) which are also formed in cultures of these fungi. Such

hyphae are absent from cultures and carpophores of both Trametes roseola and

Daedalea quercina. Because it is not known at present whether the formation

of thick-walled, nodose-septate hyphae, in species of which the carpophores

consist mainly of aseptate fibre hyphae, takes place as a result of the influence of

208

environmetal factors or genetic factors, it appears to be advisable not to group

Trametes roseola with these species of Coriolellus Murr. Donk (1966) recently

transferred these Coriolellus spp. described by Sarkar (1959), to the genus Antrodia

Karsten. The type species of Antrodia Karst., Trametes mollis (Sommerf.) Fr.,

however, lacks nodose-septate hyphae with irregularly thickened walls in its

cultures. Nobles (1958 b) placed cultures of this species in her Group 48 which

differ from cultures of Group 25 by the presence of a brown mycelial mat and the

production of extra-cellular oxidase enzymes.

Although the cultural characters and hyphal characters of the fruit-bodies of

Trametes roseola resemble those of Daedalea quercina in so many respects, their

fruit-bodies do not appear to be so markedly similar in gross morphology. Fruit-

bodies of Trametes roseola are smaller, of different colour and softer in texture

than those of Daedalea quercina and have small pores rather than daedaloid

dissepiments. They also lack tramal or skeleto-cystidia. On the other hand, the

fruit-bodies of Trametes roseola are constructed in the same way and of the same

types of hyphae as those of Daedalea quercina and further have the same thick

dissepiments, anoderm surface, cylindrical spores and context darkening in KOH.

Furthermore, fruit-bodies of Trametes roseola with a daedaloid hymenial surface

were figured and reported by Lloyd (1922, p. 1145) from North Borneo. The

differences in gross morphology of the carpophores of these two species thus appear

to be of minor importance. Many workers (e.g. Teixeira, 1962 b; Furtado, 1965

a, b) regard hyphal characters as important at the generic level. As the carpophores

of these two species agree in so many hyphal and micromorphological characters,

the differences between them appear to be of interspecific nature only and Trametes

roseola appears to be congeneric with Daedalea quercina, the type species of the

genus Daedalea Fr.

Fic. 17.—Fomes cajanderi. (a) Carpophores of DAOM 31973; (b) culture of DAOM

31973 at 6 weeks.

Fomes cajanderi Karsten, Finska Vet.-Soc. Ofv. Forh. 46 (11), 8, 1904.

Trametes subrosea Weir, Rhodora 25, 217, 1923;

Fomitopsis subrosea (Weir) Bond. & Sing., Ann. Mycol. 39, 55, 1941;

Fomitopsis cajanderi (Karst.) Kotlaba & Pouzar, Ceska mykologie 9, 157, 1957.

Cultural characters

Growth moderately fast to slow the colony reaching a radius of about 40 mm

in two weeks covering the plates in 4 to 5 weeks. Margin even to slightly bayed

with mycelium raised to limit of growth. Behind the margin mat is thin, cottony,

raised but collapsing to sub-felty, or, more compact to almost velutinate around

the inoculum, white at first but soon developing “seashell pink” to ‘“‘pale salmon

color” tints near the inoculum. Later the areas of more compact mycelium develop

irregular, pellicular-felty patches which gradually enlarge and coalesce to form

smooth or vaguely, radially, grooved patches of raised, felty mycelium on which

angular pores, labyrinthiform at first, develop. These patches gradually become

seashell pink”, “pale congo pink” or “hydrangea pink’’, expanding continually,

with the older, coloured areas darkening gradually to “‘vinaceous pink’ or

“vinaceous fawn’, and ““Roods brown”’ in the oldest tubes. Fruiting areas enlarge

gradually by the formation of new tubes around the periphery. The reverse

remains unchanged and a faint, sweet odour is emitted. No diffusion zones are

formed on gallic acid and tannic acid agar but colonies op to 3.0 cm on the former

medium and up to 1.5 cm on the latter, are formed. A negative reaction is

obtained when gum guaiac solution is applied to the culture.

Advancing mycelium: hyphae hyaline, nodose-septate, branching at or near the

septa, with contents staining deeply, 1.5 — 3.5u in diameter (Fig. 18 a).

Aerial mycelium: (a) thin-walled hyphae as in the advancing zone, 1.2 — 3.0 —

(3.5) u; (b) nodose-septate hyphae with walls irregularly thickened and refractive

and occasionally with refractive projections, lumina irregularly narrowed and

staining deeply 1.5 — 4.5u in diameter, branching freely, numerous in the pellicular

areas (Fig. 18 b); (c) fibre hyphae long, straight, unbranched, sub-hyaline or hyaline,

solid with lumina visible at the narrower thin-walled ends or sub-solid with very

narrow lumina, aseptate, up to 3.0u in diameter along middle portion (Fig. 18 c):; (d)

solid, refractive, branching hyphae with hyaline walls and prominent, solid clamp

connections, the lumina lacking or reduced to an interrupted line 1.5 — 3.0u in

diameter arising from thin-walled, nodose-septate hyohae or nodose-septate hyphae

with irregularly thickened walls (Fig. 18 d).

Fructifications: (a) basidia clavate 10.5 — 18.0 x 4.2 — 5.lu with four slender

sterigmata 2.4 — 3.lu (Fig. 18 e); (b) basidiospores long-cylindrical or allantoid,

obliquely apiculate, hyaline, smooth, thin-walled 4.8 — 6.0 x 1.6 — 2.1u (Fig.

18 f); (c) fibre hyphae as in the aerial mycelium; (d) nodose-septate hyphae with

irregularly thickened walls as in aerial mycelium rare.

Submerged mycelium: (a) thin-walled, nodose-septate hyphae as in the advancing

zone 1.5 — 3.0u in diameter; (b) nodose-septate hyphae with irregularly thickened

walls as in the aerial mycelium 1.5 — 6.0u; (c) chlamydospores rare, intercalary

or terminal, ovoid to ellipsoidal, walls slightly thickened, 8.0 — 20.0 x 6.0 — 8.0u

Carpophore characters

Carpophore annual or reviving a second season, lignicolous, solitary or com-

pound, sessile or effused-reflexed; pileus conchate to applanate, imbricate, often

laterally connate, coriaceous to corky, drying rigid, up te 5.0 — 10.5 x 1.7 cm;

upper surface at first velvety tomentose but later radially fibrillose or nearly

glabrous, smooth or somewhat rugose often zonate, pinkish red at first but soon

pinkish brown in age and occasionally with a thin, dark, brittle crust; margin acute.

thin, entire, concolourous with upper surface; pore surface “‘vinaceous”’ to “orange

vinaceous”’, poroid; pores rounded or angular 3 — 5 mm, dissepiments even;

tubes whitish, up to 3 mm deep, stratified; context soft corky, “hydrangea pink

to “congo pink’’, indistinctly zonate, up to 12 mm thick.

Fic.

eas ew |

FIGURE 18.

18.— Fomes cajanderi. a - f. Structures from cultures: (a) hyphae from ad-

vancing zone: (b) nodose-septate hyphae with irregularly thickened walls; (c)

fibre hyphae; (d) thick-walled or subsolid, nodose-septate hyphae; (e) basidia;

(f) basidiospores.

g-n. Structures from carpophores: (g) thin-walled, nodose-septate hyphae;

(h) nodose-septate hyphae with irregularly thickened walls; (k) fibre hyphae;

(m) basidia; (n) basidiospores.

Zul

Hyphal characters: (i) nodose-septate hyphae thin-walled, hyaline, branching at or

near the septa, with deeply staining contents, 1.5 — 3.0u in diameter (Fig. 18 g);

(ii) nodose-septate hyphae with irregularly thickened walls hyaline, branched or

unbranched, rare, 2.4 — 4.5u in diameter (Fig. 18 h): (iti) fibre hyphae long,

unbranched, or with one or two branches towards the tip, straight or tortuous,

thick-walled, sub-hyaline to pale brownish, lumina narrow, aseptate, occasionally

occluded, always prominent towards the ends which are thin-walled and often

collapsed, 1.5 — 5.0u in diameter (Fig. 18 k).

Hymenium: basidia narrowly clavate, 12.0 — 15.0 x 4.0 — 5.5u bearing four

short sterigmata 2.1 — 2.4u (Fig. 18 m); basidiospores long narrow-cylindrical to

somewhat allantoid, apiculate, hyaline, smooth, thin-walled. 5.0 — 7.0 x 1.8 — 2.4u

(Fig. 18 n).

Construction. At the margin the carpophore consists mainly of long, straight,

unbranched, sub-hyaline fibre hyphae with prominent lumina and_ thin-walled

ends sometimes collapsed, arranged parallel to or slightly intertwined with each

other and with numerous, narrow, branching, thin-walled. nodose-septate hyphae

from which they arise. Behind the margin and in upper context the fibre hyphae

have thicker and darker walls and turn upward towards the upper surface but are

still more or less parallel to and slightly intertwined with one another and thin-

walled, branching, nodose-septate hyphae, mostly empty and collapsed. Occasional

lengths of nodose-septate hyphae with irregularly thickened walls are also present,

intertwined with the others. At the upper surface the ends of the fibre hyphae

may be arranged parallel to each other and packed at a common level to form

a finely pubescent surface or, frequently the ends may be tangled and intermingled

with numerous, tortuous, thin-walled, nodose-septate hyphae and agglutinated

with a thin layer of lacquer-like substance into a glabrous trichoderm with resinous

crust (Lohwag, 1940). Subsequent growth of the hyphae may result in a succession

of similar layers which may be up to 500u thick. In the lower context the fibre

hyphae are similar to those in the upper context but turn downwards into the trama

of the dissepiments. Towards the tramal tissues the fibre hyphae become generally

more tortuous, one or two branches are often formed, the branches being long

and similar to the parent hyphae. All hyphae become more tortuous and tightly

interwoven and bound into a dense tissue. Below the context towards the trama,

thin-walled nodose-septate hyphae with deeply staining contents and branching

frequently and repeatedly, become increasingly numerous and tightly intertwined

with the fibre hyphae. Nodose-septate hyphae with irregularly thickened walls,

some apparently solid, are also fairly numerous in this region and intertwined

with the other hyphae. The thin-walled, nodose-septate hyphae branch repeatedly

towards the surface of the dissepiments where the basidia are borne in clusters

on their numerous branches, 1.8 — 2.4u in diameter. No accessory structures are

present.

Decay and hosts

Fomes cajanderi causes a brown rot of coniferous wood.

Specimens examined

Herb. DAOM: *10278, on Picea mariana, Champlain Co., Que.; 17029, on coniferous stump.

St. Aubert, Que.; 17164, on Pseudotsuga taxifolia, Saanichton, B.C., No. 1959; *17522, on

Picea mariana slash, Lake Sasiginigate. Man., Aug. 1947; 17528, on Picea glauca log,

Wasagaming, Man., Sept. 1947; *17529, on Picea glauca log, Wasagaming, Man., Sept.

1947; 17572, on Picea sp., Rocky Mt. House, Alta., Oct; 22380, on Picea sp., Harricanaw

Riv., Que., June 1946; 22729, on Abies balsemea, Tweedie Brook, N.B., July 1949; 30061,

on Picea glauca, Riding Mt. Nat. Park, Man., July 1950; 31849, on Pseudotsuga taxifolia,

Cathedral Grove, B.C., May 1948; *31973, on Tsuga canadensis, Warrensburg, N.Y., Oct.

212

1955; 53725, on Picea sp., Victoria Park, N.S.; *72322, on Pseudotsuga taxifolia, Beacon

Hill, Vict., B.C.; *72652, on Picea mariana, Warrensburg, N.Y., Sept. 1961; 72742, on

Picea glauca, Laird River, N.T.; *73183, on coniferous log, S. Santion Highway, Oregon,

Aug. 1962.

Discussion

Cultural characters of this species as described bere, agree well with the

descriptions by Campbell (1938), Davidson et al. (1938), Cartwright & Findlay

(1946) and Nobles (1948, 1958 b). It fits well into Group 25 and its cultures

differ from those of other species in this group mainly in the presence of pinkish

colours.

The hyphae of the carpophores of this species have been described by Overholts

(1953) who stated: “hyphae pale brown in KOH, long and flexuous, simple, with

no cross walls or clamps 2.5 — 5.0u in diameter’. Lowe (1957) later stated that

these hyphae were “mixed with a small amount of thin-walled, clamped hyphae,

3 — 5u in diameter’. Farinha (1946) reported clamped hyphae with walls very

slightly thickened up to 6u in diameter in addition to thick-walled, aseptate

occasionally branching hyphae from carpophores of this species. None of these

authors mentioned the presence of nodose-septate hyphae with irregularly thickened

walls in the carpophores as described here and first reported from cultures by

Nobles (1948).

From the above descriptions it is evident that the structures formed in the

cultures are also present in the carpophores with the exception of the chlamydo-

spores. These may probably be found in decayed wood associated with the

carpophores as in the case of Daedalea quercina (Cartwright & Findlay, 1946).

As in the other three species described here in Group 25, the nodose-septate hyphae

with irregularly thickened walls were not very abundant in carpophores of this

species either but were nevertheless present in sufficient numbers in the older

tissues above the pores to ensure their rapid detection. Basidia and spores are

virtually identical in both cultures and carpophores.

The hyphal characters and construction of the carpophores of Fomes cajanderi

Karst. resemble those of Daedalea quercina L. ex Fr. as described above, quite

closely. This resemblance indicates a close phylogenetic relationship between these

species although morphological differences between them are evident. Fomes

cajanderi has a poroid hymenium, rose-coloured context and a type of upper

surface not found in Daedalea quercina. In these characters, Fomes cajanderi and

Daedalea quercina appear to be of interspecific importance only and are outweighed

by the similarity in hyphal characters and construction with the carpophores of

Daedalea quercina.

Fomes cajanderi has been described as Trametes subrosea by Weir (1923)

but comparison of its hyphal characters with those of Trametes suaveolens (L.

ex Fr.) the type of Trametes Fr. (Donk, 1960) reveals important differences.

Carpophores of Trametes suaveolens have much branched, fibre hyphae with short

tortuous branches, (binding hyphae, Corner, 1932 a; Cunningham, 1946, 1954) in

addition to the thin-walled, nodose-septate hyphae and unbranched fibre hyphae.

Carpophores of Fomes cajanderi lack ‘“‘binding hyphae” and instead have thick-

walled, nodose-septate hyphae with irregularly thickened walls which are not present

in the carpophores of Trametes suaveolens. Any similarity between the two

species, is thus entirely superficial.

Overholts (1953) transferred Fomes cajanderi (as Trametes subrosea Weir)

to the genus Fomes (Fr.) Kickx on account of its stratified pores. Teixeira (1962 a)

in a study of three species of Fomes (Fr.) Kickx which included the type, Fomes

fomentarius (L. ex Fr.) Kickx, showed that the carpophores of species of this

genus have solid or sub-solid, aseptate ‘‘binding hyphae” in addition to unbranched

fibre hyphae and thin-walled nodose-septate hyphae ir the context, but lack

the nodose-septate hyphae with irregularly thickened walls which are present

in carpophores of Fomes cajanderi. Furthermore, the upper surface of Fomes

fomentarius is completely different from that of Fomes cajanderi. Because of these

differences, Fomes cajanderi thus cannot be regarded as congeneric with Fomes

fomentarius.

Bondartsev & Singer (1941), Bondartsev (1953) and Kotlaba & Pouzar (1957)

included Fomes cajanderi Karst. in the genus Fomitopsis Karsten together with the

type species Fomes pinicola (Sw. ex Fr.) Cooke. There are however great differ-

ences in hyphal characters between these two species. Carpophores and cultures

of Fomes pinicola lack the nodose-septate hyphae with irregularly thickened walls

which are present in the carpophores and cultures of Fomes cajanderi. Further-

more, the fibre hyphae of Fomes pinicola are seldym tranched and very slightly

intertwined. The rigidity of the carpophore is due to a certain amount of

agglutination of the hyphae. This is not evident in carpophores of Fomes cajanderi

which are thus more complex in construction and of different texture. A certain

amount of similarity in the nature and construction of the crustose upper surface

of certain specimens of Fomes cajanderi and those of Fomes pinicola is evident.

Lohwag (1940), however, reported similarities in the upper surfaces of Fomes

pinicola and a number of other species of the genus Ungulina Pat. which were

later placed in different groups on the basis of their hyphal and cultural characters

by Nobles (1958 b). Therefore. the similarity in the upper surfaces of the

carpophores of Fomes cajanderi and Fomes pinicola must be regarded as of lesser

importance than the dissimilarity in their hyphal characters and Fomes cajanderi

cannot be regarded as congeneric with Fomes pinicola.

Kotlaba & Pouzar (1957) suggested that Fomes cajande7i is transitional between

Fomitopsis Karst. and Coriolellus Murr. From the above descriptions and

discussions it is clear that there are few similarities between Fomes cajanderi and

Fomes pinicola. Sarkar (1959) showed that six species of the genus Coriolellus

Murr., including the type species, Coriolellus sepium (Be1k.} Murr. which also have

the cultural characters of Group 25 (Nobles, 1958 b) have hyphal characters

and carpophores constructed very much like those of Fomes cajanderi, but she

described the thick-walled. nodose-septate hyphae, which she called “incompletely

differentiated fibre hyphae”, in the carpophcres of these species of Coriolellus.

Such hyphae, which appear to be sclerified generative hyphae (Donk, 1964), were

not found in the carpophores of Fomes cajanderi.

The carpophores of Fomes cajanderi thus differ from those of these species

of Coriolellus Murr. in respect of the types of hyphae present in them. On the

other hand, it was shown before. that many similar characters and structures

exist in carpophores of Daedalea quercina L. ex Fr., the type of the genus Daedalea

Pers. ex Fr., and Coriolellus sepium (Berk.) Murr., the type of the genus Coriolellus

Murr. It was suggested that the six species of Coriolellus Murr. as described by

Sarkar (1959), and which included the type species. should be included in the

genus Daedalea Pers. ex Fr. The carpophores of Coriolcllus sepium (Berk.) Mutr..

however, differ from those of Daedalea quercina L. ex Fr. in the same characters

as carpophores of Fomes cajanderi differ from those of Daedalea quercina. In

cultural characters too, there is as much similarity between cultures of Fores

cajanderi and Daedalea quercina as exists between cultures of the latter species

and those of Coriolellus sepium. For these reasons it seems safe to suggest that

Fomes cajanderi Karst. should be included in the genus Daedalea Pers. ex Fr.

214

Resumé.

The four species of polypores included in Group 25 in the present study have

many characters in common. In the cultures and carpophores of all four species,

thin-walled, nodose-septate hyphae and nodose-septate hyphae with irregularly

thickened walls are present together with aseptate, thick-walled fibre hyphae which

are mostly unbranched. Their basidiospores are cylindrical. Their relatively

thick, anoderm carpophores have thick dissepiments and are similar in construction.

Their cultures do not produce extra-cellular oxidase enzymes and they all cause

brown rots in their respective hosts. They differ from each other in respect of

hymenial configuration, carpophore texture and host preferences. These appear

to be minor differences however which are overshadowed by the many similar

characters in these species. It thus appears that these species may be regarded

as congeneric with the type species of the genus Daedalea Pers. ex Fr.

5.6 Group 32

Cultures of species in this group form white mycelial mats which produce

extra-cellular oxidase enzymes. Their thin-walled hyphae have simple clamp

connections at the septa and are undifferentiated except for occasional swellings or

incrusted portions. Their basidiospores are sub-globose to ovoid or ellipsoid and

less than 8u in length. Their interfertility is of the tetrapolar type.

Fic. 19. — Polystictus subiculoides. (a) Carpophore of PRE 35331; (b) culture

of PRE 42155 at six weeks; (c) vesicle on thick-walled, nodose-septate

hypha from carpophore, xX 500 phase contrast.

Polystictus subiculoides Lloyd, C. G., in Mycological Notes No. 73, 1331, 1924.

Cultural characters

Growth is moderately fast the mat reaching a radius of 50 mm after two

weeks and covering the plates in 3 to 4 weeks; margin even to slightly bayed,

hyphae raised to limit of growth; mat white, thin, downy te cottony, azonate, with

215

a smooth, even, radially combed appearance and remaining so for many weeks;

reverse bleaching slowly after two weeks. Oxidase reaction positive with gum

guaiac solution; weak diffusion zone and slow growth, colony diameter 10 mm after

one week on gallic acid agar, no growth or diffusion zone on tannic acid agar.

Advancing mycelium: hyphae hyaline, unbranched or branching, nodose-septate,

thin-walled, 2.2 — 4.5u in diameter (Fig. 20 a).

Aerial mycelium: (a) hyphae as in the advancing zone; (b) vesicles obovate or

pyriform, hyaline, 7 — 10 x 15 — 20u arising terminally on short lateral projections

from hyphae (Fig. 20 b); (c) narrow thin-walled, hyaline nodose-septate hyphae,

1.54 in diameter, with sub-globose, terminal vesicles 3.0 — 5.0u in diameter

(Fig. 20 c).

Submerged mycelium: hyphae as in the advancing zone bui more frequently septate.

Crystals hyaline, amorphous, numerous in the medium.

Carpophore characters.

Carpophore annual, lignicolous; pilei small, sessile to effused-reflexed,

imbricate, often connate and arising from broad subiculum, soft corky, drying to

hard corky 0.1 — 0.4 x 0.3 — 1.2 x 0.1 — 0.2 cm; surface minutely pubescent,

smooth or slightly rugose, azonate, acuticulate, ‘“‘cream color’ or with small

“cinnamon” patches; margin thin, acute, entire, concolorous with surface; pore

surface concolorous or slightly buff coloured, poroid; pores angular, 4 — 7 per

mm, mouths entire, dissepiments thin, decurrent on the subiculum, tubes 0.5 — 3

mm deep. Context white to pale “cream color” drying isabelline, up to 3.0 mm

thick.

Hyphal characters: hyphae hyaline, branched nodose-septate with clamp connections

often on one side only, thin-walled in young parts (Fig. 20 e); vesicles thin-walled,

ovoid or sub-globose 4.5 — 12.0u in diameter borne terminally on short lateral

or terminal hyphae, arising at clamp connections (Fig. 19 c, 20 f).

Hymenium: basidia hyaline, broadly clavate, 8 — 12 x 4.5 — 6.0u with four short

straight sterigmata 1.5 — 2.2u (Fig. 20 h); basidiospores hyaline sub-globose

to ovoid, smooth, thin-walled 3.0 — 3.5 x 3.2 — 4.2u (Fig. 20 k): vesicles as in

the context, occasionally incrusted, small; acicular, encrusted hyphal tips projecting

into hymenium narrow, 1.5 — 2.5 x 12 — 20 (Fig. 20 n).

Construction. The cream-coloured subiculum which is up to 5 mm thick consists

mainly of hyaline, thick-walled, nodose-septate hyphae together with thin-walled.

nodose-septate hyphae, somewhat intertwined, and growing perpendicularly out

of the substrate, with vesicles scattered throughout the subiculum and hyphal

contents discoloured at different levels to form the darker zones visible in vertical

section.

The pilei are formed by thick-walled, nodose-septate hyphae which grow out

beyond the level of the subiculum and by repeated branching form intertwining

hyphae which turn upward towards the upper surface of the pilei where their ends

are densely packed at a common level to form the pubescent upper surface. In

the same way these thick-walled, nodose-septate hyphae turn downward and by

repeated branching produce intertwining hyphae which form the lower context.

In the trama of the tubes the branches of these hyphae are mostly thin-walled,

short and form a dense, even layer which bear the small basidia on the hymenial

surfaces lining the cavities of the tubes only. In the lower context vesicles are

more numerous than in the upper context. The margins of the pilei consist of

the terminal sections of thick-walled. nodose-septate hyphae from the context,

mostly unbranched. often somewhat elongated to over 100u in length and resembling

short, fibre hyphae with thin-walled tips (Fig. 20 g).

Fic.

216

» FIGURE 20.

20.— Polystictus subiculoides. a - c. Structures from cultures: (a) thin-walled,

nodose-septate hyphae from advancing zone; (b) lateral vesicles; (c) terminal

vesicles.

d-m. Structures from carpophores: (d) thin-walled, nodose-septate hyphae from

margin; (e) thick-walled, nodose-septate hypha from context; (f) hyphae with

terminal vesicles; (g) thick-walled nodose-septate hypha with lengthened, ter-

minal cell; (h) basidia; (k) basidiospores; (m) acicular and vesicular encrusted

hyphal tips.

ANY

Decay and hosts

Polystictus subiculoides causes a white rot of living hard-wood trees and dead

deciduous and coniferous wood.

Specimens examined.

Herb. PRE: 1357, on Grevillea robusta, Pietermartizburg, Natal, April 1911, (TYPE); 27651,

on bark of dead stump, Pietermaritzburg, Natal, 1934; 27775, Pietermaritzburg, Natal, 1934:

28649, indigenous wood, Pietermaritzburg, Natal, Sept. 1934; 28472, on Grevillea robusta,

Pietermartizburg, Natal, Sept. 1934; 28556, on gum tree, Pietermartizburg, Natal, Sept. 1934;

28559, on Acacia mollissima, Impolweni, Natal Sept. 1934; 28772, on Quercus sp., Johannes-

burg, 1936; 30171, on indigenous wood, Umgeni Forest. Natal, March 1935; 30179, on

indigenous wood, Umgeni Forest, Natal, March 1935; 30742, on bark, Deepwalls. Knysna.

Apr. 1939; 35331, on Quercus sp., Pietermaritzburg, Natal, 1943; 36692, on Acacia sp. stump.

Lions River, Natal, May 1948; 39195, Pietermartizburg, Natal, Jan. 1946; *42155, on Bridelia

macranthra, F. C. Erasmus Nat. Reserve, Feb. 1961; *42157, on dead stump, F. C. Erasmus

Nat. Res. Tvl., Feb. 1961; 42199, on dead wood, Pretoria, Nov. 1961; 42291, on dead Pinus

sp., Saasveld, C.P., Nov. 1962; *42359, on dead Pinus sp., Entabeni Forest Reserve, Tvl.

Apr. 1964.

Discussion

Polystictus subiculoides has not been described in culture before. Cultures

of this species are distinguished by the weakly positive oxidase reaction, hyaline-

white, radiating, silky, mycelial mat, and thin-walled, nodose-septate, undiffer-

entiated hyphae bearing thin-walled vesicles on short, lateral branches. As the

terminal vesicles may be conveniently regarded as swellings of the hyphae a basis

for the inclusion of Polystictus subiculoides in Group 32 is provided, but this

species is not well placed in Group 32 because the Jateral vesicles are not simple

swellings on hyphae and as no provision is made other than for undifferentiated

nodose-septate hyphae and differentiated fibre hyphae, nc other alternative group

is available in which this species may be better placed.

Vesicles are regarded as gloeocystidia by Lentz (1954), Talbot (1954 a) and

Van der Westhuizen (1958) but the characteristic staining reaction for gloeocystidia

was not seen when mycelium from a growing culture of Polystictus subiculoides

was mounted in sulphuric-anisaldehyde and sulphuric-benzaldehyde as used by

Slysh (1960) for species of Peniophora. The vesicles of Polystictus subiculoides

thus do not appear to be gloeocystidia. Allocysts were described from cultures

of Flammula alnicola and Flammula conissans by Denyer (1960). The vesicles

of Polystictus subiculoides resemble these allocysts very closely.

Nobles (1965) included Flammula alnicola and Flammula conissans under

Key Code 2.3.26 of a key devised as an aid for the identification of cultures of

Basidiomycetes isolated from decayed wood. As cultures of Polystictus subiculoides

have a positive reaction for extra-cellular oxidase, consistently nodose-septate,

thin-walled hyphae, and ‘‘swellings on hyphae” similar to those of Flammula

alnicola and Flammula conissans, Polystictus subiculoides may be included in Key

Code 2.3.26. Cultures of Polystictus subiculoides may be distinguished from cultures

of these two Flammula spp. by the higher growth rate and more uniform texture

of the mat together with the geographical distribution of the species.

Cultures of Collybia velutipes, Aporpium caryae, Polyporus volvatus and

Polyporus fumosus were also included by Nobles (1965) in Key Code 2.3.26.

Cultures of these species may however be readily distinguished from those of

Polystictus subiculoides as they lack the characteristic vesicles of this fungus.

Cultures of Polystictus subiculoides also resemble those of Odontia bicolor (Nobles,

1953) in many ways but differ by having thinner, silky mycelium while the vesicles

of Polystictus subiculoides are larger than the cystidia of Odontia bicolor and lack

the latter’s typical, large crystalline incrustations.

218

The carpophores of Polystictus subiculoides are most interesting morphologic-

ally as well as anatomically. The pilei arise as reflexed portions of a well-developed

subiculum. The pilei are integral parts of the subiculum and, except for the

presence of hymenial layers in the minute tubes, anatomically undifferentiated and

virtually indistinguishable from it microscopically. Both structures are remarkable

for their simple construction of nodose-septate hyphae with partly thickened walls,

arranged more or less parallel to one another and perpendicular to the substrate.

Of all the species included in this study, Polystictus subiculoides has the simplest

construction of its pilei and the least differentiation in morphology and function

of its hyphae.

From the descriptions it is evident that only one type of hypha is found in

both the cultures and carpophores. The vesicles which are so characteristic in

the cultures, are found throughout the tissues of the carpophores and subiculum.

Differences in the thickness of the walls of hyphae from the cultures and hyphae

from the carpophores had been seen in other species as well (cf. Groups 7 & 9)

and appears to be a modification which usually occurs under natural conditions

during fruit-body formation. All structures formed in cultures are thus present

in the carpophores as well.

Few species of poroid Hymenomycetes are known to have similar vesicles or

gloeocystidia in their carpophores. Notable among these are Poria versipora

(Pers.) Romell (Cunningham, 1946; Lowe, 1946) and Polyporus borealis Fr. (Ames,

1913; Overholts, 1953). These structures are found in Poria versipora in the

hymenium only but they are present in the context of specimens of Polyporus

borealis. Nobles (1958 b) placed cultures of these species in Group 32, but did

not indicate whether vesicles were formed. Since the other species in Group

32 are known to form vesicles in culture, it is assumed that Poria versipora and

Polyporus borealis agree in this respect with the other species and that the vesicles

formed in their carpophores are ontogenetically or physiologically different from

those of Polystictus subiculoides.

Pilat (1946) described conical, immersed cystidia in carpophores of Poria

fissiliformis and Nobles (1958 b) included cultures of this species in Group 32,

but this species differs from Polystictus subiculoides by having fibre hyphae in its

carpophores.

Vesicles and gloeocystidia of various forms are present in the cultures and

carpophores of many species of lower Hymenomycetes notably in the genera

Corticium Pers ex Fr. Odontia Fr., Peniophora Cke. and Stereum Pers. ex Gray

(Lentz, 1954; Talbot, 1954 a; Nobles, 1948, 1965; Cunningham, 1963). In Odontia

bicolor capitate vesicular gloeocystidia are formed in cultures (Nobles, 1953) and

in the spines of the carpophore (Talbot, 1958 b). In Peniophora utriculosa G. H.

Cunn. the small deeply-staining vesicles are borne on short lateral branches of the

intermediate layers of the fruit-body, and are not subtended by clamp connections

(Cunningham, 1963). In Peniophora vesiculosa G. H ‘Cunn. and Peniophora

utriculosa G. H. Cunn. the vesicles are larger and subtended by clamp connections

and situated in the intermediate layers of the carpophores. In both species encrusted

metuloids are also present in the tissues (Cunningham, 1963). The vesicles of

Stereum purpureum are also present in the intermediate zone of the fruit-body

but this fungus differs from the others by having fibre hyphae in culture (Van der

Westhuizen, 1958) and in the carpophore (Talbot, 1954; Cunningham, 1963). These

species of these three genera agree in the morphology and positions of the vesicles

in the tissues of their carpophores. The vesicles of Polystictus subiculoides are

similar in morphology and disposition in the tissues to the vesicles of these four

species, The carpophores also show similarities with those of the thelephoraceous

NG)

species by virtue of their simple censtruction of one type of hypha only. Polystictus

subiculoides also occurs mainly on angiosperm wood and has a weakly positive

oxidase reaction. These similarities appear to indicate that Polystictus subiculoides

has affinities with these species and should be regarded as a poroid member of a

group of Hymenomycetes with thelephoraceous carpophores.

5.7 Group 45

The cultures of species in this group form mycelial mats which mostly remain

white or develop patches of pale, bright colours. Extra-cellular oxidase enzymes

are produced. Their thin-walled hyphae have simple clamp connections at the

septa and usually remain thin-walled but thick-walled, aseptate, fibre hyphae are

formed in large numbers. Their basidiospores are cylindrical and their interfertility

is of the tetrapolar type.

Fic. 21— Polyporus versicolor. (a) Carpophores of DAOM 83052; (b) culture of PRE

42370 at six weeks; (c) nodose-septate hypha with irregular projections from culture,

> 500 phase contrast.

Polyporus versicolor L. ex Fr., Syst. Mys. 1, 368, 1821;

Coriolus versicolor (L. ex Fr.) Quél., Ench. Fung., 175, 1886;

Trametes versicolor (L. ex Fr.) Pilat, Atl. Champ. Eur. III, 261, 1939.

Cultural characters

Growth is rapid to moderately rapid the colonies reaching radii of 30 — 50

mm in one week and covering the plates in two to three weeks. The margin is

even, appressed, thin, hyaline. Behind the margin the young mat may be raised,

cottony, with vague radiating grooves, or floccose, or finely farinaceous, white, but

becoming collapsed over the older part where thin, pellicular patches of dense

mycelium start forming after 2 — 3 weeks. The pellicular areas become sub-felty,

increase in size and coalesce while in some parts they change from white or

“cream color’ to smooth, hard, crustose areas of various shades of brown, turning

finally ‘‘saccardo’s umber’ or ‘‘sepia.”” In some isolates these pellicular areas

may become more felty and increase in thickness eventually developing ‘“‘natal

220

brown” or “‘saccardo’s umber” patches. In others the pellicular areas never

develop but the mat remains thin, white, downy-farinaceous with fine, white,

farinaceous striae radiating from the inoculum. Fruiting may occur in some

cultures. Shallow, smooth depressions appear en dense, pellicular areas of mycelium

or on rounded lumps of dense mycelium. Minute, acicular projections develop in

these depressions, bearing normal, fertile basidia and basidiospores. A white

spore deposit is soon formed under these structures in inverted cultures. The

reverse is bleached after two to four weeks but patches of ““wood brown’’, “army

brown” or “natal brown” may develop in the agar under the coloured areas.

Odour may be strong mushroomy or somewhat unpleasant, fishy.

On gallic and tannic acid agars the diffusion zones are dark and wide while

growth of mycelium extends up to 2.0cm and 3.0cm in diameter on gallic acid

and tannic acid media respectively. When an alcoholic solution of gum guaiac

is applied to the mycelium the colour changes rapidly to bright blue.

Advancing mycelium: hyphae hyaline, branching, thin-walled, nodose-septate,

with deeply staining contents, 2.0 — 4.0u in diameter (Fig. 22 a).

Aerial mycelium: (a) hyphae as in the advancing zone: (b) fibre hyphae long,

unbranched hyaline, sub-solid to solid with the lumina visible mostly at the

tapering ends only, up to 4.5u in diameter at the widest part, (Fig. 22 b); (c) fibre

hyphae long, narrow, hyaline, branching repeatedly the branches long and flexuous

and tapering towards the ends, 1.2 — 3.0u in diameter, (Fig. 22 c); (d) nodose-

septate hyphae with slightly thickened, hyaline walls and without contents, 2.5 —

3.5u in diameter, and with many, short, lateral branches, either thick-walled or

solid and refractive, and stained brown by a lacquer-like substance secreted in the

brown areas (Fig. 21 c, 22d); (e) nodose-septate hyphae with thickened, brown

walls, 2.5 — 3.5u in diameter, embodied in brown, resin-like material present in

the brown areas (Fig. 22 e); (f) very narrow, hyaline hyphae, 0.5 — 0.8u in

diameter and profusely, dichotomously branched, forming a network among the

other hyphae in the pellicular areas.

Fructifications: basidia clavate, hyaline, 12.0 — 20.0 x 3.6 — 4.6u with four long,

slender, somewhat curved sterigmata, 2.8 — 3.3u; basidiospores hyaline, cylindrical,

slightly curved, rounded at the ends, obliquely apiculate, smooth, thin-walled,

4.2 — 5.4 x 1.8 — 2.2u (Fig. 22 f); occasionally branched cystidioles present among

the basidia.

Submerged mycelium: (a) nodose-septate hyphae as in the advancing zone; (b)

nodose-septate hyphae with thickened, brown walls and numerous, short, lateral

branches as in the aerial mycelium.

Carpophore characters

Carpophore annual often reviving, lignicolous, grouped or compound; pileus

dimidiate, sessile, often with a reduced base, or, effused-reflexed, occasionally

imbricate, laterally connate or forming rosettes, up to 6.0 x 8.0 x 0.1 — 0.3 cm;

tough, coriaceous, drying to hard coriaceous; surface velutinate to villose, concen-

trically zonate with alternate zones finally glabrous, and zones variously coloured

Fic. 22.— Polyporus versicolor. L. ex Fr. a - f. Structures from cultures: (a) hyphae

from advancing zone; (b) unbranched, fibre hyphae; (c) fibre hyphae with

long, flexuous branches; (d) nodose-septate hyphae with numerous, thick-walled

or solid, lateral branches; (e) thick-walled, brown, nodose-septate hyphae;

(f) basidia and basidiospores.

g-p. Structures from carpophores: (g) thin-walled, nodose-septate hyphae;

(h) tuft of agglutinated, thick-walled, nodose-septate hyphae; (k) fibre hyphae;

(m) fibre hyphae with short, tortuous, lateral branches; (h) fibre hyphae with

long, tapering branches; (p) basidia and basidiospores.

Ss — JF

_————

yay s,

from white to yellow brown, reddish, greenish, blueish and blackish; margin acute,

entire, occasionally undulate or lobed, white or pale yellowish; pore surface white

or pale cream drying to deep cream or brownish yellow, often glistening: pores

3 — 5 per mm, angular, entire; dissepiments even, thin-walled, tubes concolorous

or pale up to 2 mm deep. Context white or pale cream-coloured, floccose, thin,

0.5 — 2.5 mm.

Hyphal characters: (i) nodose-septate hyphae hyaline, branching freely, thin-walled,

contents staining deeply, 2.0 — 3.5u in diameter (Fig. 22 g); (ii) nodose-septate

hyphae mostly agglutinated into tufts or strands the walls at first hyaline, soon

pale brown, thickened, the lumina narrowed to two thirds or one half the total

diameter of the hyphae, and with deeply staining contents, 1.5 — 2.4u in diameter

(Fig. 22 h); (iii) fibre hyphae long, straight, unbranched, hyaline with thick

refractive walls and lumina wide at the tips, but narrowed along the middle sections

where they are visible as thin interrupted lines only, aseptate or with an occasional

simple septum near the tips, contents usually deeply staining often discoloured

to brownish or dark brown towards the distal ends, 3.0 — 8.0u in diameter (Fig.

22 k); (i) fibre hyphae short, with numerous short, intricately flexuous lateral

branches, hyaline, with walls thickened and lumina narrowed, with deeply staining

contents, aseptate, or solid, 1.5 — 3.0u in diameter and arising from thin-walled,

nodose-septate hyphae (Fig. 22 m), binding hyphae sensu Corner (1932 b); (v) tibre

hyphae with fairly numerous branches, the branches mostly long and tapering

rather abruptly at the tips, or similar to the “‘binding hyphae’’, the walls thickened,

hyaline, the lumina narrow but prominent with deeply staining contents, aseptate,

2.2 — 3.2u in diameter (Fig. 22 n).

Hymenium: basidia short, hyaline, narrowly clavate, 9.0 -— 15.0 x 3.6 — 4.51, with

four, straight, slender sterigmata, 1.5 — 2.5u (Fig. 22 p); basidiospores long,

cylindrical, slightly curved or allantoid, hyaline, smooth, thin-walled, 4.5 — 6.5 x

1.2 — 2.0u (Fig. 22 p); hyphal pegs broadly conical, sterile, projecting up to 40u

beyond the level of the hymenium, not numerous.

Construction. The margin consists mainly of hyaline, long, unbranched, fibre

hyphae mostly with fairly wide lumina and arranged more or less parallel to and

slightly intertwined with one another. In between them are numerous fibre hyphae

with partly thickened walls and a profusion of short, contorted branches growing

across the direction of growth of the fibre hyphae and binding them together into

a tough tissue. Both these types of hyphae arise from branches of narrow, thin-

walled, nodose-septate hyphae with deeply staining contents, which are interwoven

with the other two types. Behind the margin the upper context is similar in

construction but the fibre hyphae develop thicker walls and lumina are narrower.

Much branched fibre hyphae (binding hyphae), mostly solid, are abundant towards

the upper surface, where they bind the long fibre hyphae and numerous branching,

thin-walled, nodose-septate hyphae intc a tough, dense, layer of tissue up to 60u

thick and coloured brownish by a resinous or lacquer-like substance. Through

this layer the ends of long fibre hyphae from the context protrude to form the

pubescence of the upper surface. Most of these fibre hyphae have brownish

contents of the lumina which widen gradually towards the rounded ends (Fig. 22 k).

Arising from thin-walled, nodose-septate hyphae in this dense layer, are numerous

narrow, thick-walled, nodose-septate hyphae, with the walls pale brown and

luminal contents staining deeply. These hyphae are either agglutinated into tufts

or strands (Fig. 22 h) or closely appressed to and agglutinated with fibre hyphae

into tufts by means of a brown, lacquer-like substance. The smooth, brown zones

of the upper surface are formed by this brown, lacquer-like substance agglutinating

the fibre hyphae and nodose-septate hyphae into a brown, smooth cuticle of

resupinate elements.

223

Below this dense upper layer the context tissues are less dense consisting

of long, unbranched, hyaline, solid or sub-solid, fibre hyphae in more or less parallel

arrangement, small numbers of thin-walled, nodose-septate hyphae and intertwined

with hyaline, fibre hyphae (binding hyphae) with long, flexuous branches with

characteristically tapering ends (Fig. 22 n).

The lower context is like the middle context but the fibre hyphae (binding

hyphae) with numerous short, hyaline, solid, branches (Fig. 22 m) become very

numerous and form a dense layer. From the lower context, long unbranched

fibre hyphae turn downward into the dissepiments and become flexuous, narrower

and tightly intertwined, with numerous “binding hyphae” of the short, much

branched type as well as the other type with longer branches and which may

become indistinguishable from the short type, into a very tough and dense tissue.

In between the fibre hyphae, numerous thin-walled, nodose-septate hyphae, with

deeply staining contents and branching repeatedly, are present. At the surfaces

of the dissepiments, the branches of the nodose-septate hyphae bear the basidia

in a dense, even stand.

The hyphal pegs in the hymenium consist of the ends of fibre hyphae, in

parallel arrangement, projecting into and beyond the hymenium from the under-

lying tissue.

Decay and hosts.

Polyporus versicolor causes a white rot of dead wood of a wide variety of

species of deciduous trees.

Specimens examined

Herb. DAOM: *F8183, on Juglans sp. Saanichton, B.C., June 1938; 11781, on decayed wood,

Burnet, Que., Aug. 1944; 11782, on hardwood, Gatineau Park, Que., Aug. 1944; *11783, on

Betula sp. Gatineau Park, Que., Aug. 1944; *21150, on Betula papyrifera log, Dorset, Ont.,

July 1948; 21196, on decayed Acer log, Dorset, Ont., July 1948; *21767, on decayed

Podocarpus spectatus, Rotorua, N.Z., Nov. 1948; *22296, on Alnus rubra, Cowichan Lake,

B.C., June 1948; 22342, on decayed hardwood, Ottawa, Sept. 1949; 22348, on deciduous

host, ex herb. J. Pinto-Lopes, April 1950; 22357, on Eucalyptus sp. ex herb. J. Pinto-Lopes,

April 1950; *22586, on Eucalyptus, Seven Oaks, Surrey, Sept. 1950; *22794, on roots of

Betula lutea, Bells Corners, Ont., July 1949; *30588, on Quercus robur stump, Norway, Nov.

1953; *30589, on Acer sp., Norway, Nov. 1953; *31926, on Betula papyrifera, Calabogie,

Ont., Aug. 1955; 52102, Gainesville, Fla., Sept. 1954; *53899, on dead wood, Wakefield,

Que., July 1952; 53900, on Acer sp., Cantley, Que, July 1952; *69694, on hardwod, Walker,

La., Aug. 1960; 72326, on deciduous wood, Sargent Camp, N.H., Aug. 1956.

Herb, PRE: 1332, on Acacia decurrens, Pietermartizburg, Natal, Apr. 1911; 14838, Kirsten-

bosch, C.P., June 1929; 20603, Knysna, C.P., Jan. 1925; 21877, on Fagus sp., Krieger,

Schidliche Pilzen, Sept. 1905; 22072, Boschberg, C.P. Sept. 1876; 22857, Falkenburg, Germany,

Sept. 1873; 23482, Mont-aux-Sources, Natal, 1937; 24202, Hollos, Hungarian Fungi No. 320;

24830, dead logs, Kirstenbosch, C.P., June 1929; 24847, on stumps, Kirstenbosch, C.P.,

June 1929; 27278, Groote Schuur, C.P., Aug. 1933; 27608, Town Bush Valley, Pietermartizburg,

1934: 28754, dead wood, Pilgrim’s Rest. Tvl., Oct. 1936; 30522, Pretoria, Tvl., Feb. 1939;

30722, dead wood, Pretoria, Tvl., 1939; 30726, Xumeni Forest, Natal, 1937; 30739, Deepwalls,

C.P., April 1939; 30848, Margawa Forest, Natal, Tune 1939; 31336, Town Bush Valley,

Pietermartizburg, June 1939; 31427, old logs, Stellenbosch, C.P., Sept. 1919; 31429, Stellenbosch,

C.P., Sept. 1919; 31550, Moodies, Natal, August 1915; 34071, dead wood. Mariepskop., Tvl.

May 1943; 35651. dead wood, ex Herb. Hort. Bot. Reg., Kew, Oct. 1938; 36874, dead wood.

Umtali, S.R., July 1948; 41536, dead wood, Hogsback, C.P., May 1956; *42370, on maple

log, Packenham, Ont., June 1962; *42813, on dead wood, Stellenbosch, C.P., Aug. 1959;

*42956, on hardwood stump, Warrensburg, N.Y., Aug. 1962.

Herb. STE: 124, Kirstenbosch; 159, on Salix sp. log, Nottingham Rd., Natal; 199, on dead

stumps and logs, Natal Midlands; 291, on dead stumps and logs, Barberton, Tvl.; 714, op

dooie hout, Knysna, Jan. 1922; 882, on Alnus tenuifolia ex Herb. J. R. Weir; 1480, droé

hout, Houtbos, Tvl., Julie 1924; 2647, on Alnus tenuifolia, ex Herb. C. J. Humphrey.

224

Discussion

The cultural characters as described above, agree well with the descriptions

by Fritz (1923), Jay (1934), Refshauge & Proctor (1936). Davidson, Campbell &

Vaughn (1942), Cartwright & Findlay (1946) and Nobles (1948, 1965). The thick-

walled, brownish, nodose-septate hyphae and the short hyphae with numerous, short,

lateral branches have not been reported before. These hyphae do not form

wrinkled, pseudoparenchymatous, crustose areas which are characteristic of cultures

of the species in Nobles’ Group 53 (1958 b) although there appears to be some

superficial resemblance to “‘hyphae with interlocking projections’ (Nobles 1948,

1958 b). Instead, these hyphae from: the cultures of Polyporus versicolor are

found mostly in the mycelial mat among the fibre hyphae or occasionally agglutin-

ated with brown, lacquer-like material in brown areas of some isolates of this

species (Nobles, 1965). In view of their position, morphology and development

and the fact that the binding hyphae in the carpophores of Polyporus versicolor

apparently develop in the same way, they are regarded as homologous with the

binding hyphae, which are so numerous in the carpophores of Polyporus versicolor.

These structures are quite characteristic and may serve tc distinguish this species

from others in this group of which the cultural characters are otherwise very

similar.

From the description it is evident that the carpophores of Polyporus versicolor

consist of five kinds of hyphae. Thin-walled, nodose-septate hyphae are present

in the growing regions and hymenial areas while thick-walled, nodose-septate

hyphae are associated with the unbranched fibre hyphae on the upper surface

and in other parts of the context where they bind other hyphae into tough, dense

tissue. The branched fibre hyphae of the lower context together with the branched,

tortuous, binding hyphae constitute the binding system in the lower part of the

tissues. The carpophores thus have a trimitic hyphal system with generative,

skeletal and binding hyphae as reported by Cunningham (1948 c), Teston (1953 b),

Kotlaba & Pouzar (1957), Teixeira (1960) and Farinha (1964), but these authors,

with the exception of Teixeira (1960) and Farinha (1964), who described thick-

walled, nodose-septate hyphae from carpophores of Polyporus versicolor, mentioned

only three kinds of hyphae in the carpophores. Corner (1932 a), however, stated

that such thick-walled, nodose-septate (or generative) hyphae may contribute to the

binding hyphal system of some species, thus confirming the above observations.

The fibre hyphae with branches towards the distal ends have not been reported

from Polyporus versicolor before but similar hyphae, termed “‘arboriform hyphae”

by Teixeira (1962 b) were reported from the fruit-bodies of Ganoderma spp. by

Hansen (1958) and Furtado (1965 a) where they also assist in binding the tissues

of the carpophores. In the carpophores of Polyporus versicolor, morphologically

and ontogenically different hyphae thus contribute to the different hyphal systems

of the trimitic fruit-bodies.

From the above descriptions, it is clear that most of the structures formed

in cultures of Polyporus versicolor are also present in the carpophores from which

the cultures were made. Only very narrow hyphae which form a network in the

pellicular areas of the cultures, were not found in the carpephores. It is not known

whether such hyphae are present in wood decayed by Polyporus versicolor. No

decayed wood was available for study.

Polyporus versicolor is regarded as the type species of the genus Coriolus

Quél. (Cooke, 1959; Donk, 1960). This species was transferred to the genus

Trametes Fr. by Pilat (1936) who was followed by Kotlaba & Pouzar (1957)

in this. There is indeed great similarity in hyphal characters and carpophore

construction between Polyporus versicolor and Trametes suaveolens Fr. the type

of the genus Trametes Fr. (Donk, 1960). These similarities and their implications

will be discussed below.

Fic. 23.— Yrametes suaveolens. (a) Carpophore upper surface and (b) hymenial surface ol

DAOM 31500; (c) branched fibre hypha from upper context of fruit-body; (d) same

from context above tubes, X 1000 phase contrast; (e) culture of DAOM F1964 at six

weeks.

226

Trametes suaveolens (L. ex Fr.) Fries, Epicr. Syst. Myc., 491, 1838.

Cultural characters

Growth is moderately fast, the mat reaching a radius of 20 — 30 mm after

one week and covering the plates in 2 to 4 weeks. Margin even, mycelium raised

to limit of growth or appressed. Mycelium white, dense cottony to woolly in newest

growth occasionally somewhat lacunose but collapsing in older parts, becoming

more cottony and forming irregular sub-felty or woolly areas interspersed in some

cultures by patches of sodden mycelium, or, mat fairly evenly thin, woolly, with

elongated lumps or ridges of more compact, felty mycelium developing pale,

yellowish or brownish tints and placed on vague ridges radiating from the inoculum.

Around the inoculum may be an area of farinaceous-downy or farinaceous, thin,

sub-felty mycelium. A ring of woolly mycelium may gradually develop against the

side of the dish in some isolates. The reverse is bleached rapidly and a sweet

fragrant, odour is emitted. On gallic acid and tannic acid media, strong diffusion

zones are formed and a trace of growth takes place on tannic acid agar only. A

strong blue colour is formed when a drop of alcoholic gum guaiac solution is

applied to the culture.

Advancing mycelium: hyphae hyaline, thin-walled, with deeply staining contents,

nodose-septate, branching between the septa, 2.4 — 4.5u in diameter (Fig. 24 a).

Aerial mycelium: (a) hyphae thin-walled, nodose-septate, hyaline as in the advancing

zone; (b) fibre hyphae hyaline, unbranched, solid or sub-solid, with lumina aseptate

and visible only at the extremities, 2.0 — 4.0u in diameter (Fig. 24 b), or, branching

freely, often from a short main stem and tapering towards the ends, narrow,

thick-walled to sub-solid in parts, 1.5 — 3.0u (Fig. 24 c). Some fibre hyphae

may have spear-shaped ends (Fig. 24 d).

Submerged mycelium: (a) nodose-septate hyphae as in the advancing zone: (b)

chlamydospores terminal and intercalary, broadly ovoid to somewhat cylindrical,

thin-walled or thick-walled, 7.5 — 16.0 x 4.0 — 6.0u (Fig. 24 e).

Carpophore characters

Carpophore annual, occasionally reviving, lignicolous, solitary or compound,

sessile, dimidiate or occasionally effused-reflexed; pileus convex above, occasionally

imbricate or laterally connate, soft spongy and watery when fresh drying to tough

or corky, anise-scented when fresh, up to 10 x 16 x 4 cm; surface velutinous, to

villose-tomentose or glabrous, azonate or occasionally slightly radially rugose, mat,

white or greyish to isabelline or drying yellowish; margin obtuse, entire, somewhat

involute and concolorous with upper surface; pore surface white at first then

greyish-brownish or smoky, drying yellowish or dark smoky; pores entire, rounded

occasionally angular or elongated, 1 — 3 per mm; dissepiments thick, even or

somewhat dentate; tubes 0.3 — 1.5 cm long, sometimes stratified, concolorous

with context or with dark regions around their mouths; context white or pale

cream, 0.5 — 2 cm thick, tough-fibrous, concentrically zonate.

Fic. 24.— Trametes suaveolens. a - e. Structures from cultures: (a) nodose-septate

hyphae from the advancing zone; (b) unbranched fibre hyphae; (c) fibre hyphae

with tapering branches; (d) spear-shaped end of fibre hypha; (e) chlamydo-

spores.

t n. Structures from carpophores: (f) thin-walled, nodose-septate hypha; (g)

unbranched fibre hypha; (h) fibre hyphae with one to three branches toward

the tip; (k) fibre hyphae with numerous short, tortuous branches; (m) sub-solid,

nodose-septate hypha with numerous short, tortuous branches; (n) basidia and

basidiospores.

228

Hyphal characters: (1) nodose-septate hyphae hyaline. branching, thin-walled,

with deeply staining contents, 1.8 —3.0u in diameter (Fig. 24 f); (ii) fibre hyphae

unbranched, long, more or less straight, hyaline, refractive, thick-walled, lumina

narrow, aseptate often with staining contents, or reduced to a thin, interrupted

line, but widening at the ends, 3.0 — 6.0u in diameter (Fig. 24 g); (iii) fibre hyphae

with one to three branches towards the tips, the branches long, tapering, irregularly

beaded in outline, the walls thick, refractive, lumina reduced to an interrupted

line, widening only at the tapering tips, 2.5 — 6u in diameter (Fig. 24 h); (iv) fibre

hyphae hyaline, repeatedly branched, the branches short. or very short, tortuous,

lumina narrow or occluded, aseptate, 1.8 — 3.0u in diameter, arising at clamped

septa from lateral, thin-walled branches or thin-walled, nodose-septate hyphae

(Fig. 24k); (v) nodose-septate hyphae with numerous short, tortuous branches,

thick-walled or solid with solid clamp connections, 3.0 — 4.0u in diameter (Fig.

24 m).

Hymenium: basidia long clavate, hyaline 16.0 — 24.0 x 5.0 — 6.0u bearing 4

thick sterigmata, 3.6 — 4.2u long; basidiospores hyaline, long ellipsoidal to cylin-

drical, obliquely apiculate, thin-wailed, smooth, 7.5 — 9.8 x 3.2 — 4.0u (Fig. 24 n).

Construction. The margin consists mainly of long, unbranched, hyaline, fibre

hyphae mostly straight, arranged parallel to one another and intertwined to a

small extent. Intertwined with the fibre hyphae are numerous thin-walled,

branching, nodose-septate hyphae, from which the fibre hyphae arise. Behind

the margin the fibre hyphae become sub-solid, or solid. and turn upward into the

upper context or downward towards the pores. In the upper context the fibre

hyphae are very loosely arranged, mostly parallel to each other with little inter-

twining, and with their thin-walled ends arranged at a common level and free to

form the velutinous upper surface, or, agglutinated into tufts by brownish, resin-like

material to form the villose-tomentose upper surface. Intertwined with the

unbranched fibre hyphae are thin-walled, nodose-septate hyphae in small numbers

as well as small numbers of fibre hyphae with irregularly beaded walls and two

or three branches (Fig. 23 c, 24 h). the branches running diagonally across the

unbranched fibre hyphae and interwoven with them. In the upper part of the

context, aseptate fibre hyphae with numerous tortuous branches, (Fig. 24 k)

intertwined with the other hyphae, are present in smal! numbers. In the lower

context the tissues become much more dense and more compact. Small numbers

of long, unbranched, fibre hyphae, turn downwards into the trama of the dissepi-

ments; fibre hyphae with somewhat beaded walls and branches towards their ends,

become more numerous, and their branches are extensively intertwined. Fibre

hyphae with many short, contorted, sub-solid or solid branches (Fig. 23 d), inter-

twined with the other hyphae, and binding them into a compact tissue, are present

in large numbers. Thin-walled, nodose-septate hyphae with deeply staining contents

and branching frequently are intertwined and interwoven with the fibre hyphae.

In the dissepiments, fibre hyphae with fairly long branches. tortuous and intertwined,

and binding hyphae with numerous, short, tortuous branches, interwoven with the

fibre hyphae and binding them across their direction of growth, constitute the

bulk of the dense tissue. Thin-walled, nodose-septate hyphae, intertwined with

the fibre hyphae, branch repeatedly and turn outwards towards the hymenial

surfaces where their numerous, short, intertwined branches form a sub-hymenial

layer of small isodiametrical cells about Su thick. From this layer the basidia

are produced in a dense even stand. No accessory structures are present in the

hymenium.

Decay and hosts

Trametes suaveolens causes a white, mottled rot of the heartwood of Salix

spp. but is found occasionally on Populus and Betula spp.

229

Specimens examined

Herb, DAOM: FSO, on Salix sp., Ottawa, Ont., Oct. 1929; F911, on Salix sp.. Gaspé, Que.,

Sept. 1927; F948, on Salix sp., Ottawa, Ont., Sept. 1928; F994, on Salix sp., Ottawa, Ont.,

July 1929; F1297, on Salix nigra, Syracuse, N.Y., Nov. 1929; F1393, on Salix sp.. Ottawa,

Ont., Sept. 1930; F1633, on Salix sp., Woodpecker, B.C., Sept. 1927; F1954, Syracuse, N.Y.,

Aug. 1931; *F1964, on Salix sp., Hopewell, N.J., Sept. 1931; F2249, on living Salix sp.,

Winnipeg, Man., Apr. 1932; F2919, on living Populus balsamifera, Edmonton, Alta., Oct.

1932; F2994, on Salix, Greenwich, N.S., Sept. 1930; F3500, on Salix sp., Ottawa, Ont., Sept.

1933; *F3523, on Salix sp., Ottawa, Ont., Sept. 1933; F3669, on Alnus incana, Ste. Philomiene.

P.Q., Apr. 1931; F3704, on dead Populus balsamifera, Edmonton, Alta., Nov. 1933; F5031,

on Salix sp., Ottawa, Ont.. Sept. 1934; F5032, on Salix sp., Ottawa, Ont., Sept. 1934; F5641,

on Salix sp., Nov. 1941; *7654, on Salix sp., Matapedia, Que., Aug. 1937; F8043, on Salix

sp., Ottawa, Ont., Oct. 1937; F8331, on living Salix sp., Matapedia, Que., Aug. 1938; 10812,

on Salix sp., Kentville, N.S., Nov. 194), 21570, on Populus trichocarpa, Quesnel, B.C., 1948;

30803, on living Salix alba, Leighton Buzzard, Gt. Brit., Jan. 1953; 31500, on Salix sp.,

Fredericton, N.B., Noy. 1954.

Herb, PRE: 21920, on Salix alba, Krieger, Schadliche Pilze, Apr. 1904; 10731, on Populus

trichocarpa, Priest River, Idaho, July 1913.

Discussion

The description of the cultural characters, agrees well with those of Hirt

(1932) and Nobles (1948). The cultures agree with those of other species in this

group in many ways, but the fibre hyphae with short, tortuous branches or nodose-

septate hyphae with solid, branching processes formed in some other species of

this group as described below, were not present in cultures of Trametes suaveolens.

The absence of these kinds of hyphae together with the soft, cottony-woolly texture

of the mycelial mat and the fragrant, anise-like odour given off may serve to

distinguish cultures of this species from the others with otherwise similar characters

in this group.

The fruit-bodies of Trametes suaveolens are rather unusual in that their soft

spongy feel belie their complex construction. From the descriptions it is clear

that five kinds of hyphae could be distinguished in the carpophores. Besides

thin-walled, nodose-septate hyphae or generative hyphae (Corner, 1932 a) two

kinds of fibre hyphae of the skeletal system (Corner, 1932 a) are present while

the binding system (Corner, 1932 a) consists of fibre hyphae with short tortuous

branches and solid, nodose-septate hyphae with solid clamps and tortuous branches.

These “binding hyphae” are very numerous only in the lower context above and

among the tubes. Towards the upper context their numbers decrease rapidly

so that their binding action is less pronounced and the upper context attains the

soft texture so characteristic of the type species of Trametes Fr.

Bourdot & Galzin (1928) and Overholts (1953) reported that the hyphae of the

carpophore of Trametes suaveolens are sparingly branched and_ thick-walled.

Pinto-Lopes (1952) reported that the primary hyphae are thin-walled and nodose-

septate while the secondary hyphae are thick-walled and aseptate. Donk (1933)

described three types of hyphae from the carpophores of Vrametes suaveolens.

Teston (1953 b) after studies of species of Polyporaceae in the Bourdot herbarium

in the Museum of Natural History in Paris, reported that the carpophores of

Trametes suaveolens have a trimitic hyphal system, according to Corner’s (1932 a)

concepts, with branching, thin-walled, nodose-septate, generative hyphae and

thick-walled or solid, branching, skeletal hyphae, staining in Giemsa, present in the

context. In the tubes, extremely contorted, branching, binding hyphae, not staining

with Giemsa, were present as well. The distribution of hyphae according to

Teston (1953 a) however gives the impression that the context is dimitic, consisting

of generative hyphae and branching and unbranched skeletal hyphae only whilst

230

the trama is trimitic with binding hyphae present as well. This trimitic hyphal

system was later also reported by Kotlaba & Pouzar (1957) and O. Fidalgo (1957).

These descriptions agree quite well with that given abeve, but do not distinguish

clearly between the different kinds of hyphae in the different hyphal systems. As

in the fruit-bodies of species of Ganoderma Karst. (Hansen, 1958), the fibre

hyphae with branches towards the distal end contribute to the binding hyphal

system in carpophores of Trametes suaveolens. The solid or thick-walled, nodose-

septate hyphae, considered by Corner (1932 a) and Cunningham (1946, 1954) to

be generative hyphae, similarly contribute to the binding system. In the fruit-body

of Trametes suaveolens which has a trimitic hypnal system sensu Corner (1932 b),

morphologically and ontogenicaliy different hyphae thus contribute to the different

hyphal systems. In this respect the fruit-bodies of Trametes suaveolens (L. ex Fr.)

Fr. are similar to those of Polyporus versicolor L. ex Fr.

Comparison of the anatomy and hyphal characters of the fruit-bodies of

Trametes suaveolens wiih those of Polyporus versicolor, shows great similarity

in hyphal characters but a great difference in construction. Where the thin fruit-

bodies of Polyporus versicolor abound with tortuous binding hyphae, the thick

fruit-bodies of Trametes suaveolens have a poorly developed binding system with

binding hyphae almost entirely absent from the upper context; but this difference

appears to be one of degree of development of a particular hyphal system in a

particular species and not a difference in the types of hyphae present which could

indicate a phylogenetic difference in the carpophores of the two species. These

two species thus appear to be related and in fact apparently represent two extremes

of a series of species with similar hyphal, anatomical and micromorphological

characters but varying in construction and texture. Other species described below,

especially Polyporus pubescens Schum. ex Fr. and Lenzites palisoti (Fr.) Fr. appear

to be intermediate between Polyporus versicolor and Trametes suaveolens in

construction. This appears to be strong evidence in favour of the view held

by Pilat (1936), and Kotlaba & Pouzar (1957) that the two species are congeneric;

but it appears to be desirable that the hyphal characters and construction of the

fruit-bedies of many more species in this complex of species be studied carefully

before a final and satisfactory conclusion can be reached.

Comparison of the hyphal characters and carpophore construction of Trametes

suaveolens (L. ex Fr.) Fr. with those of Daedalea quercina Fr. shows that the

binding hyphae are not present in carpophores of the latter. The nodose-septate

hyphae with irregularly thickened walls present in carpophores and cultures of

Daedalea quercina are absent from those of Trametes suaveolens. This difference

in the kinds of hyphae present in the carpophores indicates a phylogenetic difference

between these two species so that Daedalea quercina and Trametes suaveolens

cannot be regarded as being congeneric as suggested by O. Fidalgo (1957).

In Trametes suaveolens the structures formed in cultures are, with the exception

of the chlamydospores, also present in the carpophores from which they were made.

It is not unlikely that chlamydosvores may be found in wood decayed by Trametes

suaveolens as reported in other species by Cartwright & Findlay (1946).

Zell

Fic. 25.—bLerzites betulima. (a)

Carpophore, upper surface and

(b) lower surface of PRE

43158: (c) culture of PRE

42447 at six wecks.

Lenzites betulina (L. ex Fr.) Fr., Epicr. Syst. Myc., 405. 1836 — 1838;

Daedalea betulina L. ex Fr., Syst. Myc. 1, 333, 1821:

Trametes betulina (L. ex Fr.) Pilat, Atl. Champ. Eur. III, 327, 1936.

Cultural characters

Growth is rapid to moderately rapid, the colony reaching a radiius of up to

30 mm in one week and covering the plate in 2 — 4 weeks. The margin is even,

appressed, in some isolates over a narrow zone only, mycelium then raised, mostly

cottony to woolly in a zone behind the advancing zone, then somewhat collapsed

and more densely woolly with a pebble-like surface over the older part of the

white mat. In time the mat gradually thickens developing raised areas of tough,

felty or woolly-felty mycelium somewhat lacunose or developing a warty surface

or rounded lumps of mycelium giving it a pebbly appearance, the intervening areas

mostly thin appressed, sub-felty to felty. Mat very tough but separating easily

from the agar. After about 4 weeks the pebbly mycelium may develop fruiting

areas, at first warty, of pale ““cream color” developing short, blunt, spines, rounded

or somewhat flattened, glabrous or chamois. The reverse bleaches rapidly while a

faint, pleasant, sweet, mushroomy odour is given off during the first two to four

weeks. A strong positive reaction for extra-cellular oxidase results when alcoholic

gum guaiac solution is applied to the mat.

232

Advancing mycelium: hyphae hyaline, thin-walled, nodose-septate, branching, often

opposite the clamp connections, 2.0 — 6.0u in diameter (Fig. 26 a).

Aerial mycelium: (a) nodose-septate hyphae hyaline, richly branched, narrow,

1.0 — 2.5u in diameter (Fig. 26 b); (b) fibre hyphae long, straight, unbranched,

hyaline, widest along the middle part, walls thickened and refractive, lumina narrow

or reduced to a series of interrupted spaces, widening only towards the narrower,

thin-walled ends, aseptate, 3.5 — 4.0u in diameter at the widest parts (Fig. 26 c);

aseptate lumina with deeply staining contents, or lumina narrow or occluded, and

hyphae solid, 1.5 — 3.0u in diameter (Fig. 26 d); (d) nodose-septate hyphae with

numerous short branches, walls thickened, lumina much reduced or hyphae solid,

1.5 — 3.0u in diameter (Fig. 26 h).

Submerged mycelium: (a) nodose-septate hyphae as in the advancing zone.

Carpophore characters

Carpophore annual or perennial, lignicolous, solitary, sessile to effused-reflexed;

pileus dimidiate, occasionally imbricate to laterally connate, coriaceous, up to

6.0 x 9.0 x 1.0 cm; surface tomentose or hirsute, concentrically zonate often multi-

coloured, whitish or pale greenish grey or greyish brown; margin acute, entire,

concolourous with upper surface, or lighter; pore surface white to ‘‘cream color”

darkening somewhat on drying, usually lamellate, lamellae frequently branched

or anastomising, occasionally poroid or labyrinthiform, about | mm apart, edges

even or interrupted, decurrent behind; context white 0.5 3.0 mm thick, fibrous.

Hyphal characters: (i) nodose-septate hyphae hyaline, branched, thin-walled, with

conspicuous clamps, 1.2 — 3.0u in diameter (Fig. 26 e); (ii) fibre hyphae long,

more or less straight, unbranched or with an occasional short branch near the distal

end, widest along the middle portion, thick-walled, refractive, lumina aseptate.

narrow or reduced to an interrupted line, or hyphae solid except near the ends,

3.0 — 7.5u in diameter (Fig. 26 f); (iii) fibre hyphae with numerous, short, tortuous

branches, walls thickened, lumina narrow or occluded, aseptate 1.0 — 3.0u in

diameter (Fig. 26 g): (iv) nodose-septate hyphae with numerous flexuous branches,

solid, occasionally sub-solid, 1.0 — 3.0u in diameter (Fig. 26 h).

Hymenium.: basidia hyaline, narrowly clavate, small, 20.0 — 26.0 x 4.0 — 5.0u,

with four, short, curved sterigmata, 2.5 — 3.0u long (Fig. 26 k); basidiospores

hyaline, cylindrical to slightly curved, smooth, thin-walled, with a small, oblique

apiculus, 4.5 — 6.0 x 2.0 — 2.5u (Fig. 26 m); tramal cystidia projecting 6.0 — 20.0u

above the hymenium, upper part broadly subulate, thick-walled, hyaline, lumina

prominent, aseptate, mostly with brownish contents, and arising as lateral branches

of fibre hyphae or directly from thin-walled, nodose-septate hyphae in the trama

(Fig. 26 n).

Construction. At the margin the carpophore consists of long unbranched fibre

hyphae, mostly with prominent lumina, more or less parallel in arrangement and

Fic. 26.—Lenzites betulina. a - d. Structures from cultures: (a) nodose-septate

hyphae from advancing zone; (b) narrow, branched, nodose-septate hyphae

from aerial mycelium; (c) unbranched fibre hyphae; (d) fibre hyphae with

long, tapering branches. .

e-n. Structures from carpophores: (e) thin-walled, nodose-septate hyphae; (f)

unbranched, fibre hyphae; (g) fibre hyphae with numerous tortuous branches;

(h) sub-solid or solid nodose-septate hyphae with numerous flexuous branches

also found in cultures; (k) basidia; (m) basidiospores; (n) tramal cystidium

with dark-coloured contents in sagittate terminal part.

234

interwoven with the numerous branching, thin-walled, nodose-septate hyphae from

which they arise. In the context behind the margin the fibre hyphae are sub-solid

and fibre hyphae with numerous, flexuous, lateral branches (binding hyphae)

appear, with their branches interwoven with the unbranched hyphae across the

direction of growth. The context consists mainly of long, unbranched, hyaline

fibre hyphae, parallel to each other, slightly intertwined and small numbers of

binding hyphae, tightly interwoven with the long fibre hyphae. In the upper

context the long fibre hyphae turn upwards and the numbers of binding hyphae

increase rapidly and thin-walled, nodose-septate hyphae, interwoven with the other

hyphae, appear. This merges into a dense layer 100 — 200p thick, at the upper

surface which consists of numerous, branching, thin-walled, nodose-septate hyphae

and binding hyphae all tightly interwoven with the long unbranched fibre hyphae

which project beyond this layer to form the tomentose upper surface of the pileus

(Fig. 26 f). These “hairs” of the upper surface may be sub-solid or solid, hyaline,

or may have wide, aseptate, prominent lumina occasionally with brownish contents,

and may be free or agglutinated into tufts. In the lower context the tissues are

similar to the upper context but the numbers of binding hyphae increase rapidly

towards the dissepiments and bind the straight fibre hyphae and _ thin-walled,

nodose-septate hyphae into a dense, tough tissue from which individual elements

can be dissected out only with difficulty. Many of these binding hyphae have

solid or sub-solid clamps and appear to develop as a result of thickeneing of the

walls of branched, nodose-septate hyphae (Fig. 26 h). Into the trama of the

dissepiments, fibre hyphae from the lower cont2xt turn downwards where many

may have one or more short, lateral branches near thew tips (Fig. 26 f). The

dissepiments consist mainly of binding hyphae with solid or sub-solid, aseptate,

tortuous branches and nodose-septate hyphae with short, flexuous, sub-solid or

solid branches, tightly interwoven with the unbranched fibre hyphae and branching,

thin-walled, nodose-septate hyphae in a dense, tough tissue. In the dissepiments

the thin-walled, nodose-septate hyphae branch frequently and turn towards the

hymenial surfaces of the pores where the basidia are borne on their numerous,

short, terminal branches. From the tramal tissues the ends of short, unbranched,

fibre hyphae or lateral branches of fibre hyphae, project into and beyond the

hymenium as tramal cystidia (Fig. 26 n).

Decay and hosts

Lenzites betulina causes a white rot of hardwoods.

Specimens examined

Herb, DAOM: FS5115, on Betula occidentalis, Aleza Lake, B.C., July 1934; F5121, on Betula

sp., Frankfurt am Main, Germany, Oct. 1934; F5229, on Alnus incana, Edmonton, Allta.,

Sept. 1931; F7205, on Quercus stump, Halifax, N.S., Jan. 1937; F7375, on Populus tremuloides,

Oslo, Norway, Apr. 1937; F7462, on Betula sp. Ottawa, Ont., Aug. 1937; F8021, on Fagus

grandifolia, Iberville, Que., Sept. 1938; F9071, on Quercus acutissima, Tokyo Science Museum

No. 200645; F9156, Trinity Valley, B.C., Oct. 1938; F9909, on Acer saccharum, Petawawa,

Ont., Sept. 1937; F9934, on Acer saccharum, Petawawa, Ont., Sept. 1939; *F10199, on Betula

sp., dead branch, Montreal Is., Que., Aug. 1941; F10609, on Betula sp. dead branch, Montreal

Is., Que., Aug. 1941; 10713, on Betula sp., Petawawa, Ont., Aug. 1941; 22291, on Betula sp.

Quesnel, B.C., Aug. 1949; 22362, on Eucalyptus sp., Portugal, Herb. '. Pinto-Lopes No. 983;

22936, ex Herb. Hort. Bot. Reg., Kew; 30501, on Betula papyrifera pole, Parry Sound,

Ont., Sept. 1951; 30504, Acer rubrum log, Horseshoe Lake, Ont., Aug. 1951; 30879, on

Acer rubrum, Lake Rossignol, N.S., Sept. 1953; 30963, on Betula pubescens, Viljo Kujala,

Fungi Fennici 663; 43109, Ithaca, N.Y., April 1953; 46598, on Lithocarpus densiflorus,

Darlingtonia, Calif., Feb. 1944; 62359, on Betula papyrifera, Stone Creek, B.C., Aug. 1956;

52399, decaying trunk, Rio de Janeiro, Brazil, Sept. 1955; 53772, on Betula sp. stump, Flitwick,

England, Oct. 1959; 69205, on Betula papyrifera, Naney, B.C., Aug. 1960; 69960, on Betula

papyrifera, Dawson, Yukon, Terr. July 1959.

Herb. PRE: 2339, Knysna, C.P., June 1912; 3869, Rabenhorst-Winter, Fungi Europi No.

3529; 6616, on Eucalyptus diversicolor, Fort Cunynghame, C.P., May 1913; 13875, Herb.

235)

J. R. Weir No. 654; 14833, Kirstenbosch, C.P., June 1921; 15485, on dead wood, George C.P.,

May 1922; 15576, on Acacia mollisima, Schwarzwald, Natal, May 1915; 17802, on Podocar pus

sp., Knysna, C.P., May 1923; 20465, on Podocarpus sp., Knysna, C.P., Jan. 1925; 23478, on

Podocarpus sp., Mount-aux-Sources, Naial, July 1928; 27277, on Podocarpus sp., Newlands,

C.P., Aug. 1933; 27716, on Podocarpus sp., Donnybrook, Natal, Jan. 1935; 27967, Fungi

Columbiani, E. Batholomeas No. 4935; 28604, on dead wood, Mooirivier, Natal, April 1936;

28878, on dead wood, Drakensberg, Natal, July 1937; 30189, on dead wood, Nkandhla Forest,

Natal, March 1935; 30696, on dead wood, Deepwalls, Knysna, C.P., Apr. 1939; 31306, on

Quercus sp., Kirstenbosch, C.P., Apr. 1939; 31333, on Quercus sp., Knysna, C.P., Apr. 1939;

31535, on living Celtis kraussiana, C.P., June 1921; 31558, on Celtis kraussiana, Katberg, C.P.,

Aug. 1915; 31879, Eucalyptus sp. logs, Knysna, C.P., Apr. 1917; 34926, on Eucalyptus sp.,

Melrose, Johannesburg, Apr. 1945; 34994, on Acacia mearnsii logs, Qudeni Forest, Natal,

March 1935; 37481, on Quercus sp., Mycotheca generalis, Petrak, No. 574; 41526, on dead

wood, Knysna, C.P., May 1956; 41527, on dead wood, Hogsback, C.P., May 1956; *42339,

on dead wood, Dorset, Ont., Sept. 1962; *42363, on Betula sp., Dorset, Ont., Sept. 1962;

*42434, on dead wood, Barberton, Tvl., June 1959; *42447, on Acacia mearnsii stumps,

Kaapse Hoop, Tvl., Feb. 1961; 43158, on dead wood, George, C.P., March 1966.

Herb. STE: 352; 353, Eastern Cape Province; 422, Karkloof, Jan. 1922; 478, oak stump,

Stellenbosch, Aug. 1921; 2156, oak stump Kirstenbosch, L. Bolus, July 1925; 2810; ou hout,

Stellenbosch, A. J. le Roux, Oct. 1944; 43, ou hout, (as Lenzites guineensis Fr.); 474, old

logs, Stellenbosch, Oct. 1921; 488, ou hout, J. P. Leslie, (nm vorm van L. betulina); 2212,

L. aspera, on dead Olea laurifolia, Knysna, J. F. V. Phillips, Dec. 1923.

Discussion

The description of cultural characters of Lenzites betulina agrees well with

earlier descriptions by Davidson, Campbell & Blaisdell (1938), Cartwright & Findlay

(1946) and Nobles (1948, 1965). This species resembles Trametes suaveolens and

Polyporus versicolor in cultural characters and the structures produced in culture.

The characteristic pebble-like mounds of mycelium forming on top of the very

tough mat, may serve to distinguish cultures of this species from other closely

similar species in Group 45.

From the descriptions it is evident that four kinds of hyphae are present in

the carpophores of Lenzites betulina. The fifth kind, the long fibre hyphae with

branches near the tip (arboriform hyphae, Teixeira, 1962 b) which are present in

carpophores of Polyporus versicolor and Trametes suaveolens are absent from

those of Lenzites betulina. Although short branches were seen on some fibre

hyphae, these branches were so short as to be almost inconspicuous and not

comparable to those in the carpophores of Trametes suaveolens and Polyporus

versicolor so that these hyphae do not merit special designation. In carpophores of

Lenzites betulina, the binding hyphal system also consists of branched, aseptate

fibre hyphae and branched, thick-walled, nodose-septate hyphae which are mor-

phologically and ontogenically distinct as in the carpophores of Polyporus versicolor

and Trametes suaveolens; but since nodose-septate hyphae are regarded as generative

hyphae by Corner (1932 a, b), Cunningham (1946, 1954) and Teixeira (1962 b)

the carpophores of Lenzites betulina must possess a trimitic hyphal system as

reported by Cunningham (1948 h), Teston (1953 6), Kotlaba & Pouzar (1957) and

Fidalgo (1957). None of these authors however reported the nodose-septate,

thick-walled binding hyphae.

From the above it is clear that the structures formed in cultures are also

present in the carpophores from which they were made. The branched fibre hyphae

or binding hyphae in cultures were less sinuous than those from the carpophores

but as these hyphae are interwoven with the unbranched fibre hyphae in both the

cultures and also in the carpophores, they must be regarded as homologous

structures.

Comparison of the hyphal characters and construction of Lenzites betulina

(L. ex Fr.) Fr. with those of the “brown species of Lenzites’’ (Overholts, 1953)

236

now generally referred to Gloeophyllum Karst., and desscribed in Group 13, reveals

important differences. Apart from the yellowish-brown pigment in the walls

of the fibre hyphae of these brown species, aseptate binding hyphae, which are so

numerous in the carpophores of Lenzites betulina, are completely absent from

those of Lenzites trabea Pers. ex Fr. In the trama of Lenzites sepiaria (Wulff. ex

Fr.) Fr. the aseptate binding hyphae are present in the older parts of the context

and have pale umber-brown walls and longer, less tortuous branches than those

of Lenzites betulina. In cultural chatacters Lenzites betulina also differs markedly

in respect of its white mat, hyaline hyphae and positive oxidase reaction from

cultures of Lenzites trabea and Lenzites sepiaria with their brown coloured mats,

pale brown fibre hyphae and negative reaction when tested for extra-cellular

oxidase. Lenzites betulina therefore cannot be regarded as congeneric with the

two brown species, Lenzites sepiaria and Lenzites trabea

In a discussion of the nomenclatural status of the genus Daedalea Pers. ex Fr.

and related genera, Fidalgo (1957) concluded that no real distinction could be

found between the genera Daedalea Pers. ex Fr., Lenzites Fr. and Trametes Fr.

other than in hymenial configuration as expounded by Fries (1838). He named

Lenzites palisoti (Fr.) Fr. as an example in which these three types of hymenial

surfaces may often be seen combined in one fruit-body thus illustrating the

artificiality of even this distinction. He therefore regarded the genera Lenzites

Fr. and Trametes Fr. as synonymous with Daedalea Pers. ex Fr., the oldest genus.

A comparison of the descriptions of the type species of the genera Lenzites,

Trametes and Daedalea given above, shows that the nodose-septate hyphae with

irregularly thickened walls which are found in cultures and carpophores of Daedalea

quercina L. ex Fr., the type of ihe genus Daedalea (Donk, 1960), are absent from

the cultures and carpophores of Lenzites betulina (L. ex Fr.) Fr. and Trametes

suaveolens Fr., the type species of the genus Lenzites Fr. and Trametes Fr. respect-

ively (Donk, 1960). Furthermore, the binding hyphae which are characteristic

of the carpophores of Lenzites betulina and Trametes suaveolens are absent from

those of Daedalea quercina. The carpophores cf Daedalea quercina thus have

dimitic hyphal systems, in the sense of Corner (1932 a, b) while those of Lenzites

betulina and Trametes suaveolens have trimitic hyphal systems. There are thus

distinct and fundamental differences in the hyphal systems and construction of

the carpophores of the type species of the genus Daedalea Pers. ex Fr. on the

one hand and the genera Lenzites Fr. and Trametes Fr. on the other. Therefore,

the latter two genera cannot possibly be regarded as being congeneric with Daedalea

quercina despite similarities in gross morphological characters.

From the above descriptions it is evident that cultural characters, hyphal

characters and the construction of the carpophores of Polyporus versicolor, Trametes

suaveolens and Lenzites betulina are very similar in many respects. This similarity

caused Donk (1933) to express the view that the genera Coriolus Quél., Trametes

Fr. and Lenzites Fr., of which these three species are the respective types, (Cooke,

1959) may be congeneric. This was also the basis for Pilat’s (1936) inclusion of

the genera Coriolus Quél. and Lenzites Fr. in Trametes Fr. A careful comparison

of the hyphal characters of the three species shows that the fibre hyphae with

one or two long branches near their ends are absent from the carpophores of

Lenzites betulina but are present in those of Polyporus versicolor and Trametes

suaveolens. In carpophores of Trametes suaveolens, fewer binding hyphae are

present than in those of the other two species. Carpophores of Trametes suaveolens

thus differ from those of the other two species in the numbers of one kind of

hypha present while carpophores of Lenzites betulina differ from those of Polyporus

versicolor and Trametes suaveolens in the kinds of hyphae present. Since the

kinds of hyphae present in carpophores are considered to be important at the

237

generic level (Bondartzeva, 1961; Teixeira, 1962 b), it appears that Lenzites betulina

cannot be considered to be congeneric with Polyporus versicolor and Trametes

suaveolens. The importance of the absence of these hyphae from carpophores of

Lenzites betulina can however be confirmed only by examination of a large number

of specimens and many different species with similar anatomical and micromor-

phological characters; but from the above descriptions and descriptions of the

following species in this group, it appears as if this character, if considered in

combination with the presence of a predominantly lamellate hymenium, may be

the characters which distinguish the genus Lenzites Fr. from the closely related

genus Trametes Fr.

Fic. 27—Polyporus pubescens. (a) Carpophores of DAOM_ 17530. upper surfaces and

(b) hymenial surfaces; (c) narrow, dichotomously branched hyphae and unbranched

fibre hyphae from culture, « 500; (d) culture of DAOM 17577 at six weeks.

Polyporus pubescens Schum. ex Fries, Syst. Myc. 1, 367, 1821;

Coriolus pubescens (Schum. ex Fr.) Quélet, Fl. Myc. Fr. p. 391, 1881;

Trametes pubescens (Schum. ex Fr.) Pilat, Atl Champ. Eur. III, 268, 1939.

Cultural characters

Growth is moderately rapid, the mat reaching a radius of up to 35 mm after

1 week and covering the plate after 2 to 3 weeks. Margin even, raised, white,

cottony to woolly; mat raised behind margin, becoming appressed, compact, woolly.

238

towards inoculum. Farinaceous areas appear and increase in size resulting in a

finely pebbled appearance of the surface; or, mat felty with thin, pellicular areas

appearing, which increase in size and coalesce to form irregular, pellicular areas

occasionally involving the entire mat, the surface irregularly, radially grooved, or,

smooth between the grooves, or, roughly lacunose with small raised, anastomosing

ridges and wart-like protruberances in the older parts of the pure, white mat.

Pale “‘cream color” irregular, raised, spongy areas may appear along the side of

the dish, with similar warted surfaces, giving rise to minute, acicular spines, bearing

basidia and basidiospores. The reverse bleaches after one to two weeks and a

faint mushroomy odour is given off. A strong positive reaction for extra-cellular

oxidase is given with gum guaiac solution. Strong diffusion zones are formed

on gallic acid agar and tannic acid agar. No growth occurs on gallic acid agar

but colonies up to 40 mm in diameter on tannic acid agar after one week.

Advancing mycelium: hyphae hyaline, branching, thin-walled, nodose-septate, 2.0 —

3.5u in diameter and with deeply staining contents (Fig. 28 a).

Aerial mycelium: (a) thin-walled, nodose-septate hyphae as in the advancing zone;

(b) fibre hyphae hyaline, long, unbranched, walls thick, refractive, sub-solid to solid,

up to 4.5u in diameter (Fig. 28 b); (c) fibre hyphae hyaline, branched, the branches

long, flexuous, tapering gradually towards the tips, the lumina prominent, aseptate,

20 3.0u in diameter (Fig. 28 c); (d) hyaline, much branched, solid, refractive

processes, arising from nodose-septate hyphae, 1.5 — 4.0u in diameter (Fig. 28 d):;

(e) hyaline, very narrow, reticulately branching hyphae, solid, 0.5u in diameter

(sie, Zi! ©),

Fructification: basidia long, clavate, 10.0 — 17.0 x 3.6 — 4.5u, with 4 straight

sterigmata, 3.0 — 3.6u long (Fig. 28 h); basidiospores hyaline, cylindrical, straight

or slightly curved, apiculate, smooth, thin-walled, 4.5 — 5.0 x 1.5 — 2.0u (Fig.

28 h).

Submerged mycelium: (a) nodose-septate hyphae as in the advancing zone; (b)

hyaline, branched, nodose-septate hyphae with slightly thickened walls, empty,

1.0 — 3.0u in diameter with short, much-branched processes either with thickened

walls or solid and refractive (Fig. 28 f); (c) chlamydospores hyaline, ovoid to

irregularly ellipsoid or fusoid, 6 — 14 x 4.5 — 9.0u (Fig. 28 g).

Carpophore characters

Carpophores annual, lignicolous. grouped or compound, sessile, or in circular

clusters attached at the centre, often imbricate and laterally connate; pileus

applanate to conchate or flabellate; coriaceous and somewhat watery when fresh,

drying to rigid 1 — 5 x 2 — 6 x 0.3 — 1.0 cm; surface villose-tomentose to almost

hirsute at base, to finely tomentose or glabrescent, often radially striate towards the

margin, creamy white to “cartridge buff” or “‘tilleul buff’; margin obtuse, entire

to lobate, creamy white; pore surface poroid, white, drying to “‘cream buff” or

Fic. 28.—Polyporus pubescens. a - h. Structures from cultures: (a) thin-walled,

nodose-septate hyphae from advancing zone; (b) unbranched fibre hyphae,

from nodose-septate hyphae; (f) nodose-septate, thick-walled hyphae with solid,

refractive, branched processes; (g) chlamydospores; (h) basidia and basidio-

spores.

ee s. Structures from carpophores: (k) thin-walled, nodose-septate hyphae; (m)

unbranched fibre hypha; (n) fibre hypha with branches towards the distal end;

(p) fibre hypha with short, tortuous branches; (q) solid, contorted, branched,

nodose-septate hyphae; (s) basidia and basidiospores.

240

“warm buff”; pores angular, 3 — 4 mm; dissepiments thick at first, | — 4 mm

deep: context white, zonate and with fine radiating fibres, | — 6 mm thick.

Hyphal characters: (1) nodose-septate hyphae hyaline, branching at or between

the septa, thin-walled, with deeply staining contents, 2.2 — 3.6u in diameter (Fig.

28 k); (2) fibre hyphae long, unbranched, hyaline, walls somewhat thickened,

lumina prominent, aseptate often collapsed towards the thin-walled extremities,

3.0 — 4.5u in greatest diameter, or, walls much thickened, lumina very narrow,

aseptate or occluded and hyphae sub-solid to solid with lumina visible at the

extremities only, 4.0 — 6.0u in diameter (Fig. 28 m); (3) fibre hyphae with thick

hyaline walls and aseptate lumina, branching repeatedly over a short distance

towards their tips the branches long. somewhat beaded in appearance and 3.5 —

6.0u in diameter (Fig. 28 n); (4) contorted fibre hyphae with many short thick, solid

branches, 3.0 — 6.0u in diameter (Fig. 28 p); (5) thick-walled solid nodose-septate

hyphae twisted and contorted and resembling (4) above but with solid clamps,

2.5 — 5.0u in diameter (Fig. 28 q).

Hymenium: basidia clavate, 11.0 — 16.0 x 3.6 — 4.5u, bearing four straight

sterigmata, 3.0 — 3.5u long; basidiospores hyaline, long cylindrical, occasionally

somewhat curved, obliquely apiculate, smooth, thin-walled, 4.2 — 7.2 x 1.8 —

2.4u (Fig. 28 s). Hyphal pegs broadly conical rising up to 50u above level of

hymenium.

Construction. At the margin the carpophore consists mainly of long, straight,

unbranched fibre hyphae with relatively thick walls and tips often collapsed,

the hyphae slightly interwoven and orientated parallel to the direction of growth

of the pileus. Immediately behind the foremost marginal fibre hyphae and inter-

twined with other fibre hyphae, are numerous branching, thin-walled, nodose-septate

hyphae from which the fibre hyphae arise. Intertwined and interwoven with

these hyphae are the numerous short, contorted branches of thick-walled or solid,

branched fibre hyphae which bind these other hyphal elements into a tough tissue.

Behind the margin and towards the upper part of the context the fibre hyphae

bend upwards towards the upper surface. In the upper context the unbranched

fibre hyphae are orientated upwards and lightly intertwined. Interwoven with these

hyphae are binding hyphae which are either short, tortuous branches of long fibre

hyphae or short, much-branched, fibre hyphae with sub-solid or solid branches, or,

thick-walled or solid much-branched, nodose-septate hyphae. These hyphae bind

the tissues into a tough, homogeneous mass. Thin-walled, nodose-septate hyphae

with deeply staining contents are numerous in the upper context especially just

below the upper surface. The upper surface is acuticulate and consists of the

ends of long fibre hyphae, mostly solid and projecting somewhat above the level

of the tissues, bound together by the binding hyphae to form the tomentose upper

surface of the younger part of the pileus. In the older parts nodose-septate hyphae

with deeply staining contents grow upwards beyond this layer and become closely

associated and agglutinated with the ends of fibre hyphae to form the hirsute to

fibrillar trichoderm of the older parts of the context. In the lower context the long

fibre hyphae turn downwards into the tramal tissues. Here the fibre hyphae with

terminal branches become more numerous and the short, much-branched fibre

hyphae with solid branches and thick-walled nodose-septate branched hyphae

become very numerous to form a denser and tougher tissue than the upper context.

All hyphae are more flexuous and narrower than in the upper context and are

tightly interwoven to form the tramal tissues. Here, the thin-walled, nodose-

septate hyphae with deeply staining contents are inextricably interwoven with the

other hyphae, branching profusely, the branches short and emerging on the surfaces

of the pores where they bear fasicles of basidia. On the pore surfaces are conical

hyphal pegs consisting of small bundles of fibre hyphae of which the ends project

beyond the hymenial surfaces.

Decay and hosts

Polyporus pubescens causes a white rot of hardwood logs (Nobles, 1948).

Specimens examined

Herb. DAOM: *17530, on Quercus maciocarpa, Carberry, Man., Sept. 1947; *17542, on

Betula papyrifera, Blue Lake, Duck Mi. For. Res., Sept. 1947; 17561, on dead B. papyrifera,

Buffalo Narrows, Sask., Sept. 1947; *17577, on Fagus grandifolia log, Chelsea, Que., Oct.

1947; *17578, on Acer sp., Gatineau Park, Que., Oct. 1947; *52833, on Populus sp., Belfast,

New York, Oct. 1956; 53503, Pack Forest, Warrensburg, N.Y., Oct. 1959; *73309, on dead

Alnus stump, Deux Rivieres, Ont., Sept. 1955; *94017, on dead Acer sp., Dorset, Ont., Sept.

1962; *94026, on dead Acer sp., Mouse Lake, Ont., Sept. 1962; *94039, on yellow birch,

Dorset, Ont., Sept. 1962.

Interfertility studies

Nobles (1965) reported that this species has the tetrapolar type of interfertility.

It was attempted to establish the conspecificity of some new collections of Polyporus

pubescens with older collections by means of the “Buller phenomenon.” For this

purpose, two mycelia, each obtained from a single basidiospore of collection DAOM

94039, were used as the haploid mycelium. The dikaryotic mycelia of the

collections to be examined were inoculated on the plates four days after the plates

had been inoculated with the haploid mycelia. Two days later the mycelia from

the two inocula met and after three more days the haploid mycelia were examined

at their peripheries for clamp connections. None were found. After 7 more days’

incubation the colonies were again examined and again after a further 7 days, Le.

up to 14 days after the mycelia met on the plates. No clamp connections were

ever observed on these haploid mycelia. It is thus not possible to use the “Buller

phenomenon” to test conspecificity among isolates of Polyporus pubescens. The

cause for the failure of clamps to form on these mycelia, is not known but was

not due to the disintegration of hyphae as a result of oidium formation by the

haploid mycelium as found in Merulius americanus by Hwang (1955). A similar

failure of dikaryotic mycelia to dikaryotise a haploid mycelium of the same species

was reported by Nobles (1967) in Basidioradulum radula |(Fr.) Fr.| Nobles.

The collections examined in this way are listed in TaBLe 5.

Discussion

The cultural characters as described above, agree well with those of other

species in Group 45. In an earlier description Nobles (1948) described the fibre

hyphae formed in cultures as having branches “‘ending in whiplash-like ends.”

This description, as well as a later one (Nobles, 1965) in which she mentioned the

“network of narrow hyphae” formed in cultures, agree well with the above descrip-

tion. The mat formed by Polyporus pubescens in culture resembles those of

Polypores versicolor and Lenzites betulina in many respects but may be distinguished

from these two species by careful observation and comparison of all characters.

In the carpophore five kinds of hyphae could be distinguished. It was found

that three types of hyphae. viz. nodose-septate hyphae with thickened walls, lateral

branches of long fibre hyphae and fibre hyphae with numerous short, aseptate

branches together constitute the hyphae of the binding system. These hyphae arise

in different ways and are morphologically and ontogenically different; but since

the thick-walled, nodose-septate hyphae are regarded as homologous with thin-

walled, nodose-septate hyphae by Corner (1932 a), Cunningham (1946) and

Teixeira (1962 b) and since branched fibre hyphae are regarded as hyphae of the

skeletal system like unbranched fibre hyphae by Teixeira (1962 b), the carpophores

of Polyporus pubescens have a trimitic hyphal system consisting of nodose-septate

generative hyphae, skeletal hyphae and binding hyphae as reported by Teston

(1953 b).

242

From the above description it is evident that the structures formed in cultures

are also present in the carpophores from which they were made with the exception

of the narrow, reticulately branching hyphae. These hyphae could not be located

in the carpophores. The hyphae with solid, branched processes formed in culture,

do not form pseudo-parenchymatous areas over the cultures in the manner of hyphae

with interlocking projections found in cultures of Group 53. Instead, these processes

are interwoven with fibre hyphae in the mat, which they bind into tough, felty

parts which are teased out only with great difficulty. These hyphae thus appear

to be binding hyphae which are also found in such large numbers in the carpo-

phores, interwoven with long, unbranched, fibre hyphae.

Carpophores of Polyporus pubescens resemble those of Polyporus versicolor

and Trametes suaveolens in hyphal characters and, to a lesser extent, in construction.

The same types of hyphae present in the carpophores of Polyporus pubescens are

also present in carpophores of the other two species but in apparently different

numbers, so that the carpophores differ in construction and texture. Although

carpophores of Polyporus pubescens have relatively fewer binding hyphae than

those of Polyporus versicolor, especially in the middle context, and are consequently

somewhat softer and less leathery, the binding hyphae are concentrated in the

trama and at the upper surface. In this respect the carpophores of Polyporus

pubescens differ somewhat in construction from those of Polyporus versicolor; but

they differ even more from those of Trametes suaveolens in which binding hyphae

are almost absent from the upper context. Therefore Polyporus pubescens seems

to be more closely related to Polyporus versicolor than to Trametes suaveolens

a fact recognized by Quélet (loc. cit.) when he included Polyporus pubescens in

his new genus Coriolus.

Trametes meyenii (Klotzsch) Lloyd, Myc. Writ. 5, Lett. No. 67, 14, 1918;

Polyporus meyenii Klotzsch, Nova Acta Acad. Leop. — Carol. 19, Suppl. 236,

1843;

Polystictus meyenii Klotzsch in Fungi Orb. terv. circ. a Meyen. Coll. p. 236,

1843;

Coriolus meyenii (Klotzsch) G. H. Cunningham, Proc. Linn. Soc. N.S.W., 75,

214 — 249, 1950:

Cerrena meyenti (Klotzsch) Hansen, Na. Hist. Rennell Isl., Brit. Sol. Isl. 3,

129, 1960.

Cultural characters

Growth is rapid the mat reaching a radius of 60 — 70 mm in 7 days and

covering the plate in 10 — 12 days. Advancing zone even, thin, appressed for

1 —2mm behind the margin, then raised, the young mycelium hyaline and evenly

thin, woolly, the mat increasing in thickness towards the inoculum. The thickest

parts develop faint tints of “‘cream color” after two weeks. Surface at first smooth,

even or faintly, radially sulcate and remaining so with the mat gradually thickening

to an extremely tough, sub-felty texture, or, later becoming roughened or irregularly

lacunate or granular over the thickest part and around the inoculum. The mat

remains white except over the thickest parts where colours gradually deepen to

“light buff’? or ‘“‘pale ochraceous buff’. Occasionally thin, short, erect, “‘light

buff’? spines, up to 1.0 mm in height and bearing basidia and basidiospores, develop

in small depressions in lacunose areas over the thickest part. The reverse is

bleached after 2 weeks and a faint, fragrant, mushroomy odour is given off up

to the fourth week. On gallic acid and tannic acid media colonies of up to 40

4 ~~ Minas.

Fic. 29—Trametes meyenii. (a) Carpophore of PRE 42446, upper surface and (b)

hymenial surface; (c) culture of PRE 42446 at six weeks; (d) fibre hypha with numerous

long, tapering branches from fruit-body, x 250; (e) “Buller phenomenon” plates of

PRE 42459, bottom, diploid, with PRE 42446 — 14 and PRE 42446 — 16, top, haploid.

244

mm and 60 mm respectively and strong diffusion zones, are formed in seven days.

A strong positive reaction results when cultures are tested for extra-cellular oxidase

by means of gum guaiac solution.

Advancing mycelium: hyphae hyaline, thin-walled, with simple clamp connections

at the septa, branching near the septa. 2.5 — 4.0u in diameter (Fig. 30 a).

Aerial mycelium: (a) hyphae as in the advancing zone; (b) fibre hyphae hyaline,

long, unbranched, narrow towards the ends. widening to up to 6u in diameter

near the middle part, walls thick, refractive, lumina aseptate, narrow or occluded,

widening towards the ends (Fig. 30 b); (c) fibre hyphae narrow, 1.5 — 3.0u in

diameter branching repeatedly over a short distance of main stem, the branches

sub-solid to solid, flexuous, tapering gradually towards the tips (Fig. 30 c); (d)

oidia numerous in some isolates, hyaline, ovoid to cylindrical with rounded ends,

2.0 — 3.0 x 3.0 — 6.0u (Fig. 30 d).

Fructifications: basidia short, broadly clavate, 10.0 — 15.0 x 4.5 — 6.0u, with

four slender, straight, sterigmata 3.04 long (Fig. 30 e}; basidiospores hyaline,

smooth, ellipsoid-cylindrical to cylindrical with rounded ends, obliquely apiculate.

thin-walled, 4.5 — 6.0 x 2.0 — 2.5u (Fig. 30 f).

Submerged mycelium: (a) nodose-septate hyphae as in the advancing zone, often

without contents and walls thickened; (b) oidia as in the aerial mycelium.

Carpophore characters

Carpophores annual or reviving a second season, lignicolous, solitary; pileus

sessile, dimidiate, occasionally imbricate, to effused-reflexed, often concave above,

woody or corky up to 30 x 12 x 6 cm; upper surface yellowish buff to brownish

grey or grey, velvety to tomentose often glabrous with age or rimose, often

tuberculate and concentrically sulcate, zonate or azonate, mostly with a hard,

bay-brown layer under the tomentum in the older part; margin obtuse, rounded,

entire or lobed, finely velvety, creamy white, drying yellowish; pore surface creamy

white when fresh, drying to pale buff or creamy buff; pores usually elongate,

sinuous to labyrinthiform 1 — 3/mm, dissepiments thin; tubes 0.5 — 2.0 mm deep,

concolourous with surface, occasionally indistinctly stratified, becoming bleached

in age. Context creamy white to pale creamy or buff, fibrous, zonate, 5.0 — 40.0

mm thick, often with a dark, hard zone under the tomentum in older part.

Hyphal characters: (1) thin-wallzd, nodose-septate hyphae with simple clamp con-

nections, frequently branched, 2.0 — 3.0u in diameter (Fig. 30 g); (2) fibre hyphae

hyaline, long, unbranched, narrow, thin-walled, 1.5 — 2.0u in diameter near their

origin, widening to 61 maximum diameter with narrow or occluded, aseptate lumina

(Fig. 30 h); (3) fibre hyphae as above but with two or three branches towards

the distal ends, branches long, tortuous and tapering towards the ends, 1.5 — 3.5u

in diameter (Fig. 30 k); (4) fibre hyphae unbranched as above but with short,

barblike, lateral projections or branches towards the distal ends, 4.0 — 6.0u in

diameter (Fig. 30m); (5) hyaline fibre hyphae usually solid, without clamp

connections or septa, with numerous branches, the branches short and tortuous or

long, flexuous and tapering towards their ends, 0.7 — 3.0 in diameter (Fig. 30 n);

(6) nodose-septate hyphae with numerous short, tortuous, sub-solid to solid

branches, 1.2 — 3.0u in diameter (Fig. 30 p).

Hymenium: basidia short, broadly clavate, 10.0 — 15.0 x 4.5 — 6.0u, with four

sterigmata, 3.0u long (Fig. 30 p); basidiospores hyaline, smooth, ellipsoid-cylindrical

or short cylindrical with rounded ends, obliquely apiculate, thin-walled, 4.5 — 6.0 x

2.0 — 2.5u (Fig. 30 q). Hyphal pegs small, conical, projecting up to 604 beyond

the hymenium.

245

Construction. The margin of the carpophore consists mainly of long unbranched

fibre hyphae with narrow lumina and arranged parallel to one another. Intertwined

with these are numerous thin-walled, nodose-septate, frequently branching hyphae,

from which they arise. Behind the margin the fibre hyphae are mostly solid.

Hyaline, solid or sub-solid, aseptate binding hyphae with their branches interwoven

with the other hyphae, appear in the context tissue. The middle and upper context

consist mainly of more or less parallel, intertwining fibre hyphae, unbranched or

with one to three branches towards the tips, which turn upwards towards the upper

surface. Intertwined with these are thin-walled, nodose-septate hyphae just below

the upper surface, and numerous narrow, solid, branching fibre hyphae, binding the

tissues together, in the middle part of the context. At the upper surface numerous

branched fibre hyphae, mostly with the upper parts of their lumina fairly wide,

project beyond the context tissues to form the hairy upper surface present in many

specimens. In the older parts of the specimen, these fibre hyphae often become

agglutinated by a hard, lacquer-like substance into a hard, rimose crust up to 1.5

mm thick. In the younger parts, the fibre hyphae project very little beyond the

upper level of tissue composed of nodose-septate hyphae, fibre hyphae and binding

hyphae to form the smooth, finely pubescent, upper surface. In the lower context

and towards the tubes, the tissues become increasingly dense and consist of solid,

unbranched, fibre hyphae and fibre hyphae with one to three long branches near

their ends, often somewhat flexuous, and turning downwards towards the dissepi-

ments. Intertwined with these are occasional fibre hyphae with short lateral

projections. Numerous “binding hyphae”, some with solid clamp connections,

tightly interwoven with the other fibre hyphae, bind them into a tough, woody

tissue. In the dissepiments the fibre hyphae are indistinguishable from each other,

are very tortuous and tightly interwoven but orientated in a downward direction.

Here they have slightly wider lumina and are narrower in diameter. Intertwined

with the fibre hyphae are numerous, thin-walled, nodose-septate hyphae which

branch profusely towards the surfaces of the dissepiments where they bear clusters

of basidia on numerous, short, lateral branches. On the hymenial surfaces, conical

hyphal pegs.consisting of bundles of parallel ends of fibre hyphae, project beyond

the hymenium.

Decay and hosts

This fungus causes a white rot of hardwoods. It is common in subtropical

regions where it grows on dead wood but is frequently found fruiting on the trunks

of living Acacia species in South Africa.

Specimens examined

Herb. DAOM: 30792, on Cassia siamea, Njala, Sierra Leone, Aug. 1953; 69924, on Cassia

siamea Rennell Island, Brit. Solomon Isl.. Oct. 1954.

Herb, PRE: 1873, on Citrus sinensis, Pretoria, Tvl., July 1911; 2114, on dead hardwood,

Pretoria, Tvl., Feb. 1912; 5184, on Acacia decurrens, Ixopo, Natal, Sept. 1912; 6917, on Acacia

decumens, Ixopo, Natal, July 1913; 8859, on Acacia horrida, Pretoria, Tvl.. March 1915;

8877, on Acacia horrida, Pretoria, Tvl., March 1915; 9545, on dead Celtis thamnifolia, Ngadu

Forest, C.P., Jan. 1916; 11437, in Fungi Malayana, Mt. Maquiling, Philippines, July 1916;

12183, on dead Acacia sp. stumps, Adelaide, C.P., May 1919; 13040, on stumps, Mulanga

For., Uganda, July 1919; 13943, as Dacdalea hobbsii v.d. Bijl, Howick, Natal; 14651, on

dead tree, Pretoria, April 1921; 14840, on dead tree, Kirstenbosch, Cape Prov., June 1921;

14904, on dead tree, Entebbe, Uganda; 15569, on dead tree, Ginginhlovu, Natal, July 1915;

24874, on dead tree, Pretoria, March 1915; 26402, on dead tree, Pietermaritzburg, Natal,

Aug. 1932; 27564, dead stump, Pretoria, Tvl., April 1934; 27705, dead stump, Donnybrook,

Natal, Jan. 1935; 28563, as Daedalea hobbsii, Winters Kloof, Pietermaritzburg, Sept. 1934;

28875, on Acacia sp., Louis Trichardt, Tvl., Aug. 1937; 30267, on dead wood, Donnybrook,

Natal, Feb. 1936; 30725, on dead woo, Rustenburg, Tvl., March 1939; 30804, on dead

wood, Xumeni Forest, Natal, Dec. 1936; 31311, on dead wood, Knysna, C.P., Dec. 1936;

31543, on dead wood, Pretoria, Tvl., July 1915; 31548, on dead wood, Grahamstown, C.P..

247

Aug. 1915; 31586, on dead wood, Pretoria, Tvl. 1915; 31623, on rotten logs, Knysna, C.P.,

Jan. 1916; 31685, on rotten logs Pietermaritzburg, Natal, Aug. 1916; 31712, on rotten

logs, Pietermaritzburg, Natal, Dec. 1916; 31729, on rotten logs, Pietermaritzburg, Natal.

Dec. 1916; 31820, on rotten logs, Umgeni, Natal, April 1917; 31974, on living Trema sp.,

Durban, Nov. 1917; 32030, on Celtis kraussiana, Durban, Noy. 1917; 33072, on Acacic

mollissima, Donnybrook, Natal. June 1940; 33905, on dead wood, Scottburgh, Natal, Feb.

1943; *42446, on Acacia mearnsii stump, Graskop, E. Tvl., Feb. 1961; *42449, on Acacia

mearnsii stump, Kaapse Hoop, Tvl., Feb. 1961; *42459, decaying hardwood log, Blouberg.

Tvl., Apr. 1961; *42723, on living Acacia sp., Rustenburg, Tvl., Aug. 1953.

Herb. STE: 262, at base of Acacia trez. Pretoria; 575. at base of Acacia tree, Howick Natal;

2267, at base of Acacia tree, Pietermartizburg, March 1930; 2755, at base of Acacia tree,

Maclear, C.P.

Interfertility studies

In order to determine the type of interfertility of this species, a set of 16

cultures, each grown from a single basidiospore, was made from spores collected

from a fructification formed in a culture of PRE 42446. By pairing these cultures

in all possible combinations, it was found that Trametes meyenii has the tetrapolar

type of interfertility with allelomorphs for heterothallism at two loci. The results

showing the distribution of mating types among the single spore mycelia are

presented below in TABLE 6.

The conspecificity of other collections of Trametes mieyenii of which cultures

were available, was tested by means of the “Buller phenomenon”. Two mycelia

of single spores of opposing mating types, PRE 42446 —- 3 and PRE 42446 — 14,

were used. Seven days after placing the dikaryotic mycelia on the plates on

which the above haploid mycelia were growing, the haploid mycelia were examined

for clamp connections at three different positions along their periphery. Clamp

connections, indicating dikaryotization of the single spore mycelium by the added

dikaryotic mycelia, were found in very position. Collections of Trametes meyenii

numbered PRE 42449, PRE 42459 and PRE 42732 were demonstrated to be

conspecific with PRE 42446 (Fig. 29 e).

Discussion

As described above, the cultural characters agree with an earlier description

(Van der Westhuizen, 1958) but the hyphae are described in greater detail. Besides

the thin-walled, nodose-septate hyphae, two types of fibre hyphae, one unbranched,

and resembling the “‘vermiculiform skeletal hyphae’’ figured by Teixeira (1962 b),

the other branched and resembling the ‘‘branched fibre hyphae with whiplash-like

ends” described by Nobles & Frew (1962) from Pycnoporus spp., are present.

These characters, as well as the strong positive oxidase reaction of the culture,

agree well with the characters of other species in this group, especially with those

of Polyporus versicolor and Lenzites betulina; but the cultures of Trametes meyenii

form a thicker, smoother and more woolly mat, than those of Polyporus versicolor

and Lenzites betulina, and develop colours not found or the other species. This

combination of characters may serve to distinguish cultures of this species from

cultures of other species in this group which are otherwise very similar.

Fic. 30.— Trametes meyenii. a - f. Structures from cultures: (a) thin-walled

nodose-septate hyphae from advancing zone; (b) unbranched fibre hyphae; (c)

fibre hyphae with long tapering branches; (d) oidia; (e) basidia; (f) basidio-

spores.

ae Structures from carpophores: (g) thin-walled, nodose-septate hyphae;

(h) unbranched fibre hyphae; (k) fibre hypha with branches toward the distal

end; (m) fibre hypha with lateral projections; (n) fibre hypha with numerous

long, flexuous branches; (p) nodose-septate hypha with thickened walls and

numerous short, tortuous branches; (q) basidia and basidiospores.

248

The presence of the tetrapolar type of interfertility in Trametes meyenii,

which causes a white rot of hardwood and produces extra-cellular oxidase enzymes

in culture, is in full agreement with Nobles’ (1958 b) thesis that tetrapolarity is

correllated with the production of extra-cellular oxidase in a large group of poly-

pores with simple clamp connections on their thin-walled hyphae.

In the carpophores cf Trametes meyenii five kinds of hyphae could be

distinguished, viz.: thin-walled and thick-walled nodose-septate hyphae, unbranched

fibre hyphae, fibre hyphae with one or two branches towards the tip and fibre

hyphae with many short, flexuous branches. These latter are the binding hyphae

in Corner’s (1932 a) terminology but the thick-walled, nodose-septate hyphae as

well as the branches of some of the long fibre hyphae also act as binding hyphae.

Hyphae which are morphologically and ontogenically different thus contribute to

the binding hyphal system. Since the nodose-septate hyphae are regarded as

generative hyphae and the unbranched or sparingly branched hyphae as skeletal

hyphae (Corner, 1932 a, b; Cunningham, 1946; Teixeira, 1962 b) carpophores of

Trametes meyenii have a trimitic hyphal system in the terminology of Corner

(1932 a, b) and Cunningham (1954).

From the above descriptions it is evident that the structures formed in culture

are also present in the carpophores from which they were made. A greater variety

of structures are present in the carpophores than in the cultures, a phenomenon

also observed in most other species studied. The branched fibre hyphae or binding

hyphae from cultures develop longer, less flexuous branches but in morphology

approach the binding hyphae of the carpophores more closely than the branched

skeletal hyphae.

The characters of the hyphae and construction of the carpophores of Trametes

meyenii resemble those of Polyporus versicolor and Trametes suaveolens very

closely. In construction the thin forms of Trametes meyenii resemble the carpo-

phores of Polyporus versicolor by having large numbers of binding hyphae in the

upper and lower context. The thick forms, some of which were found to be

conspecific with a thin form by means of the “Buller phenomenon” technique, on

the other hand have relatively fewer binding hyphae in the upper context and

resemble carpophores of Trametes suaveolens more nearly. Such thick carpophores

approach the softer upper surface and feel of those of Trametes suaveolens but

are never as soft or have so few binding hyphae in the upper context as to be

closely similar to the carpophores of Trametes suaveolens. [rametes meyenii thus

appears to be a transitional species between Polyporus versicolor L. ex Fr., the

type species of the genus Coriolus Quél., on the one hand and Trametes suaveolens

(L. ex Fr.) Fr., the type species of the genus Trametes Fr., on the other. This

indicates a close relationship between the two generic type species and is evidence

in support of Pilat’s (1936) and Kotlaba & Pouzar’s (1957) inclusion of the genus

Coriolus in the genus Trametes Fr.

Carpophores of Trametes meyenii also resemble those of Lenzites betulina

in the usually lamellate hymenium and to a large extent in hyphal characters and

construction, but the carpophores of Lenzites betulina do not have fibre hyphae

with one or two branches near the tips which are present in the carpophores

of Trametes meyenii, Polyporus versicolor and Trametes suaveolens. In respect

of the characters of the other hyphae and in construction of their carpophores there

are close similarities between Trametes meyenii and Lenzites betulina. It thus

appears that Trametes meyenii has characters of all three genera, Coriolus Quél.,

Lenzites Fr. and Trametes Fr., combined in its carpophores. This supports the

view expressed by O. Fidalgo (1957) that the genera Trametes and Lenzites are

congeneric but until the importance of the absence of these fibre hyphae with one

or two branches can be determined, it appears desirable te regard them as distinct

genera.

249

The relationship of Trametes meyenii to the genus Coriolus Quél., was recog-

nized by Cunningham (1950 b) and Imazeki (1952) who independently transferred

this species to the genus. Hansen (1960) transferred this species to the genus

Cerrena Mich. ex S. F. Gray but the carpophores of the type species of this genus,

Cerrena unicolor (Bull. ex Fr.) Murr. was shown to consist entirely of nodose-septate

hyphae, mostly thick-walled, while nodose-septate, thick-walled hyphae were formed

in its cultures (Van der Westhuizen, 1963). ranetes meyenii thus has no affinities

with the genus Cerrena Mich. 2x S. F Gray.

Murrill (1907 b) cited Polyporus meyenii Klotzsch as a synonym of Coriolus

maximus (Mont.) Murrill based on Irpex maximus Montagne of which the descrip-

tion antedates that of Polyporus meyenii by about six years. Overholts (1953) ac-

cepted Murrill’s synonymy but made the combination Polyporus maximus (Mont.)

Overholts which was later accepted by Lowe & Gilbertson (1961 b). Lloyd (1918),

Imazeki (1952) and Hansen (1960) however regard /rpex maximus and T. meyenii

as distinct species because of differences in the shape of their pores. This matter

could not be investigated as the type specimens of these two species were not

available for study. For this reason the name Trametes meyenii (Klotzsch) Lloyd

which is well established in South Africa for the specimens studied, is used here.

Fic. 31.—Lenzites palisoti. (a) Carpophore of PRE 42442, upper surface and (b) hymenial

surface; (c) fibre hypha with numerous short branches and unbranched fibre hyphae

from context of fruit-body, x 1000 phase contrast; (d) culture of PRE 42442 at six

weeks.

250

Lenzites palisoti (Fr.) Fries, Epic. Syst. Mycol. 404, 1838:

Daedalea palisoti Fr., Syst. Mycol. 1, 335, 1821;

Trametes palisoti (Fr.) Imazeki, Bull. Tokyo Sci. Mus. 6, 73, 1943;

Coriolus ambiguus (Berk.) G. H. Cunn., Proc. Linn. Soc. N. S. Wales 75,

216, 1950.

Cultural characters

Growth is moderately fast to slow the mat reaching a radius of 22 mm after

one week and covering the plate after three to five weeks. The advancing zone

is even at first but becomes uneven due to irregular growth of parts of the margin

which result in radially elongate patches projecting beyond the margin. Mycelium

mostly submerged, forming irregular milky patches in the agar. Aerial mycelium

scanty, at first thin, downy, white becoming submerged but having irregular,

radially elongated “islands” of white, farinaceous, downy mycelium which may

gradually become tough, sub-felty or coalesce to form irregular, tough, white,

sub-felty areas especially in the older parts of the cclony. Fertile areas form

occasionally on some of these “islands” but are not visibly differentiated. The

reverse bleaches slowly and the cultures emit a strong, penetrating, sweetly fragrant

odour. On gallic acid and tannic acid media strong diffusion zones are formed

and colonies grow to diameters of 32 mm on gallic and 10 mm on tannic acid

media in seven days. A strong blue colour is quickly formed when cultures are

ested for the presence of extra-cellular oxidase by means of alcoholic gum guaiac

solution.

Advancing mycelium: hyphae hyaline, branching, thin-walled, nodose-septate with

simple clamp connections, 2.5 — 3.5u in diameter (Fig. 32 a).

Aerial mycelium: (a) hyphae as in the advancing zone; (b) fibre hyphae hyaline,

long, unbranched, refractive, mostly solid or with lumina reduced to an interrupted

line, and suddenly expanding at the apex, 2.5 — 4.6u in diameter (Fig. 32 b);

2.5 —4.0u in diameter (Fig. 32 c); (d) fibre hyphae hyaline, mostly solid, tortuous

with numerous short, solid, tortuous branches, 2.0 — 3.5u in diameter intertwined

with the long fibre hyphae and binding them into a tough tissue (Fig. 32d); (e)

nodose-septate hyphae with walls thickened or solid. empty or with staining

contents and numerous short, tortuous, sub-solid or solid branches, 1.5 — 2.5u in

diameter, often inflated terminally up to 8u in diameter or with clusters of short

lateral projections, mostly solid, at the apex (Fig. 32 e).

Fructifications: basidia long, clavate almost cylindrical 10.0 — 15.0 x 3.5 — 4.0u

with four slender sterigmata 2.5 — 3.0u long (Fig. 32 f); basidiospores ellipsoid-

cylindrical to cylindrical, obliquely apiculate, hyaline, smooth, thin-walled, 3.5 —

6.5 x 2.0 — 2.5u (Fig. 32 g).

Submerged mycelium: (a) hyphae as in the advancing zone; (b) chlamydospores

thick-walled, fusiform, ovoid to sub-globose, 3.5 — 10.0 x 5.0 — 12.0u.

Carpophore characters

Carpophore annual or reviving, lignicolous, solitary. sessile or sub-stipitate

or with reduced, peltate base; pileus reniform to flabelliform or orbicular, variable,

tough coriaceous to firm, rigid or woody, up to 20 x 35 x 3 cm; upper surface

minutely velutinate or glabrous, tubercular or smooth or radially and concentrically

sulcate, white when fresh drying to cream coloured, occasionally dark brown to

blackish near the base; margin thin, acute, rounded, undulate or entire, white to

cream coloured, sterile underneath; pore surface white to cream coloured poroid

743)

to daedaloid or lenzitoid, 2 — 3 per mm, entire; tubes 2 — 4 mm deep, con-

colourous; context pure white or creamy white, even-textured, floccose, punky or

corky.

Hyphal characters: (1) thin-walled hyphae nodose-septate with simple clamp

connections, branching frequently, 1.8 — 3.0u in diameter (Fig. 32 k); (ii) fibre

hyphae hyaline unbranched, narrow near origins, with wide lumina, widening

towards middlepart and simultaneously thickening of walls and narrowing of lumina,

the latter widening towards the thinner-walled distal ends, aseptate, 3.0 — 6.0u

in diameter, or mostly solid with luminalacking or reduced to a narrow interrupted

line which widens suddenly toward the tips (Fig. 32 m); (iii) fibre hyphae with

numerous branches, the branches short, very tortuous or longer and_ tapering

somewhat, hyaline, with thickened walls and conspicuous lumina or sub-solid

with lumina lacking, aseptate, 1.5 — 3.5u in diameter (Fig. 32 n); (iv) nodose-septate

hyphae sub-solid or solid with short, tortuous branches, 1.5 — 3.5u in diameter

(Fig. 32 p); (v) fibre hyphae sub-solid or solid with two or three long branches

arising over a short distance of main stem, 3.0 — 6.0u in diameter (Fig. 32 q);

(vi) short fibre hyphae with terminal clusters of branches with prominent lumina

projecting into the hymenium, 1.0 — 3.0u in diameter (Fig. 32 s).

Hymenium: basidia narrowly clavate, 18.0 — 24.0 x 4.5 —6.0u with four slender

sterigmata up to 3.0u long (Fig. 32 t); basidiospores cylindrical, smooth, hyaline,

thin-walled, 5.0 — 7.0 x 2.0 — 2.4u (Fig. 32 x); cystidioles occasional, narrowly

cylindrical, hyaline, thin-walled, with deeply staining contents, 20.0 30.0 x

1.0 — 2.0u and arising from the basidial fascicles (Fig. 32 y).

Construction. At the margin the carpophore consists of long, hyaline, unbranched,

fibre hyphae more or less parallel to each other and slightly intertwined. Numerous

branching and anastomosing, thin-walled, nodose-septate hyphae from which the

fibre hyphae arise are intertwined and interwoven with them deeper in the tissues.

Behind the margin in the upper context, the unbranched fibre hyphae turn towards

the upper surface. Branched fibre hyphae (binding hyphae) with fairly wide

lumina and the branches fairly long and interwoven with the fibre hyphae across

their direction of growth, become increasingly numerous towards the upper surface.

In the upper context the long, unbranched, fibre hyphae are mostly solid and run

parallel to each other towards the upper surface where their ends are packed at

a common level to form the finely pubescent surface of the younger part of the

carpophore. At the upper surface, thin-walled, hyaline branching and anastomos-

ing, nodose-septate hyphae become very numerous. They are tightly intertwined

and interwoven with the fibre hyphae to form the smooth, glabrous, upper surface

which becomes covered with a very thin, hyaline layer of lacquer-like material

over the older parts of the carpophore The dark patches on the upper surface

of some specimens, consist of the ends of fibre hyphae with their lumina dilated

at the apex and the luminal contents discoloured to reddish brown or dark grey

brown. The ends are embedded in a dark-coloured, resin-like substance.

In the lower context the unbranched fibre byphae are arranged parallel

to the direction of growth of the fruit-body while some turn downward toward the

dissepiments. Fibre hyphae with long or short, tortuous branches and sub-solid

or solid, nodose-septate hyphae with tortuous branches are interwoven with the

unbranched fibre hyphae forming a tough dense tissue. In the lower parts of these

tissues and in the dissepiments, thin-walled, nodose-septate hyphae become

increasingly numerous, branching and anastomising and bearing basidia in fascicles

on the surfaces of the dissepiments. In the dissepiments short fibre hyphae arise,

bearing numerous, short, terminal branches which project into the hymenium as

pseudocystidia or paraphyses as described by Overholts (1953) (Fig. 32 s).

Decay and hosts

This species causes a white rot of the sapwood of hardwood logs. It is widely

distributed throughout the warmer regions of the world.

Specimens examined

Herb. DAOM: F5255, on Quercus log, Endora, Arkansas, Aug. 1931; F2093, on Quercus sp.,

Buzzards Roost, Alachua Co., Florida, Sept. 1954; *69696, on dead wood, Walker, Louisiana,

Aug. 1960.

Herb, Berkeley in Herb. K: Daedalea ambigua Berk., from Ohio, (Holotype).

Herb. NY: Lenzites repanda Fr., Sicrra Leone, Africa, 1889, North Am. Fl. 6003; Lenzites

palisoti Fr., Somerset East, C.P., Feb. 1876, North Am. Fl. 804; Lenzites deplanata Fr.,

Plants of New Guinea No. 208, Crane Expedition, May 1929; Lenzites palisoti Fr., Sydow,

Fungi exotica exsicati, 304, Luzon, Mt. Isarog, Dec. 1913; Dacdalea amanitoides Beauv.

23714, Hulgra, Ecuador, Aug. 1918; Daedalea amanitoides Beauy. Dutch Guiana, Herb.

NYBG 12; Daedalea amanitoides, British Guiana, 13529; Lenzites applanatus Fr., Novo

Petropolis, Rio Grande du Sul, Brazil, 1924; Lenzites applanata Fr., Porto Novo, Sta.

Catharine, Brazil, 1928; Lenzites palisori Fr., Baker, Fungi Malayana 242; Daedalea aesculi,

ex Herb. A.P. L. V. Morgan, No. Am. FI.; Daedalea ambigua Berk, on dead, standing oak

trees, Perryville, Mo., Aug. 1885; Trametes lactea, Florida, N. W. Calkins 185, Feb. 1886;

Trametes ambigua, Fern, Putnam Co., Ind., Sept. 1891, L.M. Underwood Coll.; Daedalea

aesculi, near New Orleans, L.A., F. S. Earle, Aug. 1908, NYBG 113; Daedalea aesculi,

Batesville, Arkansas, Oct. 1908, 2829; Daedalea ambigua Berk., Middlebrove IIl., Sept. 1907,

2518; Daedalea amanitoides, Panama Canal Zone, S. L. Meyer, Jan. 1945, 17303.

Herb. PRE: 41, Lenzites repanda, Zululend; 156, dooie hout, Salisbury, S. Rhodesia; 354,

Lenzites repanda, dead logs, Zululand; 372, Lenzites applanata, Knysna, Jan. 1922; 757,

Lenzites repanda, ou hout, Knysna; 911, Lenzites applanata on dead Acacia melanoxylon,

Grootvadersbosch, Swellendam; 1235, Hectorspruit, Tvl., July 1916; 1318, Zoutpansberg,

Transvaal; 1322, on Acacia decurrens, Harden Heights, Natal, April 1911; 1323, on wattle

stumps, Cramond, Natal, April 1911; 1648, Lenzites applanata, on Eucalyptus poles, Tzaneen,

July 1924; 1878, on dead tree stump, S. Rhodesia, Aug. 1923; 2370, Lenzites repanda, dead

stump, Salisbury, S. Rhodesia, Dec. 1924; 2444, Lenzites repanda, on old stump, Zimbabwe,

Rhodesia, July 1927; 2532, Lenzites repanda, Mycological Herbarium, Dept. of Agriculture,

S. Rhodesia per J. C. Hopkins, May 1928; 6749, Sydow, Fungi exotica exsiccati No. 101;

8819, Pietermartizburg, Natal, Feb. 1915; 8877, on dead Acacia karroo, Pretoria, Tvl., March

1915; 9750, Hectorspruit, Tvl., July 1916; 9957, Fungi Malayana, No. 242; 11550, Buccleugh,

Natal, July 1918; 11917, Kyiwaga, Uganda, Aug. 1915; 13155, Flora of the Philippines

No. 45; 14932, Victoria Nyanza, Uganda; 15050, on dead Ocotea bullata, Hankey, C.P.,

Nov. 1921; 15488, on dead wood, George, C.P., May 1922; 17805, on Olea laurifolia, Knysna,

C.P., June 1923; 22055, Somerset East, C.P., 1875; 23350, Alexandra Forest, C.P., July 1927;

27269, Donnybrook, Natal, Aug. 1933; 27707, Donnybrook, Natal, Jan. 1935; 28941, Xumeni

Forest, Natal, July 1935; 30101, on Acacia mollisima, Butterworth, C.P., Aug. 1937; 30733,

Knysna, C.P., Apr. 1949; 31624, Knysna, Cape Province, Jan. 1916; 31671, Ginginhlovu, Natal,

May 1916; 33378, on dead wood, Grahamstown, C.P., Sept. 1941; 33923, Mufulira Copper

Mines, Zambia, Feb. 1943; 34578, Lothair, E. Tvl., Apr. 1945; 34993, on Olea laurifolia,

Qudeni Forest, Natal, Feb. 1945; 38738, Petrak, Mycotheca generalis No. 1831; *41534,

Hogsback, C.P., May 1956; *42094, oa wood, Potgietersrus, Tvl., March 1960; 42432, on

dead hardwood, Graskop, E. Tvl., March 1957; *42442, on Acacia sp. log, Bushbuckridge,

E. Tvl., Feb. 1961; *42531, on hardwood log, Louis Trichardt, N. Tvl., April 1964; *42748,

on Olea laurifolia, Knysna, C.P., Oct. 1955; 43157, on dead wood, George, C.P.. March

1966.

Fic. 32.— Lenzites palisoti a - h. Structures from cultures: (a) thin-walled, nodose-

septate hyphae; (b) unbranched fibre hyphae; (c) solid fibre hypha with long

branches; (d) fibre hyphae with short branches; (e) thick-walled, nodose-

septate hyphae with short, tortuous branches; (f) basidia; (g) basidiospores;

(h) chlamydospores.

k-y. Structures from carpophores: (k) thin-walled, nodose-septate hyphae;

(m) unbranched fibre hyphae; (n) fibre hyphae with numerous, tortuous, short

branches; (p) thick-walled, nodose-septate hyphae with tortuous branches; (q)

fibre hyphae with branches toward the distal end; (s) short fibre hyphae with

terminal clusters of branches; (t) basidium; (x) basidiospores; (y) cystidiole.

254

Interfertility studies

Spores were obtained from a fruiting area of the culture of PRE 42442. In

order to determine the type of interfertility of this species. 16 cultures, each grown

from a single basidiospore, were paired in all possible combinations. The results

indicate that Lenzites palisoti possesses the tetrapolar type of interfertility with

allelomorphs for heterothallism at two loci. The mycelia from five single spores

did not mate with any one of the other mycelia. Where they were used in

pairings, the two mycelia never met even after prolonged incubation but showed

signs of mutual aversion and inhibition. The results, showing the distribution of

mating types among the single spore mycelia are presented below in TABLE 7.

The cause of the failure of these five monospore cultures to mate with the

others is not known and no attempt was made to determine it. It may be due

to the “‘barrage phenomenon” as described by Vandendries and Brodie (1933)

and Brodie (1935, 1936) or to staling effects since it was noticed that these

monospore mycelia grew rather poorly in culture and always formed restricted

colonies.

Discussion

The presence of nodose-septate hyphae, fibre hyphae and extra-cellular oxidase

in its cultures and cylindrical basidiospores in the carpophores, places Lenzites

palisoti in Group 45. In most respects it agrees in cultural characters with other

trametoid species in this group but the rather penetrating fragrant odour given off

by the ragged cultures, which produce so much submerged mycelium, serve to

distinguish this species from others in this group.

Nobles (1958 b) placed Lenzites repanda (Pers.) Fr., (synonymous with Lenzites

palisoti, Fidalgo & Fidalgo, 1966) in Group 27, among species with simple septate

hyphae and a negative oxidase reaction in culture, but she placed Daedalea

ambigua Berk. (synonymous with Lenzites palisoti, Fidalgo & Fidalgo, 1966)

in Group 45. Davidson et al. (1938) reported th> formation of strong diffusion

zones by cultures of Daedalea ambigua on gallic acid and tannic acid media and

the formation of a white rot by the fungus on wood. Van der Westhuizen (1958)

also reported a strong, positive, oxidase reaction in culiures of Lenzites palisoti

and a white rot of wood decayed by this common and widely distributed species.

Carpophores of Lenzites palisoti agree in anatomical characters with other

species of this group as described above. The nodose-septate hyphae may be

thin-walled generative hyphae or may be thick-walled “‘sclerified generative hyphae”

with tortuous branches which form part of the binding system. The fibre hyphae

may be unbranched, solid, or have a number of long, flexuous branches which

contribute to the binding system or they may be short binding hyphae with

numerous, short, tortuous branches. Hyphae which are morphologically and

ontogenically distinct thus comprise the binding hyphal system. Because nodose-

septate hyphae are regarded as generative hyphae, and because aseptate binding

hyphae are present in the carpophores, the fruit bodies of Lenzites palisoti have

a trimitic hyphal system in Corner’s (1932 a, b) terminology.

From the above descriptions it is clear that the structures formed in cultures

are also present in the carpophores of Lenzites palisoti from which they were made.

The chlamydospores are however again the exception but this appears to be the

general condition. It is possible that chlamydospores may be formed in the wood

decayed by this fungus, which was not available for examination.

255

Despite its wide distribution, little was known about the anatomical characters

of Lenzites palisoti. Cunningham (1950 b) placed this species in the genus Coriolus

Quél, which he had characterized earlier (Cunningham, 1948 c) as having a trimitic

hyphal system with nodose-septate generative hyphae, unbranched, aseptate, skeletal

hyphae and aseptate, branched, binding hyphae of the bovista type. Overholts

(1953) stated that the hyphae in the carpophore of Daedalea ambigua Berk.

(— Lenzites palisoti (Fr.) Fr.) are ‘‘mostly simple, thick-walled with no cross-walls

or clamps, 3 — 7u in diameter; sometimes much branched hyphal complexes

present.”

Fidalgo & Fidalgo (1966) reported this species to possess a trimitic hyphal

system with ‘‘generative hyphae thin-walled, hyaline . . . with clamp connections”’,

“skeletal hyphae thick-walled, . . . unbranched, straight or wavy, never twisted . .

with no clamps or simple septa; binding hyphae thick-walled to solid, . . . much

branched, very tortuous, curled, with no clamps or simple septa’. These reports

agree in most respects with the descriptions given above but these authors do not

mention the presence of either the branched, thick-walled nodose-septate ‘binding

hyphae,” which are numerous in the lower context and trama of the fruit-bodies,

or the branched fibre hyphae which are present in the lower parts of the

carpophores.

Fidalgo & Fidalgo (1966) cited the full synonymy of this widely distributed

species. From this list it is clear that the numerous specific epithets have been

combined with the genera Daedulea Fr., Trametes Fr. or Lenzites Fr. by most

authors. Fidalgo & Fidalgo (1966) regard Daedalea elegans Spreng. ex Fr. as the

correct name for this fungus in view of O. Fidalgo’s (1957) earlier argumentation

that the genera Lenzites Fr. and Trametes Fr. are synonymous with Daedalea Fr.

In that argumentation Lenzites palisoti was quoted as the species in which the

morphological characters of all three genera were combined. ‘Comparison of the

cultural characters and hyphal characters of Lenzites palisoti with those of the

type species of Daedalea Fr., Daedalea quercina, however reveal many differences.

Cultures of Daedalea quercina do not produce extra-cellular oxidase like those

of Lenzites palisoti, and form nodose--septate hyphae with irregularly thickened

walls which are absent from the cultures of Lenzites palisoti. These nodose-

septate hyphae with irregularly thickened walls are also present in the carpophores

of Daedalea quercina but absent from those of Lenzites palisoti. In carpophores

of the latter species, fibre hyphae with numerous tortuous branches (or binding

hyphae) are present in large numbers but are absent from those of Daedalea

quercina. Since such differences in the kinds of hyphae present in carpophores

indicate phylogenetic dissimilarities and are regarded by Corner (1953), Cunningham

(1954), Bondartzeva (1961) and Teixeira (1926 b) as of generic importance, Lenzites

palisoti cannot be regarded as being congeneric with Daedalea quercina. On the

other hand, it is evident from the descriptions given above that Lenzites palisoti

resembles the type species of the genera Coriolus Quel., Lenzites Fr. and Trametes

Fr. very closely in respect of cultural and micromorphological characters and

construction of their carpophores. Not only are the same types of hyphae present

in the carpophores of their respective type species [with the exception of the

absence of long fibre hyphae with one or two branches near the distal ends from

carpophores of Lenzites betulina (L. ex Fr.) Fr.| but the hyphae as well as the

basidia and basidiospores show strong similarities in morphology. It is thus evident

that Lenzites palisoti Fr. has close phylogenetic relationships with these three genera.

In order to determine the taxonomic position of Lenzites palisoti the con-

struction of its carpophores and their variability in texture should be considered

in relation to these characters of the carpophores of the type species of these

256

three genera. Different specimens of Lenzites palisoti may vary considerably in thick-

ness and texture. In thin, leathery carpophores of Lenzites palisoti, the construction

may resemble that of typical, thin, leathery specimens of Polyporus versicolor

with very numerous binding hyphae in the tissues. In the thicker specimens the

construction approaches that of carpophores of Lenzites betulina where binding

hyphae are relatively less numerous in the middle context, or even Trametes

suaveolens where binding hyphae are absent from the upper context. In view of

this variability in construction together with the well known variability in the

configuration of its hymenial surface, it appears that Lenzites palisoti is an

intermediate form between Lenzites betulina, Polyporus versicolor and Trametes

suaveolens. This is further evidence in support of the view that these latter three

species are congeneric and that the genera of which they are the type species,

are synonymous with Trametes Fr. If this view is accepted, then Lenzites palisoti

should be included in the genus Trametes Fr. as Trametes elegans (Spreng. ex Fr.)

Fr., (cf. Fidalgo & Fidalgo, 1966).

Polyporus occidentalis Klotzsch, Linnaea 8, 486, 1833:

Trametes occidentalis (Klotzsch) Fr., Epicrisis, p. 491, 1838:

Coriolopsis occidentalis (K1.) Murr., Bull. Torrey Bot. Club, 32, 358, 1905;

Coriolus occidentalis (K1.) G. H. Cunn., Proc. Linn. Soc. N.S. Wales, 75,

233, 1950.

Cultural characters

Growth is rapid, the mat reaching a radius of about 40 mm in one week

and covering the plates in two weeks. The advancing zone is even or slightly

bayed with the mycelium appressed for about 1 mm bebind the extreme margin,

then raised to form a woolly ridge about 1 cm wide across the plate, but becoming

somewhat more felty and less raised behind this ridge towards the inoculum.

A second and third woolly ridge may be formed across the mat between the first

ridge and the edge of the plate. The mat is white at first but the ridges turn

“light buff’ to “‘cream color” within a week. As the mat ages, the raised ridges

become more compact and tough, felty and darken in colour to “‘naples yellow” or

almost “mustard yellow” after four weeks, becoming lacunose and uneven or,

may develop irregular, raised lumps of tough, compact mycelium. Thinner areas

of the mat become more farinaceous woolly to somewhat felty and remain white

but eventually develop the “cream colour” to ‘light buff’ colour of the raised

ridges. At six weeks the mat around the inoculum is mostly farinaceous felty,

white or with patches of “maize yellcw” and traversed by thick, raised, felty

ridges, lacunose or roughened, mostly ‘“‘cream buff’ or “‘ochraceous buff” in

colour and with irregular, smooth, velvety lumps of the same colour on them or at

the sides of the dish. Occasionally depressions develop on these lumps from

which acicular spines, up to 2 mm high, arise, bearing basidia and spores. The

reverse of the colony bleaches quickly and remains so; a faintly fragrant odour

is given off. A strong blue colour is quickly produced when the culture is tested

for the presence of extra-cellular oxidase enzymes by means of alcoholic gum

guaiac solution.

Advancing mycelium: hyphae hyaline, simple or branching, nodose-septate with

simple clamp connecticns, thin-walled, with deeply staining contends, 2.5 — 5.0u

in diameter (Fig. 34 a).

Aerial mycelium: (a) hyphae as in the advancing zone; (b) fibre hyphae long,

unbranched, thin-walled and narrow at the origins, widening towards the middle

Jessi

Fic. 33.— Polyporus occidentalis. (a) Upper surface and (b) hymenial surface of

PRE 42450: (c) culture of PRE 42450 at six weeks; (d) fructification in culture;

(e) “Buller phenomenon” plates of PRE 42863 x PRE 42445; (f) branched and

unbranched fibre hyphae from culture of PRE 42863.

258

part with simultaneous increase in thickness of the walls which become faintly

yellowish brown and narrowing of the aseptate lumina, often with contents

staining in phloxine, but widening toward the thinner walled tips, 4.5 — 6.5u in

diameter at widest part (Fig. 34 b); (c) fibre hyphae hyaline, narrow, branching,

the branches long, flexuous, of even diameter, walls thick, lumina prominent or

reduced to interrupted lines, 0.7 — 1.5u in diameter (Fig. 34 c); (d) oidia hyaline,

cylindrical, ends rounded, 2.5 — 7.5 x 1.5 — 4.0u (Fig. 34 d).

Fructification: basidia hyaline, narrowly clavate, 14.5 — 20.0 x 3.6 — 5.5u and

bearing four, straight, sterigmata, 3.0 — 3.6u long (Fig. 34 e); basidiospores hyaline,

cylindrical, smooth, thin-walled, 4.5 — 7.0 x 2.0 — 2.5u (Fig. 34 f).

Submerged mycelium: (a) nodose-septate hyphae as in the advancing zone; (b)

oidia as in the aerial mycelium.

Carpophore characters

Carpophore annual, lignicolous, solitary or grouped; pileus sessile to effused-

reflexed, applanate or dimidiate frequently concave above and imbricate, tough

coriaceous to firm and rigid, 6 — 13 x 1.5 — 8 x 0.1 — 1.0 cm; surface tomentose

or hirsute, strongly concentrically sulcate, occasionally tuberculate, “‘olive brown’,

“buffy brown’, “‘tawny olive” or ‘‘chamois’” with zones of “‘ochraceous buff”,

“cinnamon buff’ and “‘verona brown’; margin acute, undulate concolourous

with upper surface or darker; pore surface “‘light ochraceous buff” drying to

“cinnamon buff’’, or “ochraceous buff’, pores rounded at first but later angular to

elongate in older parts, | — 3 per mm; dissepiments mostly even, thick or thin;

tubes “ochraceous buff’, 1 — 2 mm deep, decurrent on the bases of the pileus,

sometimes stuffed; context “pale ochraceous buff” to “ochraceous buff’, zonate,

fibrous corky to sub-woody, 0.5 — 8 mm thick.

Hyphal characters: carpophores consist of: (i) nodose-septate hyphae hyaline,

branching, thin-walled, 2.2 — 3.0u in diameter (Fig. 34 g); (ii) fibre hyphae long,

unbranched straight or somewhat flexuous, sub-hyaline to pale yellowish-brown,

thick-walled, the lumina aseptate, wide at the extremities but narrow or partly

occluded near the middle portion, with or without staining contents, 3.0 — 7.5u

in diameter and arising from thin-walled, nodose-septate hyphae (Fig. 34 h);

(iii) fibre hyphae long, straight or flexuous with one to three branches towards

the distal end, the branches flexuous, thick-walled, suh-hyaline to pale yellowish

brown, lumina prominent, aseptate, 2.0 — 5.2u in diameter (Fig. 34 k); (iv) fibre

hyphae hyaline with numerous long, tortuous, tapering branches mostly solid or

lumina much reduced, arising from a short length of main stem, the branches

interwoven with other hyphae, 1.0 — 4.0u in diameter (Fig. 34 m).

Hymenium. basidia hyaline, long clavate almost cylindrical with four sterigmata,

14.5 — 18.0 x 3.6 — 4.5; sterigmata 3.0 — 3.6u (Fig. 34 n); basidiospores

cylindrical, hyaline, smooth, thin-walled, 4.5 — 7.0 x 2.0 — 2.5u (Fig. 34 p):;

hyphal pegs conical, projecting 40 — 50u beyond the level of the basidia.

Construction. The margin consists of long, unbranched fibre hyphae with thickened

walls and prominent lumina, and arranged more or less parallel or slightly inter-

twined. Also intertwined with the fibre hyphae are numerous deeply staining,

branching, nodose-septate, thin-walled hyphae from whick they arise. Behind the

margin the context consists mainly of fibre hyphae. In the upper context the

fibre hyphae are mostly straight with partly thickened, pale yellowish-brown walls

and prominent lumina. They are parallel in arrangement and loosely intertwined.

Occasional thin-walled, nodose-septate hyphae are present among them. At the

upper surface a somewhat denser layer of tissue is present which consists of

branching, thin-walled, nodose-septate hyphae and hyaline, fibre hyphae with

259

many thick-walled or solid branches, intertwined with the unbranched fibre hyphae.

The unbranched fibre hyphae project beyond this layer to form the dense, tomentose

upper surface of the carpophore. In the lower context the construction of the

tissues is similar but the tissues become more dense towards the pores. Long

unbranched fibre hyphae, 3.5 — 4.5u im diameter, with the lumina narrow or partly

occluded, turn downwards into the trama of the dissepiments where they become

interwoven with hyaline much-branched fibre hyphae with solid or sub-solid

branches, and branched, thin-walled, nodose-septate hyphae to form a dense, tough,

homogeneous tissue. In the dissepinients, thin-walled, nodose-septate hyphae

ramify throughout the tissues, branching repeatedly and forming numerous

short branches bearing clusters of basidia at the hymenial surfaces of the tubes.

Decay and hosts

Polyporus occidentalis causes a white rot of hard-wood logs in dry, exposed

positions in sub-tropical areas.

Specimens examined

Herb. DAOM: 31731, on Cola cordifolia, Jasikan, Tongoland, May 1949: 38997, on wood,

Rest Pew, Manchester, Jamaica, Feb. 1945; 52393, Municipio Benjamin Constant, Brazil.

Aug. 1955.

Herb. PRE: 1372, on wattle stump, Cramond, Natal, Apr. 1911; 1697, on fence post, Letaba

Drift, Zoutpansberg, Tvl., Aug. 1911; 5645, on fence post. Winkelspruit, Natal, Feb. 1912:

6685, on Citrus simensis stump, Nelspruit, Tvl., May 1913; 6926, on Celtis rhamnifolia,

Lusikisiki, Transkei, Sept. 1913; 8818, on Celtis rhamnifolia, Pietermaritzburg, Natal, Feb.

1915; 9204, on Celtis rhammifolia, Pietermaritzburg, Natal, Dec. 1915; 9482, on Celtis

rhamnifolia, Inanda, Natal, Dec. 1915; 11249, on Celtis rhamnifolia, Durban, Natal, Nov.

1916; 11254, on Pyrus malus, Wolhuters Kop, Tvl., Feb. 1919; 13169, Flora of the Philippines

No. 491, Nov. 1916, 13367, Flora of the Philippines No. 19101, Nov. 1916; 14072, Flora of

Victoria Nyanza, Uganda, Nov. 1916; 14892, Flora of Kenya, Nov. 1916; 15016, on stump,

Pretoria, Tvl., April 1921; 15051, on wood, Selati River, E. Tvl., Nov. 1921; 15573, on wood,

Durban, Natal, Nov. 1916; 15582, on Rhus viminalis, Branders High Forest, Aug. 1915; 23349,

on wood, Alexandria, July 1927; 25491, on dead trunk, Kasane, Bechuanaland, July 1930;

25917, on tree stump, Zoutpansberg, Tvl., Dec. 1929; 26324, on tree stump, Elim Mills.

Zoutpansberg, Tvl., March 1932; 26380, on tree stump, Eshowe, Natal, Jan. 1916; 26385,

Cn tree stump, Margate, Natal, Feb. 1911, 27606, on tree stump, Pietermartizburg, Natal, Feb.

1934; 28250, Fungi Venezuelani No. 418, H. Sydow; 28707, on Prunus domestica, Pieter-

maritzburg, Natal, Nov. 1934; 30102, on Acacia mollissima, Willowvale Plant., C.P., July

1937; 30283, East Afr. Agric. Res. Stat.. Amani, No. 1084; 30820, on dead wood, Port

St. Johns, Aug. 1937; 31631, on dead wood, Eshowe, Natal, Jan. 1916; 31649, on dead

wood, Durban, Natal, May 1916; 31675, on dead wood, Gingindhlovu, Natal, June 1915;

31686, on Albizzia fastigiata, Durban, Natal, Sept. 1916; 31693, on dead log, Durban Natal,

Sept. 1916; 31701, on dead log, Botanical Garden, Durban, Natal, Oct. 1916; 31856. on dead

log, New Germany, Natal, April 1917; 31857, on dead log, New Germany. Natal, Apr. TONG

31945, on dead log, Stella Bush, Durban, Natal, Apr. 1917; 31980, on dead log, Pietermaritz-

burg, Natal, Nov. 1917; 35561. on Neiium oleander, dead branch, Pietermaritzburg, Sept.

1946; 33563, on Brachystegia log, Mufilira, Rhodesia, March 1942; 33754, on Afzelia

quanzensis, Mocambique, Nov. 1942; 33991, on Salix sp. wood, Pietermaritzburg, Natal,

1943; 34373, on dead wood, Lake St. Lucia, Natal, 1935; 34376, on dead wood, Lake St.

Lucia, Natal, 1935; 34461. on dead wood, Kazungulu, July 1930; 34558, on dead log, Kirsten-

bosch, Cape Prov., June 1934; 41357, on Citrus aurantifolia, Herb. C.M.I. no. 49391;

*42144, on hardwood log, Honeydew, Johannesburg, Jan. 1961; *42445, decaying hardwood,

F. C. Erasmus Nat. Reserve, E. Tvl., Feb. 1961; *42450, decaying hardwood, White River,

Tvl., Feb. 1961; *42863, on Pinus sp. log, Bushbuckridge, E. Tvl., Feb. 1961.

Herb. STE: 104, on Pinus, near Durban, Natal; 151, on old stump, Durban, Natal; 275,

on dead logs, Durban, Natal; 284, on dead logs, Zululand, Natal; 289, on dead logs, Zululand,

Natal; 558, on dead Persea gratissima; 73), on dead Persea gratissima, Natal; 751, on dead

Persea gratissima, Durban, Natal; 764, on dead Persea gratissima, Durban, Natal; 797, on

old stump, Durban, Natal; 1043, old stumps, Ngare Mutoni, E. Africa, July 1922; 2415, on

dead stump in bush, Umtali, Rhodesia, 2446, old stumps, Rhodesia, July 1927; 2553, old

stumps, F. Eyles No. 4226, Feb. 1926; 2617, old stumps, Pietermaritzburg, Natal, May 1931.

Interfertility studies

GOUD

A CNY OG ES

3 ee ae

CC, See!

261

Single basidiospores were collected from a carpophore of collection PRE

42863 kept in a moist chamber. In order to determine the type of interfertility

of this species, 16 cultures, each grown from a single basidiospore, were paired

in all possible combinations. It was found that Polyporus occidentalis has the

tetrapolar type of interfertility with allelomorphs for heterothallism at two loci.

The results, showing the distribution of mating types among the single spore

cultures are presented in TABLE 8.

To test the conspecificity of collections of which cultures were available by

means of the “Buller phenomenon’, two mycelia from single spores of different

mating types, PRE 42363 — 1 and PRE 42863 — 5 were used as haploid mycelia.

Dikaryotic mycelia of collections FRE 42445 and PRE 42450 were tested by means

of this technique. A set of six plates were prepared for each dikaryotic culture

to be tested. Three plates in each set were then inoculated with each haploid

monospore culture and incubated for five days. Each set of plates were then

inoculated with a small piece of dikaryotic mycelium which was placed near the

periphery of the growing haploid colony. After further incubation for four days,

the haploid mycelium on each plate was examined for the presence of clamp

connections at three positions along its periphery (Fig. 3? e).

Clamp connections were found in every case thus showing that dikaryotization

of the single spore haploid mycelia by the added dikaryotic mycelia had taken

place and confirming that the collections of Polyporus occidentalis numbered, PRE

42144, PRE 42445, PRE 42450 and PRE 42863 are interfertile and therefore

conspecific.

Discussion

With the formation of thin-walled, nodose-septate hyphae, thick-walled fibre

hyphae and extra-cellular oxidase enzymes in its cultures and its cylindrical

basidiospores, Polyporus occidentalis fits well into Group 45 (Nobles, 1958 b).

The cultures have the general and micromorphological characters of other species

described above in this group. Cultures resemble those of ‘rametes meyenii most

nearly but the buff coloured areas which develop on the mat and fruiting areas,

together with the sub-hyaline to faintly yellowish walls of the fibre hyphae, serve

to distinguish cultures of this species from others in the group. This fungus has

not been described in culture before but it was included by Nobles (1958 b) in

Group 45.

The tetrapolar type of interfertility present in Polyporus occidentalis also

agrees with that of other species of Group 45 as well as Nobles’ (1958 b) thesis

that this type of interfertility is present in polypore species of which the thin-walled

hyphae are nodose-septate and whose cultures produce extra-cellular oxidase.

The carpophores of this species consist of four types of hyphae, viz. thin-

walled, nodose-septate hyphae, unbranched fibre hyphae, fibre hyphae with branches

towards the distal ends and fibre hyphae with a number of long tapering branches

often arising from a short length of main stem. The latter hyphae appear to be

binding hyphae of the bovista type as described by Cunningham (1946) and the

Fic. 34— Polyporus occidentalis. a - f. Structures from cultures: (a) thin-walled

nodose-septate hypha from advancing zone; (b) unbranched fibre hypha; (c)

narrow fibre hyphae with numerous long branches; (d) oidia; (e) basidia;

(f) basidiospores.

g-p. Structures from carpophores; (g) thin-walled, nodose-septate hypha: (h)

unbranched fibre hypha; (k) fibre hypha with one to three branches towards

the distal end; (m) sub-solid or solid fibre hypha with numerous long, flexuous

tapering branches; (n) basidia; (p) basidiospores.

262

branches of the fibre hyphae with bianches towards the distal ends appear to

assist in the binding function. The other hyphae correspond to Corner’s (1932 a)

definition of generative and sieletal hyphae so that carpophores of Polyporus

occidentalis have a trimitic hyphal system in the terminology of Corner (1932 a)

and Cunningham (1946).

The hyphal characters of Polyporus occidentalis were described recently by

Fidalgo & ridalgo (1966) who reported a trimitic hyphal system with thin-walled,

nodose-septate generative hyphae, “‘skeletal hyphae thin- to thick-walled usually

with a distinct lumen, walls hyaline to yellowish, unbranched, not septate . . .;

binding hyphae thick-walled to solid, hyaline, much branched, non-septate”. This

description agrees well with that given above but these authors did not mention

the presence of thick-walled fibre, or skeletal hyphae, with branches toward the

distal end which were fairly abundant in the carpophores examined by me. No

other description of the hyphal characters of Polyporus occidentalis had been

published but both Imazeki (1943) and Cunningham (1950 b) included this species

in the genus Coriolus Quél. thereby implying similarities in hyphal and anatomical

characters between this species and Polyporus versicolor. From the descriptions

it is evident that many similarities in cultural and carpophore characters exist

between Polyporus occidentalis, Polyporus versicolor and the other species of

Group 45 described above. Certain differences, however, exist. Polyporus

occidentalis is the only specics studied in this group in which the fibre hyphae

have faintly yellowish brown walls. All the other species have hyphae with hyaline

walls. Carpophores of Polyporus occidentalis lack the solid, branched, nodose-

septate hyphae which are present in the binding hyphal system of carpophores of

Polyporus versicolor, Lenzites betulina, Trametes suaveolens and other species

in this complex. The solid, branched processes formed on nodose-septate hyphae

in cultures of Polyporus versicolor were not found in cultures of Polyporus

occidentalis. The binding hyphae of Polyporus occidentalis do not have numerous

short, curled branches like those of Polyporus versicolor, Lenzites betulina and

Trametes suaveolens but instead have long, tapering branches resembling those of

Trametes cingulata and the Pycnoporus spp. described by Nobles & Frew (1962).

Indeed, in respect of hyphal characters, Polyporus occidentalis resembles Trametes

cingulata and Pycnoporus spp. more closely than Polyporus versicolor. Some

of the kinds of hyphae present in carpophores of the type species of Trametes Fr.,

Lenzites and Coriclus Quél. are thus lacking from carpophores of Polyporus

occidentalis. Because Bondartseva (1961) and Teixeira (1962 b) regard the

absence or presence of different kinds of hyphae as taxonomically important at the

generic level, it appears that Polyporus occidentalis should not be regarded as

congeneric with these three genera.

Murrill (1905) segregated the genus Coriolopsis with Polyporus occidentalis

Klotzsch as type species, from the trametoid group of species on the basis of

its dark-coloured context. For reasons advanced above, Polyporus occidentalis

appears best placed in this genus which however is closely related to the trametoid-

corioloid complex of species. It is however not impossible that future studies

may show that other hitherto unknown species may reveal a combination of

characters common to Polyporus cecidentalis and other species of the trametoid-

corioloid complex, thus offering evidence of congeneric relationship in a series

of species.

From the descriptions it is evident that the structures formed in culture are

also present in the carpophores from which they were made. The nodose-septate

hyphae, fibre hyphae and hymenial structures formed in culture were identical

to those of the carpophores but the oidia, which were abundant in cultures, were

not found in the carpophores.

263

Fic. 35.—Trametes cingulata. (a) Carpophores of PRE 27506; (b) culture of

PRE 42455 at six weeks; (c) thin-walled hypha with dark-brown resin-like

contents from upper surface of carpophore, x 1000.

Trametes cingulata Berkeley, in Hook, Journ. Bot. 6, 164, 1854;

Coriolus cingulatus (Fr.) G. H. Cunningham, Proc. Linn. Soc. N. S. Wales

Tl, Hall, WSO,

Cultural characters

Growth is moderately fast the mat reaching a radius of 40 mm in one week

and covering the plate after two to three weeks. The advancing zone is even with

the hyphae appressed, for one or two millimetres, then raised slightly at the edge

of the mat. Mat behind margin at first thin, downy but becoming gradually more

dense, somewhat raised, then collapsing somewhat toward the inoculum. Mat

smooth at first but developing faint radiating grooves after two weeks with

transverse ridges of dense, raised, thin, woolly to felty mycelium which abutt

sharply on the thin, downy areas of younger mycelium. The mat is hyaline or

white at first and remains so while thickening until, at 6 weeks, it is mostly

characterized by areas of tough, dense, somewhat pellicular mycelium around

the inoculum, radially sulcate and bordering sharply on thin, downy or sodden

mycelium which gradually increases in density to form a transverse zone of dense,

felty mycelium over the thin, subfelty mycelium of the newest growth. These

zones of dense mycelium may develop irregular, granular-woolly patches which

may form fruiting areas bearing minute, waxy, acicular spines, or, irregular lumps

of dense, smooth, chamois-like mycelium may form on the sides of the dish or on the

areas of thin mycelium and eventually form fruiting areas of minute, erect spines.

The reverse of the culture bleaches gradually and a faint, sweetish, fragrant odour

is given off. On gallic acid and tannic acid media no growth takes place but a

strong diffusion zone is formed on gallic acid medium and a weaker one on

264

tannic acid medium. A strong blue colour is produced when an alcoholic gum

guaiac solution is applied to the mat.

Advancing mycelium: hyphae narrow, hyaline, branching, thin-walled, nodose-

septate, 3.0 — 4.0u in diameter (Fig. 36 a).

Aerial mycelium: (a) hyphae as in the advancing zone; (b) fibre hyphae hyaline

more or less straight, unbranched, the walls thick, refractive and lumina prominent,

widening at their tips, aseptate, or, occluded and reduced to a thin, interrupted

line, 2.5 — 4.5 in diameter (Fig. 36 b); (c) fibre hyphae as above but with a

number of branches over a short length, the branches long, flexuous and tapering,

1.5 — 3.0u in diameter (Fig. 36 c); (d) oidia hyaline, smooth, cylindrical with

rounded ends, 5.0 — 7.0 x 3.0 — 4.0u (Fig. 36 d).

Fructifications: basidia broadly clavate 12.0 — 25.5 x 4.5 — 6.0u with four short,

straight sterigmata, 2.2 — 3.0u (Fig. 36); basidiospores ovoid to short cylindrical,

hyaline, smooth, thin-walled, obliquely apiculate 4.5—6.0 x 3.0—3.5u (Fig.

36 f).

Submerged mycelium: (a) hyphae as in the advancing zone but more frequently

nodose-septate; (b) chlamydospores intercalary or terminal. sub-globose to ellipsoid,

hyaline, thick-walled, 6.0 — 10.0 x 8.0 — 12.0u (Fig. 3€ g).

Carpophore characters

Carpophore annual, lignicolous, mostly solitary, sessile to dimidiate, rarely

imbricate, occasionally laterally connate; pileus coriaceous to woody up to 9.0 x

5.0 x 0.3 — 0.7 cm; upper surface glabrescent, concentrically sulcate to tubercular

and rough, with fine, irregular cracks or smooth, matt, dark grey to black, azonate

or alternating dark and lighter-coloured zones; margin soft velutinate, entire, thick

and rounded or thin, acute, pale cream to dark ‘“‘cream color’, sterile below: pore

surface pale ‘“‘cream color” drying darker, glistening, poroid; pores, rounded,

3 — 6 per mm; dissepiments thin, edges entire; tubes concolorous, 0.5 — 2.5 mm

deep, not stratified; context white to pale ‘‘cream color’, even textured, floccose

punky to corky, 1.0 — 5.0 mm thick.

Hyphal characters: (i) nodose-septate hyphae hyaline, branching, thin-walled and

with deeply staining contents, 2.5 — 3.5u in diameter, some inflated terminally with

contents hard, resin-like, dark-brown (Fig. 36 m); (ii) fibre hyphae hyaline,

long, unbranched, walls thick and refractive, widest over middle portion, aseptate,

often with dark brown contents at the distal end, or, occluded, and reduced to

an interrupted line, 3.5 — 6.0u in diameter (Fig. 36 k, n); (iii) fibre hyphae

thick-walled, hyaline, with one to three branches towards the distal end, lumina

prominent, aseptate, 2.5 — 4.5u (Fig. 36 p); (iv) branched fibre hyphae hyaline,

thick-walled, with lumina aseptate, prominent or occluded, branches few or many,

long, flexuous, tapering or short, flexuous and arising from a short distance of

main stem, 1.5 — 3.5u in diameter (Fig. 36 q).

Fic. 36.— Trametes cingulata. a - g. Structures from culture: (a) thin-walled,

nodose-septate hyphae from advancing zone; (b) fibre hypha, unbranched; (c)

fibre hyphae with numerous long, flexuous branches; (d) oidia; (e) basidia;

(f) basidiospores; (g) chlamydospores.

h-t. Structures from carpophores: (h) thin-walled, nodose-septate hyphae; (k)

unbranched fibre hyphae; (m) inflated thin-walled hypha with dark-coloured

contents from upper surface; (n) inflated terminal portion of fibre hypha with

dark-coloured contents from upper surface; (p) fibre hypha with one to three

branches toward the distal end; (q) solid fibre hypha with numerous long,

flexuous, tapering branches; (s) basidia; (t) basidiospores.

265

e>5 COC

266

Hymenium. basidia hyaline, long clavate to cylindrical, 12.0 — 22.0 x 4.5 — 6.0u

with four sterigmata 2.5 — 3.0u long (Fig. 36 s); basidiospores ovoid to short

cylindrical, hyaline, smooth, thin-walled, 4.0 — 6.0 x 2.5 — 3.5u (Fig. 36 t).

Construction. The margin consists mainly of long unbranched fibre hyphae, mostly

thick-walled and with aseptate lumina, orientated parallel to the direction of

growth of the pileus. Loosely intertwined with them are branching, thin-walled,

nodose-septate hyphae from which the fibre hyphae arise Immediately behind the

margin in the context, fibre hyphae with one or two long branches or with a larger

number of shorter branches, the branches tapering, flexuous and interwoven with

the other hyphae, become abundant.

The older part of the context consists mainly of unbranched, sub-solid or

solid fibre hyphae in more or less parallel arrangement and slightly intertwined,

turning gradually upward towards the upper surface. Branched fibre hyphae,

mainly solid, the branches long or short, tortuous, and tapering towards the ends

are interwoven with the unbranched fibre hyphae across their direction of growth

and bind them into a firm, homogeneous tissue. Occasional lengths of thin-walled,

nodose-septate hyphae, mostly empty and collapsed, are present among the others.

At the upper surface the ends of the unbranched fibre hyphae are closely packed

at a common level. Their lumina are wide and mostly filled with dark-brown,

resin-like contents. At the upper surface thin-walled, nodose-septate hyphae are

very numerous, intertwined with the fibre hyphae and with ends projecting, often

distended and at the same level as the fibre hyphae and filled with dark-brown,

resin-like, hard contents (Fig. 36 m). The dark contents of the nodose-septate

and fibre hyphae form a zone of about 90u thick at the upper surface of the pileus.

Below this zone the nodose-septate hyphae and many fibre hyphae have deeply

staining luminal contents. In the lower context, the long, unbranched fibre hyphae

turn downwards into the trama. Fibre hyphae with one or two branches towards

the distal part become more numerous, the branches becoming tortuous and fibre

hyphae with many tapering branches over a short length of main stem increase

in numbers towards the dissepiments where their branches are interwoven with the

other hyphae, binding them into a tough tissue. The fibre hyphae are mostly

tortuous and tightly intertwined and interwoven into the dense, homogeneous

tissue of the lower trama and dissepiments. Most of these fibre hyphae have

prominent lumina. Intertwined with the fibre hyphae in the lower trama and

dissepiments are the narrow, thin-walled, nodose-septate hyphae, branching freely,

the branches short, with numerous clamp connections and becoming very numerous

at the hymenial surfaces where they bear basidia in small clusters. The edges

of the dissepiments are sterile and consist of fibre hyphae with prominent lumina.

Decay and hosts

Trametes cingulata causes a white sap rot of various species of hardwoods

(Banerjee & Naha, 1960 b).

Specimens examined

Herb. PRE: 2127, Pretoria, Jan. 1917; 8799, on Eucalyptus globulis, Pietermartizburg, Jan.

1915; 9145, Pietermaritzburg, Natal, Oct. 1915; 11246, on Acacia mollissima, Cramond, Natal,

Jan. 1916; 12004, on dead log, Kyagwe, Uganda, Jan. 1916; 12465, Limpopo Valley,

Transvaal, June, 1919; 14489, Pretoria, Transvaal, Apr. 1921; 14691, Glen, O.F.S., Apr. 1921;

14903, Samu, Kenya; 20289, Malay Peninsula, No. 10858; 20467, Knysna, C.P., Jan. 1925;

20610, dead wood, Pretoria, Transvaal, Aug. 1929; 25492, on dead tree, Kasane, Bechuanaland,

July 1930; 26407, on dead wood, Pretoria, Transvaal, Aug. 1929; 26614, on dead tree,

Mariental, S.W.A.; 27506, on dead log, Nelspruit, Transvaal, July 1934; 28970, on dead wood,

Duiwelskloof, Transvaal, May 1937; 36521, on dead Eucalyptus sp., Swartruggens, Tvl.,

Feb. 1939; 30751, Rustenburg, Transvaal, Jan. 1939; 30876, on dead wood, Xumeni Forest,

267

Natal, Jan. 1938; 31458, Lobatsi, Bechuanaland, Apr. 1929; 31564, on Acacia sp.. Balfour.

Transvaal; 31628, Krantzkloof, Natal, Jan. 1916; 31629, Cramond, Natal, Jan. 1916; 31655,

Ngoye, Natal, May 1916; 34984, on dead stump, Qudeni Forest, Natal, Feb. 1965: 36875,

on fallen tree trunks, Amatongas Forest, Mocambique, June 1948; 41356, on Albizzia zygia,

ex Herb. C.M.I. No. 37382; 41522. on Olea laurifolia, Knysna, C.P., Apr. 1965; 41737, on

Acacia mollissima stump, Richmond, Natal, June 1951; 40297, on dead bark, Potgietersrust.

Transvaal, March 1960; 42254. on fallen log, Sabie, Transvaal, Apr. 1962; *42433. on dead

hardwood, Blouberg, Transvaal, Jan. 1959; *42448, on dead Eucalyptus sp., Bosbokrand,

Transvaal, Feb. 1961; *42455, on Eucalyptus sp. log, Johannesburg, Tvl. Jan. 1961; *42456,

on dead wood, Magaliesberg, Transvaal, Jan. 1961.

Herb. STE: 108, Acacia mollissima stwnp, Krantzkloof, Natal; 750, Acacia mollissima stump.

Pinetown, Natal; 794, Pretoria; 805, on branch of apricot tree, Pretoria; 817, Lobatsi;

1074, Waterberg, Transvaal, Feb. 1923; 1464, on dry Eucalyptus pole, Tzanezn plantation,

Transvaal, July 1924; 1671, Potgietersrust, Transvaal. July 1924: 1711, on old wood,

Pietersburg, Transvaal, July 1924.

Interfertility studies

Single spores were collected from a fructification formed in a culture of

PRE 42448. In order to determine the type of interfertility of this species, 16

cultures, each grown from a single basidiospore, were paired in all possible

combinations. Clamp connections formed in the paired mycelia in a manner

indicating the tetrapolar type of interfertility with allelomorphs for heterothallism

at two loci, in this species. The results showing the distribution of mating types

among the single spore cultures are presented below. These results confirm Naha’s

(1957) report that Trametes cingulata has the tetrapolar type of interfertility. This

distribution of mating types among the basidiospores is set out in TABLE 9.

Single basidiospore cultures of other collections of Trametes cingulata from

South Africa were later obtained from the respective dikaryotic cultures, viz:

PRE 42433, PRE 42455 and PRE 42456. In order to determine the conspecificity

of these collections with PRE 42448, four cultures, each obtained from a single

basidiospore, from each collection, were paired in all possible combinations with

four single spore cultures from PRE 42448. In all the paired mycelia, clamp

connections developed thus proving that collections PRE 42433, PRE 42448, PRE

42455 and PRE 42456, are interfertile and therefore conspecific.

Discussion

The presence of nodose-septate, thin-walled hyphae and fibre hyphae in

cultures which produce extra-cellular oxidase and the possession of cylindrical

basidiospores, place Trametes cingulatc in Group 45 (Nobles, 1958 b). In appear-

ance and texture the mat resembles cultures of Polyporus versicolor and Pycnoporus

Sanguineus. From the former species it differs in the presence of the thinner,

more fragile, radially striate mat in which nodose-septate hyphae with thickened

walls and solid branched processes are lacking. These characters also distinguish

Trametes cingulata from cultures of other species in this group. Its cultures differ

from those of Pycnoporus sanguineus as described by Nobles & Frew (1962) by

the absence of orange yellow colours. In respect of general appearance of the

mat and the structures formed in culture, however, there are many similarities.

This description agrees with previous descriptions by Naha (1957), Van der

Westhuizen (1958) and Banerjee & Naha (1960 a).

In the carpophores four kinds of hyphae were found, viz: nodose-septate,

thin-walled hyphae and aseptate fibre hyphae without branches, or with one to

three branches towards the tip or with numerous long tapering branches. These

latter appear to be “binding hyphae of the bovista type” as described by Cunning-

ham (1946); but the fibre hyphae with one to three branches also contribute to

the binding hyphal system. The carpophores of Trametes cingulata thus have a

trimitic hyphal system as reported by Cunningham (1950 b) and Farinha (1964).

268

From the descriptions it is evident that the structures formed in the cultures

are also present in the carpophores from which they were made. The nodose-septate

hyphae, fibre hyphae and hymenial structures from the cultures are identical to

those from the carpophores. Chlamydospores, which were fairly numerous in the

cultures, were not present in the carpophores. This discrepancy had also been

recorded for other species.

In descriptions of the hyphal characters of Trametes cingulata, Banerjee &

Naha (1960 b) reported the presence of clamp connections on the thin-walled

hyphae of its carpophores. Farinha (1964) reported that secondary hyphae of the

carpophore were thin-walled, nodose-septate and, the tertiary hyphae, aseptate,

thick-walled, branched and up to 7 in diameter, while others were much-branched

and narrow. Cunningham (1950 b) placed this species in the genus Coriolus Quél.,

which he had characterized as having a trimitic hyphal system with thin-walled,

nodose-septate, generative hyphae, thick-walled aseptate, skeletal hyphae and

thick-walled, aseptate, much-branched binding hyphae. These reports thus partially

confirm the above observations but certain differences are apparent between

carpophores of Trametes cingulata and those of Polyporus versicolor L. ex Fr.

and Trametes suaveolens (L. ex Fr.) Fr. the accepted type species of the genera

Coriolus Quél. and Trametes Fr. respectively. In carpophores of Trametes

cingulata, solid or sub-solid, nodose-septate binding hyphae are not present as in

the carpophores of these two species. Also, the binding hyphae of Trametes

cingulata have long, tapering branches which arise over a short length of main

stem. Binding hyphae of the other twe species are short with fairly short, thick,

branches. Hyphae with tapering branches are present in the carpophores of

Lenzites betulina (L. ex Fr.) Fr., the type species of the genus Lenzites Fr. (Cooke,

1959) but they are lateral binding processes or branches of solid, nodose-septate

hyphae. Carpophores of Lenzites betulina do not possess long fibre hyphae with

branches near the end which contribute to the binding system. Carpophores of

Trametes cingulata thus differ in respect of the types of hyphae present in them

from the carpophores of the type species of the genera Coriolus Queél., Trametes

Fr. and Lenzites Fr. On the other hand, the binding hyphae in carpophores of

Trametes cingulata resemble those in carpophores of Pycnoporus cinnabarinus

(Jacq. ex Fr.) Karst., the type species of the genus Pycnoporus Karst. as described

by Nobles & Frew (1962) much more closely. Indeed, in cultural characters and

hyphal characters Trametes cingulata appears to resemble species of the genus

Pycnoporus Karst. more than they do those of the genera Trametes, Coriolus and

Lenzites but lack the characteristic orange-red colours which distinguish species of

the genus Pycnoporus. Species of the genus Pycnoporus, however, apart from

their orange-red colours, have cultural characters which agree with those of Group

45 (Nobles, 1958 b) while their carpophore characters agree in many respects with

those of the type species of the genera Coriolus, Trametes and Lenzites.

The cultural and carpophore characters of Trametes cingulata thus agree in

many respects with those of the type species of the genera Coriolus, Trametes

and Lenzites, but lack certain of the types of hyphae which are present in their

carpophores. The differences in morphology of the binding hyphae of these species

may be of specific significance only but the absence or presence of types of

hyphae in carpophores are regarded as of generic importance by Bondartseva

(1961), Teixeira (1962 b) and Donk (1964). It thus appears best to regard Trametes

cingulata as generically distinct from Trametes suaveolens until detailed studies

of the hyphal characters and hyphal morphology of more species in this group

can clarify the significance of such differences in hyphal morphology. The genus

in which Trametes cingulata Berk. will be more suitably placed, cannot be indicated

at present, however.

269

| Hovoryee = Here Kew

Poly pov. s YIngses Berk

. Ale “e 3

E St Domingo a

Fic. 37.—PolYporus vinosus. (a) Holotype: (b) carpophores from British Hon-

duras; (c) pore surface of holotype: (d) culture of PRE 42154: (e) fructifica-

tion in culture.

Polyporus vinosus Berk., Ann. Mag. Nat. Hist. 11, 9, 195, 1852:

Coriolus vinosus (Berk.) Pat., Ess. Taxon. 94, 1900:

Nigroporus vinosus (Berk.) Murr., Bull. Torrey Bot. Club 32, 361, 1905:

Fomitopsis vinosa (Berk.) Imazeki, Bull. Gov. For Expt. Stat., Tokyo, Japan.

ING; Sik Ibe eby

270

Cultural characters

Growth moderately slow to slow, the mycelium reaching a radius of about

10 mm in one week and covering the plate in four weeks. The advancing zone is

very uneven with the mycelium mostly submerged and forming hyaline or white

plumose outgrowths radiating out from prominent strands which originate from

the inoculum. After three weeks small farinaceous pustules appear, scattered

over the culture near the inoculum or along the main strands of submerged

mycelium, white at first but gradually becoming “‘light greyish vinaceous” and

slowly increasing in size. In an old culture a pad of pubescent “‘deep livid brown”

mycelium developed on the side of the dish, from which thin, lamellar structures

grew out laterally, uniting at various points to form daedaloid slits, which in turn

rounded off to form minute tubes. A white spore deposit appeared below these

tubes five months after inoculation of the plate. The reverse bleaches slowly and

a faint mushroomy odour is given off. A strong blue colour is formed when a

drop of alcoholic gum guaiac solution is applied to the culture. Strong diffusion

zones are formed on gallic acid and tannic acid media but no growth occurs in

seven days.

Advancing mycelium: hyphae hyaline, branching, nodose-septate, thin-walled,

2.2 — 3.5u in diameter (Fig. 38 a).

Aerial mycelium: (a) hyphae as in the advancing zone; (b) fibre hyphae short,

mostly unbranched, occasionally with one branch, thick-walled, the walls faintly

reddish brown and thickest along the middle part, the lumina prominent, widening

towards the extremities, aseptate, 2.2 — 5.0u in diameter (Fig. 38 b).

Fructification: basidia hyaline, short clavate, 6.6—9.0 x 3.6—4.2u with four

straight, slender sterigmata 1.8 — 2.4u long (Fig. 38 c); basidiospores hyaline,

allantoid, smooth, thin-walled, 3.0 — 3.6 x 1.2 — 1.6u (Fig. 38 d).

Submerged mycelium: hyphae hyaline, branching, nodose-septate, thin-walled,

1.5 — 5.5 in diameter (Fig. 38 e).

Carpophore characters

Carpophore annual, lignicolous, solitary, sessile; thin, dimidiate to reniform,

narrowly attached by a scutate disc or laterally connate and broadly decurrent,

woody and brittle when dry, 1 — 4 x 3 — 7 x 0.3 — 0.7 cm; surface finely

velutinate in young part, then glabrous, concentrically sulcate, mat, “dark vinaceous

brown” to “‘hays brown” becoming “‘brownish drab” in age; margin acute, rounded,

occasionally somewhat lobate, concolorous; pore surface “pale vinaceous drab”

to “dark vinaceous brown” or “‘sorghum brown” poroid; pores angular, 6 — 8/

mm; dissepiments thin, even; tubes up to 2 mm deep, occasionally stratified;

context up to 5 mm thick, “‘sorgium brown’, 2ven, homogeneous.

Hyphal characters. Carpophores consist of: (i) nodose-septate hyphae hyaline,

branching, thin-walled, with deeply staining contents, 2.2 3.54 in diameter

(Fig. 38 f); (ii) nodose-septate hyphae tortuous and branched with walls pale

brownish and slightly thickened, with staining contents, or, empty and often with

simple septa, 2.5 — 4.0u in diameter (Fig. 38 g); (iti) fibre hyphae arising from

nodose-septate hyphae, straight or flexuous, unbranched, pale smoky brown,

thick-walled, lumina aseptate, wide at the extremities narrow or occluded in the

middle parts, 2.5 —6.0u in diameter (Fig. 38 h).

Hymenium; basidia hyaline, clavate. 6.0 — 9.0 x 3.6 — 4.2u, with four, straight

sterigmata, 1.8 — 2.4u (Fig. 38 k); basidiospores hyaline, allantoid, smooth, thin-

walled, 3.0 — 4.0 x 1.2 — 1.6u (Fig. 38 m).

‘FIGURE 38.

Fic. 38.— Polyporus vinosus. a - e. Structures from cultures: (a) thin-walled,

nodose-septate hypha from advancing zone; (b) unbranched fibre hyphae; (c)

basidia; (d) basidiospores; (e) hypha from submerged mycelium.

f-m. Structures from carpophores: (f) thin-walled, nodose-septate hypha; (g)

thick-walled, tortuous, branching nodose-septate hyphae with simple septa;

(h) unbranched fibre hyphae; (k) basidia; (m) basidiospores.

272

Construction. At the margin the carpophores consist mainly of fibre hyphae with

prominent lumina and staining contents, arranged parallel to and somewhat inter-

twined with one another and with branching, hyaline, thin-walled, nodose-septate

hyphae. The context consists mainly of unbranched fibre hyphae with pale,

smoky-brown, thickened walls, intertwined with one another and with nodose-

septate hyphae with pale-brown, thickened walls. In the young parts behind the

margin the ends of the fibre hyphae form the pubescent upper surface. Immediately

behind this region the hyphal ends are bent over, flattened on to the surface

in all directions and agglutinated by a thin layer of a clear, lacquer-like substance

up to 30u thick, into the smooth soft trichocutis (Lohwag, 1940) of the glabrous

upper surface. Immediately below the trichocutis numerous hyaline, thin-walled,

nodose-septate hyphae are present. The lower context consists mainly of pale,

smoky-brown fibre hyphae intertwined with one another and turning downwards

towards the trama of the dissepiments, becoming more tortuous and intricately

intertwined. Just above the dissepinients the pale brown, nodose-septate hyphae

with slightly thickened walls, become more numerous and tortuous, branching

frequently and are tightly interwoven with the fibre hyphae, binding them into a

tough tissue of which the elements are separated out with difficulty. In the

dissepiments the fibre hyphae are tightly intertwined with numerous, hyaline,

thin-walled, nodose-septate hyphae which branch freely to bear the basidia at the

hymenial surfaces.

Decay and hosts

This species causes a white rot of hardwood logs and stumps in sub-tropical

climates.

Specimens examined

Herb. PRE: 12022, on decayed log, Kyagwa, Uganda, July 1916; 14885, on decayed log,

Victoria Nyanza, Uganda, July 1916; 27791, on gum tree, Pietermaritzburg, Natal, July

1916; 33125, on rotting log, Mt. Silinda Forest, S. Rhodesia, Jan. 1939; 34092, on old dead

tree, Sichele For. Reserve, Zambia, Apr. 1944; 36585, on old dead tree, Sierra Leone, (Deighton

No. 2571), March 1947; 40074, on logs in woods, Fungi Cubens., Wright No. 241, March

1947; *42154, on Eucalyptus sp. stump, Wilgeboom Plantation, E. Transvaal, Feb. 1961.

Herb. K: Berkeley Herbarium, 1879, Polyporus vinosus Berk. No. 43, St. Domingo. (Holotype).

Herb, STE: 139, on old log, Pietermaritzburg, Natal.

Herb. NY: 132, on prostrate log, Lamao River, Mt. Mariveles, Bataan, Luzon, 1903; 739,

Alto Cedro, Cuba, 1903; 764, Cooper’s Ranch, El Yunque, Mt. Baracoa, Cuba, 1903; 764,

pine log, Gainesville, Fla., coll. Weber, May 1938; 873, on dead wood, Plants of Trinidad,

Caroni, North Beach Road, May 1938; 873, on dead wood, Reinkliaar no. 305, St. Domingo,

April 1906; 2148, on Dipterocarpus vcrnicifluus, Bosoboso, Rizal, Luson, 1907; 2148, on

Dipterocarpus, Camp Keithley, Lake Lanao, Mindanao, 1907; 3695, on prostrate log, Mt.

Mariveles, Bataan, Luzon, 1904; 7212, on half decayed logs and stumps, Palo, Leyte, Jan.

1906; 16469, Mt. Bulusan, Sorsogon, Luzon, June 1916; 17335, Mt. Bulusan, Sorsogon, Luzon,

June 1916; 18444, Los Banus (Mt. Maquiling) Laguna, Luzon, 1917; 18444, British Honduras,

1906; 18444, Troye and Tyre, Cockpit County, Jamaica, Jan. 1909; 18444, Montgomery Co.,

Alabama, Jan. 1915; 19236, on prostrate log, Attapulgus Station, Decatur Co., Ga., 1903;

178527, Florida Agr. Expt. Station, Planera Hammock, Fla., Feb. 1938; 380, Porto Rico,

1923; 581, Porto Novo, St. Catharines, Brazil, 1928; 581, Herbarium, Expt., Station, Porto

Rico Sugar Growers Association, No. 1404, El Dugue, 1914.

Interfertility studies

In order to determine the type of interfertility of Polyporus vinosus, sixteen

cultures, each obtained from a single basidiospore from a small fructification

formed in a culture of PRE 42154, were paired in all possible combinations on malt

agar slopes. It was found that Polyporus vinosus has the tetrapolar type of inter-

fertility with allelomorphs for heterothallism at two loci. Only three mating types

were presented in the mycelia used. The distribution of mating types among

the single basidiospore cultures are given in Tasie 10.

Pye)

Discussion

Polyporus vinosus had not been described in culture before but with the positive

reaction for extra-cellular oxidase, the presence of fibre hyphae and clamp con-

nections on its thin-walled hyphae in culture, it agrees in most respects with Nobles’

(1958 b) characters of Group 45. It differs from other species in this group

by having fibre hyphae with coloured walls and basidiospores which are allantoid

rather than cylindrical. Because no separate group for species with allantoid

spores was available this species is placed in Group 45. The cultures of Polyporus

vinosus differ from those of other species in this group because of the slow growth

rate, scanty mycelium and reddish-purple colours of its fibre hyphae. These

features, which serve to distinguish cultures of this species, also indicate that it

is not well placed in this group and that its phylogenetic relationships may be with

species outside this group.

The carpophores of Polyporus vinosus consist of three kinds of hyphae, viz:

thin-walled, nodose-septate hyphae, thick-walled, nodose-septate hyphae and fibre

hyphae. The small number of hyphal types present, suggest a simple construction

of the carpophores of Polyporus vinosus, but, it was seen that thick-walled, nodose-

septate hyphae were interwoven with the fibre hyphae of the lower context, binding

them into a dense and very tough tissue. These hyphae cannot be regarded as

binding hyphae in the sense of Corner’s (1932 a, 1953) and Cunningham’s (1946,

1954) definitions as they seem to be part of the generative hyphal system and are

continuous with it. In this respect they do not resemble the thick-walled, nodose-

septate, branching hyphae of the binding hyphal system seen in carpophores of

Lenzites betulina and Polyporus versicolor in this group. These thick-walled,

nodose-septate hyphae in the carpophores of Lenzites betulina, Polyporus versicolor

and Polyporus vinosus, may be described as “‘sclerified generative hyphae”? (Donk,

1964) with a binding function. Their presence in carpophores of Polyporus

vinosus establishes a much more complex construction of these carpophores than

in those of Fomes pinicola, which also has a dimitic hyphal system. This com-

plexity of construction is not conveyed by the phrase, ““carpophores with dimitic

hyphal system” in the sense of Corner (1932 b, 1953), Cunningham (1946, 1954),

Teixeira (1962 b) and Fidalgo & Fidalgo (1967).

From the descriptions it is clear that the structures found in the cultures of

Polyporus vinosus are also present in the carpophores from which they were made.

Fibre hyphae formed in culture were found to be much shorter than those of the

carpophores. This appeared to be due to the very slow rate of growth of these

hyphae in culture. In all other characters, the hyphae from these two sources

were similar. No nodose-septate hyphae with pale-brown and slightly thickened

walls were formed in the cultures although they were numerous in the carpophores.

They may be expected to form in cultures under the right conditions since they

wete often seen to be continuous with the thin-walled nodose-septate hyphae in

the carpophores.

When compared with other species of Group 45 described above, important

differences in hyphal characters and carpophore construction are evident between

their carpophores and those of Polyporus vinosus. The other species all have

carpophores in which branched, aseptate, binding hyphae (Corner, 1932 a) bind

the skeletal hyphae together. The presence or absence of different types of hyphae

in carpophores is regarded by Teixeira (1962 b), Bondartseva (1961) and Fidalgo

& Fidalgo (1966) as important at the generic level. Since branched, aseptate.

binding hyphae (Corner 1932 a, b) are not present in the carpophores of Polyporus

vinosus, this species cannot be regarded as congeneric with any genus in which

such hyphae are present. For this reason Patouillard’s (loc. cit.) transfer of this

species to the genus Coriolus Quél., is untenable.

274

Imazeki (1952) transferred Polyporus vinosus to the genus Fomitopsis Karsten

of which Fomes pinicola (Sw. ex Fr.) Cooke is the type species (Cooke, 1959).

Carpophores of both these species have dimitic hyphal systems while their upper

surfaces are covered by resinous or lacquer-iike layers. In carpophores of

Polyporus vinosus however, the fibre hyphae are dark-coloured and more closely

interwoven than the hyaline fibre hyphae in carpophores of Fomes_pinicola.

Thin-walled, nodose-septate hyphae in carpophores of Fomes pinicola do not turn

dark or develop thickened walls and bind the fibre hyphae in the tramal tissues

as in carpophores of Polyporus vinosus. The carpophores of Fomes pinicola

arc thus simpler in construction than those of Polyporus vinosus. Furthermore,

Polyporus vinosus has the tetrapolar type of interfertility and its cultures produced

extra-cellular oxidase enzymes, whereas Foines pinicola has the bipolar type of

interfertility (Mounce, 1929) and its cultures lack extra-cellular oxidase. It

appears therefore that Polyporus vinosus and Fomes pinicola cannot be regarded

as being congeneric.

Murrill (1905) created the genus Nigroporus with Polyporus vinosus Berk.

as the type and only species. In view of the above descriptions it appears that

this genus may be retaincd tor species with dark-coloured carpophores consisting

of hyaline, thin-walled, nodose-septate hyphae, brown, thick-walled, nodose-septate

hyphae and unbranched fibre hyphae with brown walls, hyaline, allantoid

basidiospores and which cause a white rot of hardwoods. No other species

possessing this combination of characters are known so that the relationships of

this species are obscure at present.

Resumé.

The species included here in Group 45 have all those characters in common

which are required for their inclusion in this group. On the basis of differences

in the micromorphology of their carpophores however. three smaller sub-groups

may be distinguished, viz.: (i) a sub-group in which the binding hyphal system

consists of thick-walled, aseptate fibre hyphae with short, tortuous branches and

branching, thick-walled or solid, nodose-septate hyphae and which includes

Polyporus versicolor, Trametes suaveolens, Lenzites betulina, Polyporus pubescens,

Trametes meyenti and Lenzites palisoti; (ii) a sub-group in which the binding

hyphal system is composed of aseptate, thick-walled fibre hyphae with long flexuous

tapering branches and which includes Polyporus occidentalis and Trametes cingulata

end (iii) a sub-group without a binding hyphal system which includes Polyporus

VINOSUS.

The species in these three sub-groups thus share a number of correlated

characters, viz.: production of extza-cellular oxidase, association with white rots,

nodose-septate hyphae, fibre hyphae and the tetrapolar type of interfertility. It

appears that these species share a ccmmon ancestry but show diversity in the

elements of their carpophores and in their construction.

5.8 Group 51

Cultures of species in this group form white to cream coloured mycelial mats

which soon develop extensive, appressed, brown, pseudoparenchymatous areas.

Extra-cellular oxidase enzymes are produced. Their thin-walled hyphae have

simple clamps at the septa and may remain so or may develop thick-walled,

irregular projections and cuticular cells in the pseudoparenchymatous areas. Thick-

walled, aseptate fibre hyphae are also formed. Their basidiospores are cylindrical.

Interfertility is of the tetrapolar type in those species of which this character is

known.

t h b

ff I] ier

Fic. 39.— Daedalea confragosa. (a) Carpophores, upper surface, of PRE 42386 and (b)

hymenial surface; (c) unbranched fibre hyphae and fibre hyphae with numerous

branches from lower context, « 400 phase contrast; (d) cuticular cells from upper

surface of DAOM 30121, x 1000, squash preparation; (e) culture of PRE 42345 at six

weeks; (f) cuticular cells and hyphae with interlocking projections from culture, x 500

phase contrast.

276

Daedalea confragosa Bolt. ex Fr. in Syst. Myc. 1, 336, 1821;

Daedaleopsis confragosa (Bolt. ex Fr.) Schroet. in Cohn Kryptog.-Fl. Schles.

Pilz. p. 493, 1888;

Trametes confragosa (Bolt. ex Fr.) Jorstad, Kgl. Norske Videnskab. Selskabs.

10, 28, 1936.

Cultural characters

Growth moderately fast to slow the mats attaining radii of 8 — 15 mm in

one week and covering the plates in three to six weeks. Advancing zone even,

closely appressed, hyaline or white becoming more raised and somewhat cottony

to woolly towards the inoculum or remaining sub-felty and appressed with isolated

felty patches. After about 2 weeks sunken areas of collapsed mycelium appear,

bordering abruptly on the white aerial mycelium and on dark, crustose areas

of “hazel”, “russet’’, ‘‘avellaneous’, “wood brown” to ‘‘army brown’ colour

which develop in some isolates. In others, the mat remains thin, sub-felty to sodden,

with little or no aerial mycelium, developing patches of submerged mycelium, or,

patches of raised, felty, aerial mycelium covered with iregular, crustose areas of

“cinnamon buff’, “‘Saccardo’s umber” or ““mummy brown” which gradually

increase in size. White, aerial mycelium may darken gradually to “light buff”,

“light pinkish cinnamon” or “tawny”. After six weeks the plates may be covered

with thin, tough, felty mycelium, white in some parts or mostly in shades of

brown varying from “‘light buff’, “light pinkish cinnamon” to ‘‘avellaneous’’,

“wood brown” or “cinnamon brown” and oozing droplets of dark brownish

liquid, some covered by irregular, crustose areas of “‘natural brown” or “Mars

brown’’, somewhat sunken and sharply demarcated from the felty mycelium. Or,

the mat may be sub-felty and sodden with irregular, crustose areas with character-

istic, sunken margins in “cinnamon brown’ to “Saccardo’s umber” along the

margins, occasionally incompletely covered in their central parts. The reverse

darkens gradually in reddish brown colours, mostly more deeply coloured under

the crustose areas and presenting a marbled appearance. No odour is emitted

by most isolates but a slight, pepper-like odour may be present in some.

On gallic acid and tannic acid agar strong diffusion zones are formed but no

growth takes place on gallic acid and only a trace on tannic acid agar.

Advancing mycelium: hyphae hyaline, thin-walled, nodose-septate, branching at or

near the septa often with numerous short branches from a short section of hyphae.

2.0 — 4.0u in diameter (Fig. 40 a).

Aerial mycelium: (a) hyphae as in the advancing zone; (b) fibre hyphae hyaline

at first, darkening later, branched, the branches long, tapering, solid or nearly

so with the lumina narrow, aseptate or with one or two simple septa near the

thin-walled tip, 1.5 — 3.0u in diameter (Fig. 40 b); (c) cuticular cells ferruginous

brown, thin-walled or thick-walled, of irregular shape, often distended into a

number of irregular projections up to 20u in diameter and arising from thin-walled,

Fic. 40.— Daedalea confragosa. a - d. Structures from cultures: (a) hyphae from

advancing zone; (b) fibre hyphae; (c) cuticular cells; (d) nodose-septate hyphae

with thick, brown walls and irregular projections.

e - q. Structures from carpophores: (e) thin-walled, branching, nodose-septate

hyphae; (f) brown, thick-walled, nodose-septate hyphae; (g) unbranched, fibre

hyphae; (h) fibre hyphae with branches towards the distal end; (k) fibre hyphae

with numerous, short, tortuous branches; (m) thick-walled, nodose-septate

hyphae with tortuous branches; (n) basidia; (p) basidiospores; (q) basidiole.

278

nodose-septate hyphae, (Fig. 40 c) not present in all cultures; (d) nodose-septate

hyphae with thickened brown walls, 2.0 — 4.5u in diameter, branched, often

with irregular projections and often agglutinated into strands, apparently inter-

grading into the cuticular cells and forming crustose areas (Fig. 40 d).

Submerged mycelium: hyphae as in the advancing zone.

Carpophore characters

Carpophore annual or reviving, lignicolous, solitary, sessile, dimidiate,

applanate, plane to somewhat convex above, occasionally imbricate, or laterally

connate; leathery and watery when fresh, drying to hard, rigid, woody, 2 — 10 x

3 — I5 x 0.2 — 3.0 cm; upper surface greyish, smoky, umber or sometimes with

a reddish brown crust, finely pubescent to glabrous or nearly so, radiately rugose,

often concentrically grooved; margin acute, thin; pore surface whitish to avellaneous

drying to isabelline or pale brown, poroid to daedaloid or lamellate, pores 0.5 — 1.5

mm wide, dissepiments entire but often becoming lacerate or dentate; tubes

concolorous 0.1 — 1.5 cm deep; context floccose to corky, whitish to pale brownish,

zonate, 0.2 — 1.5 cm thick.

Hyphal characters. Carpophores consist of: (i) hyaline, thin-walled, nodose-septate

hyphae branching close to the septa, 2.0—3.0u in diameter (Fig. 40e); (ii)

nodose-septate hyphae with thickened walls and narrow or occluded lumina, the

walls sub-hyaline or brownish and lumina empty or with deeply staining contents,

1.0 — 3.0u in diameter (Fig. 40 f); (ii) fibre hyphae long, unbranched, straight

or flexuous, smooth or somewhat uneven to almost moniliform towards the tapering

ends, sub-hyaline to pale brown, the walls thick and lumina narrow or occluded

to form a series of deeply staining dots along the moniliform parts, aseptate,

widening only at the extremities, 3.0 — 9.0u in diameter (Fig. 40 g); (iv) fibre

hyphae, hyaline or sub-hyaline, flexuous or fairly straight, sub-solid to solid,

aseptate, branching towards the distal end, the branches short tortuous and tapering

towards their tips, 3.0 — 6.0u in diameter (Fig. 40 h); (vj fibre hyphae with short,

tortuous branches, sub-hyaline, thick-walled, lumina narrow or occluded, aseptate,

1.0 — 3.0u in diameter (Fig. 40 k); (vi) nodose-septate hyphae thick-walled, sub-

hyaline, with short, tortuous branches. 1.0 — 3.0u in diameter (Fig. 40 m); (vii)

cuticular cells brown, irregularly distended, often with irregular projections, thin-

walled or thick-walled and occasionally with deeply staining contents, 8 12u

in diameter (Fig. 39 d).

Hymenium: basidia long, clavate, hyaline, 12.0 — 24.0 x 2.5 — 4.5u, bearing

4 straight sterigmata 2.8 — 3.2u in length (Fig. 40 m); basidiospores hyaline, long

cylindrical to allantoid, smooth, thin-walled, 5.4 — 7.8 x 1.2 — 1.8u (Fig. 40 p);

basidioles sub-hyaline, thin-walled or walls slightly thickened, 28.0 — 36.0 x

1.5 — 3.0u with one to four irregular, short, terminal branches up to 15.0 x

1.0 — 3.0u (Fig. 40 q).

Construction. At the margin the fruit-body consists of long, unbranched fibre

hyphae straight or somewhat flexuous and more or less parallel to each other,

arising from thin-walled, branching, nodose-septate hyphae with deeply staining

contents, intertwined with the fibre hyphae.

Behind the margin in the upper part of the context the tissues consist almost

entirely of solid or sub-solid, faintly brownish, unbranched fibre hyphae arranged

more or less parallel to one another, bending towards the upper surface where

their tips, with slightly dilated lumina, end at a common level to form the finely

pubescent upper surface. Interwoven with these fibre hyphae and running across

their direction of growth are long flexuous or tortuous fibre hyphae binding the

parallel hyphae into a firm tissue.

DD)

At the upper surface in the older parts the terminal portions of the fibre

hyphae are closely intertwined and interwoven in all directions to form a cortex

with a finely pubescent upper surface which soon becomes agglutinated into a

thin, glabrous, cuticular layer by a hyaline lacquer-like substance. In this cortex

narrow, hyaline, branching, thin-walled, nodose-septate hyphae 1.0 — 1.8u in

diameter and with deeply staining contents, are very numerous and interwoven

with the fibre hyphae. On some fruit-bodies short, narrow, hyaline, solid hyphae

grow upward from these nodose-septate hyphae and become lightly entangled and

interwoven to form patches of pubescent tissues on the upper surface. Rarely,

nodose-septate hyphae with pale brown walls may grow upwards from these

thin-walled hyphae in the cortex and may develop irregular projections or become

expanded into cuticular cells in some fruit-bodies. All these elements eventually

become agglutinated together with fibre hyphae by a brown, amorphous, lacquer-

like material into hard, crustose masses over the upper surfaces. The middle and

lower context, consist of long fibre hyphae, fairly tightly packed, more or less

parallel and bending downward towards the dissepiments. Just above the dis-

sepiments the fibre hyphae become more tortuous especially towards their tips

and many develop short, tortuous, lateral branches. In this region short, tortuous,

fibre hyphae with many short, tortuous branches (binding hyphae, Corner, 1932 a)

become very numerous and tightly interwoven with the other hyphae, binding

them into a tough, homogeneous tissue. Also present in this region are narrow,

branching, hyaline, thin-walled, nodose-septate hyphae with deeply staining contents

interwoven with the fibre hyphae. The tissues of the dissepiments consist of

tightly interwoven, branched, thick-walled, nodose-septate hyphae and fibre hyphae

inextricably interwoven and narrow, thin-walled, nodose-septate hyphae branching

repeatedly and ramifying among the fibre hyphae towards the hymenial surfaces

where they bear the basidia in clusters on short branches.

Decay and hosts

Daedalea confragosa causes a white rot of dead sapwood of various hardwood

trees but had been noticed on wounds as well (Overholts, 1953).

Specimens examined

Herb. DAOM: *F1577, on Fagus grandifolia, Meach Lake, Que.; F6307, on Salix nigricotini-

folia, Ottawa, Ont., Nov. 1931; F6457. on Betula alba var Papyrifera, Cartier Lake, Petawawa,

Ont., Aug. 1935; F7757, on Betula sp., Chalk Riv., Ont., Sept. 1937; F7763, on Betula sp.,

Chalk Riv., Ont., Sept. 1937; F8018, on Betula sp., Iberville, Que., Jan. 1938; F8063, on

Acer saccharum, Petawawa, Ont., Aug. 1937; F8080, near Ludlow, Shropshire, Sept. 1937;

F8340, on Betula papyrifera, Gatineau, Que., Aug. 1938; *F8997, on Acer sp., Ottawa,

Ont.; F9111, on Alnus incana, Notakim Depot, Que., Sept. 1939; *F9210, on Betula papyrifera,

Chelsea, Que., May 1939; F9411, on Betula papyrifera, Horseshoe Bay, Ont., Aug. 1939;

F10783, on Prunus avium, Caledon East, Ont., Oct. 1941; *17555, on Betula lutea, Gatineau

Park, Que., Sept. 1947; 22399, on Beinla occidentalis, Kaslo, B.C., Aug. 1948; 22546, on

Salix bibbiana, Steen River, Alberta, July 1950; 30121, on Prunus sp., Vancouver, B.C., 1948;

30269, on Populus trichocarpa, Kaslo, B.C., Oct. 1951; 30270, on dead Salix sp., Candle Lake.

Sask., Aug. 1949; 31089, on Salix bibbiana, Riding Mountain, Man., May 1949; 52911, on

Betula sp., Sicamous, B.C., Aug. 1912; 53773, on Betula sp., Esher, Surrey, Sept. 1959;

69975, on Salix sp., Agassiz, B.C., Sept. 1959; 72334, on Betula papyrifera, Petawawa, Ont..

Sept. 1946; *94045, on dead yellow birch, Dorset, Ont., Sept. 1962; *94052, on dead hardwood

branches, Dorset, Ont., Sept. 1962; *94054, on dead wood, Dorset, Ont., Sept. 1962.

Discussion

The cultural characters of Daedalea confragosa as described above, agree well

with the requirements for its inclusion in Group 51. The description also agrees

well with earlier descriptions by Davidson et al. (1938, 1942) and Nobles (1948,

1965).

280

Cultures of Daedalea confragosa develop dark-coloured, skin-like or crustose

areas which are formed in cultures of species of stipitate polypores as described by

Nobles (1958 b) in Group 53. Nobles (1948) stated that cultures of Daedalea

confragosa may be confused with those of Polyporus tuberaster and Polyporus

brumalis but that cultures of these stipitate species may be distinguished from those

of Daedalea confragosa by having more extensive pseudoparenchymatous areas.

The presence of cuticular cells, which are frequently found in cultures of Daedalea

confragosa but not in those of Polyporus tuberaster and Polyporus brumalis, appears

to be an aditional diagnostic character. Furthermore, cultures of Daedalea

confragosa tend to have brownish compact, tough, felty or sub-felty mycelial mats

while those of Polyporus brumalis and other stipitate species in Group 53 (Nobles,

1958 b) mostly possess white, woolly, aerial mycelium around the pseudo-parenchy-

matous areas. The distinctions together with consideration of host records may

serve to distinguish cultures of Daedalea confragosa from those of the other two

species.

The carpophores of Daedalea confragosa consist of five different types of

hyphae, viz. nodose-septate hyphae, which may be thin-walled generative hyphae,

or, thick-walled hyphae with tortuous branches which form part of the binding

hyphal system, unbranched fibre hyphae or skeletal hyphae (sensu Corner, 1932 a),

fibre hyphae with flexuous branches towards the distal ends, which also contribute

to the binding hyphal system, and fibre hyphae with numerous short, tortuous

branches (binding hyphae, sensu Corner, 1932 a, 1953). Since generative, skeletal

and binding hyphae are present in the fruit-bodies, these fruit-bodies have a

trimitic hyphal system (sensu Corner, 1932 a; Cunningham, 1946); but morphologic-

ally and ontogenically different hyphae comprise the binding hyphal system of

the fruit-bodies.

Cuticular cells and brown, thick-walled nodose-septate hyphae with irregular

projections were present on the upper surfaces of a very small proportion of the

fruit-bodies examined. These structures arise as modified terminal parts of the

thin-walled, nodose-septate hyphae present near the upper surfaces of the fruit-

bodies. Numerous narrow, thin-walled, nodose-septate hyphae were present at

the upper surfaces of all the fruit-bodies examined. It therefore appears that the

growth and modification of these hyphae into cuticular cells, occur in nature under

certain conditions only.

From the descriptions it is evident that structures formed in cultures of

Daedalea confragosa may also be present in the carpophores from which they

were made; but some differences in morphology are evident in certain structures.

The fibre hyphae formed in culture are of one type only being narrow and branched

with the branches long, narrow and tapering. These fibre hyphae differ in their

manner of branching from the fibre hyphae present in the carpophores of Daedalea

confragosa and appear to be interniediate between the unbranched and much-

branched fibre hyphae of the fruit-bodies. Cuticular cells were not formed in all

the cultures examined but thick-walled hyphae with irregular projections were

mostly present. These structures developed even in cultures made from sporo-

phores from which they were absent. It thus appears that the development of

these structures depends on the conditions under which the mycelium is growing

rather than its genetic complement. Conditions favourable for their development

thus appear to exist more frequently in culture than in nature. Their presence in

carpophores thus represents a character of doubtful taxonomic value.

The hyphal characters and construction of carpophores of Daedalea confragosa

had been described before by different workers. Cunningham (1948 h) included

the genus Daedaleopsis Schroet., of which Daedalea confragosa is the type species

281

(Donk, 1960), in the genus Daedalea Pers. ex Fr., which he characterized as having

a trimitic hyphal system with skeletal hyphae unbranched, aseptate, some shade

of brown; binding hyphae aseptate, commonly of the bevista type, some shade of

brown and nodose-septate, hyaline, generative hyphae. Overholts (1953) reported

that the hyphae of Daedalea confragosa were mostly simple, aseptate and_thick-

walled while some narrow hyphae were branched to form ‘‘a simple type of hyphal

complex”. Teston (1953 b) also reported nodose-septate generative and unbranched,

thick-walled, skeletal hyphae in fruit-bodies of Trametes erubescens Alb. & Schw.

ex Fr. (= Daedalea confragosa Bolt. ex Fr.), with much-branched, sinuous, thick-

walled binding hyphae also present in the trama of the tubes. The hyphal characters

and construction of the fruit-bodies ef Daedalea confragosa as described above

thus agree with reports by earlier workers but more detail is presented here.

Comparison of the above descriptions with those of the type species of other

genera to which Daedalea confragosa had been assigned by earlier workers, reveals

important differences. Fries (1821) placed this species in the genus Daedalea

Pers. ex Fr. and was followed in this by many later workers (Pilat, 1936); but in

cultures and carpophores of Daedalea quercina L. ex Fr., the type of the genus

Daedalea Pers. ex Fr., nodose-septate hyphae with irregularly thickened walls are

present. Its cultures do not produce extra-cellular oxidase and its carpophores

lack binding hyphae. Nodose-septate hyphae with irregularly thickened walls

are absent from the cultures and carpophores of Daedalea confragosa. Instead,

cuticular cells and hyphae with irregular projections are present. Furthermore,

cultures of Daedalea confragosa produce extra-cellular oxidase and binding hyphae

are present in its carpophores. Daedalea confragosa thus cannot be regarded to

be congeneric with Daedalea quercina despite many superficial similarities.

Ames (1913), in her study of the structure of polypore fruit-bodies, included

Daedalea confragosa in the genus Daedalea Fr. of which she stated. ‘“This genus

differs from Trametes only in the form of the hymenial surface’. She found no

difference in the structure of different species in these two genera. Later, Jorstad

(loc. cit.) transferred Daedalea confragosa to the genus Trametes Fr. and comparison

with Trametes suaveolens (L. ex Fr.) Fr., the type species of that genus, reveals

many similarities. The general plan of construction of the carpophores is similar

in both species. Binding hyphae are numerous in the lower context and rare

in the upper context of carpophores of both species. Unbranched fibre hyphae

are more closely packed in the carpophores of Daedalea confragosa resulting in a

corky texture of the fruit-bodies while those of Trametes suaveolens are loosely

packed to form the soft, spongy tissue of carpophores of this species. The same

types of hyphae occur in carpophores of both species but the cuticular cells and

hyphae with irregular projections are never formed in carpophores of Trametes

suaveolens of which the upper surfaces never become crustose. Also, the binding

hyphae of Trametes suaveolens are more translucent than those of Daedalea

confragosa which resemble the unbranched fibre hyphae. The basidiospores of

Daedalea confragosa are allantoid in shape unlike the cylindrical spores of Trametes

suaveolens. In cultural characters the differences between the two species are

much more striking because of the presence of brown, skin-like or crustose areas

and their associated modified hyphal elements are never found in carpophores or

cultures of Trametes suaveolens so that their presence in those of Daedalea

confragosa constitute a genetic difference between the two species. Because the

absence or presence of different types of hyphae are considered to be of importance

at the generic level (Bondartseva, 1961; Teixeira, 1962 b) and because of the

differences in cultural and carpophore characters, these two species do not appear

to be congeneric despite the presence of many similar characters.

282

Daedalea confragosa had also been referred to the genus Lenzites Fr. by various

workers (Pilat, 1936). Many similarities in hyphal characters and construction

exist between fruit-bodies of Daedalea confragosa and Lenzites betulina (L. ex Fr.)

Fr. the type of the genus Lenzites Fi.; but Daedalea confragosa differs from

Lenzites betulina in the same characters in which it differs from Trametes suaveo-

lens so that these two species cannot be considered to be congeneric.

Because of these differences it seems best to maintain Daedalea confragosa

in the genus Daedaleopsis Schroet. of which it is the type species (Donk, 1960).

Future studies, however, may well revea. transitional species to the genus Trametes

with which Daedalea confragosa has strong affinities.

Trametes corrugata (Pers.) Bresadola in Hedwigia 51, 316, 1912;

Polyporus corrugatus Pers. in Gaudichaud, Voy. Freyc. Uranie Bot. 172,

1827;

Earliella corrugata (Pers.) Murill in Bull. Torrey Bot. Club 34, 468, 1907;

Coriolus corrugatus (Pers.) G. H. Cunningham in Proc. Linn. Soc. N.S.W. 75,

Mita, NOX02

Formitopsis corrugata (Pers.) Imazeki in Bull. Tokyo Sci. Mus. 6, 92, 1943.

Cultural characters

Growth is moderately rapid the mat reaching a radius of 25 mm in one week

and covering the plate in 34 weeks. The margin is even, mycelium appressed

or submerged for a short distance, then raised, floccose-woolly, pure white at first

but becoming somewhat collapsed, more woolly to felty with faint, brownish

colours developing in spots on the felty mycelium, after two weeks. The mat

gradually becomes more dense with balls of woolly, white mycelium forming on

its surface and on the sides of the dish. One or two concentric, sulcate zones

appear over the cultures and within two to three weeks crustose areas, at first

smooth and “hazel” or “‘cinnamon’’, appear and increase in size, their margins

contrasting sharply with the white, woolly mat. Crustose areas remain “‘hazel”

or “cinnamon” or become “cinnamon rufous” or “cinnamon brown” and

roughened, somewhat papillate and rugose in the older parts. Shiny, smooth,

“cinnamon rufous” or “‘cinnamon brown”’ laccate areas form in the crustose areas

against the glass sides of the dish. After four to five weeks lumps of woolly

mycelium may form against the sides of the dishes and gradually develop waxy

or pasty, slightly sunken areas on which low, labyrinthiform lamellae, from which

spores are discharged in inverted cultures, are formed. The reverse bleaches after

two weeks and a faint mushroomy odour is given otf. A strong positive reaction

is obtained when the culture is tested for extra-cellular oxidase by means of gum

guaiac solution. Strong diffusion zones are formed on gallic acid and tannic

acid media with colonies reaching up to 15 mm in diameter on tannic acid agar

after one week.

Advancing mycelium: hyphae hyaline, branching, nodose-septate, thin-walled,

2.2 — 4.5u in diameter (Fig. 42 a).

Aerial mycelium: (a) hyphae as in the advancing zone; (b) fibre hyphae hyaline,

straight, unbranched or occasionally branched, thick-walled, the lumina narrow,

aseptate, 1.5 — 3.5u in diameter (Fig. 42 b); (c) narrow hyphae repeatedly

dichotomously branched, hyaline 0.7 — 1.5u in diameter, forming a net-like

structure just above the agar (Fig. 42 c); (d) hyphae with brown, thickened walls

283

Fic. 41.— Trametes corrugata. (a) Upper surface and (b) hymenial surface of carpophore

of PRE 34446; (c) cuticular cells from red stain on upper surface of carpophore of

type specimen of Earliella cubensis, < 1000; (d) unbranched fibre hyphae and fibre

hyphae with numerous branches from context of type specimen of Earliella cubensis,

x 500; (e) culture of PRE 42454 at six weeks; (f) fructification in culture; (g) nodose-

septate hyphae with irregular projections from culture, x 100; (h) cuticular cells from

culture, x 100; (k) narrow, hyaline, branching hyphae from culture, < 1000.

284

bearing short lateral projections up tc 7u long, walls thickened to sub-solid or

solid, 2.5 — 3.5. in diameter and arising from nodose-septate hyphae (Fig. 42 d);

(e) cuticular cells pale brown, sub-globose to ovoid or irregular in shape with fairly

thin walls up to 25u in diameter and arising from thin-walled pale or hyaline

nodose-septate hyphae, and often embedded in a brown lacquer-like substance

(Fig. 42 e).

Fructification: basidia cylindrical or narrowly clavate, 22.5 — 34.0 x 5.2 — 6.7u,

bearing four straight sterigmata 5.2 — 6.0u (Fig. 42 f); basidiospores hyaline,

smooth, thin-walled, long-elliptical to cylindrical, 6.7 — 10.5 x 3.7 — 5.2u (Fig.

42 g).

Submerged mycelium: hyphae as in the advancing zone.

Carpophore characters

Carpophore annual, lignicolous, sessile, effused-reflexed, laterally extended, or

conchate, connate, occasionally imbricate, woody, up to 15 x 3 — 6 x 0.2 — 2.5

cm; surface at first smooth, finely pubescent to glabrous. becoming rugose, zonate

in older parts, creamy to “‘light buff’? and thinly encrusted in “‘cinnamon brown”

to dark “liver brown” or almost black areas in the oldest parts; margin obtuse at

first, thick, later thin, creamy white, undulate, drying to “‘light buff’; pore surface

pale pinkish when fresh drying to creamy white or “‘light buff’, pores 2 — 4 per

mm, poroid to daedaloid, rounded; dissepiments even, thick at first, later thin;

tubes | — 8 mm deep; context white, corky to fibrous, zonate, drying to “‘light

buff’, 2 — 20 mm thick.

Hyphal characters. Carpophorzs censist of (i) hyaline, branching, thin-walled.

nodose-septate, hyphae, 1.5 — 3.54 in diameter (Fig. 42 h); (ii) fibre hyphae hyaline,

straight or tortuous, unbranched or with an occasional long branch, the walls

thickened, lumina prominent or narrow or occluded, aseptate or occasionally with

one or two simple septa near the apex, 2.5 — 5.0u in diameter (Fig. 42 k); (iii)

fibre hyphae hyaline, sub-solid, repeatedly branched, the branches short or fairly

long, flexuous, 2.5 — 3.5u in diameter (Fig. 41 d, 42 m); (iv) narrow, hyaline,

sub-solid hyphae, repeatedly branched, 0.5 — 0.7u in diameter (Fig. 41 k); (v)

cuticular cells with thickened, brownish walls distended into irregular shapes,

4.0 — 10.0u in the widest parts and borne on thin-walled, nodose-septate hyphae

(Fig. 41 c); (vi) nodose-septate hyphae with brownish thickened walls with irregular

projections, 2.5 — 3.5u in diameter (Fig. 41 c).

Hymenium: basidia hyaline, long, clavate 18.0 — 34.0 x 6.0 — 7.5u and bearing

four prominent sterigmata 4.5 — 6.0u (Fig. 42 n); basidiospores hyaline, long

cylindrical, smooth, thin-walled, obliquely apiculate, 8.0 — 12.0 x 4.5 — 5.5u

(Fig. 42 p).

Construction. At the margin the carpophore consists of long, more or less straight,

fibre hyphae with thick, hyaline walls often thin-walled, and collapsed towards the

extremities with lumina narrow, aSeptate or with one or two simple septa, tightly

intertwined with one another and with the branching, thin-walled, hyaline, nodose-

septate hyphae from which they arise. Behind the margin the context consists

of straight and tortuous, unbranched, fibre hyphae with hyaline walls partly

thickened or sub-solid, and tightly intertwined. Nodose-septate hyphae, mostly

empty, thin-walled and sometimes collapsed, are present in small numbers inter-

twined with the fibre hyphae. Interwoven with these hyphae are numerous hyaline,

branching fibre hyphae, their branches long and tortuous, which bind all the hyphae

into a tough, dense tissue. In the upper part of the coniext the fibre hyphae turn

towards the upper surface where theit ends are packed at a common level to

285

FIGURE 42.

Fic. 42.— Trametes corrugata. a - g. Structures from cultures: (a) hypha from

arrow, dichotomously branching hyphae;

advancing zone; (b) fibre hyphae; (c) n )

ular projections; (e) cuticular

(d) hyphae with brown, thickened walls and irreg

cells; (f) basidium; (g) basidiospores.

h-p. Structures from carpophores: (h) thin-walled, nodose-septate hyphae;

(k) fibre hyphae, unbranched or with occasional long branch; (m) fibre hypha

with many short, flexuous branches; (n) basidium; (p) basidiospores.

286

from the finely pubescent upper surface not covered by the deep, reddish-brown,

crustose structure. At the upper surface below the dark-coloured, crustose areas,

numerous thin-walled, nodose-septate hyphae, mostly narrow and _ frequently

branched are present, intertwined with the ends of the fibre hyphae and forming

a thin-walled, pseudoparenchymatous layer up to about 50u thick over the ends

of the fibre hyphae. Immediately above this layer are larger cuticular cells and

nodose-septate hyphae with interlocking projections on their thickened brown

walls. These structures are agglutinated by a pale brown lacquer-like substance

into the hard crust in a layer up to 150u thick over the older part of the upper

surfaces of the fruit-bodies. In the older parts of the context of some of the

sporophores examined a delicate network of very narrow, hyaline, repeatedly

branched hyphae was visible, interwoven with the wider hyphae. Their origin

could not be determined (Fig. 41 k).

In the lower context the fibre hyphae turn downwards towards the dissepiments.

They are narrow and with more prominent lumina than in the upper context.

The long fibre hyphae become more tortuous and ase tightly interwoven with

branching “‘binding hyphae’, some with solid clamps, which bind them into a

tough, dense tissue. Thin-walled, nodose-septate hyphae become more numerous

in the trama and dissepiments where they branch repeatedly, the branches

ramifying between the fibre hyphae in the direction of the pore surfaces. At the

pore surfaces the nodose-septate liyphae form numerous, very short branches on

which the basidia are borne.

Decay and hosts

Trametes corrugata causes a white rot of hardwood logs in sub-tropical areas.

Specimens examined

Herb. PRE: 15623, on living Albizzia sp.. Durban, Natal, March 1917; 28968, on dead wood,

Rooikoppies Plantation, Duiwelskloof, Transvaal, May 1937; 30230, indigenous wood, Krantz-

kop, Natal, December 1935; 31684, indigenous wood, Ifafa, Natal, August 1916; 34446,

rotting logs of Kukin trees, Hawaii, July 1930; 31736, on dry branch, Chinizina, Beira,

Mocambique, April 1957; *42454, on decaying lidchi stem, Tzaneen, Transvaal, January

1961.

Herb. NY: Earliella cubensis Murrill, on dead wood, Herradura, Pinar del Rio Province,

Cuba, March 7 — 12, 1905, (TYPE).

Interfertility studies

In order to determine the type of heterothallism present in Trametes corrugata,

16 mycelia, each obtained from a single basidiospore produced from a small

fruit-body formed in culture, were paired in all possible combinations. The

formation of clamp connections on the paired mycelia took place in a manner

which proved that Trametes corrugata has the tetrapolar type of interferitility.

The distribution of mating types among the single spore mycelia is given in

TaBLE 11 in the abbreviated form used by Yen (1950).

Discussion

The presence of cuticular cells, fibre hyphae and nodose-septate hyphae, in

cultures which produce extra-cellular oxidase enzymes, places Trametes corrugata,

which had not been described in culture before, in Group 51. Its cultural characters

agree in many respects with those of other species in this group but cultures of

Trametes corrugata may be recognized by the reddish colours of the crustose

areas which contrast sharply with the pure white, woolly mycelium, and the very

large, thin-walled, cuticular cells present in these crustose areas.

287

The carpophores of Trametes corrugata agree in construction and hyphal

characters with those of other species in this group and consist of six types of

hyphae. Of these, the cuticular cells and thick-walled hyphae with irregular

projections are modified portions of the thin-walled, nodose-septate hyphae but

because of the presence of clamp connections in them, they must be regarded as

generative hyphae sensu Corner (1953) and Cunningham (1946). The unbranched

fibre hyphae and fibre hyphae with many long, tortuous branches agree with

Corner’s (1932 a) and Cunningham’s (1946) definitions of skeletal and binding

hyphae respectively. The very natrow, branched, aseptate hyphae present in some

carpophores appear to serve a binding function and may be regarded as part of

the binding system although their origin and true nature could not be determined.

Carpophores of Trametes corruvata thus have a trimitic hyphal system sensu

Corner (1932 a) but hyphae which differ in morphology and ontogeny are present

in the generative and binding systems. The trimitic hyphal system in carpophores

of this species was also reported by Fidalgo & Fidalgo (1966).

The very narrow, branched hyphae were not present in all the carpophores

examined. They were present in the context of the carpophores of PRE 42454

and in the trama of the pores of the type specimen of Earliella Murr., but they

are visible only after prolonged and extremely careful search of the carpophore

tissues. It appears that they become very brittle on drying and disintegrate when

tissues from old specimens are teased out for examination. Similar hyphae had

been noted in cultures of this and other species but have not been found in their

carpophore tissues.

Although very large, thin-walled cuticular cells were present in the crustose

areas of the cultures, the dark reddish-brown, crustose areas of the fruit-bodies

of Trametes corrugata were found to consist almost entirely of small, distorted cells

with thickened, brownish walls which closely resemble the hyphae with interlocking

projections, present in the cultures. It was, however, found in the cultures that

some of these projections on the brown, thick-walled hyphae were distended into

thin-walled vesicles resembling small cuticular cells. It thus appears that cuticular

cells and hyphae with interlocking projections are different structures that develop

from the same hyphae probably under different conditions. If this is correct, then

the hyphae with interlocking projections found in the carpophores must be regarded

as homologous structures to the cuticular cells formed in cultures. All the

structures formed in cultures are thus present in the carpophores from which they

were made.

Although Trametes corrugata fits well into Group 51 in cultural characters,

its fruit-bodies differ in hyphal chatacters from those of other species in this

group. In the fruit-bodies of Daedalea confragosa, Hexagona tenuis and Fomes

fomentarius, the long, unbranched fibre hyphae (skeletal hyphae, ‘Corner 1932 a)

are readily distinguishable from the tortuous, much-branched fibre hyphae (binding

hyphae, Corner 1932 a). In fruit-bodies of Trametes corrugata, the binding hyphae

mostly have long branches which are usually narrower and more tortuous than

the unbranched hyphae in the upper context and may be recognized by their

smaller diameter, branching and more tortuous appearance; but in the lower

context where the skeletal hyphae are narrower and more tortuous than in the

upper context, the two types of hyphae are very similar and portions of the

branched hyphae are often indistinguishable from flexuous portions of the

unbranched hyphae. Furthermore the fibre hyphae of Trametes corrugata are

consistently hyaline under the microscope while those of Daedalea confragosa,

Hexagona tenuis and Fomes fomentarius (Teixeira, 1962 b) are sub-hyaline to pale

brown. The hyphal characters of Trametes corrugata thus differ from those

288

of Daedalea confragosa, Hexagona tenuis and Fomes fomentarius, and Trametes

corrugata cannot, because of these differences as well as other differences in

carpophore morphology, be considered to be congeneric with these species.

Imazeki (1943) placed Trametes corrugata in the genus Fomitopsis Karst.

but comparison with the cultural characters and micromorphological characters

of the carpophores of the type species, Fomes pinicola (Sw. ex Fr.) Cooke, shows

this transfer to be untenable. The hyphal characters and construction of the

context and crusts of the carpophores of the two species are completely different.

Furthermore, extra-cellular oxidase enzymes are produced by cultures of Trametes

corrugata but not by those of Fomes pinicola. The latter species has the bipolar

type of interfertility (Mounce & Macrae, 1938) whilst Trametes corrugata has the

tetrapolar type of interfertility.

Cunningham (loc. cit.) transferred Trametes corrugata to the genus Coriolus

Quél. thereby implying similarity in hyphal characters and carpophore construction

between this species and Polyporus versicolor; but from the above descriptions

it is clear that Trametes corrugata differs from Polyporus versicolor in respect

of the morphology of the binding hyphae in their carpophores. Furthermore, the

hyphae with irregular projections, present in the crustose areas of carpophores

of Trametes corrugata, are absent from those of Polyporus versicolor although

somewhat similar structures have been found in its cultures. Because the absence

or presence of different types of hyphae in carpophores is regarded as significant

at the generic level by various workers (Teixeira, 1962 b; Donk, 1964; Fidalgo &

Fidalgo, 1966) it appears that Trametes corrugata cannot be regarded as being

congeneric with Polyporus versicolor L. ex. Fr., the type species of the genus

Coriolus Quél. Trametes corrugata also differs from Trametes suaveolens in the

same characters in which it differs from Polyporus versicolor so that Bresadola’s

(loc. cit.) combination also appears to be untenable.

The long, branched binding hyphae of Trametes corrugata resemble those of

carpophores of Trametes cingulata. Other hyphal characters and the construction

of the carpophores of these two species are also similar but hyphae with irregular

projections and cuticular cells are absent from carpophores and cultures of

Trametes cingulata.

Carpophores and cultures of Trametes corrugata differ in respect of hyphal

morphology from some other species in Group 51 and from the type species of

the genera Coriolus Quél., Fomitopsis Karst. and Trametes Fr. to which it had

been referred by various workers. The presence of cuticular cells and hyphae

with irregular projections and trimitic hyphal system in its carpophores, indicate

affinities with the type species of the genera Daedaleopsis on the one hand

and Coriolus and Trametes on the other. If the construction of the carpophores

of Daedalea confragosa, Fomes fomentarius, Hexagona tenuis, Trametes acupunctata

and Trametes corrugata is considered, however, it becomes evident that the presence

of cuticular cells and hyphae with irregular projections in their carpophores is the

main character common to them all while differences in the hyphal characters

and construction of their carpophores exist.

It thus appears that these species may have acquired this character by

convergent evolution and that their phylogenetic relationships are less intimate than

the presence of cuticular cells and hyphae with inter-locking projections in their

cultures and carpophores imply.

The general morphology of the carpophores of Trametes corrugata, their

hyphal characters, construction, texture and spore characters agree with those

of some species included in Group 45 (e.g. Trametes cingulata, Trametes meyenii).

289

It thus appears that this species has strong affinities with others in that group;

but the hyphal characters and construction of the carpophores of a larger number

of species in both Group 45 and Group 51 will have to be examined before any

definite conclusions can be reached, and the validity of the genus Earliella Murrill,

of which Trametes corrugata (Pers.) Bres. (— Earliella cubensis Murr.) is the

type species (Murrill, 1907 a), be established or rejected.

Fic. 43.—Hexagona tenuis. (a) Carpophores of PRE 43116; (b) culture of PRE

42159 at six weeks; (c) cuticular cells in red stain on upper surface of

carpophore, x 500.

Hexagona tenuis Hooker ex Fr., Epicr. Syst. Mycol., 498, 1838;

Daedaleopsis tenuis (Hooker ex Fr.) Imazeki in Bull. Tokyo Sci. Mus., 6,

78, 1943.

Cultural characters

Growth is moderately rapid, the mat covering the plate in 3—4 weeks. The

advancing zone is bayed or even with hyphae raised to the limit of growth, the

young mat thin, downy to floccose-cottony. Towards the inoculum the mat becomes

more dense, woolly, white, radially sulcate, and, at about 10 — 20 mm around

the inoculum, sunken and compacted into thin, appressed, pellicular, crustose areas

of “‘cinnamon brown” to “wood brown” often with a thin, downy overgrowth

of white mycelium after 2 — 3 weeks. With advancing age the white mycelium

becomes increasingly woolly and dense, gradually becoming compacted into leathery,

skin-like, wrinkled, crustose areas with colours ranging from “‘light ochraceous

buff” to “tawny” to “‘‘russet,” ‘‘natal brown’? and patches of ‘“‘mummy brown,”

irregular in outline and extending gradually until most of the surface is covered

at six weeks. The reverse darkens with age, dark brown patches developing

under the crustose areas often traversed by very dark, irregular lines and presenting

a marbled appearance. A faint mushroomy odour is given off after two to three

weeks but disappears later. A positive reaction is obtained when cultures are

290

a UO

Fic. 44.—Hexagona tenuis. a - e. Structures from cultures: (a) hypha from ad-

vancing zone; (b) unbranched, fibre hypha; (c) fibre hyphae with long tapering

branches; (d) nodose-septate hyphae with irregular projections and cuticular

cells; (e) swellings on hypha from submerged mycelium.

f-n. Structures from carpophores: (f) thin-walled, nodose-septate hyphae;

(g) fibre hyphae, unbranched or occasionally branched; (h) fibre hyphae with

numerous tortuous branches; (k) cuticular cells; (m) basidia; (n) basidiospores.

FIGURE 44.

291

tested for extra-cellular oxidase by means of alcoholic gum guaiac solution. On

gallic acid and tannic acid media, strong diffusion zones were formed but no

growth occurred.

Advancing mycelium: hyphae hyaline, branching or simple, thin-walled, nodose-

septate, with simple clamps and staining in phloxine, 2.2 — 3.7u in diameter

(Fig. 44 a).

hyaline to pale brown, walls thickened and lumina narrow or occluded, aseptate,

2.2 — 3.5u in diameter (Fig 44 b); (c) fibre hyphae with long tapering branches,

hyaline or pale brown, walls thickened, lumina aseptate, narrow or occluded or

hyphae solid, .2 — 3.0u in diameter (Fig. 44 c); (d) nodose-septate hyphae with

walls slightly thickened, pale brownish to reddish brown and distorted into irregular

projections and swellings or cuticular cells up to 15 in diameter, tightly packed

to form the dark-coloured, crustose areas (Fig. 44 d).

Submerged mycelium: hyphae as in the advancing zone and often developing

swellings resembling the cuticular cells (Fig. 44 e).

Carpophore characters

Pileus annual, lignicolous, solitary, sessile, effused-reflexed, occasionally

resupinate, applanate, conchate or flabelliform, free or laterally connate, coriaceous,

3.5 — 7.0 x 2.0 — 4.5 x 0.05 — 2 cm; surface glabrous, radially sulcate, rugose

or smooth, “‘snuff brown” or ‘“‘cinnamon brown” to “hazel” or “chestnut brown”

often with dark “blackish brown” to “‘seal brown” areas towards the base; margin

thin, acute, entire or rarely lobate, often undulate, ‘‘snuff brown,” ‘‘cinnamon

brown” to “hazel” or “chestnut brown’; pore surface “‘buckthorn brown’ to

“cinnamon brown,” poroid; pores large, angular, 0.5 — 1 mm in diameter;

dissepiments thin, even; tubes shallow, 0.5 — 1 mm deep; context rusty brown,

darkening in KOH, fibrous, up to 1 mm thick.

Hyphal characters. Carpophores consist of: (i) hyaline, branching, thin-walled,

nodose-septate hyphae with staining contents, 1.5 — 3.0u in diameter (Fig. 44 f):

(ii) fibre hyphae long, more or less straight or flexuous, unbranched or occasionally

branched, the branches few, long, sub-hyaline to yellowish brown, the walls thickened

or solid, lumina narrow or occluded widening at the extremities, aseptate or with

One or two simple septa near the tip 3.0 —6.0u in diameter (Fig. 44g); (ii) fibre

hyphae hyaline to sub-hyaline, very tortuous, branching repeatedly over short

distances, the branches short, tortuous, thick-walled, lumina narrow or occluded

widening at the hyphal tips, aseptate, 1.0 — 3.0u in diameter (Fig. 44 h); (iv)

cuticular cells subglose to clavate with irregular, lobate projections, dark reddish-

brown, thin-walled 5 — 15u in diameter, arising from thin-walled, nodose-septate

hyphae (Fig. 43 c, 44 k).

Hymenium: basidia hyaline, long clavate, almost cylindrical, 15.0 — 24.0 x 3.0 —

4.5u with four slender, straight sterigmata 2.8 — 3.2u (Fig. 44 m); basidiospores

hyaline cylindrical smooth, thin-walled 10.0 — 15.0 x 4.0 — 6.0u (Fig. 44 n).

Construction. At the margin the carpophore consists of long, straight, unbranched,

pale-brown fibre hyphae with prominent lumina, arranged parallel to the direction

of growth and slightly intertwined with one another and with the hyaline, thin-

walled, branching, nodose-septate hyphae from which they arise. Behind the

margin the fibre hyphae have thicker walls and bend upwards towards the upper

surface. Few nodose-septate hyphac are present in the upper context, which

292

consists mainly of parallel or intertwined, unbranched, fibre hyphae with yellow-

brown, thickened walls and narrow or occluded lumina, and, numbers of hyaline

or sub-hyaline, thick-walled or solid, fibre hyphae with short, tortuous, branches

interwoven with long, unbranched, fibre hyphae. At the upper surface the ends

of the fibre hyphae are packed at a common Jevel and are bent over to lie flat

on the surface. These elements are covered by a thin, transparent, lacquer-like

substance to form the characteristic glabrous surface of the carpophores. In carpo-

phores with dark, reddish-brown stains over their upper surfaces, thin-walled,

nodose-septate hyphae with deeply staining contents are present in large numbers

among the fibre hyphae below these areas. From these hyphae, dark-brown,

swollen, cuticular cells and hyphae with irregular projections, extend into the

dark, stained area. Here, these elements are agglutinated by means of a dark-brown,

lacquer-like substance into a hard, brittle crust, up to 120u thick, over the ends of

the fibre hyphae (Fig. 43 c).

In the lower context, long, unbranched, yellow-brown, fibre hyphae are

arranged more or less parallel to the direction of growth but some turn downwards

into the trama of the dissepiments. Other unbranched, yellow-brown, fibre hyphae

with prominent lumina, mostly unbranched, but very tortuous, are tightly inter-

woven with these straight fibre hyphae and with numerous, hyaline, short, much-

branched, thick-walled or solid, fibre hyphae and branching, thin-walled, nodose-

septate hyphae, to form the dense, tough tissues of the trama. In this tissue,

thin-walled, nodose-septate hyphae ramify among the thick-walled hyphal elements,

branching frequently towards the surfaces of the pores where they bear the basidia

on numerous, short branches.

Decay and hosts

This species causes a diffused white rot of dead branches of hardwood trees.

Specimens examined

Herb. PRE: 11521, on decayed wood, Kentani, C.P., May 1918; 11545, on living branch,

Buccleugh, Natal, July 1918; 15541, on Albizzia gummifera, Stellenbosch, C.P., Sept. 1916;

17099, Wilderness, C.P., May 1923; 25477, Mount-aux-Sources, Natal, July 1928; 23689,

Margate, South Coast, Natal, Feb. 1931, 28259, on dead wood, Pretoria, Transvaal, March

1935; 28889, on dead branches, Drakensberg, Natal, Tuly 1937; 31549, on dead branches,

Ivy Range, Moodies, Natal, Aug. 1915; 31669, Ginginhluvu, Natal, May 1916; 31674,

Ginginhluvu, Natal, July 1915; 31702, dead stump, Stellabush, Durban, Natal, Oct. 1916;

31850, New Germany, Natal, April 1917; 31865, on dead wood, Bluff, Durban, Natal, May

1917; 31894, on dead wood, Mazoe, Rhodesia, May 1917; 31900, Stellabush, Durban, Natal,

July 1917; 31920, Bluff, Durban, Natal, Aug. 1917; 33066, on dead wood, Xumeni Forest,

Donnybrook, Natal, Dec. 1940; 33207, or dead wood, Rustenburg, Transvaal, May 1939;

35329, on Quercus sp., Pietermaritzburg, Natal, 1943; 36422, on dead wood, Chiradzulu,

Malawi, Sept. 1944; 36848; 36866, on dead wood, Vumba Mts., Umtali, Rhodesia, July 1948;

39112, on dry branch, Isipingo, Natal, Oct. 1950; 42067, Senanga, Barotseland, Aug. 1952;

*42159, on dead hardwood branch, Bushbuckridge, Tvl., Feb. 1961; *42161, on dead hardwood

branch, Bushbuckridge, Tvl., Feb. 1961; 43116, on Acacia karroo stump, Tongoland, Natal,

March 1965.

Herb. STE: 222, old rotting wood, Durban; 223, Rhodesia; 224, old log, Durban; 1044,

on wild Ficus sp., Kyrassa, E. Africa, July 1922; 1485, op droé hout, Houtbos, Transvaal,

Julie 1924; 2414, in thick forset, Umtali, Rhodesia, No. 1926.

Discussion

Although the cultural characters of Hexagona tenius have not been described

before, it is evident that, with its cultures which produce extra-cellular oxidase

enzymes, and form fibre hyphae, cuticular cells and clamp connections on its thin-

walled hyphae, this species fits well into Group 51. Indeed, it resembles a number

of other species also present in this group (Nobles, 1958 b, 1965) so that confusion

293

may arise when it is attempted to identify isolates from unknown decays; but when

the absence of secondary spores in cultures, the presence of fibre hyphae and the

possession of white, woolly to felty mat which becomes largely covered by a reddish-

brown crust, are considered together with the geographical origin of the specimen,

cultures of Hexagona tenuis may be recognized with a fair degree of certainty.

The carpophores of Hexagona tenuis were seen to consist of thin-walled,

nodose-septate hyphae, aseptate, branched and unbranched fibre hyphae or binding

hyphae. The cuticular cells present in some carpophores are modified terminal

portions of thin-walled, nodose-septate hyphae in which the septa and clamp

connections are often involved. Since hyphae with clamp connections are regarded

as generative hyphae by Corner (1953) and Cunningham (1946, 1954) only three

types of hyphae are present in carpophores of Hexagona tenius which thus have

trimitic hyphal systems, sensu Corner (1953).

The cuticular cells, which are the structural elements of the dark, reddish-

brown, crustose areas of some carpophores, were not always present in all

carpophores. They were often absent from some carpophores of a collection in

which crustose areas were present on others, an inconsistency also reported by

Van der Bijl (1922 a). No satisfactory reason for this sporadic appearance can be

given although it is possible that this may be influenced by the conditions under

which the carpophores develop. Because of this sporadic presence, however, the

value of cuticular cells as a useful diagnostic character in carpophores is reduced

considerably.

From the above description it is evident that the structures formed in cultures

of Hexagona tenuis are usually also present in the carpophores from which they

were made. It is noteworthy that the branched fibre hyphae of the cultures did not

resemble those of the carpophores but similar differences were also observed

in other species. Also, the cuticular cells formed in cultures were larger, had

thinner walls and appeared more regularly and extensively than in the carpophores,

but were undoubtedly homologous structures. Their more extensive development

in cultures can be asscribed only to the existence of more favourable conditions

for their development.

The anatomical characters of Hexagona tenuis have not been described before

and little is known about them in apparently closely related species. Lloyd (1910)

mentioned context colour but no hyphal characters in his Synopsis of the genus

Hexagona Fr. Van der Bijl (1922 a) stated that hyphae of Hexagona tenuis were

4u in diameter. Overholts (1953) described hyphae of Hexagona variegata, a species

which he considered to be closely related to Hexagona tenuis, as “pale brown in

KOH, long and flexuous, simple or nearly so, mostly with partly thickened walls,

with no cross-walls or clamps, 4 — 6u in diameter.” Pinto-Lopes (1952) reported

that secondary hyphae of Hexagona nitida Mont., are hyaline, nodose-septate and

the tertiary hyphae are brownish, thick-walled or solid and septate. Fidalgo &

Fidalgo (1962) reported that carpophores of Hexagona apiaria Pers. ex Fr. and

Hexagona hirta (Beauv. ex Fr.) Fr., have trimitic hyphal systems. The above

descriptions of the hyphal characters of Hexagona tenuis thus generally agree with

observations by other workers on related species.

The description also agrees with Fidalgo & Fidalgo’s (1962) report of the

hyphal characters and hyphal systems of Hexagona apiaria Pers. ex Fr., the type

species of the genus according to some authors (Cooke, 1959); but considerable

uncertainty exists about the identity of the type species of the genus Hexagona Fr.

This problem was discussed by Donk (1960) who concluded that Favolus hirtus

P. Beauv. should be regarded as the type species. Until this problem is solved

294

however, the affinities of Hexagona tenuis with the genus Hexagona Pollini per

Fr. cannot be determined with any degree of certainty.

Imazeki (1943), however, transferred Hexagona tenuis to the genus Daedaleop-

sis Schroet. with the remark that ‘ic has no affinity to Hexagona apiaria the type

of the genus Hexagona sensu stricto. This species is unique but it is safe for the

writer that it would be placed under Daedaleopsis at least, if we do not erect

a new genus for the species. The species connects with the genus Daedaleopsis

through D. conchifer or D. corrugata.”’ This is contradicted by the descriptions

given here and the report by Fidalgo & Fidalgo (1962) who described the trimitic

hyphal system in carpophores of Hexagona apiaria. In comparison with Daedalea

confragosa L. ex Fr., the type species of Daedaleopsis Schroet., Hexagona tenuis

differs from it mainly by the presence of binding hyphae throughout the context

tissues of its thin carpophores, its large, cylindrical basidiospores and the large

regular, shallow pores of its carpophcres. These differences are of a similar nature

to those between Trametes suaveolens and Lenzites betulina and because these are

considered to be distinct genera, it appears best, at this stage, to regard Daedalea

confragosa and Hexagona tenuis as species of separate genera although the two

species are similar in many respects.

Comparison of Hexagona tenuis, with Trametes corrugata, presumably

Imazeki’s (1943) “D. corrugata,” which is also included and described above in

this group, shows the latter to have construction of the carpophores and

binding hyphae which are unlike those of the fruit-bodies of Daedalea confragosa

and Hexagona tenuis. It thus seems to be extremely unlikely that Hexagona tenuis

can be related to Daedalea confragosa through Trametes corrugata.

More information, however, is required on the hyphal and anatumical characters

of more species of the genera Daedaleopsis Schroet. as well as Hexagona Fr. before

a satisfactory conclusion can be reached about the nature of their relationships.

The solution of the problem of the type species of the genus Hexagona Fr. will

be an important step in the determination of this relationship.

Note added in proof.

In later works K. Fidalgo advanced reasons for accepting the designation by

Clements and Shear of Hexagona crinigera Fr. as lectotype of the genus Hexagona

(Taxon 1968: 37-43) and excluded Hexagona tenuis Hooker ex Fries from this

genus (Mem. N.Y. Bot. Gard. 1968: 100).

Trametes acupunctata Berkeley, Jour. Lin. Soc. 13, 164, 1873;

Coltricia acupunctata (Berk.) G. H. Cunningham, Proc. Linn. Soc. N.S.W.

IS, ZNO; WSO.

Cultural characters

Growth is moderately rapid, the colonies reaching radii of about 115 mm in

one week and covering the plates in 3—4 weeks. ‘The advancing zone is even

or slightly bayed, mat appressed for short distance, then raised, becoming thin,

cottony and gradually passing into a thin, felty zone, or, bordering abruptly on a

clear zone of submerged mycelium around the inoculum. After 2 — 3 weeks

the mat becomes zonate with increasingly woolly texture towards the inoculum,

with crustose areas of “‘verona brown,” “natal brown’ or “‘bister’” developing

over the clear areas and along the side of the plate. With increasing age the myce-

lial mat thickens and the dark-coloured, crustose areas increase in size until at six

weeks the mat consists of thin, downy, white mycelium over the youngest parts,

295

eA eae: ali mieasle 4

Fic. 45.— Trametes acupunctata. (a) Upper surface and (b) lower surface of carpophore of

PRE 42440; (c) culture of PRE 42440 at six weeks; (d) cuticular cells from culture,

x 100; (e) hyphae with partly thickened walls and irregular projections from dark-

brown, crustose areas on upper surface of carpophore, squash mount in KOH, x 500.

often bordering abruptly on clear areas of submerged mycelium over which

crustose areas of “*Prout’s brown,” “‘cinnamon brown,” ‘“Verona brown” or “‘Natal

brown” are forming, or, passing over successive zones of increasingly woolly to

felty mycelium mostly fairly smooth or becoming pebbly towards the inoculum,

white or “pale ochraceous buff’’ or “pinkish buff’ in the older parts and bordering

abruptly on the irregular, sunken “‘hazel,” ‘ochraceous tawny,” “‘cinnamon brown,”

“sayal brown,” “‘Natal brown” or “‘Prout’s brown,”’ crustose areas which may cover

up to half the area of the mat. The reverse bleaches at first, then darkens to deep,

brown colours in which sharp, darker lines indicate the limits of the crustose

areas. No odour or faint, mushroomy odour is given off. Cultures give a weak

positive reaction when tested for extra-cellular oxidase enzymes with gum guaiacum

solution (Nobles, 1958 a). On gallic acid and tannic acid media no diffusion

zones are formed and no growth or a trace of growth occurs.

Advancing mycelium: hyphae hyaline, simple or branching, nodose-septate, thin-

walled and staining deeply in phloxine 2.5 — 4.5u in diameter (Fig. 46 a).

296

Aerial mycelium: (a) hyphae as in the advancing zone; (b) fibre hyphae hyaline

with thick, refractive walls and prominent, aseptate lumina, widening towards

the thin-walled ends, unbranched, or branching occasionally, 3.0 — 5.0u in

diameter (Fig. 46 b); (c) narrow hyphae repeatedly branched, the branches

tapering, walls hyaline partly thickened, the lumina prominent and _ aseptate,

0.5 — 1.2u in diameter (Fig. 46 c); (d) nodose-septate hyphae with numerous

short, stout branches or projections, thick, brown walls and narrow or partly

occluded lumina, or, blown out into thin-walled processes or cuticular cells,

3.0 — 20 in diameter (Fig. 45d, 46d).

Schmerged mycelium: hyphae as in the advancing zone.

Carpophore characters

Carpophore annual, lignicolous, sessile, dimidiate, occasionally imbricate, often

slightly concave, coriaceous, firm, rigid, almost woody, 3.5 — 20 x 2.0 — 9.0 x 0.2

— 2.0 cm; surface glabrous with fine radiating grooves and ridges traversed by

concentric grooves, ridges, and tubercles, fawn to olive brown, often with patches

of soft, dark cream-coloured or pale buff mycelium over the surface, or, dark

reddish-brown crustose areas over the older parts; margin undulate and sinuose,

concolourous with upper surface; pore surface dark olive-brown, poroid; pores

rounded to somewhat daedaloid, 2 — 4 mm; dissepiments even, thin, tubes short,

.02 — 0.5 mm deep becoming olivaceous brown inside; context olive-brown, fibrous,

zonate, darkening in KOH, 0.1 — 12.0 mm thick.

Hyphal characters. Carpophores consist of: (i) hyaline, branching, nodose-septate,

thin-walled hyphae with deeply staining contents, 2.0 — 3.0u in diameter (Fig.

46 e); (ii) fibre hyphae long, unbranched, straight or somewhat tortuous, yellowish-

brown, walls thickened, lumina narrow widening towards the thinner-walled ends,

aseptate, often with staining contents, 2.0 — 5.0u in diameter and arising from

thin-walled, nodose-septate hyphae (Fig. 46 f); (iii) dark, yellow-brown, thick-walled

hyphae, 3.0 — 4.0u in diameter with short, thick, lateral projections or inflated

portions up to 10u in diameter (Fig 45 e).

Hymenium: basidia hyaline, long clavate, 16.0 — 22.0 x 6.0u bearing four straight

sterigmata, 3.0u long (Fig. 46 g); basidiospores hyaline, long ellipsoidal to cylindrical,

smooth, thin-walled, 5.0 — 8.0 x 2.5 — 3.5u (Fig. 46 h).

Construction. At the margin, the carpophore consists of long, unbranched fibre

hyphae, more or less straight, with the walls thickened and lumina prominent,

aseptate, more or less parallel to one another and loosely intertwined. Intertwined

with them are the narrow, branching, thin-walled, nodose-septate hyphae with

deeply staining contents, from which they arise. In the upper context, behind the

margin, fibre hyphae with much thickened walls and narrow, aseptate lumina,

which constitute the bulk of the tissues, lie more or less parallel to one another

and slightly intertwined, bending upward towards the upper surface where their

ends are closely packed at a common level and become agglutinated by a thin layer

of transparent lacquer-like material into a glabrous, cuticular surface. Narrow,

branching, thin-walled, nodose-septate hyphae are intertwined with the fibre hyphae

in the upper context and below the surface cuticle. On some specimens dark-brown,

irregular, crustose areas may be present on the older parts of the upper surface.

These consist of dark, yellow-brown hyphae with walls partly thickened or solid

and with short, irregular, lateral projections and small, irregular swellings all

agglutinated by brown, resin-like material into a hard, brittle mass forming a layer

up to 50u thick in which individual elements are distinguished with difficulty

(Fig. 45 e). The lower context is similar to the upper but in the trama of the

297

FIGURE 46.

Fic. 46— Trametes acupunctata. a - d. Structures from cultures: (a) thin-walled,

nodose-septate hyphae from advancing zone; (b) fibre hyphae; (c) narrow,

repeatedly branched hyphae; (d) nodose-septate hyphae with irregular projec-

tions and cuticular cells.

e-h. Structures from carpophores: (e) thin-walled, nodose-septate hyphae:

(f) fibre hyphae; (g) basidia; (h) basidiospores.

298

tubes the fibre hyphae are narrower, more flexuous and with thinner walls and wider

lumina than in the context. These hyphae become tightly interwoven to form

a dense homogeneous tissue distinct from the context tissues. In the trama the

thin-walled, nodose-septate hyphae become very numerous, narrow and branch

frequently, the branches interwoven with and ramifying among the fibre hyphae,

finally emerging at the hymenial surfaces where the clavate basidia are borne

on short, terminal branches of these nodose-septate hyphae.

Decay and_ hosts

This fungus is fairly common on dead hardwood logs on which it causes

a pale brown rot slightly lighter in colour than the wood.

Specimens examined

Herb, PRE: 11283, on Acacia mollissima stump, Cramond, Natal, April 1911; 13938, on

Acacia mollissima, stump, Cramond, Natal, April 1911; 15546, on Acacia mollissima stump,

New Germany, Natal, April 1917; 15591, on dead Vepris lanccolatis, Buxton, C.P., Aug. 1916;

27554, on dead wood, Table Mountain, Natal, Oct. 1929; 30821, on indigenous wood, Port

St. Johns, C.P., Aug. 1937; 31555, on Gead wood, Branders Main Forest, Natal, Aug. 1915;

31573, on dead wood, Branders Main Forest, Natal, Aug. 1915; 31587, on dead wood,

Branders Main Forest, Natal, Aug. 1915; 31663, on dead wood, Ngoye, Natal, May 1916;

31733, on dead wood, Ngoye, Natal, May 1916; 31754, on dead wood, Durban, Natal, March

1917; 33208, on dead wood, Rustenburg, Tvl., May 1939; 34368, on dead wood, Hluhluwe,

Natal, Oct. 1935; 42034, on dead wood, Knysna, C.P., 1959; *42171, decayed hardwood log,

F. C. Erasmus Nature Res., Tvl., Feb. 1961; 42253, on dead log, Sabie, Tvl., Apr. 1962;

*42440, decayed hardwood log, near Bushbuckridge, Tvl., Feb. 1961.

Herb. STE: 158, on dead logs, Bluff, Durban; 200, on dead logs, Krantzkloof, Natal, Jan.

1921; 350, on dead logs, Krantzkloof, Natal, Jan. 1921.

Discussion

The presence of nodose-septate, thin-walled hyphae, fibre hyphae and

cuticular cells in its cultures, partly qualify Trametes acupunctata for inclusion in

Group 51; but the weakly positive reaction of its cultures when tested for extra-

cellular oxidase is in striking contrast to the strong positive reactions of other

species in this group. The negative oxidase reaction of this species on gallic

acid and tannic acid media together with the fact that this species causes a brown

rot suggests that its inclusion in the large number of species of which the cultures

do not produce extra-cellular oxidase may be justified. In that case it would

then constitute a new group beyond Group 25, in which cuticular cells, fibre

hyphae and nodose-septate hyphae are formed in cultures which lack extra-cellular

oxidase. A group with this combination of characters is not provided in Nobles’

keys, (Nobles, 1958 b, 1965); but because of the weak positive reaction for extra-

cellular oxidase when its cultures are tested with gum guaiac solution, Trametes

acupunctata must be included in Group 51 with which it also agrees in hyphal

characters.

In carpophores of Trametes acupunctata, only two types of hyphae are present,

viz.: thin-walled, hyaline nodose-septate hyphae and yellow-brown, thick-walled,

unbranched fibre hyphae. No branched fibre hyphae or binding hyphae were

found in any carpophore and the somewhat rigid, woody and fibrous texture of

the fruit-bodies is due to the presence of these tightly packed and intertwined

fibre hyphae. This species thus have carpophores with dimitic hyphal systems

sensu Corner, (1932 b, 1953) and Cunningham (1946, 1954). In this respect

fruit-bodies of Trametes acupunctata differ in hyphal characters and construction

from those of all other species of which these characters are known, in this group.

299

Also present in some of the carpophores are dark-brown, thick-walled hyphae

with the terminal parts inflated or distended into irregular projections which

constitute the dark-brown crustose areas. These structures appear to be the

counterparts in the fruit-bodies of the cuticular cells in the cultures. They occur

in the same relative position as similar structures in other species in this group

and apparently arise from thin-walled, nodose-septate hyphae in the upper surface

of the carpophores. They are aggtutinated into a very hard and very brittle

structure which could not be prepared satisfactorily for proper examination. In

view of their character and position in the fruit-bodies however, it seems extremely

likely that they are the ends of thin-walled, nodose-septate hyphae modified into

cuticular cells an dhyphae with irregular projections.

From the descriptions it is clear that most of the structures formed in cultures

are also present in the fruit-bodies frcm which they are made. Only the narrow,

hyaline, branched hyphae which are present in the cultures, over the agar, could

not be located in the carpophores. Similar hyphae had been noticed in cultures

of other species too e.g. Polyporus versicolor, but were absent from their carpo-

phores. The absence of these hyphae from the carpophores could be due to their

formation inside the wood on which the carpophores are formed, or to the existence

of conditions in cultures which allow their formation and the absence of these

conditions in growing carpophores. It was not possible to investigate either

alternative.

Cuticular cells formed in cultures with greater regularity than in the relevant

carpophores. The structures were more readily recognizable and could be traced

to their origins with ease in the cultures. It appears that conditions which favour

their formation occur more frequently in cultures than in carpophores. Their

presence in cultures may thus be useful as a diagnostic feature when cultures

from unknown decays have to be identified but their sporadic appearence on

carpophores in nature diminishes their value as a character of taxonomic im-

portance.

The hyphal characters and construction of the carpophores of Trametes

acupunctata differ strikingly from those of other species in this group of which

these characters are known. Much branched fibre hyphae or binding hyphae

(Corner, 1932 a) are absent from carpophores of Trametes acupunctata but are

present in those of Daedalea confragosa, Hexagona tenuis, Trametes corrugata,

and Fomes fomentarius (Teixeira, 1962 b). As the absence or presence of types

of hyphae in carpophores is considered to be of importance at the generic level

by a number of workers (Teixeira, 1962 b; Bondartzeva, 1961), Trametes acupunc-

tata cannot be considered to be corgeneric with any of these species.

Although originally placed in the genus Trametes Fr. by Berkeley (loc. cit.)

it is clear that Trametes acupunctata has little in common with Trametes suaveolens,

(L. ex Fr.) Fr. the type of this genus. It differs from Trametes suaveolens by

having a brown context of simple construction, by the presence of cuticular cells

in its cultures and carpophores, by causing a brown rot. and in the weak production

or absence of extra-cellular oxidase enzymes in its cultures.

Cunningham (1950 b) transferred Trametes acupunctata to the genus Coltricia

S. F. Gray which he characterized as having pileate fruit-bodies with a ““monomitic

hyphal system, hyphae long, ribbon-like, branched and septate. without clamp

connections.” Since this description does not fit the hyphal characters and

construction of the fruit-bodies of Trametes acupunctata as described above, this

species cannot be assigned to the genus Coltricia.

Trametes acupunctata thus differs in respect of cultural characters, carpophore

characters and type of decay from the type species of genera to which it had

300

been assigned. It is not well placed in Group 51 either, because of the inconsistent

oxidase reactions of its cultures while its carpophores differ in construction from

those of other species in this group of which these characters are known. The

combination of characters found in its cultures and carpophores are not known to

exist in any other species at present, largely because of the limited knowledge of

hyphal characters and fruit-body construction of poroid Hymenomycetes. Its

taxonomic position is thus uncertain but a more suitable position cannot be

suggested. Description of a new genus based on this species may thus be justified

but because future studies may reveal a genus to which Trametes acupunctata may

be satisfactorily assigned it is proposed not to transfer it to a new genus which

may well become an addition to an already long list of generic synonyms.

Resumé

From these descriptions it is evident that the four species of Group 51 included

in this study possess the cultural characters which justify their inclusion in this

group. With the exception of Trametes acupunctata, they share a number of

correlated characters, viz.: the production of extra-cellular oxidase, association

with white rots, the presence of nodose-septate hyphae, fibre hyphae and cuticular

cells. It appears that these species share a common ancestry but show diversity

in the elements of their carpophores. Its association with a brown rot, uncertain

extra-cellular oxidase production in culture and absence of binding hyphae from

its carpophores, suggest that the characters which Trametes acupunctata has in

common with the other three species, may have developed as a result of convergent

evolution.

5.9 Group 53

Cultures of species in this group have white mycelial mats covered by extensive,

wrinkled, brown, pseudo-parenchymatous areas. Extra-cellular oxidase enzymes

are produced. The thin-walled hyphae have simple clamps at the septa and may

remain thin-walled or give rise to thick-walled, brown hyphae with interlocking

projections. Thick-walled, aseptate fibre hyphae are also formed. Their basidio-

spores are large, cylindrical or ellipsoid-cylindrical and the interfertility for species

of which this character is known, is the tetrapolar type. Carpophores of these

species are alike in being stipitate.

Polyporus sacer Afz. ex Fries, Epicr., 436, 1836.

Cultural characters

Growth is moderately rapid the colony reaching a diameter of 15 mm after

1 week and covering the plate in 3 to 4 weeks. Advancing zone even, hyphae

raised almost to the limit of growth Mat at newest growth white, cottony to

woolly, thin, towards the inoculum at first appressed and becoming woolly-felty

with faint, radiating grooves or woolly-felty streaks, then suddenly pale “cream

color’ with slightly uneven, lacunose surface and fine droplets of colourless liquid

on it, around the inoculum. At three weeks the margin straightens as growth

proceeds more rapidly adjacent to the sides of the dish. Mat becomes more

appressed to sub-felty with pellicular areas developing at concentric grooves of

previous week’s growth and coalescing into pellicular areas which soon become

301

covered by raised, crustose areas. Mat remains white but crustose areas are at

first “‘cinnamon buff’ and bordering abruptly on the white mat, later darkening to

“cinnamon.” Or, zones of clear, submerged mycelium develop after 2-3 weeks

over which crustose areas of “ochraceous tawny,” raised mycelium, smooth at first

but later wrinkled, and becoming “clay colour,’ soon form. At six weeks the

mat is usually white or pale “cream color,’ downy or pellicular in the younger

parts with smooth or somewhat wrinkled crustose areas in a wide zone around the

inoculum and with scattered, raised, crustose patches over the older parts of the

mat. Colours on these range from ‘“‘ochraceous tawny” to “‘clay colour’’ or “‘light

pinkish cinnamon” to “‘cinnamon.” The reverse is bleached after 3 weeks but

dark brownish colours gradually develop in the agar. A pleasant, fragrant odour

is given off till about the fourth week but then gradually diminishes. A weak

positive reaction is obtained when the culture is tested for extra-cellular oxidase

enzymes. No growth takes place on gallic acid and tannic acid media but small

diffusion zones are formed on both media within one week.

Advancing mycelium: hyphae hyaline, thin-walled branched or unbranched with

simple clamp connections at the septa, 2.0 — 4.5u in diameter (Fig. 48 a).

Aerial mycelium: (a) hyphae as in the advancing zone; (b) fibre hyphae unbranched

or branching, long, more or less straight, hyaline, the walls thickened, lumina

narrow or occluded for most of their length, widening at the extremities, aseptate,

1.0 — 3.0u in diameter (Fig. 48 b); (c) narrow fibre hyphae, hyaline with numerous

short tapering branches and mostly solid, 0.5 — 1.0u in diameter (Fig. 48 c);

(d) nodose-septate hyphae with interlocking projections and short, thick, lateral

branches, sub-hyaline to pale yellowish brown, the walls thickened and lumina

narrow or occluded, 3.0 — 8.0u in diameter with projections up to 10u long

(Fig. 48 d).

Submerged mycelium: hyphae as in the advancing zone but more tortuous and

wider, 2.5 — 6.0 in diameter.

Carpophore characters

Carpophore annual, terrestrial, solitary rarely grouped; pileus orbicular, tough

coriaceous to woody, velutinate to glabrous, rugulose, slightly furrowed, concentric-

ally sulcate or zoned, ‘“‘tawny olive,” ‘“‘snuff brown” to “Verona brown,” “‘Prout’s

brown” or ‘‘bister’ and ‘‘dark olive’ in concentric zones, centrally stipitate,

6.0 — 10.0 cm in diameter, 0.15 — 0.3 cm thick; margin acute, thin, entire, white

or concolorous; pore surface white drying to “pale ochraceous buff”, “warm buff”

or “clay color,” poroid; pores daedaloid, rounded or angular, 2/mm; dissepiments

thin, even in younger parts but in older parts, radially raised; tubes concolorous

up to 2.5 mm deep; context tough, fibrous, white to pale “cream color,” 0.75 — 1.5

mm thick: stipe erect, tapering apically mostly slender, smooth, velutinate, “tilleul

buff” to ‘wood brown’’ 0.3 — 1.4 x 4.5 — 15 cm, subtubular to tubular, context

white, arising from basal sclerotium, sclerotium ovoid to irregular, rugose to rimose,

horny, hard, concolorous with stipe, 1.5 — 5.0 x 2.5 — 7.5 cm; context white

or ‘“‘pale tilleul buff,” firm, woody.

Hyphal characters. Carpophores consist of: (i) hyaline, thin-walled, nodose-septate

hyphae, branching at the septa, 1.8 — 3.5u in diameter (Fig. 48 e); (ii) fibre

hyphae long unbranched or with an occasional long branch, more or less straight

or tortuous, widest near middle, the walls hyaline, thick, refractive, lumina narrow

or occluded, aseptate, 1.5 — 6.0u ir. diameter (Fig. 48 f); (iti) fibre hyphae with

many lateral branches, walls hyaline, thickened, lumina narrow or occluded,

aseptate, 1.5 — 3.5u in diameter (Fig. 48 g); (iv) nodose-septate hyphae with

Fic. 47.—Polyporus sacer. (a) Carpophore of PRE 31545 showing upper suface and

(b) hymenial surface; (c) dark-coloured, thick-walled, nodose-septate hyphae forming

hairs on upper surface of carpophore, x 500; (d) cuticular cells and “hairs” from

upper surface, * 1000; (e) culture of PRE 42163 at six weeks.

FIGURE 48.

Fic. 48.—Polyporus sacer. a - d. Structures from cultures: (a) thin-walled, nodose-

septate hyphae from advancing zone; (b) fibre hyphae; (c) narrow hyphae with

numerous short, tapering branches; (d) nodose-septate hyphae with interlocking

projections.

e-m. Structures from carpophores: (e) thin-walled, nodose-septate hyphae:

(f) unbranched fibre hyphae; (g) fibre hyphae with numerous short, lateral

branches; (k) basidia; (m) basidiospores.

304

thick brown walls and narrow or partly occluded lumina and solid or sub-solid

clamp connections, 4.5 — 10.0u in diameter (Fig. 47 c); (v) nodose-septate hyphae

with thick hyaline walls distended into interlocking projections, 4.0 — 8.5u in

diameter (Fig. 47 c, 48 h).

Hymenium: basidia hyaline long clavate 14.0 — 18.0 x 5.0 — 17.0u bearing four

sterigmata 2.0 — 3.0u long (Fig. 48 k); basidiospores hyaline, ovoid to ellipsoid,

flattened on one side, smooth, thin-walled, 5.0 — 7.0 x 3.0 — 4.5u (Fig. 48 m);

hyphal pegs numerous, white, long-conical up to 200u long.

Construction: At the margin the pileus consists of fibre hyphae, hyaline and

straight or flexuous, unbranched and with narrow or occluded lumina up to 6.0u

in diameter, intertwined with one another and interwoven with numerous hyaline,

branching, thin-walled, nodose-septate hyphae 1.8 — 3.5u in diameter, from

which they arise. At the upper part of the margin, numerous nodose-septate

hyphae are present, their walls distended into irregular, inter-locking projections

which thicken and turn brownish simultaneously. A very short distance

from the margin these hyphae are agglutinated into a hard, brittle, brown cuticle

30 — 50u thick over the upper surface. The uppermost layers of this cuticle

consists of nodose-septate hyphae with thick, brown walls lying parallel to the

direction of growth of the pileus and agglutinated by a brownish, lacquer-like

substance onto the surface to form glabrous zones (Fig. 47 c). Or, free, nodose-

septate hyphae with thick, brown walls project upward from the cuticle to form

the velutinate zones (Fig. 47 c). Below the cuticle and behind the margin the

context consists mainly of intertwining hyaline, unbranched, straight or flexuous,

fibre hyphae, solid or sub-solid with a thin layer of numerous, thin-walled, nodose-

septate hyphae below the cuticle, freely intertwined with the ends of the fibre

hyphae, some of which are agglutinated into the cuticle. Interwoven with the long

fibre hyphae are fibre hyphae with short, sub-solid or solid, lateral branches which

bind them together into the tough tissues of the pileus.

In the lower context and trama of the tubes the tissues are more dense and

consist of long, fibre hyphae, more tortuous, generally narrower and more fre-

quently branched than in the upper context, the branches long and tapering, and

tightly interwoven with fibre hyphae with many short branches binding them to-

gether into a dense, even, tough tissue. Also interwoven with these hyphae are thin-

walled, nodose-septate hyphae with deeply staining contents, branching repeatedly

and becoming increasingly numerous towards the hymenial surfaces where they

bear the basidia on numerous short branches. From the tissues of the dissepiments,

hyphal pegs, each consisting of a bundle of parallel ends of fibre hyphae, project

into the pore space for up to 200u.

The tubular stipe consists mainly of long, unbranched, sub-solid to solid,

hyaline fibre hyphae 3.0 — 6.0u in diameter at the widest middle part, intertwined

and parallel to the length of the stipe. Interwoven with them are narrower,

hyaline fibre hyphae with numerous short tortuous branches with thick walls and

prominent lumina, 1.0 — 1.5u in diameter. Thin-walled, nodose-septate hyphae

are interwoven with the fibre hyphae and become more numerous towards the

outer and inner surfaces where they give rise to plectenchymatous layers of thick-

walled cells 40 — 90 in thickness on the inner and outer surfaces.

The sclerotium consists mainly of interwoven branched and unbranched

hyaline fibre hyphae apparently without directional orientation and thin-walled

nodose-septate hyphae in a homogeneous context and covered by a hard rind

9) — 180u thick, apparently consisting of thick-walled, nodose-septate hyphae.

305

Decay and hosts

This fungus does not cause decay of timber but grows in humus rich soil

in damp sub-tropical areas.

Specimens examined

Herb. PRE: 9112, Elandshoek, Tvl., Aug. 1915; 11519, Kentani Distr., C.P., May 1918;

31545; 36590, in soil in forest, Njala, Sierra Leone, Feb. 1947; *42163, on ground, F. C.

Erasmus Nature Reserve, Tvl., Feb. 196i

Discussion

The cultural characters of Polyporus sacer have not been described before

but there can be no doubt that the presence of thin-walled, nodose-septate hyphae,

fibre hyphae and hyphae with interlocking projections in cultures which produce

extra-cellular oxidase enzymes, places this species in Group 53. This group includes

11 other species of stipitate polypores, all of which display very similar characters

in culture (Nobles, 1958 b). In cultures of these species ‘‘the dark-brown, wrinkled,

pseudoparenchymatous areas contrast sharply with the white cottony or woolly

parts of the mats’? (Nobles, 1958 b). Cultures of Polyporus sacer are lighter in

colour over the crustose areas tending towards brownish yellow and the aerial

mycelium is sub-felty rather than woolly. These characters serve to distinguish

cultures of Polyporus sacer from those of other species in this group.

The carpophores of Polyporus sacer were shown to consist of five types of

hyphae. Of these, the two types of hyphae with thickened walls have clamps at

their septa and must be regarded as generative hyphae, sensu Corner (1932 a)

and Cunningham (1946) and of the same type as the thin-walled nodose-septate

hyphae. The unbranched and branched hyphae correspond to the skeletal and

binding hyphae respectively so that the fruit-body of Polyporus sacer has a trimitic

hyphal system sensu Corner (1932 b, 1953).

In this species it is again evident that the structures formed in its cultures

are also present in the fruit-bodies from which they were made. There are no

differences between the thin-walled, nodose-septate hyphae and fibre hyphae from

the cultures and fruit-bodies except that fibre hyphae from cultures are somewhat

less flexuous than those from the fruit-bodies; but the nodose-septate hyphae

with thickened wails which form the “hairs” of the upper surface of the fruit-bodies,

were not found in the cultures. In the fruit-bodies, these hyphae are closely

associated with the nodose-septate hyphae with inter-locking projections. Both

types arise from thin-walled, nodose-septate hyphae; but nodose-septate hyphae

with interlocking projections did develop in the cultures. It thus appears that

these hyphae may be variations of the same modification of nodose-septate hyphae

but which develop under different conditions of growth. Conditions favourable

for their development probably did not exist in the cultures.

The hyphal characters of Polyporus sacer have not been described before but

Furtado (1965 a) stated that ‘the hyphal system and generai habit of the species

suggest that Polyporus sacer may belong to the genus Amauroderma, but basidio-

spores were not seen.” Furtado (1965 a) further stated that “‘the arboriform

skeletal hyphae (Teixeira 1956, 1962 a, b) are commonly found in the ganodermoid

polypores and it seems probable that they are characteristic of the sub-family

Ganodermoideae.”’ He included the two genera Ganoderma Karsten and Amauro-

derma Murrill in this sub-family and distinguished between them on the basis of the

shape of the thick-walled echinulate spores; but the specimens of Polyporus sacer

examined for this study have thin-v.alled, short-cylindrical spores and arboriform

hyphae were not found in their fruit-bodies. This species thus does not have the

characters of the genus Amauroderma Murr. as described by Furtado (1965 a) and

can therefore not be included in that genus.

306

Of the eleven species of stipitate polypores included by Nobles (1958 b) in

her Group 53, the hyphal characters of only one species, Polyporus squamosus

Fr., is known in detail from the description by Corner (1953). In this description,

Corner stated that the young fruit-body of Polyporus squamosus has a monomitic

hyphal system consisting of clamped, generative hyphae only. After a certain

stage of maturity is reached, the generative hyphae develop 2 — 4 lateral, branching

processes, which Corner designated as ‘binding hyphae.” By growing laterally

between the generative hyphae and developing thickened walls, these hyphae bind

the generative hyphae into the tough, dry tissues of the mature fruit-body which

has a dimitic hyphal system consisting of generative and binding hyphae. The

hyphal characters and construction of the fruit-body of Polyporus squamosus are

thus completely different from those of Polyporus sacer as described here.

The pileus of another stipitate species, Polyporus arcularius, was found to

consist of nodose-septate hyphae with short inflated cells from which fibre-like,

thick-walled processes arise and generally resemble those of the mature fruit-bodies

of Polyporus squamosus as described by Corner (Dr. D. D. McLain, personal

communication and demonstration).

Overholts (1953), in his brief descriptions of the hyphae of the fruit-bodies

of the stipitate polypores included by Nobles (1958 b) in Group 53, mentioned

the attenuated, whip-like ends of the branches of the hyphae in the pilei of

Polyporus arcularius, Polyporus brumalis, Polyporus squamosus and Polyporus

tuberaster. In the fruit-bodies of Polyporus elegans, Polyporus melanopus and

Polyporus varius the hyphae are thick-walled and much branched, while those

of Polyporus fagicola and Polyporus radicatus are thin-walled and tend to collapse.

The hyphal characters of these stipitate species thus also differ among species

of this group and all of them differ from the hyphal characters of Polyporus sacer.

This species can therefore not be regarded as congeneric with any of the species

included by Nobles (1958 b) in Group 53 of which two had been designated as

generic types. No other species of stipitate polypore of which the hyphal characters,

construction and general morphology of the fruit-bodies resemble those of Polyporus

sacer, is known at present.

Donk (1960) showed that Fries: indicated some affinity between Polyporus

sacer and Polyporus versicolor by placing each of these species as the first species

in two of his nine stirpes of his genus Polystictus. Donk (1962) stated later that

Fries had conceived the taxon already before he decided to treat it as the separate

genus Polystictus, and that Fries’ remarks tend to show that “‘the genus Polystictus

started with the conception of a stirpes typified by Polyporus perennis L. per Fr., in

the first place, and a stirpes typified by P. sacer Afz. ex Fr.” This close affinity

with Polyporus perennis is not evident when the hyphal characters of these two

species are compared. Cunningham (1948 e) described the hyphal characters of

Polyporus perennis as “hyphal system monomitic, hyphae long, ribbon-like,

branched and septate without clamp connections.” ‘This observation was confirmed

by Overholts (1953). Because Polyporus sacer has a trimitic hyphal system in

Corner’s (1932 a, b) and Cunningham’s (1946, 1954) terminology, the two species

cannot be regarded as closely related at all.

The affinity between Polyporus versicolor and Polyporus sacer first indicated

by Fries (in Donk, 1960) received additional support when Nobles (1965) included

Polyporus versicolor together with three species of stipitate polypores as well

as Daedalea confragosa and Fomes scutellatus in her Key Code 2.3.8.11 on the

basis of their cultural characters. This group includes species of which the cultures

produce extra-cellular oxidase and the thin-walled, nodose-septate hyphae are

differentiated to form fibre hyphae and hyphae with interlocking projections.

307

Because cultures of Polyporus sacer agree with these characters, this species may

also be included in Key Code 2.3.8.11 thus revealing similarities with cultures of

Polyporus versicolor. The simiiarities in hyphal characters and construction

of the pilei of Polyporus sacer and Polyporus versicolor are even more striking

and suggest a much closer relationship between them than between Polyporus

sacer and the other species of stipitate polypores discussed above. An important

difference between them, however, exists in the nature of the construction of

the upper surface of their pilei. While the upper surface of both species may

be described by the term “‘trichoderm” (Lohwag, 1940; Furtado, 1965 a) the “‘hairs”’

of the trichoderm of Polyporus versicolor consist of the ends of fibre hyphae

(skeletal hyphae) which project from a dense layer of agglutinated hyphae (Fig.

47 c), whilst the “hairs” of the trichoderm of Polyporus sacer are thick-walled,

nodose-septate hyphae which arise from the upper parts of a layer of agglutinated,

thick-walled, nodose-septate hyphae and fibre hyphae. Furthermore, the binding

hyphae in the pileus of Polyporus sacer have fairly long flexuous, tapering branches

while those of Polyporus versicolor kas short, tortuous branches. These differences

in the nature of the trichoderm and the character of the binding hyphae together

with the presence of an orbicular pileus borne on a well differentiated stipe arising

from a hypogeous sclerotium, must separate Polyporus sacer from Polyporus

versicolor at the generic level.

Differences between Polyporus sacer and Daedalea confragosa are of a similar

nature to those between Polyporus sacer and Polyporus versicolor. The cultural

characters and fruit-bodies of Fomes scutellatus could not be included in the present

study.

The fruit-bodies of Polyporus sacer possess a combination of hyphal and

morphological characters that have not been found in any other species till now.

Its relationships and systematic position cannot be determined at present

but future studies of more species of poroid Hymenomycetes may confirm Fries’

idea of a taxon typified by Polyporus sacer.

6. DISCUSSION

The object of this study was to determine to what extent the structures formed

in cultures of poroid Hymenomyecetes are also present in their carpophores in order

to determine whether their carpophores reveal the same relationships as their

cultures. From the descriptions of cultural and carpophore characters of the species

studied, it is evident that the structures formed in cultures are mostly present in

the carpophores but that certain exceptions and discrepancies were observed. These

concerned the different types of hyphae and hyphal modifications.

In all the species studied, nodose-septate hyphae were found to be present in

both the cultures and carpophores of all specimens examined, but in some species

the nodose-septate hyphae became thick-walled. In Polyporus dichrous, Polyporus

adustus and Polyporus subiculoides these thick-walled, nodose-septate hyphae

make up the bulk of the carpophore tissues although they are rarely found in the

cultures. In other species, e.g. Polyporus versicolor and Lenzites sepiaria, thick-

walled as well as thin-walled, nodose-septate hyphae were present in both the

cultures and carpophores. In still other species, e.g. Trametes cingulata and

Polyporus occidentalis, thick-walled, nodose-septate hyphae were never seen in

the cultures or carpophores. It therefore appears that in some species the nodose-

septate hyphae may be modified by thickening of their walls, under conditions

308

prevailing in the formation of fruit-bodies; but hyphae modified in this way were

not present in all the species in which thin-walled, nodose-septate hyphae were

present in both cultures and carpophores. It thus appears that the modification

or differentiation of thin-walled, nodose-septate hyphae into thick-walled, nodose-

septate hyphae or “‘sclerified generative hyphae’ (Donk, 1964) can occur in

certain species only. This character must therefore be recognized and _ these

hyphae must be regarded as a distinct type of hypha. Their presence or absence

in carpophores must therefore be taken into consideration in studies involving

hyphal characters of fruit-bodies.

In the species studied in Group 25, another type of nodose-septate hypha of

which the walls are irregularly thickened, were shown to be present in both the

carpophores and cultures. These species were also shown to have other carpophore

characters in common which would allow their inclusion in one genus. Nobles

(1958 b) showed that a number of species, in which this type of hypha is present

in their cultures, also have other hyphal and basidiospore characters in common

which allow their inclusion in Group 25. These nodose-septate hyphae with

irregularly thickened walls are present in certain species only so that it is clear

that this character must be genetically constant. For these reasons, nodose-septate

hyphae with irregularly thickened walls must be regarded as constituting a

morphologically distinct type of hypha and should be recognized as such in

studies involving hyphal characters of fruit-bodies. It is well-known that these

hyphae are often found with difficulty in fruit-bodies of species in which they

are present in cultures but they cannot be ignored for this reason. Careful

search in parts where fibre hyphae are not numerous, will usually reveal their

presence.

In those species in which fibre hyphae were present in cultures, fibre hyphae

also occurred in their carpophores but differences in morphology of the fibre

hyphae were noticed in some cases. In general, these differences were observed

mainly in the extent of branching of the fibre hyphae and in their diameter. In

Daedalea confragosa it was seen that the fibre hyphae in cultures were of one

kind only, viz. narrow and branched with long branches while the fibre hyphae

in the carpophores were either unbranched or had numerous short, twisted

branches. In most species included in Group 45, with the exception of Polyporus

vinosus, the fibre hyphae of the carpophores consist of unbranched skeletal hyphae

and of binding hyphae with numercus, short, tortuous branches (sensu Corner,

1932 a, b); but the fibre hyphae of their cultures are mainly unbranched whilst

some fibre hyphae have a number of fairly long branches, often fairly straight.

In other species, mainly those of Group 25, where short-branched binding hyphae

were not present in the carpophores, the fibre hyphae of the cultures were also

mostly unbranched or had an occasional long branch. It can be concluded,

however, that the fibre hyphae of the carpophores of all the species included

in this study, agree with those of the fibre hyphae of the carpophores.

In cultures of Polyporus versicolor, Polyporus pubescens, Trametes acupunc-

tata and Trametes corrugata a network of very narrow, dichotomously branched

hyphae were observed in very tough parts of the mycelial mat. These hyphae,

which were less than 1.0u in diameter, could not be traced to their origin and

were just barely discernible under the oil immersion lens. Such hyphae had been

reported by Nobles (1965) in cultures of Polyporus versicolor and Polyporus

pubescens but their presence in carpophores had not been reported by other

workers. In this study they have been found only in the tissues of carpophores

of two collections of Trametes corrugata including the carpophores of the type

specimen of Earliella. Since these hyphae are so narrow and inconspicuous, their

nature could not be determined and because the tissues are torn apart with needles,

309

to dissect out the different types of hyphae, they are subject to destruction because

of their dichotomous branching habit. Small pieces may easily be overlooked

as debris which is often present in the mounts; but their presence in the two

carpophores indicate that such hyphae may also be present in carpophores of

species in which they are formed in culture. The very tough nature of those parts

of the mat in which these hyphae are present, suggests that these hyphae may

serve as binding hyphae in the tissues.

In the five species studied in Group 51 and Group 53, nodose-septate hyphae

of which the terminal portions were differentiated into cuticular cells or irregular

projections, were present in the cultures. The cuticular cells were usually well-

developed, mostly with thin walls and present in large numbers. In the carpophores,

however, these structures were either not easily seen or were absent. Of the large

number of carpophores of Daedalea confragosa that were examined, cuticular

cells could be found in two only and they were smaller than those of the cultures.

In the cultures of Hexagona tenuis and Trametes acupunctata the cuticular cells

were well-developed but in their carpophores the corresponding structures did not

resemble cuticular cells. Instead they resembled the ‘hyphae with irregular

thick-walled branches, nodules or protuberances’? described by Nobles (1965)

under Code Symbol 11. Similar hyphae were also present in the cultures of

Daedalea_ confragosa, Hexagona tenuis, Trametes acupunctata and Trametes

corrugata as well as in their carpophores where they were present as the only

specialized cuticular structures. Therefore, it seems that ‘‘cuticular cells’ and

“hyphae with irregular thick-walled branches, nodules or protuberences” are

different manifestations of the same hyphal modification which develop under

different conditions of growth. These structures, which form dark-coloured

patches over the older part of the upper surface of the carpophores, occurred

sporadically on the carpophores. They were often absent from some carpophores

but present on others in the same collection. Because they are formed more

frequently and regularly in cultures than in carpophores, it appears that their

formation is influenced by the conditions of growth of the relevant mycelia. Their

more regular presence in cultures of various species is thus of greater value in

the recognition of cultures than in the identification and classification of the

carpophores of these species.

Fructifications which produce fertile basidia and basidiospores were formed

by a number of species in cultures. In every case the basidia and basidiospores

were identical in respect of dimensions and morphology to those of the carpophores

found in nature. This confirms the statements by Teixeira (1962 b) and Kotlaba

(1964) that the characters of the basidia and basidiospores are fixed and constant

for each species and emphasizes the great taxonomic importance accorded to

these structures by all workers.

An interesting aspect of the formation of fructifications in culture is the fact

that basidia and spores may be borne on structures which bear no resemblance

to the corresponding fruit-bodies formed in nature. Further, the fructifications

formed in cultures were seen to develop in distinctly different ways. In Daedalea

spp. the fertile areas consisted of irregular, low, anastomosing ridges which grew

out from the areas of compact mycelium. In Fomes cajanderi the fruiting areas

developed as gradually deepening tubes in areas of felty mycelia. In Lenzites

trabea flat, antler-like processes which united laterally to form large tubes, grew

out of the mat. In Polyporus versicolor, Polyporus occidentalis, Trametes meyenii

and some other trametoid species in Group 45, thin, acicular spines developed from

felty patches and gradually widened into flat processes which united laterally to

form tubes. These different ways of formation of fertile spore-bearing tubes,

310

occurred in cultures which differed in cultural characters. These observations

thus indicate that these different ways of formation of spore-bearing tubes may

represent phylogenetic differences between the species concerned. This conclusion

is supported by the fact that differences in hyphal and anatomical characters

were shown to exist between the carpophores of the. various species in which

fructifications formed in cultures. It is further supported by the fact that differences

in the method of pore formation are known to occur in carpophores of different

species under natural conditions (Ames, 1913; Corner, 1953). Savile (1955),

suggested that ontogenic studies may throw light on the origins of tubes of various

types of Hymenomycetes; but careful observations on fruit-body formation in

culture and in nature of a large number of species will have to be made before

practical use can be made of such observations in the taxonomy and phylogeny

of these fungi.

From the above it is thus evident that the structures formed in cultures of

Hymenomycetes are usually also present in the fruit-bodies from which they were

made, although some exceptions to this general rule were encountered and certain

structures were not quite identical in the cultures and in carpophores. This

conclusion agrees with the statement by Pinto-Lopes (1952) and the results of

work of Sarkar (1959), Davidson, Lentz & McKay (1960), McKay & Lentz (1960),

Weresub & Gibson (1960), Nobles & Frew (1962) and Lombard & Gilbertson

(1965, 1966). Consequently, the carpophores of the different species studied here

can also be assigned to the same groups as the cultures made from them. The

relationships between the carpophores of the different species as indicated by their

cultures must be examined now.

In all those groups in which more than one species was studied, it was found

that although the carpophores displayed the characters which allow their inclusion

in the group, differences in the morphological characters of the hyphae and

construction of the carpophores were present between the individual species or

between smaller groups of species within the group. So it was shown that

carpophores of Lenzites trabea with dimitic fruit-bodies (sensu Corner, 1932 a)

differ from those of Lenzites sepiaria with trimitic fruit-bodies (sensu Corner)

although both are included in Group 13. Of the four species studied in Group

25, the carpophores of Daedalea quercina and Trametes moesta are identical in

hyphal characters but differ only in small morphological characters. The other

two species, Trametes roseola and Fomes cajanderi, reveal similar micromorpho-

logical characters but differ from the two Daedalea spp. in respect of carpophore

colour, texture and the presence of poroid hymenia. Despite these differences,

which appear to be of interspecific importance only, it was suggested that these

two species should be included in the genus Daedalea Fr. In Group 45, Polyporus

vinosus differs from the Coriolus — Lenzites — Tranietes spp. by the absence

of binding hyphae from its carpophores. Among the other species in Group 45

a smaller group in which “‘sclerified generative hyphae” (Donk, 1964) are not

present and another group in which “‘arboriform skeletal hyphae” (Teixeira,

1962 b) are present in the carpophores, could be distinguished. The four species in

Group 51 have in common the presence of nodose-septate hyphae, fibre hyphae

and cuticular cells in their cultures and their carpophores. They were found

to differ widely in respect of carpcphore construction, carpophore morphology

and hyphal characters. Such differences are held to be sufficiently important to

regard the relevant fungi as species of separate and distinct genera. Inclusion

of some species in certain groups thus appears to depend on the presence of

common characters which arose though convergent evolution (Savile, 1954, 1955)

whilst other species may be grouped together because they are related through

many common characters in respect of the morphology of their hyphae and other

311

microstructures and construction of their carpophores. These observations thus

support Nobles’ (1958 b) suggestion that her groups may constitute taxa of generic

or higher rank but that some groups may be entirely artificial.

The characters that should be taken into account when considering affinities

at generic level in the polypores have not been clearly enumerated as yet and the

problem of delimitation of genera of the Hymenomycetes has developed as the

most important aspect of their taxonomy. A number of workers including

Corner (1948), Wakefield (1948), Pinto-Lopes (1952), Cunningham (1954) and

Teixeira (1962 b) regard spore characters, and micromorphology and anatomy

of carpophores as the most important indicators of generic affinities although

Teston (1953 a) and Smith (1966) are rather sceptical. The views of the first-

named workers are summarized by Kotlaba (1964) who stated that ‘‘the importance

of these characters lie in their particular combinations.”” He admitted that a

particular character may have different taxonomic values in different groups

and that no generalizations can be made. He also considered a complex of

characters to be the basis necessary for delimitation of genera. The problem

thus evolves as the need to determine the relative values of various characters

available for taxonomic purposes; but these relative values can be determined

only after careful observations on a very large number of species had been made.

Such observations had been made on a relatively small number of species only.

The number of species included in the present study is insufficient to allow

delimitation of genera but the observations made on them serve to emphasize

some aspects of carpophore anatomy and micromorphology of hyphae and other

structures, that should be taken into consideration in taxonomic studies of these

fungi.

With the introduction of the concept of hyphal systems, Corner (1932 a, b)

made available useful terminology to describe the construction of carpophores

of macrofungi. This concept had been applied to the study of various groups

of Hymenomycetes but without further extension or definition of hyphal types

or attempts at finer distinctions in hyphal morphology except by Teixeira (1956,

1962 b), who described different kinds of skeletal hyphae. No attempts had been

made to describe differences in carpophore construction in polypores, comparable

to the different types of texture of carpophore of resupinate Hymenomycetes as

defined by Talbot (1954 a), but the existence of similar differences is evident

from the above descriptions of the carpophores. The carpophores of Polyporus

dichrous, Polyporus adustus and Polyporus subiculoides consist entirely of nodose-

septate hyphae. These carpophoies thus have monomitic hyphal systems as

defined by Corner (1932 a, b), Cunningham (1946, 1954) and Teixeira (1962 b).

It is, however, evident from the descriptions that thin-walled, nodose-septate

hyphae and thick-walled, nodose-septate hyphae are present in different amounts

in carpophores of those species. It is further evident from the descriptions that

the thick-walled, nodose-septate hyphae occur in definite regions of the carpophores

and that they may be orientated in different directions in the tissues. These

differences result in differences in complexity of construction and of texture of

the carpophores. They are even more strikingly evident when the carpophores

of these three species are compared with carpophores of some species of Peniophora

and Corticium, with monomitic hyphal systems which consist of branched, thin-

walled, nodose-septate hyphae terminating in clusters of basidia (Slysh, 1960);

Cunningham, 1963; Talbot, 1951, 1954 a, 1958 b). It is thus clear that differences

in carpophore construction and hyphal characters can exist in carpophores with

monomitic hyphal systems and that these differences are neither recognized nor

conveyed by the expression ‘“‘monomitic hyphal system.” Characters of carpophore

construction, hyphal orientation and hyphal morphology observed in these species

are constant for each species and genetically fixed. They are therefore of

phylogenetic importance. Furthermore, it had been shown that the thick-walled,

nodose-septate hyphae, of which several types have been described, must be

regarded as morphologically distinct from thin-walled, nodose-septate hyphae.

Therefore, the presence or absence of thick-walled septate hyphae in carpophores,

the relative position of the different types of septate hyphae, and their orientation

in the carpophores must be considered in taxonomic studies and expressed in

suitably descriptive terms which can convey characteristic types of construction

of carpophores with “‘monomitic hyphal systems.” In this way generic affinities

may become more clearly apparent than is the case at present.

In species with carpophores with dimitic hyphal systems in Corner’s (1932

a, b) terminology, similar differences in construction and hyphal characters exist.

In carpopohres of Lenzites trabea, branched, thick-walled, nodose-septate hyphae

which seem to form a primitive and poorly developed binding hyphal system are

present besides the thin-walled, nodose-septate hyphae and fibre hyphae. In the

carpophores of Daedalea quercina, Daedalea moesta, Trametes roseola and Fomes

cajanderi, nodose-septate hyphae with irregularly thickened walls are present besides

the thin-walled, nodose-septate hyphae and fibre hyphae. These different types

of hyphal differentiation contrast strongly with that of the carpophores of Fomes

pinicola where only thin-walled nodose-septate hyphae and fibre hyphae are present.

They also differ from those of carpcphores of Trametes acupunctata where some

nodose-septate hyphae are differentiated into hyphae with irregular projections

and cuticular cells. In carpophores of Polyporus vinosus some of the thin-walled,

nodose-septate hyphae become thick-walled, turn brown and bind the fibre hyphae

into a dense tissue. Yet, despite the morphological differences found in the

nodose-septate hyphae they are regarded as generative hyphae by Corner (1932

a, b) and other workers and all these species are regarded as having dimitic

hyphal systems. It is thus evident that differences in hyphal characters and

carpophore construction of a similar nature to those found among species with

carpophores consisting of nodose-septate hyphae only, are also present among

species of which the carpophores ccnsist of nodose-septate hyphae and aseptate

fibre hyphae. These differences in carpophore construction and hyphal characters

were also found to be fixed and constant for the different species. They are thus

genetically constant and therefore of phylogenetic importance. Their presence

in carpophores must therefore be recognized and taken into consideration when

affinities at the generic level are being considered as had been done by Lentz

(1960) with Lopharia crassa and Lopharia cinerascens; but these differences in

carpophore construction and hyphal morphology of species whose carpophores

have “‘dimitic hyphal systems” are not apparent from this expression.

Corner (1932 a, b; 1953), Kotlaba & Pouzar (1957) and Teixeira (1962 b)

regard species of Polyporaceae having carpophores constructed of generative hyphae,

skeletal hyphae and binding hyphae as the most highly evolved group of species

in this family. All the species included in the present study in Group 45, Group

51 and Group 53 with the exception of Polyporus vinosus and Trametes acupunctata

respectively, have carpophores of this type with trimitic hyphal systems (sensu

Corner, 1932 a, b); but differences in hyphal morphology and carpophore

construction similar in many respects to those found in carpophores of species

with monomitic and dimitic hyphal systems, were also found to exist among

carpophores of these species. In these species too, the modified nodose-septate

hyphae are considered to be important because of their different forms in carpo-

phores of different species. In carpephores of some species e.g. Trametes cingulata

and Polyporus occidentalis the nodose-septate hyphae were consistently thin-walled.

In others, e.g. Lenzites betulina and Polyporus versicolor, nodose-septate hyphae

S18)

with thick walls were present in the context as more or less straight, branching

hyphae, parallel to the fibre hyphae, while in some other species, e.g. Polyporus

versicolor and Lenzites palisoti, thick-walled, nodose-septate hyphae contributed

to the binding hyphal system of the carpophores. These hyphae resemble the

aseptate binding hyphae but are recognizable by the presence of clamp connections.

In some carpophores of species in Group 51, terminal cells of nodose-septate

hyphae are differentiated into cuticular cells or brown, thick-walled hyphae with

irregular projections, whilst in Group 53, thick-walled, nodose-septate hyphae

formed the hairy upper surface of the carpophores of Polyporus sacer. There

are thus differences in the morphology and function of these hyphae in carpophores

of different species and even in different parts of the same carpophore of some

species.

Teixeira (1956, 1962 b) showed that different types of skeletal hyphae were

present in carpophores of different species with trimitic hyphal systems. Although

the skeletal hyphae found in the species mentioned above were mainly unbranched,

corresponding to Teixeira’s “‘vermiculiform skeletal hyphae” (Teixeira, 1962 b),

fibre hyphae with one to three branches towards the distal end were found in

carpophores of some species with trimitic hyphal systems in this study. These

branches were found to contribute to the binding hyphal system of the carpophores.

Morphologically they appear to correspond to Teixeira’s (1956, 1962 b) ‘‘arboriform

skeletal hyphae,” but Furtado (1966) maintained that “‘arboriform skeletal hyphae”

are found only in carpophores of Ganodermoid species. From observations made

in this study, however, it appears that they may also occur in carpophores of

species of Trametes and Coriolus. Differences in the merphology of fibre or skeletal

hyphae thus occur and must be taken into consideration in taxonomic studies of

these fungi. Differences in the binding hyphae of different species or groups of

species were also evident. ‘Cunningham (1946) recognized two types, viz.: the

“bovista’’ type and the “‘long’’ type of binding hypha but failed to distinguish

adequately between them. Morpholegical differences between binding hyphae of

different species were observed in the species studied here. In some species of

Trametes and Coriolus the binding hyphae were found to be rather intricately

branched structures with the branches short, thick, often tortuous and of a different

refractive index from that of the skeletal hyphae. In other species, e.g. Trametes

cingulata, Polyporus occidentalis and Polyporus sacer the binding hyphae resemble

skeletal hyphae but have fairly long, flexuous, tapering branches. The length and

form of branches of binding hyphae may vary according to their position in the

carpophore but differences in morphology of the branches and the difference in

refractive index may be observed with little difficulty.

Corner (1953) described the binding hyphae from the carpophores of

Polyporus sulphureus and Polyporus squamosus, species which he regarded as

having dimitic hyphal systems with generative and binding hyphae. In both these

species, the binding hyphae are formed by the evagination of the walls of intercalary

cells of generative hyphae into a number of tortuous, lateral processes which later

become thick-walled. Structures which bear some resemblance to these were seen

in cultures and carpophores of Lenzites sepiaria and in the carpophores of Polyporus

adustus (Fig. 2 p), but these structures, which have a binding function in the

carpophores, do not arise in the same way as the binding hyphae which originate

and develop as the terminal cells of lateral branches of generative hyphae. These

structures are not separated from the parent cells by septa. For this reason these

structures should be regarded as binding processes rather than binding hyphae.

Differences in morphology and ontogeny of the elements of the binding hyphal

system thus exist and should be of great value in taxonomic studies of Polyporaceae.

From the above it is thus evident that numerous differences in the morphology

and ontogeny of the hyphae which comprise the different hyphal systems exist

314

in carpophores of different species. The different types of hyphae, their

morphology, occurrence and function in fruit-bodies of polypores in the present

study may be summarized as follows:—

1. Septate hyphae

1.1 Thin-walled, nodose-septate hyphae; branching, mostly with deeply staining

contents; present in the growing regions of the upper surface, margin and hymenial

surfaces; giving rise to all other structures in the carpophore, (generative hyphae,

Corner, 1932 a, b); collapsed and empty in older parts of carpophores of some

species.

1.2 Thick-walled, nodose-septate hyphae; walls regularly thickened, with or

without staining contents, regularly septate, branching, (sclerified generative hyphae,

Donk, 1964) occur as:

(i) hyphae supporting reflexed pilei and constituting major or only hyphal type,

orientated mainly parallel to direction of growth of pileus;

(ii) hyphae as in (i) but present in small numbers in context, among fibre hyphae

in species where these are present; function unknown;

(iii) hyphae with tortuous branches orientated across the direction of growth

of the pileus and assisting in binding the tissues;

(iv) hyphae supporting pilei as in (i) but forming lateral, branched, binding

processes;

(v) short, thick-walled or solid hyphae forming “hairs” of tomentose upper

surface of pilei.

1.3 Nodose-septate hyphae with irregularly thickened walls, present in lower

parts of context of certain species; function unknown.

1.4 Hyphae with irregular projections and cuticular cells; brownish, thick-walled

elements arising from septate hyphae, present in dark-coloured incrusted areas

over the older parts of some carpophores of certain species; function unknown,

probably protective.

2. Fibre hyphae

2.1. Unbranched, straight or somewhat flexuous, hyaline to pale brown, thick-

walled or sub-solid to solid, aseptate or with one or two simple septa towards the

apex, arising from thin-walled or thick-walled regularly septate hyphae; when

present, constituting bulk of tissues of carpophore, terminating in context or at

upper surface and margin or below hymenial surfaces; arranged parallel to direction

of growth of carpophore; supporting and protecting hymenophore and forming

tomentum or pubescence or, by agglutination with lacquer-like material, incrusted

or fibrillar or glabrous upper surface. (Aciculiform and vermiculiform skeletal

hyphae, Teixeira 1956, 1962 b).

2.2 Branched fibre hyphae, as in 2.1 but with one to three branches towards the

apex; the main stem parallel to the direction of growth of the carpophore and

supporting the carpophore tissues, the branches arranged across the direction

of growth of the carpophore and binding the tissues; occurs in lower context

of certain species (arboriform skeletal hyphae, Teixeira 1956, 1962 b).

2.3 Branched fibre hyphae, the branches long and tapering, otherwise as in 2.1

branches interwoven with other hyphae across the direction of growth of the

carpophore; bind hyphae into tough tissues; present in context of some species

in which unbranched fibre hyphae are also present.

315

2.4 Branched fibre hyphae with numerous short tortuous branches, interwoven

with other fibre hyphae across the direction of growth of the carpophores, otherwise

as in 2.3.

2.5 Dichotomously branched, very narrow hyphae, forming a network in the lower

context tissues of some species; apparently aseptate, origin and function unknown.

This list is by no means complete as many types of hyphae, such as simple-

septate, thin-walled, generative hyphae (Corner, 1932 b), thick-walled, simple-septate

hyphae (Pinto-Lopes, 1952), inflated hyphae (Corner, 1953) and others were not

encountered in the carpophores of the species included in this study. It does

however serve to illustrate the diversity in hyphal morphology and hyphal function

which exists in carpophores of the poroid Hymenomycetes. It is evident from this,

that this diversity and its possible phylogenetic connotations had not been fully

utilized in taxonomic studies of these fungi. It is also evident that this diversity

in hyphal morphology and function together with the resulting differences in

carpophore construction and texture, are not adequately expressed and conveyed

by the concept of hyphal systems. Consequently, the concept of hyphal systems

had been criticized by Pinto-Lopes (1952), Teston (1953 a), Welden (1960), Smith

(1966) and others.

Bondartzeva (1963) and Smith (1966) expressed the view that hyphal systems

are indications of adaptive evolution and devices to restrict waterloss from the

carpophores and prevent dessication and damage to the hymenium. Undoubtedly,

there is strong evidence in favour of these views. Savile (1954, 1955) stated that if

an ecological niche exists, it will be filled repeatedly by different organisms which

find similar ways to achieve this. Problems involved in the extension of the

hymenial surface, protection of the hymenial surface from rain and reduction of

loss of moisture from fruit-bodies of Hymenomycetes can be overcome in a limited

number of ways only so that similar structures must have developed repeatedly.

It is thus conceivable that species of which fruit-bodies have similar hyphal

systems (sensu Corner, 1932 a, b) may have developed repeatedly and independently.

For these reasons species having similar hyphal systems cannot be regarded as

being congeneric on that basis only. Indeed, it became evident in this study that

important differences in morphological characters of the hyphae and in their

arrangement and function in the carpophores can exist in different species with

similar hyphal systems. It is, however, also evident from this study that the hyphae

present in fruit-bodies of individual species are morphologically and genetically

constant for each species unlike such characters as, habit, insertion, hymenial

configuration and texture of the upper surface of the carpophores. It also became

evident that the hyphal complement and construction of the fruit-bodies, i.e. the

placing of different types of hypha in the carpophores of different species, are

constant for each species although certain specialized structures, such as cuticular

cells, may be absent from carpophores of species which are capable of forming

them. For this reason these characters should be studizd and recorded in detail

in descriptions of carpophores of different species. All these characters must be

considered together with other constant characters, such as spore shape and size

and basidial shape and size, in the delimination and characterization of genera

of the polypores. Applied in this way these characters become valuable com-

ponents of the “complex of characters’ which must be considered for generic

delimination as stated by Ames (1913), Wakefield (1948), Pinto-Lopes (1952),

Nobles (1958 b), Teixeira (1962 b) and Kotlaba (1964) among others; but careful

observation and accurate descriptions of hyphal characters and construction of

carpophores rather than generalizations by means of collective terms are essential

prerequisites for their use in this connection.

316

7. SUMMARY

1. Twenty-four species of poroid Hymenomycetes from South Africa and

Canada were studied. Of these, twelve species occur in South Africa, four in Canada

whilst eight are found in both countries.

2. The micromorphological characters and oxidase reactions of the cultures

and the micromorphological characters and construction of the carpophores

together with the type of decay and host range of these twenty-four fungi were

studied in order to determine: (i) which microstructures are formed in cultures of

these species; (ii) their relationships as indicated by their cultural characters; (iti)

whether the structures formed in culture are also present in their carpophores,

and (iv) whether the relationships indicated by cultural characters are also revealed

by their carpophores.

3. Mycelia obtained from single basidiospores were paired in culture

in order to determine the type of interfertility of certain species or to determine

conspecificity between different collections. In some species attempts were made

to dikaryotize large haploid mycelia by pairing them with small dikaryotic mycelia

in culture in order to establish conspecificity between the different collections

from which the mycelia were obtained.

4. The literature on the classification of the poroid Hymenomycetes, the

structure and anatomy of their carpophores in relation to their taxonomy and

studies of Hymenomycetes in pure culture, was reviewed.

5. The cultural characters were studied by observations on cultures of the

fungi incubated in the dark on 1.5 per cent malt agar plates for a period of six

weeks according to the methods of Nobles (1948). Construction and micro-

morphology of the carpophores were studied by teasing apart thick sections of

carpophores to obtain undamaged structures for examination according to the

methods of Teixeira (1956). All microstructures were examined by means of the

oil immersion lens and recorded by means of camera lucida drawings or photo-

micrographs.

6. It was found that the 24 species were distributed among nine of the 36

groups proposed by Nobles on the basis of their cultural characters. In five of

these groups only one species was studied in each. Two of these species, Polyporus

dichrous in Group 9 and Polyporus subiculoides in Group 32 displayed characters

that made inclusion in their respective groups somewhat dubious.

7. The structures formed im cultures of the different species were also

found in their carpophores with the exception of chlamydospores. Although

chlamydospores were present in cultures of most species they were found in

carpophores of one species only, Fomes pinicola.

8. As the structures formed in cultures are also present in the carpophores

of the different species, the carpophores can be placed in the same groups as the

cultures but do not show the same relationships.

9. Differences in the micromorphological characters of the hyphae and in

the types of hyphae present in the carpophores of species from the same group

were found in species of three of the four groups in which more than one species

was studied.

10. Differences in construction of the carpophores and orientation and

functions of their hyphae were observed in carpophores of species in which similar

types of hyphae are present.

37)

11. Important differences between the hyphal characters and construction

of the carpophores of various species and the carpophores of type species of

genera to which they have been assigned by different authors, were demonstrated.

12. Differences in the characters and origin of the “‘hairs’’ that constitute

the trichocutis or upper surfaces of carpophores of a number of species, were

noted.

13. Cuticular cells which characterize cultures of Daedalea confragosa,

Trametes corrugata, Trametes acupunctata and Hexagona tenuis, are often lacking

from individual fruit-bodies of these different species or are present as hyphae with

irregular projections on the carpophores. Because of their sporadic appearance

on carpophores, these structures are not regarded as being reliable characters

for taxonomic purposes.

14. Three different ways of development of fruiting structures were observed

in cultures of the various species, viz.: (i) formation of low anastomising ridges;

(ii) tubules and (iii) erect acicular cr flattened spines. These are considered to be

of phylogenetic importance.

15. The type of interfertility of seven species was determined. All displayed

the tetrapolar type of interfertility. Of these, Polyporus dichrous only, is associated

with brown rot and its cultures do not produce extra-cellular oxidase enzymes.

The other six species are associated with white rots and their cultures produce

extra-cellular oxidase enzymes.

16. By pairing haploid myceliz derived from single basidiospores from

different collections, it was found that haploid mycelia from a South African

collection of Lenzites trabea were completely compatible with haploid mycelia

from a Canadian collection. The conspecificity of the two collections were thus

confirmed. The conspecificity of four collections of Trametes cingulata from

South Africa were also confirmed by means of this technique. When this technique

was used to determine the conspecificity of a South African collection of Polyporus

adustus with Canadian collections oi this species it was found that only a very

low degree of compatibility existed between the haploid mycelia from the South

African and Canadian collections although no differences in cultural and carpophore

characters could be found.

17. The technique of dikaryotizing a large haploid mycelium grown in culture

by pairing it with a small dikaryotic mycelium was used to confirm the identity

of different collections of four different species. This was successful with four

collections of Trametes meyenii anc four collections of Polyporus occidentalis.

This method failed however with five collections of Polyporus dichrous and seven

collections of Polyporus pubescens.

18. It was concluded that the micromorphological characters of the hyphae

and other microstructures as well as the construction of the carpophores are

constant for each species. All these characters should be carefully described and

recorded for each species and should be taken into consideration in taxonomic

studies of these fungi. Differences and similarity of micromorphological characters

and construction of carpophores cf species are not adequately conveyed by the

concept of hyphal systems.

318

8. TABLES

TaBLE 1. — Oxidase reactions and colony diameter (in mm) of different isolates

of Polyporus adustus on malt-gallic acid medium and malt-tannic acid medium

and oxidase reaction as indicated by gum guaiac solution applied directly to

cultures on malt agar, after 14 days’ incubation.

Oxidase reaction ig es

Isolate No.

Gum Gallic Tannic Gallic Tannic

guaiac acid acid acid acid

DOAM = 9209 + (*) —— of trace trace

DAOM 17571 — (*) _ — trace no growth

DAOM 17575 — a — 5 no growth

DAOM 22576 — — f- 5) no growth

DAOM 53500 f- — + trace trace

PRE 42039 +t (3) — — 12 no growth

PRE 42328 — — — 1 no growth

PRE 42332 + + — 10 no growth

PRE 42350 4. — — 5) trace

PRE 42365 — a= — trace none

(1) Positive reaction;

(2) Negative reaction;

(3) Strong positive reaction.

319

TaBLe 2.— Results of pairing four mycelia derived from single basidiospores of a

South African collection, PRE 42039, with single basidiospore cultures of each of

four Canadian collections PRE 42365, PRE 42328, PRE 42329 and DAOM 53500,

of Polyporus adustus.

PRE PRE PRE PRE DAOM

42039 42365 42328 42329 53500

[eee eke oe oNede Dy Sued el) 3) ae > 34

1 = gree 2 eh cy ke

PRE 2 See =

AD (3.0m nen Pn Rr ey Ss ee Der ceee tay

4 —-——4f—-4-——-—-—

1 ———— +++4+4+4+4+4+ 4444

PRE 2 ———— ++i 4444+ 4444

42365 3 ——— — p++ t+ +--+ 4+ 4444

4 ———— +++4+ 4444 4444+

1 —-~—-——+4++4++4 +++4++4+4++4+

PRE 2 ————4+4+4+4 P+t+ +4+44

42328 3 — —++4++4+4 t + 4 b +

4 ——-——+4+4+44 bf + + +

bo --—-—~+4+4++4+4+4++4++ b+ ++

PRE 2 ————+++4-—4+4+44 +4++4++4+

42329 3 —-—+—4444+4++4+44 +4+4+4

4 —4+—-4+ 44 ++4+4 + +++

1 ~-—-—~—~+4+4+4++4+4+4+4++4+4+44

DAO Qo = ee Es tc a

53500 3 —~—-——~4+4++4++4+4+4++4+4+++4+4

4 — —~+4+44+44+4+4+ 4444

A (+) indicates formation of clamps on the mycelium.

TasLe 3.— Mating types of mycelia from single spores of Polyporus dichrous

PRE 42384.

JX 188 Il, By LO, 122 ie 1B98 Sh thy JI) TS, Ue

Py Je 2 Thy Ys IOI, IO, Ize A, By: 4, 14.

Taste 4.— Results, showing the formation of clamp connections (+-), when

four single basidiospore cultures from each of two isolates of Lenzites trabea,

PRE 42457 and DAOM 72285, were paired in all possible combinations.

PRE 42457

1 2 3 4

oe a ee

DAOM U2 ce 2

72285 3 se =e ar ar

4 4 4+ 4+ +

320

TABLE 5.— Collections of Polyporus pubescens tested for conspecificity with

monospore cultures Nos. 5 and 8 of Polyporus pubescens DAOM 94039.

DAOM 17577 DAOM 94017

DAOM 52833 DAOM 94026

DAOM 53503 DAOM 94039

DAOM 73309

TaBLE 6.— Mating types of single spores of Trametes mzyenii PRE 42446.

AB leno: [Xe \858 35 4, Sp 12, 3, Se

Aye Ba: eG; Je \B58 2, O, I, 8 IO, tél

TasLe 7.— Mating types of single spores of Lenzites palisoti PRE 42442.

A Bids Wiles: A, B,: 8, 10;

A, B,: 12, 13, 16; AVE BOM:

No mating: 1, 2, 3, 7, 14.

Taste 8.— Mating types of mycelia from 16 single spores of Polyporus occidentalis

PRE 42863.

Ja IBS Ih, 2, By Wil, WA, Sse JN; 1852 4, ©, IGe

ING N888 By 1 138 A, By: 8, 9, 10, 14, 16.

TaBLE 9.— Mating types of single basidiospores of Trametes cingulata Berk.

PRE 42448.

PX 1848 Il, GD, Wl, 4 iS, ies Ja, 1858 Bo Gh Do So MO, WA, USE

A, Bz: 6, 7; A» B,: 2.

TasLe 10.— Mating types of single spore of Polyporus vinosus PRE 42154.

Jo 18318 Wy Wo th, By WA, IS}, Id’ A, B,: not present;

A, Bz: 2, 4, 5, 10, 15, 16; Az Bz: 3, 6, 11.

TaBLe 11. — Distribution of mating types in 16 single spore cultures of Trametes

corrugata, PRE 42454.

JNg Bi 2, S51 9, Ie A, Bs: 4, 12, 13:

A, Bz: 6, 10, 11, 14; INg 138 IL, 35 B IS;

9. REFERENCES

Ames, A., 1913. A consideration of structure in relation to genera of the polyporaceae. Ann.

Mycol. 11: 211—253.

Aosuima, K., 1967. Synopsis of the genus Daedalea Pers. ex. Fr. Trans. Mycol. Soc. Japan 8 : 1—6.

Aosuima, K., Lentz, P. L. & McKay, H. H., 1961. Sterewm taxodii in Japan and Formosa.

Mycologia 53: 145—153.

Baxsut, B. K. & Srncu, B., 1961. New and noteworthy records of Hymenomycetes in India.

Indian Forest Bulletin (New series, No. 235).

Banergee, S. N. & Dest, H., 1956. Variation of Polystictus xanthopus. Indian J. Mycol. Res. 2:1—28.

Banergee, S. & Nana, P. M., 1960a. Trametes cingulata in culture. Proc. Ind, Acad. Sci., Sect. B.

52: 9—18.

BANERJEE, S. & Nana, P. M., 1960b. Trametes cingulata Berk. causing decay of Sal (Shorea robusta

Gaertn.) timber. Phyton 14: 93—109.

Baxter, D. V., 1924. Fomes fraxineus in culture. Papers Mich. Acad. Sci. 4: 55—66.

Baxter, D. V., 1925a. The biology and pathology of some of the hardwood heart-rotting fungi.

I. Am. J. Botany 12: 522—552.

321

Baxter, D. V., 1925b. The biology and pathology of some of the hardwood heart-rotting fungi.

II. Am. J. Botany 12: 553—576.

Baxter, D. V., 1926. Some Porias from th= region of the Lake States. Papers Mich. Acad. Sci.

6: 67—76.

Baxter, D. V., 1929. Some Porias from the region of the Lake States. II. Papers Mich. Acad.

Sci. 9: 39—45.

Baxter, D. V., 1931. Some resupinate polypores from the region of the Great Lakes. III. Papers

Mich. Acad. Sci. 15: 191—228.

Baxter, D. V., 1933. Some resupinate polypores from the region of the Great Lakes. IV. Papers

Mich. Acad. Sci. 17: 421—439.

Baxter, D. V., 1934. Some resupinate pciypores from the region of the Great Lakes. V. Papers

Mich. Acad. Sci. 19: 305—332.

Baxter, D. V., 1935. Some resupinate polypores from the region of the Great Lakes. VI. Papers

Mich. Acad. Sci. 20: 273—281.

Baxter, D. V., 1936. Some resupinate polypores from the region of the Great Lakes. VII. Papers

Mich. Acad. Sci. 21: 243—267.

Baxter, D. V. 1937. Some resupinate polypores from the region of the Great Lakes. VIII. Papers

Mich. Acad. Sci. 22: 275—295.

Baxter, D. V., 1938. Some resupinate polypores from the region of the Great Lakes. IX.

Papers Mich. Acad. Sci. 23: 285—305.

Baxter, D. V., 1939. Some resupinate pciypores from the region »f the Great Lakes. X. Papers

Mich. Acad, Sci. 24: 167—188.

Baxter, D. V., 1940. Some resupinate polypores from the region of the Great Lakes. XI. Papers

Mich. Acad. Sci. 25: 145—170.

Baxter, D. V., 1941. Some resupinate polypores from the region of the Great Lakes. XII. Papers

Mich. Acad. Sci. 26: 107—121.

Baxter, D. V., 1942. Some resupinate polypores from the region of the Great Lakes. XIII. Papers

Mich. Acad. Sci. 77: 139—161.

Baxter, D. V., 1943. Some resupinate polypores from the region of the Great Lakes. XIV. Papers

Mich. Acad. Sci. 28: 215—233.

Baxter, D. V., 1944. Some resupinate polypores from the region of the Great Lakes. XV. Papers

Mich. Acad. Sci. 29: 85—109.

Baxter, D. V., 1945. Some resupinate polypores from the region of the Great Lakes. XVI. Papers

Mich. Acad, Sci. 30: 175—191.

Bavenpamm, W., 1928. Uber das Vorkommen und den Nachweiss von Oxydasen bei holzzerstorenden

Pilzen. Z. Pflanzenkrankheit Pflanzenschutz 38: 257—276.

Bensaupe, M., 1918. Recherches sur le cycle évolutif et la sexualité des Basidiomycetes. Rev.

gen. Bot. 30: 1—153.

Borin, J., 1964. Valeur des caractéres culturaux et cytologique pour la taxonomie des Thele-

phoraceae résupinés et étalés réfléchés (Basidiomycétes). Bull. Soc. Mycol. France III:

309—315.

Borin, J. & pes Pomeys, M., 1961. Hétéro-basidiomycétes saprophytes et Homobasidiomycetes

résupinés. IX. De Vutilization des criteres d’interfertilité et de polarité pour la reconnaissance

objective des limites spécifiques et des affinitéés. Bull. Soc. Mycol. France 78: 237—261.

Bonpartzev, A. S. 1953. Trutovye griby evropeyskoy chasti SSSR. i. Kavkaza. Leningrad.

Bonpartzev, A. S. & Sincer, R., 1941. Zur Systematik der Polyporaceen. Ann. Mycol. 39: 43—65.

Bonpartzeva, M. A., 1961. A critical review of the most recent classifications of the family

of the Polyporaceae. Botanicheskii Zhurnal, 46: 587—593. (Translation, Canada Department

of Agriculture No. 73250, 1961).

Bonpartzeva, M. A., 1963. On the anatomical criterion in the taxonomy of the Aphyllophorales.

Botanicheskii Zhurnal 48: 362—373. (Translation No. 99038, by the Translation Bureau of

the Canada Dept. of State).

Bosg, S. R., 1930. On the biology of wood-rotting fungi. J. Linn. Soc. London 48: 417—438.

Bosg, S. R., 1943. Moisture relation as a determinant factor in the transformation of the basidia

of certain Polyporaceae. Mycologia 35: 33—46.

Bose, S. R., 1952. Identity of Polystictus (= Polyporus = Trametes) cinnabarinus (Jacq.) Fr.

with Polystictus sanguineus (L.) Fr. Nature 170: 1020.

Bourpot, H. & Gatzin, A., 1928. Hymenomyceétes de France. Paris.

Bropir, H. J., 1935. The occurrence in nature of mutual aversion between mycelia of Hymeno-

mycetous fungi. Can. J. Research C, 13: 187—189.

Broprg, H. J., 1936. The barrage phenomenon in Lenzites betulina. Genetica 18: 61—73. z

Buiier, A. H. R., 1931. Researches on Fungi. 4: 187—305. London: Longmans, Green & Co.

Burt, E. A., 1917. Merulius in North America. Ann. Mo. Bot. Gard. 4: 305—362.

CampsetL, W. A., 1938. The cultural characteristics of the species of Fomes. Bull. Torrey Bot.

Club. 65: 31—69. ;

Cartwricut, K. St. G., 1929. Notes on Basidiomycetes grown in culture. Trans. Brit. Mycol.

Soc. 14: 300—305.

S22

Cartwricut, K. St. G., 1931. Further notes on Basidiomycetes in culture. Trans. Brit. Mycol.

Soc. 16: 304—307.

Cartwricnt, K St. G. & Finptay, W. P. K . 1946. Decay of timber and its prevention. London:

His Majesty’s Stationery Office.

Cooke, W. B., 1959. The genera of pore fungi. Lloydia 22: 163—207.

Corner, E. J. H., 1932a. The fruit-body of Polystictus xanthopus Fr. Ann. Bot. 46: 61—111.

Cone EK. J. H., 1932b. A Fomes with two systems of hyphae. Trans. Brit. Mycol. Soc. 17:

—8l.

Corner, E. J. H., 1947. Variation in the size and shape of spores, basidia and cystidia in

Basidiomycetes. New Phytologist 46: 195.

Corner, E. J. H., 1948. Asterodon, a clue to the morphology of funzous fruit-bodies. Trans. Brit.

Mycol. Soc. 31: 234—245,

Corner, E. J. H., 1950. Clavaria and allied genera. Oxford University Press, pp. 1—740.

Corner, E, J. H., 1953. The construction of polypores. I. Introduction: Polyporus sulphureus, P.

squamosus, P. betulinus and Polystictus microcyclus. Phytomorphology 3: 152—167.

Corner, E. J. H., 1954a. The classification of the higher fungi. Proc. Linn. Soc., London

165: 4—6.

Corner, E. J. H., 1954b. Reviews. Trans. Brit. Mycol. Soc. 87: 92—94.

Corner, E. J. H., 1966. A monograph of Cantharelloid fungi. Oxford University Press.

CunnincuaM, G. H., 1946. Notes on classification of the Polyporaceae. New Zealand J. Sci. and

Technology 28, Sect. A: 238—25]1.

CunnincHaM, G. H., 1947. New Zealand Polyporaceae. The genus Poria. D.S.I.R. Plant Dis.

Div. Bull. No. 72.

CunnincHaM, G. H., 1948a. New Zealand Polyporaceae. 2. The genus Fuscoporia. D.S.I.R.

Plant Dis. Div. Bull. No. 73.

Cunnincuam, G. H., 1948b. New Zealand Polyporaceae. 3. The genus Polyporus. D.S.I.R. Plant

Dis. Div. Bull. No. 74.

CunnincHaM, G. H. 1948c. New Zealand Polyporaceae. 4. The genus Coriolus. D.S.1.R. Plant

Dis. Div. Bull. No. 75.

CunnincHaM, G. H., 1948d. New Zealand Polyporaceae. 5. The genus Fomitopsis. D.S.1,R,

Plant Dis. Div. Bull. No. 76.

Cunnincuam, G. H., 1948e. New Zealand Polyporaseae: 6. The genus Coltricia. D.S.I.R, Plant

Dis. Div. Bull. No. 77.

CunnincHam, G. H., 1948f. New Zealand Polyporaceae. 7. The genus Inonotus. D.S.1.R. Plant

Dis. Div. Bull. No. 78.

Cunnincuam, G. H., 1948¢. New Zealand Polyporaceae. 8. The genus Fomes. D.S.1.R. Plant

Dis. Div. Bull. No. 79.

Cunnincuam, G. H., 1948h. New Zealand Polyporaceae. 9. Trametes, Lenzites and Daedalea.

D.SA.R. Plant Dis. Div. Bull. No. 80.

Cunnincuam, G. H., 1949a. New Zealand Polyporaceae. 10. Revision of New Zealand species

and records. D.S.I.R. Plant Dis. Div. Bull. No, 81.

CunnincuaM, G. H., 1949b. New Zealand Polyporaceae. 11. The genus Irpex. D.S.1.R. Plant.

Dis. Div.. Bull. No. 82.

Cunnincuam, G. H., 1950a. New Zealand Polyporaceae. 12. The genus Merulius. D.S.I.R. Plant.

Dis. Div. Bull. No. 83.

Cunnincuam, G. H., 1950b. Australian Polyporaceae in herbaria of the Royal Botanic Gardens,

Kew and British Museum of Natural History. Proc. Linn. Soc. N.S.W. 75: 214—249.

Cunnincuam, G. H., 1954. Hyphal systems as aids in identification of species and genera of the

Polyporaceae. Trans. Brit. Mycol. Soc. 37, 44—50.

CunnincuaM, G. H., 1963. The Thelephoraceae of Australia and New Zealand. New Zealand

D.S.1.R. Bull. No. 145.

Da Costa, E. W. B., Marrers, C. N. & Tamsiyn, N., 1952. The identification of Basidiomycete

fungi in culture. Progress report No. I. Methods of describing the morphological character-

istics of Basidiomycete fungi in culture. Sub-project P.11—12. Commonwealth of Australia,

Scientific and Industrial Research Organization, Division of Forest Products..

Davipson, R. W. & Camppety, W. A.. 1943. Decay in merchantable black cherry in the Alleghany

National Forest. Phytopatology 33: 9€65—985.

Davipson, R. W., Camppett, W. A. & BratspeLL, D. J., 1938. Differentiation of wood-decaying

fungi by their reactions on gallic or tannic acid medium. J. Agr. Research 57: 683—695.

Davipson, R. W., Camppett, W. A. & Vaucun, D. B., 1942. Fungi causing decay of living oaks

in the Eastern United States and their cultural identification. U.S.D.A. Tech. Bull. No. 785,

Davinson, R. W., Lenz, P. L. & McKay, H. H., 1960. The fungus causing pecky cypress.

Mycologia 52: 260—279.

Davipson, R. W. & Lomparp, F. F., 1953. Large-brown-spored houserot fungi in the United States.

Mycologia 45: 88—150.

Davipson, R. W., Lomparp, F. F. & Hirt, R. R., 1947. Fungi causing decay in wooden boats.

Mycologia 39: 313—327.

325

Denyer, W. B. G., 1960. Cultural studies of Flammula a!nicola (Fr.) Kummer and Flammula

conissans (Fr.) Gillet. Can. J. Bot. 38. 909—920.

DoincE, E. M., 1950. The South African fungi and lichens. Bothalia 5: 1—1094.

Donk, M. A., 1933. Revision der Niederlandischen Homobasidiomycetae-Aphyllophoraceae. Med.

Bot. Mus. en Herb. van de Rijks Universiteit te Utrecht, No. 9: 1—278.

Donk, M. A., 1960. The generic names proposed for Polyporaceae. Persoonia 1: 173—302.

Donk, M. A., 1962. The generic names proposed for Polyporeae. Additions and corrections.

Persoonia 2: 201—210.

Donk, M. A., 1964. A conspectus of the families of Aphyllophorales. Persoonia 4: 337—343.

Donk, M. A., 1966. Notes on European Polypores. 1. Persoonia 4: 337—343.

Eruerwce, D. E., 1957. Differentiation of white and brown rot fungi by an oxidase reaction.

Nature 179: 921—922.

Fatck, R., 1909. Die Lenzites-faule des Coniferenholzes. Hausschwamforschungen 3: 1—234.

Fartnua, M., 1964. Caracteres morfol6zicos das hifas des himendforos de Polyporaceae. Portug.

Acta Biol. Sir. B, 7: 288—346.

Fipatco, M. E. P. K., 1958. Notes on Lenzites cinnamomea Fr. Mycologia 50: 753—756.

Fipatco, M. E. P. K., 1961. The genus Phaeodaedalea. Mycologia 53: 201—210.

Fipatco, M. E. P. K., 1962. The genus Osmoporus Sing. Rickia 1: 95—138.

Fipatco, O., 1957. The nomenclature status of Daedalea Pers. ex Fr. and related genera. Taxon

7: 133—140.

Fiparco, O., 1958. Studies on Ptychogaster rubescens Boud., the chlamydosporous form of

Polyporus guttalatus Peck. Mycologia 50: 831—836.

Fipatco, O., 1962a. Anatomical observations on the genus Bornetina Mangin & viala. Rickia 1:

139—143.

Fipatco, O., 1962b. Type studies and revision of the genus Diacanthodes Sing. Rickia 1: 148—180.

Fipatco, O., 1963. Studies on the type species of Hydnopolyporus. Mycologia 55: 713—727.

Fipatco, O., 1967. Ultrasounds applied to the study of hyphal systems in Polyporaceae. Mycologia

59: 545—547.

EDanCe, O. & Finatco, M. E. P. K., 1962. Estudos anatomicos em Polyporaceae. Rickia 1:

195—205.

Fipatco O, & Finatco, M. E. P. K., 1963. A new genus based on Fistulina brasiliensis. Mycologia

54: 342—352.

Fipatco, O. & Fipatco, M. E. P. K., 1966. Polyporaceae from Trinidad and Tobago. I. Mycologia

58: 862—904.

Fipatco, O. & Finauco, M. E. P. K., 1967. Polyporaceae from Trinidad and Tobago. IH. Mycologia

59: 833—869.

Fries, E., 1821. Systema Mycologicum. 1: 326—393. Lund.

Fries, E., 1828. ..Elenchus Fungorum 1: 64—127.

Fries, E., 1838. Epicrisis Systematis Mycologici. 403—498. Uppsala.

Fries, E., 1874. Hymenomycetes Europaet, 492—594. Uppsala.

Fritz, C. W., 1923. Cultural criteria for the distinction of wood-destroying fungi. Trans. Roy.

Soc. Canada 17: 191—288.

Furrapo, J. S., 1964. Taxonomic and nomenclatural status of Porotheleum rugosum. Mycologia

56: 923—926.

Furrapo, J. S., 1965a. Relation of microstructures to the taxonomy of the Ganodermoideae (Poly

poraceae) with special reference to the cover of the pilear surface. Mycologia 57: 588—611.

Furrapo, J. S., 1965b. Ganoderma colossum and the status of Tomophagus. Mycologia 57: 974—984.

Furrapo, J. S., 1966. The significance of the clamp connection in the Basidiomycetes. Persoonia

4: 125—144.

Furrapo, J. S., 1967. Some tropical species of Ganoderma (Polyporaceae) with pale context.

Persoonia 4: 379—389.

Gitpertson, R. L., 1961. Polyporaceae of the Western United States and Canada. I. Trametes

Fr. Northwest Sci. 35: 1—20.

Gray, S F., 1821. A natural arrangement of British plants. London.

Happow, W. R., 1931. Studies in Ganoderma. J. Arnold Arboretum 12: 25—46.

Hansen, L., 1956. Two Polyporaceous fungi with merulioid hymenophore, Poria taxicola (Pers.)

Bres. and Polyporus dichrous Fr. Friesia 5: 251—256.

Hansen, L., 1958. On the anatomy of the Danish species of Ganoderma. Bot. Tidskr. 54: 333—352.

Hansen, L., 1960. Some macromyectes from Rennell and Alcaster Islands. The Natural History

of Rennell Island; British Solomon Islands 3: 127—132.

Harmsen, L., 1960. Taxonomic and cultural studies on brown-spored species of the genus

Merulius. Friesia 6: 233—277.

Harmsen, L., Baxsui, B. K. & Cuoupuury, T. G., 1958. Relationship between Merulius lacrimans

and Merulius himantoides Nature ‘6): 1011. bee

Hester, L. R. & Smitn, A. H., 1963. North American species of Hygrophorus. Univ. Tenn.

Press, Knoxville

324

Hirt, R. R., 1932. The biology of Trametes suaveolens, (LL) Fr. New York State College of

Forestry Bulletin, Technical Publication No. 37.

Hopkins, J. C. F., 1939. Report of the branch of plant pathology for the year ending 3lst

December 1938. Rhod. Agr. J. 36: 5&2.

Heres: y C. F., 1943. A descriptive lis: of plant diseases in Southern Rhodesia. Rho. Agr. J.

: 188.

Humpnrey, C. J. & Leus, S., 1931. A partial revision of the Ganoderma applanatum group with

particular reference to its oriental variants. Phillip. J. Sci. 45: 483—589.

Humpnrery, C. J. & Siccers, P. V., 1933. Temperature relations of wood-destroying fungi. J. Agric.

Res. 47: 997—1008.

Hwane, S., 1955. Compatibility and variability in Merulius americanus. Mycologia 47: 317—328.

Imazex1, R., 1943. Genera of Polyporaceae of Nippon. Bull. Tokyo Sci. Mus. No. 6: 1—111.

Imazeki, R., 1952. A contribution to the fungus flora of Dutch New Guinea. Bulletin of the

Government Forest Experiment Station. No. 57.

Jay, B. A., 1934. A study of Polystictus versicolor. Kew Bull. 10: 409—424.

JORGENSEN) E. & Vesey, K., 1953. A new polyphenol oxidase test. Physiologia Plantarum 6:

533—537.

KAARIK, A., 1965. The identification of the mycelia of wood-decay fungi by their oxidation

reactions with phenolic compounds. Siudia Forestalia Suecica, No. 31: 1—80.

Karsten, P. A., 1880. Symbolae ad mycologican fennicam. V. Soc. Faun. Fl. Fenn, Medd.

Karsten, P. A., 1881. Enumeratio Boletinearum et Polyporum Fennicarum, Systemate novo dis-

posetorum. Rev. Mycol. 3: 16—23.

Karsten, P. A., 1889. Kritisk ofversigt af Finlands Basidsvamper Helsingfors. Finska Vetenskaps-

Societaten, 1—482.

Kawamura, E., 1941. Studies on the sex of Hymenomycetous furgi, Bull. Sct. Fak. Terkult.

Kjusu Imp. Univ. Fukuoka, Japan, 9: 337—382.

Kirk, T. K. & Ketman, A., 1965. Lignin degradation as related to the phenol oxidase of selected

wood-decaying Basidiomycetes. Phytopathology 55: 739—745

Kniep, H., 1928. Die Sexualitat der niederer Pflanzen. Fischer, Jena.

Korrasa, F., 1964. Bemerkungen zur Auffassung der Gattung bei den Macromyceten. Preslia 36:

329—336. y

Korrapa, F. & Pouzar, Z., 1957. Notes on classification of Europear. pore fungi. Ceska Mykol.

9: 152—170

Lanyouw, J. & Srarteu, F. A., 1964. Index Herbariorum. The herbaria of the world, 5th ed.,

Utrecht.

Lentz, P L., 1954. Modified hyphae of Hymenomycetes. Bot. Rev. 20: 135—199.

Lentz, P. L., 1955. Stereum and allied genera of fungi in the upper Mississippi Valley, U.S.D.A.

Agr. Monograph 24.

Lentz, P. L., 1960. Taxonomy of Stereum and allied genera. Sydowia Ann. Mycol. 14: 116—135.

Lioyp, C. G., 1898—1905. Mycological Writings Vol. 1. Cincinnati, Ohio.

Lioyp, C. G., 1905—1908. Mycological Writings Vol. 2. Cincinnati, Ohio.

Lioyp, C. G., 1909—1912. Mycological Writings Vol. 3. Cincinnati, Ohio.

Lioyp, C. G., 1913—1916. Mycological Writings Vol. 4. Cincinnati, Ohio.

Lioyp, C. G., 1916—1919. Mycological Writings Vol. 5. Cincinnati, Ohio.

Lioyp, C. G., 1920—1921. Mycological Writings Vol. 6. Cincinnati, Ohio.

Lioyp, C. G., 1922—1925. Mycological Writings Vol. 7. Cincinnati, Ohio.

Louwac, H., 1941. Anatomie der Asco- und Basidiomyceten. Borntraeger, Berlin-Nikolassee (1965).

Louwac, K., 1940. Zur Anatomie des Deckgeflechtes der Polyporaceae. Ann. Myc. 38: 401—432.

Lomparp, F. F., Davinson, R. W. & Lows, J. L., 1961. Cultural characteristics of Fomes ulmarius

and Poria ambigua. Mycologia 52: 280—294.

Lomparp, F. F. & Girpertson, R. L., 1965. Studies on some Western Porias with negative or

weak oxidase reactions. Mycologia 57: 43—76.

Lomparp, F. F. & Girsertson, R. L., 1966. Poria luteo-alba and some related species in North

America. Mycologia 58: 827—845.

Lone, W. H. & Harscu, R. M., 1918. Pure cultures of wood-rotting fungi on artificial media.

J. Agr. Res. 12: 33—82.

Lowe, J. L. 1946. The Polyporaceae of New York State. The genus Poria. New York State

College of Forestry Technical Publication No. 65.

Lowe, J. L., 1947. Studies in the genus Poria. II. White and bright-colored type material.

Lloydia 10: 45—59.

Lowe, J. L., 1948. Studies in the genus Poria. IV. Brown type material. Lloydia 11: 162—170.

Lowe, J. L., 1956. Type studies of the polypores described by Karsten. Mycologia 48: 99—125.

Lowe, J. L., 1957. Polyporaceae of North America. The genus Fomes. State University of

New York, College of Forestry, Syracuse, Technical Publication No. 80.

Lowe, J. L., 1958. The genus Poria in North America. ..Lloydia 21: 100—114.

Lowe, J. L., 1961. Southern and Tropical polypores II. Papers Mich. Acad. Sci. 46: 205—207.

325

Lowe, J. L., 1963a. The Polyporaceae of the world. Mycologia 55: 1—12.

Lowe, J. L., 1963b. A synopsis of Poria and similar fungi from the tropical regions of the

world. Mycologia 55: 453—486.

Lowe, J. L., 1966. Polyporaceae of Nor:h America. The genus Poria. State University of New

York, College of Forestry, Syracuse, Technical Publication No. 90.

Lowe, J. L. & Gitpertson, R. L., 1961a. Synopsis of the Polyporaceae of the Western United

States and Canada. Mycologia 53: 474—511.

Lowe, J. L. & Gitpertson, R. L., 1961b. Synopsis of the Polyporaceae of the South Eastern

United States. Elisha Mitchell Scientitic Society Journal 77: 43—6l1.

Lurnarpt, W. & Lyr, H., 1965. Uber die Bildung von Tyrosinase bei holzzerstorenden Pilzen.

Arch. Mikrobiol. 51: 103—117.

pe 1955. Vorkommen yon Peroxydase bei holzzerstorenden Basidiomyceten. Planta 46:

8—413.

bye i 1963. Enzymatic detoxification of chlorinated phenols. PhAytopathologische Zeitschrift 47:

8—83.

Maas GeesterANuS, M. A., 1962. Hyphal structures in Hydnums. Persoonia 2: 377—405.

Maas Geesteranus, M. A., 1964. Notes ou Hydnums II. Persoonia 3: 155—192.

Macrakg, R., 1967. Pairing incompatibility and other distinctions among Hirschioporus (Polyporus)

abietinus, H. fusco-violaceus and H., laricinus. Can J. Bot. 45: 1371—1398.

Macrag, R. & Aosuima, K., 1966. Hirschioporus (Lenzites) laricinus and its synonyms L. abietes,

L. ambigua and L. pinicola. Mycologia 58: 912—926,

McKay, H. H., 1959. Cultural basis for maintaining P. cinnabarinus and P. sanguineus as two

distinct species. Mycologia 51: 465—473.

McKay, H. H., 1962. Interfertility studies of Polyporus palustris and other brown rot species

occurring on conifers and hardwoods. Plant Disease Reporter 46: 26—29.

McKay, H. H., 1967. Cultural studies of Polyporus meliae and two similar species. Mycologia

59: 1050—1058.

McKay, H. H. & Lenrz, P. L., 1960. Descriptions of some fungi associated with forest tree decay

in Colorado. Mycopath. et Mycol. Appl. 13: 265—286.

McKeen, C. G., 1952. Studies of Canadian Thelephoraceae. IX. A cultural and taxonomic study

of three species of Peniophora. Can Jj. Bot. 30: 764—787.

Mounce, I., 1929. Studies in Forest Path. II. The biology of Fomes pinicola (Sw.) Cooke. Dom.

Canada Dept. Agric. Bull. No. 111. New series.

Mounce, I. & Macrep, R., 1936. The behaviour of paired monosporous mycelia of Lenzites sepiaria

(Wulf.) Fr., Lenzites trabea (Pers.) I'r., Lenzites thermophila Falck and Trametes americana

Overh. Canad. J. Res. C, 14: 215—221

Mounce, I. & Macrea, R., 1937. The behaviour of paired monosporous mycelia of Fomes roseus

(Alb. and Schw.) Cooke and Fomes subroseus (Weir) Overh, Canad. J. Res. C, 15: 154—161.

Mounce, I. & Macrea, R., 1938. Interfertility phenomena in Fomes pinicola. Canad. J. Res.

C, 16: 354—376.

Murritt, W. A., 1903. A historical review of the genera of the Polyporaceae. J. Mycol. 9: 93.

Murritt, W. A., 1905. The Polyporaceae of North America XI. A synopsis of the brown pileate

species. Bull. Torr. Bot. Cl. 32: 353-—372.

Murrityt, W. A., 1907a. Some Philippine Polyporaceae. Bull. Torr. Bot. Cl. 34: 468.

Murrity, W A., 1907b. Polyporaceae. North American Flora 9 (1): 1—72.

Murritit, W. A., 1908. Polyporaceae. North American Flora, 9(2): 73—131.

Nana, P. M., 1957. Sexuality of Trametes cingulata Berk. Indian J. Mycol. 3: 36.

Noses, M. K., 1943. A contribution towards a clarification of the Trametes serialis complex.

Canad. J. Res. C, 21: 211—234.

Noses, M. K., 1948. Studies in forest pathology WI. Identification of cultures of woodrotting

fungi. Canad. J. Res. C, 26: 281—431.

Noses, M. K., 1953. Studies in wood inhabiting Hymenomycetes, I. Odontia bicolor. Can. J.

Bot. 31: 745—749.

Noses, M. K., 1956. Studies in wood-inhabiting Hymenomycetes. III. Stereum pini and species

of Peniophora Sect. Coloratae on conifers in Canada. Can. J. Bot. 34: 104—130.

Nostes, M. K., 1958a. A rapid test for extra-cellular oxidase in cultures of wood-inhabiting

Hymenomycetes, Can. J. Bot. 36: 91—99.

Noses, M. K., 1958b. Cultural characiers as a guide to the taxonomy and phylogeny of the

Polyporaceae. Can. J. Bot. 36: 883—926.

Nostes, M. K., 1965. Identification of cultures of wood-inhabiting Hymenomycetes. Can. J. Bot.

43: 1097—1139.

Noses, M. K., 1967. Conspecifity of Basidioradulum (Radulum) radula and Corticium

hydnans. Mycologia 59: 192—211.

Noses, M. K. & Frew, B. P., 1962. Studies in wood-inhabiting Hymenomycetes. V. The genus

Pycnoporus Karst. Can. J. Bot. 40: 987—1016.

Nostes, M. K., Macrea, R. & Tomtin, B. P., 1957. Results of interfertility tests on some species

of Hymenomycetes. Can. J. Bot. 35: 377—387.

326

Overnotts, L. O., 1953. Polyporaceae of the United States, Alaska and Canada, University of

Michigan Press, Ann Arbor, Mich.

Pantipou, M. E., 1961. Culture studies of the Boletaceae: Gyrodon merulioides and four species

of Boletinus. Can. J. Bot. 39: 1149—1162.

Panripou, M. E. & Groves, J. W., 1966. Cultural studies of Boletaceae. Some species of

Suillus and Fuscoboletinus. Can. J. Pot. 44: 1371—1392.

Parourtiarp, N., 1887. Les Hyménomyceres d’Europe. Paul Klincksieck, Paris.

Parouritiarp, N., 1900. Essai taxonomique sur les familles et les genres des Hymenomycetes.

Lucian Declume, Lous-le-Saunier.

Patourtiarp, N. & Hartor, P., 1900. Enumeration des Champignons recoltes par M. A. Chevalier

au Sénégal et dans le Soudan occidental. Journ. de Botanique 14: 234—244.

Prerersen, D. H., 1961. Polyporus spp. associated with wood decay of living peach trees in

South Carolina. Plant Dis. Reptr. 45: 186—189.

Pitat, A., 1936. Atlas des Champignons de l'Europe. Ill. Polyporaceae. Prague.

ee 1946. In “New Species of Poria,?’ OVERHOLTS, L. O. & LOWE, J. L. Mycologia 38:

206.

A 1952. Polyporaceae, contribucao para a sua bio-taxonomica. Mem. Soc. Broter.

» 1—195.

Pinto-Lopes, J. & Faruina, M., 1950. The presence and absence of clamp connections in the

species of Polyporaceae. Rev. Fac. Sci. Lisboa, 2a Ser. C., Ci. Nat. 1: 39—51

Cee L., 1886. Enchiridion Fungorum in Europa media et praesertim in Gallia vigentium.

aris.

Racas, M. A., 1953. Taxonomic studies in the Hydnaceae with reference to their hyphal systems.

Mycologia 45: 941—946.

Raper, J. R., 1953. Tetrapolar sexuality. Quart. Rev. Biol. 29: 23—259.

Raper, J. R., 1954. Life cycles, sexua!tity and sexual mechanisms in the fungi. Sex in micro-

organisms. AAAS Publication, Washington, 5, D.C.

Rea, C., 1922. British Basidiomyceteae. Cambridge.

Rersuauce, L. D. & Procror, E. M., 1936. The diagnosis of some wood-destroying Australian

Basidiomycetes by their cultural characters. Proc. Roy. Soc. Vict. 48: 105—123.

Rew, D. A., 1959. The genus Cymatoderma (Cladoderris). Kew Puli. 13: 518—530.

Rew, D. A., 1962. Notes on fungi which have been referred to the Thelephoraceae sensu lato.

Persoonia 2: 109—170.

Rei, D. A., 1963. Fungi venezuelani: Vf. New and interesting records of Australasian fungi.

Podoserpula Reid; a new genus based on Craterellus pusio Berk. (C. multiplex Cooke ex.

Massee). Kew Bull. 16: 437—445.

Rew, D. A., 1965. A monograph of the stipitate stereoid fungi. Beiheft zur Nova Hedwigia 18:

1—388 and 58 plates.

Ripeway, R., 1912. Color standards and color nomenc’ature. Washington D.C,

Ropax, H., 1942. Cultural studies in some Norwegian wood-destroying fungi. Meddelelse Nr.

25. Fra Vestlandets Forstlige Forksstation, Bergen, 1—248.

Sarkar, A., 1959. Studies in wood-inhabiting Hymenomycetes IV. The genus Coriolellus Murr.

Can. J. Bot. 37: 1251—1270.

Savitz, D. B. O., 1954. Cellular mechanics, taxonomy and evolution in the Uredinales and

Ustilaginales. Mycologia 46: 736—761.

Savite, D. B. O., 1955. A phylogeny of the Basidiomycetes. Can. J. Bot. 33: 60—104.

Sen, M. & Sencar, H. S., 1967. Heierothallism and oxidase reaction in twelve species of Indian

polypores. Can. J. Bot. 45: 635—638.

Sincer, R., 1962. The Agaricales in modern taxonomy. Second, fully revised edition. J. Cramer,

Weinheim.

Stysu, A., 1960. The genus Peniophora in New York State and adjacent regions. State University

College of Forestry at Syracuse University, Technical Publication No. 83.

Smiru, A. H., 1966. The hyphal structure of the basidiocarp. Chapter 6. The Fungi, an advanced

treatise, 2, 151—177. G. C. Ainsworth and A. S. Sussman, Editors, New York Academic Press.

Smirn, G., 1960. An introduction to industrial mycology. 5th Ed., p. 274. London: Edward

Arnold.

Snect, W. H., 1922. Studies of certam fungi of economic importance in the decay of building

timbers. United States Dept. Agr. Bull. No. 1053, 1—47.

Tarpor, P. H. B., 1951. Studies of some South African resupinate Hymenomycetes. I. Bothalia

6: 1—116.

Tatsor, P. H. B., 1954a. Micromorpholoxy of the lower Hymenomycetes. Bothalia 6: 249—299.

Tarpot, P. H. B., 1954b. The genus Stercum in South Africa. Bothalia 6: 30—338.

Tatpor, P. H. B., 1958a. The genera Craterellus, Cymatoderma (Cladoderris) and Thelephora in

South Africa. Bothalia 7: 117—130.

Tarsor, P. H. B., 1958b. Studies of scme South African resupinate Hymenomycetes. Part II.

Bothalia 7: 131—187.

Sil

Terxerra, A. R., 1956. Metodo para estude das hifas do carpoforo de fungus Polyporaceos. Sec.

Agr. Est. Sao Paulo, Inst. Bot. 1—23

TEIXEIRA, As R., 1958. Studies on microstructure of Laricifomes officinalis. Mycologia 50:

671— 676.

TerxetrA, A. R., 1960. Characteristics of the generative hyphae of polypores of North America

with special reference to the presence or absence of clamp connections. Mycologia 52: 30—40.

TerxerraA, A. R., 1962a. Microestraturas do basidiocarps e sistematica do genero Fomes Kickx.

Rickia 1: 13—93.

Terxerra, A. R., 1962b. The taxonomy of the Polyporaceae. Biological Reviews 37: 51—81.

Teixeira, A. R. & Rocers, D. P., 1955. Aporpium, a polyporoid genus of the Tremellaceae.

Mycologia 47: 408—416.

Terra, P., 1953. Determination de la polarité sexuelle de trente espéces de Basidiomycetes sapro-

phytes. Compt. Rend. Acad. 236: 115—117.

Teston, D., 1953a. Etude de la différenciation des hyphes chez les Polypores dimidiés de la flore

francaise. I. Historique. Ann. de l'Université de Lyon, Sect. C 7, 11—23.

Teston, D., 1953b. Etude de la diffécenciation des hyphes chez les Polypores dimidiés de la

flore francaise. II. Descriptions et figures. Bull. Soc. Nat. d’Oyonnax 7: 80—110.

VanpENpRIES, R., 1922. Recherches sur le sexualité des Basidiomycétes. Comp. rend. Soc. Biol.

86: 513.

VanpenpriEs, R., 1923. Nouvelles recherches sur le sexualité des Basidiomycétes. Bull. Soc. Roy.

Bot. Belg. 56: 1—25.

VANDENDRIES, R., 1924. Contribution nouvelle a l’etude de la sexualité des Basidiomycetes. La

Cellule 35: 129—155.

VaNDENDRIES, R., 1933. Nouvelles investigations dans le domaine sexuel des Hyménomycétes.

Bull. mycol. soc. France, 49: 130—165.

VANDENDRIES, R. & Bropig, H. J., 1933. Nouvelles investigations dans le domaine de la sexualité

des barrages sexuelles. La Cellule 42: 165—209.

Van DER Bry, P. A. 1922a. A contributior to our knowledge of the Polyporaceae of South Africa.

S. Afr. J. Sci. 18: 246—293.

Van DER Bry1, P. A., 1922b. A host list of Polyporaceae occurring in the Union of South Africa.

Kew Bull. 177—182.

Van ver Brigit, P. A. 1922c. Fungi of the Stellenbosch district and immediate vicinity. Trans.

Roy. Soc. S. Afr. 10: 281—288.

Van DER Biyt, P. A., 1924. Descriptions of additional South African Polyporaceae. S. Afr. Journ.

Sci. 21: 308—313.

Van ver Bryt, P. A., 1925. Preliminary list of fungi from Knysna and surroundings in the

author’s herbarium. S. Afr. Journ. Sct. 22: 191—196.

Van ver Bit, P. A., 1926. List of fungi recorded from the Pietersburg-Tzaneen area of the

Transvaal. S. Afr. Journ. Sci. 23: 285—289.

Van DER WestHUIZEN, G. C. A., 1958. Studies of wood-rotting fungi. I, Cultural characters of

some common species. Bothalia 7: 83—107.

Van ver Westuuizen, G. C. A., 1959. Polyporous sulphureus, a cause of heartrot of Eucalyptus

saligna in South Africa. J. S. Afr. For. Ass. 33: 53—56.

Van ver Westuurzen, G. C. A., 1963. The cultural characters, structure of the fruit-body and

type of interfertility of Cerrena unicolor (Bull. ex. Fr.) Murr. Can. J. Bot. 41: 1487—1499.

WakerleLp, E. M,, 1948. Taxonomic problems in Hymenomycetes. Trans. Brit. Mycol. Soc.

31: 152—160.

WaLeK-CzerNECKA, A., 1933. Sur les champignons destructeurs de traverses de chemins de fer

en Pologne. Acta Soc. Botan. Poloniae 10: 179—290.

Weir, J. R., 1923. Fomes roseus (A. & S.) Cke. and Trametes subrosea nom. novum. Rhodora

25: 214220.

Wexpen, A. L., 1960. The genus Cymatoderma in the Americas. Mycologia 52: 856—876.

Weresus, L. K. & Gipson, S., 1960. Stereum pini in North America. Can. J. Bot. 38: 833—867.

Wuirenouse, H. L. K., 1949. Heterothallism and sex in fungi. Biol. Rev. 24: 416—447.

Yen, H. C., 1950. Contribution a l'étude de la sexualité et du mycélium des Basidiomycétes

saprophytes. Ann. Univ. Lyon Sci. Sect. C. (1949): 1—131.

329

Bothalia, 10, 2: 329—334

Somatic Nuclear Division in Stemphylium botryosum

K. T. van Warmelo*

ABSTRACT

Nuclear division in an isolate of Stemphylium botryosum obtained from lucerne was

investigated using the HCI-Giemsa technique. Vegetative mycelium was predominantly

monokaryotic. Chromosome counts at metaphase gave a tentative haploid number of six.

Six chromosomes could again be distinguished at late anaphase. Migration of nuclei between

hyphae was observed. The conclusion is drawn that somatic divisions in this fungus are

strictly mitotic.

INTRODUCTION

This investigation on the somatic nuclear division in Stemphylium botryosum

was undertaken to investigate the mechanism and regularity of genome replications

at the divisions. This would give an indication of the possible degree of aneuploidy

and the stability of specific genomes which could be useful in any future investi-

gation of the pathogenicity of this fungus.

The genus Stemphylium is considered to be closely related to the genus

Alternaria and a comparison of their respective chromoscme numbers and sizes

could perhaps demonstrate the closeness of this relationship.

Hartmann (1964) investigated nuclear divisions in Alternaria tenuis and found

that these followed a typical mitotic sequence. The haploid chromosome number

was determined as five.

Despite many early investigations on somatic divisions in fungi (Olive, 1953;

Hrushovetz, 1956) it is only recently that close attention has been given to, and

success attained with, the fine structure of dividing somatic fungal nuclei.

Several possible mechanisms of division have been put forward by workers

on many different fungi. These have included amitosis (Bakerspigel, 1961, 1962;

Robinow, 1957a, 1957b; Saksena, 1961), atypical mitosis by means of a nuclear

filament (Dowding & Weijer, 1961; Dowding, 1966; Weijer & Weisberg, 1966) and

a variant of mitosis (Aist & Wilson, 1967) the interpretation of which has since

been modified (Aist & Wilson, 1968).

Typical mitosis has, in contrast, been found in different genera by different

workers (Somers, Wagner & Hsu, 1960; Hall, 1963; Ward & Ciurysek, 1961,

1962; Hartmann, 1964; Rogers, 1965; Hosford & Gries, 1966; Shatla & Sinclair,

1966; Knox-Davies, 1966, 1967).

* Plant Protection Research Institute, Private Bag 134, Pretoria. Present address: Dept.

of Botany, Rand Afrikaans University, P.O. Box 524, Johannesburg.

330

Ward & Ciurysek (1962) summarised and discussed the various mechanisms

of somatic division put forward by many different workers, both from the logical

and from the factual point of view and concluded that ‘*. . . the evidence .. .

constitute strong grounds, therefore, for the conclusion that the somatic nuclei

of fungi divide in the same manner as those of higher organisms.”

MATERIALS AND METHODS

The strain of Stemphylium botryosum used in this investigation was isolated

from lucerne. The fungus was maintained on Difco malt agar. Best growth was

obtained at 25°C while fairly satisfactory growth was seen at 16-17°C.

Unfortunately, however, a noticeable decrease in sporulation vigour at each

successive sub-culture occurred until all cultures were virtually sterile. Even

transfers using conidia as inoculum failed to regenerate the original capacity for

sporulation.

The nuclei were stained with the HCI-Giemsa stain used on a large number

of fungi by many different workers (Hrushoveiz, 1956; Ward & Ciurysek, 1961,

1962; Rogers, 1965; van Warmelo, 1966; Knox-Davies. 1966, 1967). Best results

were obtained by taking blocks of agar and mycelium cut from a culture through

the various solutions instead of using macerated air-dried mycelium (Ward &

Ciurysek, 1962) or mycelium on Cellophane (Roane, 1952).

In the following descriptions of the nuclei, the nomenclature for the nuclear

status of cells will be according to Jinks & Simchen (1966).

RESULTS

Interphase

Vegetative mycelium with the nuclei in interphase was seen to be predominantly

monokaryotic (Fig. 1). The nuclei were large, ellipsoid, usually centrally placed

in the cells and showing little or no structural differentiation. These nuclei stained

very well. Not infrequently, however, dikaryotic mycelial cells could be found

interspersed between the monokaryotic cells (Fig. 2). These nuclei were similar

in size, shape, degree of visible differentiation and staining intensity to the nuclei

illustrated in Fig. 1. The dikaryotic condition was often associated with branched

or anastomosed cells. Frequently, however, two nuclei could be found in cells where

reasons for their presence were not immediately apparent.

Prare 1.—All figures are at a magnification of 3500x.

Fig. 1. Monokaryotic mycelium with nucleus in interphase.

Fig. 2. Dikaryotic mycelium with nuclei in interphase.

Fig. 3. Very early prophase nucleus showing structural differentiation.

Fig. 4. Prophase nucleus showing condensation of chromosomal material and nucleolus.

Fig. 5. Prophase nucleus showing weakly staining chromosomes.

Fig. 6.. Metaphase chromosomes.

Fig. 7. Early anaphase. :

Fig. 8. Anaphase with chromosomal material arranged on the outside of the spindle.

Fig. 9. Anaphase at further stage than Fig. 8.

Fig. 10. Late anaphase showing six chromosomes in left-hand nucleus.

Fig. 11. Early telophase showing six chromosomes in left-hand nucleus.

Fig. 12. Condensed late telophase nuclei.

Fig. 13. Post-division interphase nuclei.

Fig. 14. Beaded nucleus in narrow mycelium.

Fig. 15. Nucleus migrating between hyphae.

332

Prophase

At very early prophase the nucleus became structurally differentiated (Fig. 3),

and a network of bands began to appear. The nucleus was, apparently, still

surrounded by the nuclear membrane at this stage. Occasionally the nucleolus

could be distinguished. At later prophase, nuclei showed a marked condensation

of chromosomal material (Fig. 4). The nuclear membrane appeared to be still

intact at this stage. Towards the end of prophase the chromosomes were still

not highly condensed and could be seen as thin, rather weakly staining strands

(Fig. 5). The nucleolus was often no longer visible and the nuclear membrane

had apparently disappeared.

Metaphase

At metaphase discrete chromosomes could be seen (Fif. 6). The chromosomes

were much contracted compared with prophase and stained fairly intensely. A

count at this stage gave a tentative haploid chromosome number of six.

Anaphase

At anaphase the chromosomes were highly condensed and considerably

smaller than at metaphase. In Fig. 7 the chromosome clumps have just begun

to move apart on a structure which is regarded as a spindle. Continued movement

of the chromosomes on the spindle (Fig. 8) produced a rhomboidal shape with the

most densely staining material arranged around ihe outs:de of the spindle. Move-

ment of the chromosomes continued until there was marked aggregation towards

the poles with the spindle showing a large clear central area (Fig. 9). Remnants

of the spindle could still be seen at late anaphase (Fig. 10). Movement of the

chromosomes towards the poles appeared to be unsynchronised as at the left-hand

pole the chromosomes were grouped fairly closely together and could be counted,

whereas the chromosomes towards the right-hand pole were still moving. A

count at the left-hand pole again gave a chromosome number of six.

Telophase

At early telophase the spindle was no longer visible (Fig. 11). The chromosomes

were closely grouped but still distinct. Nucleoli were not seen. A count of the

chromosomes in the left-hand daughter nucleus once more gave a total of six.

At late telophase (Fig. 12) the nuclei were highly condensed, small and often

of irregular shape. No structural differentiation could be observed.

Post-division interphase

Daughter nuclei going into interphase (Fig. 13) enlarged, became somewhat

diffuse, while irregular in shape, and stained as intensely as interphase nuclei

before division.

General

In addition to the wide mycelium in Figures 1 and 2, a narrow mycelium

was also observed, mainly at the surface of the culture medium. This narrow

mycelium (Fig. 14) was also monokaryotic but the nuclei were much elongated and

sometimes attentuated. Not much structural differentiation could be observed

but the nuclei appeared moniliform. This shape of the nucleus is regarded as

being due to the small hyphal diameter.

333

Although migration of nuclei from one cell to another along a hypha was not

observed, migration of nuclei through anastomosing hyphae, which were frequently

formed, was seen (Fig. 15). It is, therefore, highly likely that migration of nuclei

along individual hyphae can occur as well.

Occasionally filamentous nuclei were observed. These showed varying

numbers of granular thickenings and were very similar to the thread-like nuclei

described by Dowding (1966).

Discussion

The staining time was considerably longer than that recommended by Ward

& Ciurysek (1962) but was found to be the minimum time acceptable. After being

stained for three hours mycelial nuclei were barely visible and the material was

left to stain overnight. The fact that individual chromosomes could be distinguished

at several stages is adequate proof that the staining time was not too long.

It is interesting to note the close similarity between the chromosome number

of Alternaria tenuis (5) reported by Hartmann (1964) and the chromosome number

(6) reported here for Stemphylium botryosum. In the absence of more detailed

chromosomal data, however, no further inferences as to the relationship between

these genera can be drawn.

The regularity of the divisions appeared to be high, which led to a low

incidence of aneuploidy. This means that the stability of a specific genome will

be high, subject of course to heterokaryotic selection.

Although occasional structures similar to the filament described by Dowding

(1966) were seen they were not considered to be of any great importance, mainly

because of their scarcity. In the face of an overwhelming number of figures

suggestive of true mitosis, the “filaments” were considered as artifacts or transient

chromosomal arrangements.

Ward & Ciurysek (1962) formulated the criteria for mitosis, i.e. demonstrable

chromosomes, their alignment on a metaphase plate and the separation of

chromatids to daughter nuclei. It is considered that, in the investigation reported

here, these requirements were met, viz. chromosomes were demonstrated at several

stages of division, the alignment on a metaphase plate was perhaps not shown but

was at least suggested and movement of chromatids on a spindle was seen. It is,

therefore, believed that the somatic divisions in Stemphylium botryosum can be

accepted to be strictly mitotic.

REFERENCES

Ast, J. R. & Witson, C. L., 1967. Nuclear behavior in the vegetative hyphae of Ceratocystis

fagacearum (Bretz) Hunt. Am. J. Bot. 54: 99—104.

Aist, J. R. & Witson, C. L., 1968. Interpretation of nuclear division figures in vegetative hyphae

of fungi. Phytopathology 58: 876—877. ae. E ;

BaKersPIGEL, A., 1961. Vegetative nuclear division in the Ascomycete Ophiostoma fimbriata.

Cytologia 26: 42—49.

BakersPIGEL, A., 1962. The nucleolus in vegetative hyphae of Neurospora crassa. Can. J. Microbiol.

8: 271—272.

Dowprnc, E. Sttver, 1966. The chromosomes in Neurospora hyphae. Can. J. Bot. 44: 1121—1125.

Dowpnc, E. S. & Weer, J., 1961. Mitosis in Neurospora and Gelasinospora. I, Genetics,

Princeton 32: 339—351. f : ane

Hatt, R., 1963. Cytology of the asexual stages of the Australian brown rot fungus Monilinia

fructicola (Wint.) Honey. Cytologia 28: 187—193.

Hartmann, G. C., 1964. Nuclear division in Alternaria tenuis. Am. J. Bot. 51: 209—212.

Hosrorp, R. M. & Gries, G. A., 1966. The nuclei and parasexuality in Phymatotrichum omnivorum.

Am. J. Bot. 53: 570—579.

334

Hrusnovetz, S. B., 1956. Cytological studies of ascus development in Cochliobolus sativus. Can.

J. Bot. 34: 641—651.

Jinks, J. L. & Stwcuen, G., 1966. A consistent nomenclature for the nuclear status of fungal

cells. Nature, Lond. 210: 778—780.

Knox-Davies, P. S., 1966. Nuclear division in the developing pycnospores of Macrophomina

phaseoli. Am J. Bot. 53: 220—224.

Knox-Davies, P. S., 1967. Mitosis and aneuploidy in the vegetative hyphae of Macrophomina

phaseoli. Am. J. Bot. 54: 1290—1295.

Ouive, L. S., 1953. The structure and behavior of fungus nuclei. Bot. Rev. 19: 439—586.

Roane, C. W., 1952. A method for preparing fungi for cytological studies. (Abstr.) Phytopatho-

logy 42: 480.

Rosinow, C. F., 1957a. The structure and behavior of the nuc!ei in spores and growing hyphae

of Mucorales. I. Mucor hiemalis and Mucor fragilis. Can. J. Microbiol. 3: 771—789.

Rosinow, C. F., 1957b. The structure and behavior of the nuclei in spores and growing hyphae

of Mucorales. II. Phycomyces blakesleeanus. Can. J. Microbiol. 3: 789—791.

Rocers, J. D., 1965. The conidial stage of Coniochaeta ligniaria: Morphology and cytology.

Mycologia 57: 368—378.

Saxsena, H. K., 1961. Nuclear structure and division in the mycelium and basidiospores of

Ceratobasidium praticolum. Can. J. Bot. 39: 749—756.

SHatia, M. N. & Srvcvair, J. B., 1966. Rhizoctonia solani; Mitotic division in vegetative hyphae.

Am. J. Bot. 53: 119—123.

Somers, Carotyn E., Wacner, R. P. & Hsu, T. C., 1960. Mitosis in vegetative nuclei of Neurospora

crassa. Genetics, Princeton 45: 801—810.

Van Warne o, K. T., 1966. The cytology of Mycosphaerella pinodes. Bothalia 9: 195—202.

Warp, E. W. B. & CruryseK, KatHLeen W., 1961. Somatic mitosis in a Basidiomycete. Can. J.

Bot. 39: 1497—1503.

Warp, E. W. B. & Ciurysex, K. W., 1962. Somatic mitosis in Neurospora crassa. Am. J. Bot.

49: 393—399.

Weryer, J. & Weisperc, S. H., 1966. Karyokinesis of the somatic nucleus of Aspergillus nidulans.

I. The juvenile chromosome cycle (Feulgen staining). Can. J. Genet. Cytol. 8: 361—374.

WwW

Bothalia 10, 2 : 335-339

Conidial Nucleation in Stemphylium botryosum

K. T. van Warmelo*

ABSTRACT

Conidial nucleation in an isolate of Stemphylium botryosum obtained from lucerne

was investigated using the HCI-Giemsa technique. Conidiophores arose as anucleate buds

into which somatic nuclei migrated. After attaining varying stages of complexity the terminal

cell, which was strictly monokaryotic, developed inte a conidium. Division of the young

conidial nucleus and subsequent cell wall formation resulted in a multicellular conidium.

Conidial cells contained either one, two or four nuclei. As all the conidial nuclei are

derived from the single nucleus of the terminal conidiophore cell, the whole conidium

i pouokatyous: There is thus no mechanism for the perpetuation of heterokaryons through

the conidia.

INTRODUCTION

Stemphylium botryosum is a common fungus of worldwide distribution which

is found on a wide variety of host plants. This study was undertaken to investigate

the transmission of genotypes through the conidia.

Two closely similar genera, Helminthosporium spp. and Alternaria tenuis

(Hughes, 1953) were investigated by different workers. Hrushovetz (1956) showed

that, in Helminthosporium sativum, heterokaryosis could be perpetuated through

the conidia, as several possibly dissimilar nuclei entered the developing conidium

and continued to divide within it. Knox-Davies & Dickson (1960) showed that

the same mechanism operated in H. turcicum. Although the same mechanism

for the perpetuation of heterokaryons did exist in Alternaria tenuis (Hartmann,

1966), it was, however, also possible for only one nucleus to migrate into the young

conidium. All the nuclei within a conidium would, therefore, be identical and the

conidium would be homokaryotic.

Van Warmelo (1970) showed that nuclear migration between hyphae of

Stemphylium botryosum could occur. There is thus a mechanism for the

production and cytoplasmic maintenance of heterokaryons.

MATERIALS AND METHODS

The strain of Stemphylium botryosum used in this study was isolated from

lucerne and is the same isolate as was used in a previous investigation on somatic

divisions (van Warmelo, 1970). Cultivation of the fungus and staining of the

nuclei was done in the manner described before. The use of agar pieces cut from

a culture instead of air-dried mycelium (Ward & Ciurysek, 1962) or Cellophane

* Plant Protection Research Institute, Private Bag 134, Pretoria. Present address: Dept.

of Botany, Rand Afrikaans University, P.O. Box 524, Johannesburg.

337

Prates 1 and 2.

Fig.

Ile

Anucleate bud arising from dikaryotic hyphal cell (2000x).

Single nucleus migrating into a bud (3000x).

Two nuclei migrating into a bud (3000x).

Bud with central nucleus (3000x).

Pigmented secondary branch arising from primary branch (2000x).

Maticeliular pigmented secondary conidiophore with terminal conidial primordium

(2000x).

Multicellular pigmented tertiary conidiophore (2000x).

Primary branch developing into a pigmented conidiophore (2000x).

Conidiophore with terminal nucleus in prophase (ZO0Ox).

Bicellular developing conidium (2000x).

Young multicellular conidium (2000x).

Uninucleate conidial protoplasts (2000x).

Conidial protoplast with nucleus in early prophase (2000x).

Late prophase or metaphase nucleus (2000x).

Dikaryotic conidial protoplast (2000x).

Dikaryotic protoplast with nuclei in prophase (2000x).

Quadrinucleate conidial protoplast (2000x).

Conidial protoplast with four nuclei in interphase (2000x).

Squashed conidiophore showing protoplasts with stained nuclei (2000x).

338

films (Roane, 1952) was found to be more convenient as sporulating areas could

be identified on the surface of the agar and selectively removed for examination

after staining.

In the following descriptions the nomenclature for the nuclear status of cells

is according to Jinks & Simchen (1966).

RESULTS

Branch formation

Hyphal branches originated from relatively unpigmented hyphae as anucleate

lateral buds from individual cells which usua!ly contained two nuclei (Fig. 1).

With continued elongation of the bud, one or occasionally both nuclei migrated

into the developing hypha (Fig. 2, 3, 4). A basal septum then formed to cut off

the newly formed mono- or dikaryotic cell from its parental cell.

Subsequent development could give rise to either a new hypha or to a

conidiophore.

Development of the conidiophore

Conidiophores were found to develop in a variety of ways. The primary

branch described above could develop into a pigmented conidiophore of which

the terminal cell became the conidium (Fig. 8).

The primary branch could, however, give :ise to a secondary branch which

could then develop into a pigmented conidiophore with an apical developing

conidium (Fig. 5, 6).

Instead of developing into a conidiophore a secondary branch could form

a tertiary branch which would become pigmented and develop into a conidiophore,

as described above (Fig. 7).

At a certain stage, for which no definite stimulus was identified, the single

nucleus in the terminal cell of the conidiophore, the young conidium, began to

divide. It can be seen that cells of the conidiophore became dikaryotic, while the

young conidium, with its nucleus in prophase (Fig. 9), remained monokaryotic.

After division of the young conidial nucleus a septum was formed which separated

the two daughter nuclei (Fig. 10}. This process of nuclear division and septum

formation was repeated until a multicellular monokaryotic homokaryotic conidium

was formed (Fig. 11, 12). As development of the conidium progressed further

nuclear divisions took place which were not followed by septum formation. (Fig.

13, 14). These divisions gave rise to dikaryotic cells (Fig. 15). A further division

could follow to give rise to quadrinucleate cells (Fig. 16, 17, 18). No higher

nuclear number per cell was seen during the savestigation.

Not all the cells of a single conidium contained four nuclei at maturity and

the conidium was thus a mono-, di-, multikaryotic homokaryon.

The heavily pigmented walls of the conidiophores and conidia made observation

of the nuclei extremely difficult. It was found, however, that on squashing, the

cell walls ruptured and released the intact protoplasts with the clearly visible

stained nuclei (Fig. 16, 19).

339

DISCUSSION

From the results it is evident that the mature conidium of Stemphylium

botryosum contains several cells, each with a variable number of nuclei. All

these nuclei are, however, genetically identical as they are all derived from the one

nucleus present in the terminal cell of the conidiophore. The mechanism of

nucleation resembles that of Alternaria tenuis (Hartmann, 1966) except that only

one nucleus was present in the young conidium. Heterokaryosis cannot, therefore,

be perpetuated through single conidia.

It is not known whether these differences in nucleation are of any taxonomic

or phylogenetic importance.

The reason for the variable number of nuclei in the conidial cells must also

remain unexplained. Multinucleate cells in both sexual and asexual spores have

been reported in a number of different fungi (Carr & Olive, 1958; Hall, 1963;

Knox-Davies, 1966; van Warmelo, 1966; Rogers, 1967).

Nuclear migration between hyphae of Stemphylium has already been demon-

strated (van Warmelo, 1970). It is, therefore, probable that the mycelium is

heterokaryotic. Each conidium, with the exception of the rare parasexual diploid

or aneuploid nucleus, would contain haploid nuclei of one genotype. This would

also explain the observed reduction in sporulation vigour if the genotype favouring

sporulation differs from the genotype favouring growth in culture. Continued

culture would thus be, in effect, an unconscious selection for the “‘culture genotype.”

REFERENCES

Carr, A. J. H. & Onive, L. S., 1958. Genetics of Sordaria fimicola. Il. Cytology. Am. J. Bot.

45: 142—150.

Hatt, R., 1963. Cytology of the asexual stage of the Australian brown rot fungus Monilinia

fructicola (Wint.) Honey. Cytologia 28: 181—193.

Hartmann, G. C., 1966. The cytology of Alternaria tenuis. Mycologia 58: 694—701.

Hrusuovetz, S. B., 1956. Cytological studies of Helminthosporium sativum. Can. J. Bot. 34:

321—327.

Hucues, S. J., 1953. Conidiophores, conidia and classification. Can. J. Bot. 31: 577—659.

Jinks, J. L. & Simcuen, G., 1966. A consistent nomenclature for the nuclear status of fungal

cells. Nature, Lond. 210: 778—780.

Knox-Davies, P. S., 1966. Nuclear division in the developing pycnospores of Macrophomina

phaseoli. Am. J. Bot. 53: 220—224.

Knox-Davies, P. S. & Dickson, J. G., 1960. Cytology of Helminthosporium turcicum and _ its

ascigerous stage Trichometasphaeria turcica. Am. J. Bot. 47: 328—339.

Roane, C. W., 1952. A method for preparing fungi for cytological studies. (Abstr.) Phytopathology

42: 480.

Rocers, J. D., 1967. Hypoxylon multiforme: cytology of the ascus. Mycologia 59: 295—305.

Van Warme.o, K. T., 1966. The cytology of Mycosphaerella pinodes. Bothalia 9: 195—202.

Van Warme.o, K. T., 1970. Somatic nuclear division in Stemphylium botryosum. Bothalia

(in press).

Warp, E W. B. & Crurysex. K. W., 1962. Somatic mitosis in Neurospora crassa. Am. J. Bot.

49: 393—399.

341

Bothalia 10, 2 : 341-345

Some Members of the Sphaeropsidales from

South Africa

Karel Cejp*

ABSTRACT

Seventeen isolates of Sphaeropsidaceous fungi from South Africa are described. Of

these, nine are described as new species: Phyllosticta cussoniac, P. vangueriae, Ascochyta

grewiae, Septoria combretiana, S. cucleac, S. lanneae, S. setariaec, S. transvaaliana and S.

tripteridis.

Recently I received several specimens of Sphaeropsidales (Phyllostictales)

from the National Herbarium (PRE) (Mycological Herbarium), Pretoria, for

determination and revision. Some important works in this field are those by

Kalchbrenner & Cooke (1880), Van der Byl (1922), Verwoerd & Du Plessis

(1931), Nel (1942) and Doidge (1950).

Several of the species received were found to be unknown and are here

described as new.

Phyllosticta cussoniae Cejp, sp. nov.

Maculae orbiculares, 0.5-1 cm in diam., amphigenae, cum rubro_ limbo,

albescentes, praesertim in siccitate, paucae. Pycnitia dispersa, primum submersa,

80 — 90 wu in diam. Conidia maiore parte breviter ellipsoidea, aut globosa aut

irregulariter ellipsoidea, cum numerus oleosis guttulis minoribus, cum 2 — 3

maioribusque, in utraque parte rotundata aequali, recta, aut ad unam partem

inflexa, 4.2-5.6 x 7-12.5 (14) uw. infirme claro viridia.

Spots rounded, 0.5-1 cm. in diam., amphigenous, with red border, whitish,

especially when dry, sparse on the leaves. Pycnidia scattered, at first immersed,

black, 80-90 u in diam. Conidia mostly short and ellipsoidal, with numerous

smaller guttulae, and with 2-3 larger ones, regularly rounded at both ends, straight

or curved at one side, 4.2-5.6 x 7-12.5 (14) wu. weakly pale green.

PRE 32821 (Holotype): On leaves of Cussonia umbellifera Sond., Schagen,

Nelspruit district, Transvaal, 25. XII. 1933, leg. L. C. C. Liebenberg 3063: isotype

in herb. Dr. K. Cejp, Praha.

* Department of Mycology and Plant Pathology, Charles University, Praha 2, Benatska

2, Czechoslovakia.

342

Phyllosticta vangueriae Cejp, sp. nov.

Maculae irregulares aut fere rotundatae, amphigenae, usque 1 cm in diam.,

clare cremaceo-flavescentes, in extremitate magis obscurae. Pycnidia obscuro-

brunnea in ostiolo magis fusca, 15-20 in diam., punctiformia, copiosa, 80-90

4 in diam. Conidia breviter cylindriformia, recta, raro inflecta, cum 1 — 2 oleosis

guttuli raro eguttulata, in utraque parte rotundata, 2.5-3 x 7-8 w, claro viridia.

Spots irregular or almost rounded, amphigenous, up to 1 cm in diam., light

pale yellow, the periphery being darker. Pycnidia brown, darker at the ostiole

which is 15-20 uw in diameter, punctiform, numerous, 80-90 « in diam. Conidia

short cylindrical, straight, seldom curved, with 1-2 guttulae, seldom without, rounded

at both ends, 2.5-3 x 7-8 ut, pale green.

PRE 32874 (Holotype): On the leaves of Vangueria infausta Burch., Schagen,

Nelspruit district, Transvaal, July 1936, leg. L. C. C. Liebenberg 3544; isotype

in herb. Dr. K. Cejp, Praha. PRE 26044: Near falls at Nelspruit, Transvaal,

leg. L. C. C. Liebenberg, 1931, (with Septoria sp., conidia 2.5 x 44.5 — 60 wn).

Phyllosticta crastophila Saccardo, Michelia 1: 153 (1889); Sylloge Fung. 3: 61

(1884).

Spots indistinct, mainly oblong, about 2 x 3 mm in diam., rarely confluent

and irregular, dirty ochraceous, the edge having a broad darker border. Pycnidia

restricted into small mounds, rounded, lentiform, erumpent, black or almost

black. Conidia oblong-cylindrical, straight, rounded at both ends, with 1 or 2

guttules and several dispersed guttulae, 3.5-4(5) x 5.5-8 u, very pale green.

PRE 32080: On the leaves of Setaria sp., from Rietondale Grass Station,

Pretoria, Transvaal, March 1940, leg. L. C. C. Liebenberg 3639.

Phyllosticta richardiae Halstet, New Jersey Agric. Exp. Sta. 1893.

Large spots scattered all over the upper surface of the leaves, 2-4 cm in

diam., dark brown, irregular. Margin narrow having a dark brown border.

Pycnidia numerous on the upper surface of the spots, being densely packed in some

areas, leaving bare patches in other areas, globose, lentiform, epiphilous, dark

brown, (80)90-130(140) uw in diam. Conidia elongate-cylindrical, rounded at both

ends, one end attenuated, with (1)2 large guttulae, also with minute guttulae,

2.5-3(3.5) x 5.5-7(8.5) uw, pale green.

PRE 41655. On the leaves of Zantedeschia sp., locality not stated, 1956,

leg. V. A. Wager. Septoria callae (Lasch) Saccardo (Sylloge Fung. 3 : 569, 1884,

nomen, Bull. Soc. Mycol. de France 5:122, 1889, and Sylloge Fung. 10:382, 1892),

is described on this host genus, but is distinguished by the small spots on the leaves,

and especially by having somewhat curved stick-shaped conidia. It is known in

Europe in pare Riga (Estonia, SSSR — Aksel, 1956).

Ascochyta grewiae Cejp, sp. nov.

Maculae in foliis copiosa, clara ferrugineae, fere subrubrae, plerumque

rotundatae, raro confluentes et irregulares, saepissime 0.5 cm in diam., acute textura

incolumni divisae. Pycnidia pauca, brunnea, punctiformia, 80-90 w in diam.

Conidia cum duobus cellulis, plerumque inferior cellula parva, raro unicellularia,

cum 2 oleosis guttulis et aliquot minoribus guttulis, saepe inflectis ellipsoideis, in

parte superiori cum extremitate rotundata (2)2.8-3.5 x (5.6) 8.4-11.2 \, infirme

claro viridia.

343

Spots abundant, light ferruginous, almost sub-reddish, regularly rounded,

rarely confluent and irregular, often 0.5 cm in diam., sharply divided from the

healthy texture of the leaf, sometimes covering a greater area of the upper surface

of the leaves. Pycnidia few, dark brown, punctiform, 80-90 u in diam. Conidia

two-celled, the lower cell being usually smaller, seldom unicellular, with 2 guttulae

and several smaller ones, often curved, ellipsoidal, with rounded tips, (2)2.8-3.5 x

(5.6)8.4-11.2 u, very pale green.

PRE 25967 (Holotype): On the leaves of Grewia monticola Sond., Nelspruit,

Transvaal, 28. VIII. 1931, leg. L. C. C. Liebenberg 2773; isotype in herb. Dr. K.

Cejp, Praha.

Ascochyta chenopodii Rostrup, Botan. Tidskrift 26:311 (1905).

Diplodina chenopodii Karsten, Hedwigia 24 : 73 (1885).

Ascochyta atriplicis (Vestergr.) Diedicke, Ann. Mycol. 2 : 180 (1904).

Spots round or irregular, 1 cm in diam., light ochraceous, whitish in the centre

with a broad dark border. Pycnidia on the upper surface of the spots, sometimes

also on the lower surface, globose, umbrous, becoming black, 120-150 uw in

diam. Conidia sub-cylindrical, straight or sometimes curved, rounded at both

ends, often irregular, with 2-3 large guttulae, also many minute ones, con-

stricted, and narrowed, 2-4 x 14-22.5 u, pale brown to yellowish green.

PRE 43127: On leaves of Chenopodium murale L., in Addo National Park,

Alexandria district, C.P., July, 1962, leg. L. C. C. Liebenberg 6251.

Septoria combretiana Cejp, sp. nov.

Maculae amphigenae, rotundatae, 0.3-0.5 cm in diam., clare brunneae ferru-

gineaeque, saepe in medio magis obscurae. Pycnidia fusce ferruginea, in textura

macularum submersa, paulum ad summum emergentia, 70-90 u in diam. Conidia

bacillaeformia, magis aut infirme adunca, cum 3-5 septa, saepe cum magnis 3-5

oleosis guttulis, 3-3.5(4) x 25-65 u, infirme clare viridia.

Spots amphigenous, round, 9.3-0.5 cm in diam., light ferruginous brown, the

centre often darker. Pycnidia dark ferruginous, round, immersed, scattered, few

over the surface, erumpent, 70-90 wu in diam. Conidia bacillar, curved, or only

slightly curved with 3-5 septa, often 3-5 large guttulae, 3-3.5(4) x 25-65 u, of a

very pale green colour.

PRE 33266 (Holotype): On the leaves of Combretum erythrophyllum (Burch.)

Sond. (=Combretum glomeruliflorum var. riparium Sond.), Pelindaba, Pretoria

district, Transvaal, 30.V.1944, leg. E. M. Doidge & A. M. Bottomley; isotype in

herb. Dr. K. Cejp, Praha.

Septoria eucleae Cejp, sp. nov.

Maculae minutae, vix 2 mm in diam., irregulares, raro subrotundatae, sub-

nigrescentes, aut obscuro-brunneae, perspicuae supra parte, densae. Pycnidia

nigra, punctiformia, parva, 70-80 « in diam. Conidia_bacillaeformia, saepe

torquentia aut inflectentia, cum numerosis septis, saepissime cum 4-5, sine

oleosis guttulis, (2.5) 3-3.5 x (31)40-50(60) u, pallide viridia.

Spots minute, scarcely 2 mm in diam., irregular, seldom sub-circular, almost

black, or dark brown, distinct only on the upper surface, dense. Pycnidia black,

punctiform, small, 70-80 u in diam. Conidia bacillar, frequently curved or slightly

curved, with many septa, or with just 4-5 septa, no guttulae present, (2.5)3-3.5 x

(31)40-50(60) u, pale green.

344

PRE 32812 (Holotype). On the leaves of Euclea crispa (Thunb.) Guerke var.

crispa (—Euclea lanceolata E. Mey. ex A.D.C.), Waterkloof Ridge, Pretoria,

Transvaal, 22.X%.1928, collector unknown; isotype in herb. Dr. K. Cejp, Praha.

This species is easy to recognise by the spots.

Septoria lanneae Cejp, sp. nov.

Maculae fere rotundatae aut paulum irregulares, e textura folii erumpentes

cca O.5 cm in diam., amphigenae, obscure brunneae, sine limbo, acute dividae a

textura salva aut cum lato obscuriore limbo. Pycnidia nigra, globosa, parva,

70 wv in diam. Conidia bacillaeformia, aliqua dissimilia crassa, recta, aut saepe

distorta, cum 3-4 septis et densis magnis oleosis guttulis, 3-4.5 x 30-33.5 u, claro

infirme viridia.

Spots almost round or very slightly irregular. about 0.5 cm in diam. dark

brown, amphigenous, without a border, distinct from the healthy texture of the

leaf, or if separate, with a wide dark border. Pycnidia black, round and small,

70 wt in diam. Conidia bacillar, some unequally thick, straight or often twisted,

with 3-4 septa and large dense guttulae, 3-4.5 x 30-33.5 wt, very pale green in

colour.

PRE 32815 (Holotype): On the leaves of Lannea edulis (Sond.) Engl. (=Odina

edulis Sond.), Suidwalliskraal, Lydenburg district, Transvaal, 8.1.1935, leg. L. C. C.

Liebenberg 3367; isotype in Herb. Dr. Kk. Cejp, Praha.

Septoria setariae Cejp, sp. nov.

Maculae parviorae, longae, 1-2 mm in diam., ambiguo brunneae, obscuro

brunneo limbo praetextae, dein medio albescente. Pycnidia pauca, rotundata,

lentiformia, nigra, 80-90 « in diam. Conidia bacillaeformia, recta aut subinflecta,

quondam usque adunca, cum 3-5 indistinctis septis, sine oleosis guttulis, 2.5-3 x

(28)60-80 wu, leviter viridia.

Spots smaller, 1-2 mm in diam., brownish, surrounded by a darker border,

becoming whitish in colour in the centre. Pycnidia few in number, round, lentiform,

black 80-90 uw in diam. Conidia bacillar, straight or moderately curved, with

3-5 indistinct septa, without guttulae, 2.5-3 x (28) 60-80 u, pale green.

PRE 25866 (Holotype): On the leaves of Setaria chevalieri Stapf ex Stapf &

C. E. Hubb, Soutpansberg, N. Transvaal, 12.1.1931, leg. P. Watson; isotype Dr.

K. Cejp, Praha.

Septoria transvyaaliana Cejp, sp. nov.

Maculae copiosae in foliis, nigro-brunneae, cum angusto fere nigro limbo,

amphigenae, fere rotundatae aut irregulares, cca 2-3 mm in diam. Pycnidia parva,

70-80 « in diam., nigra pauca. Conidia extende cylindriformia, nonnulla clava,

taro directa, saepissime detorta, cum 2-3 septis atque aliquot oleosis guttulis,

(2)3-3.5 x (24)39-70 wu, claro viridia.

Spots numerous on the leaves, dark brown, with a narrow black margin, almost

round or irregular, about 2-3 mm in diam. Pycnidia small, 70-80 w in diam.,

black, few in number. Conidia elongated cylindriform, often club-shaped, seldom

straight, more frequently twisted, with 2-3 septa and several guttulae, (2)3-3.5 x

(24)39-70 w, pale green.

PRE 32895 (Holotype): On the leaves of Hypoestes verticillaris R.Br., Wolwe-

kloof, Garstfontein, Pretoria, Transvaal, 22.11.1929, leg. A. M. Bottomley; isotype

in herb. Dr. K. Cejp, Praha.

345

Septoria tripteridis Cejp, sp. nov.

Maculae irregulares, plerumque in margine folii, minus in medio, dein moderate

oblongatae, 0.5-0.8 mm in diam., amphigenae, brunneo-canae, siccicate albescentes,

in margine macularum cum lato brunneo-purpureo limbo praetexae. Pycnidia nigra,

densae, copiosae, in superiore parte folii, punctiformia, 80-90 « in diam. Conidia

oblonga cylindriformia, fere claviformia, directa, saepius moderate detortae, cum

3-6 oleosis guttulis et sine septis, dissimiliter crassa, 2.5-3(3.5) x 55-70 u, claro

infirme vriridia.

Spots irregular, usually on the edges of the leaves rather than in the middle,

moderately oblong, 0.5-0.8 mm in diam., amphigenous, brownish-grey in colour,

becoming whitish when dry, broad brownish-purple border. Pycnidia_ black,

copious, dense on the upper surface of the leaves, punctiform, 80-90 u in diam.

Conidia oblong cylindriform, straight, often moderately twisted, with 3-6 guttulae,

without septa, unequally thickened 2.5-3(3.5) x 55-70 u, of a very pale green

colour.

PRE 41849 (Holotype): On the leaves of Tripteris sp. (Asteraceae), Fauresmith

Municipal commonage, Orange Free State, April 1939, leg. L. C. C. Liebenberg;

isotype in Herb. Dr. K. Cejp, Praha.

Septoria anaxea Saccardo, Michelia | : 189 (1879); Sylloge Fung. 3:549 (1884).

Spots irregular, copious on the leaves, regularly on the upper surface, at first

ochraceous, becoming ferruginous-brown, with yellowish border, 1-2 cm in diam.

Pycnidia punctiform or lentiform, 70-100 uw in diam., with large ostiole. Conidia

bacillar, straight or slightly curved, rounded at both ends, with numerous guttulae,

with 3-5 indistinct septa, 2-2.5(3) x 28-75 u, very pale olive green.

PRE 32878: On the leaves of Senecio isatideus DC., Baviaanspoort, Pretoria

district, Transvaal, 13.111.1935, leg. L. C. C. Liebenberg 3436.

Septoria callistephi Gloyer, Phytopathology 11:50-51 (1921).

Spots amphigenous, rounded becoming irregular, brown, with a broad reddish-

brown border, 3-5 mm in diam., copious, especially on the lower leaves. Pycnidia

blackish-ferruginous, copious, occasionally dense, punctiform or lentiform, 120-140

in diam. Conidia thread-shaped, somewhat thicker, straight or moderately curved,

with 2-3 indistinct septa, 2-2.5(3) x 11-17 u, pale green.

PRE 41664: On the leaves of Chinese daisies, Callistephus chinensis Nees,

Pinetown, Natal, Dec. 1956, leg. V. A. Wager, 100.

Septoria gerberae Sydow, Ann. Mycol. 10:43 (1912).

Spots regular, circular, 3-6 mm in diam., or irregular and confluent, amphige-

nous, dark brown, with a darker brown border. Pycnidia brownish, often scattered,

occasionally in small groups. Conidia bacillar, sometimes slightly curved, straight,

with 2-3 indistinct septa, 2-2.5(3) x 14-22.5 uw, very pale green in colour.

PRE 32656: On the leaves of Gerbera jamesonii H. Bol. ex Hook f., Botanical

Research Institute Garden, Pretoria, Transvaal, 5.X11.1939, leg. A. M. Bottomley.

This species is distributed throughout South Africa: it is also Known from Europe

where the host is cultivated (Cejp & Jechova, 1967; Cejp & Deighton, 1969).

This species differs from Phyllosticta gerbericola (Ch.) Batista, which is

found in Brazil, India and Czechoslovakia (Cejp, 1965, 1966).

346

Septoria rhamni-cathart‘ca Ces. var. rhamni-saxatilis Saccardo, Mycotheca Veneta,

No. 546; Sylloge Fung. 3:482 (1884).

Spots amphigenous, ferruginous-brown, without border, fading when dry,

round, small, 0.5 cm in diam., sometimes smaller, abundant on the leaves. Pycnidia

on the upper surface of the leaves only, punctiform, 80-90 uw in diam. Conidia

bacillar, cylindrical, elongated, straight or strongly curved, club-shaped, septa not

usually present, but when present from 3-5, frequently with large guttulae, 2.5-3(3.5)

x 30-70 u, light green to hyaline.

PRE 41640: On the leaves of Phyllogeiton zeyheri (Sond.) Suesseng.

(—Rhamnus zeyheri Sond.), roadside, near Bulge River, Waterberg district,

Transvaal, 22.V1.1955, leg. L. C. C. Liebenberg.

Septoria rhoina Berk. & Curtis, North American Fungi, No. 434.

Spots amphigenous, round, 2-3 mm in diam., dense on the leaves, whitish

with black borders, especially when dry, brownish on the lower surfaces of the

leaves. Pycnidia 4-5 in the centre of the spots, black, punctiform, with a broad

ostiole, 80-100 1 in diam. Conidia elongated, vermiform, straight or twisted, some

broader than others, some only thread-like, either without septa, or with 4-5 septa,

2.8-4(5.5) x 50-85 u, very pale green to hyaline.

PRE 32728: On the leaves of Rhus sp., Schagen, Nelspruit district, Transvaal,

June 1939, leg. L. C. C. Liebenberg 3575.

Phyllosticta rhoina Kalchr. & Cooke is found on Rhus viminalis Vahl, but

differs from Septoria rhoina in the size of the conidia.

REFERENCES

AxseL, M. J., 1956. Genus Phyllosticta in Estonia. Acta Inst. Bot. V. L. Komarovii. Acad. Sc.

USSR 11:75—174.

Cryp, K., 1965. The occurrence of some Phyllosticta on ornamental plants. I. Preslia 37:330—352.

Crsp, K., 1966. The occurrence of some Phyllosticta on ornamental plants and shrubs. II. Ceska

Mykol. 20:205—214.

Ceyp, K. & Jecnovd, V., 1967. Beitrag zur Kenntnis einiger tschechoslowakischen Arten der

Gattung Septoria Fries. Acta Musei Nationalis Pragae 23, B: 101—123.

Cryp, K. & Detcnton, C. F., 1969. Microfungi III]. Some african species and redispositions of

some Hyphomycetes, mainly African. Mycol. Pap. 117: 1—131.

Doipce, E. M., 1950. The South African fungi and lichens, Bothalia 5:1—1094.

KALCHBRENNER, K. & Cooker, M. C., 1880. South African Fungi. Grevillea 9:17—34.

Net, G. C., 1942. Genera et species fungorum ex herb. P. A. van der Byl, Stellenbosch. Ann.

Univ. Stellenbosch, Je. 20, No. 2:1—110.

Van per Byt, P. A., 1922. Fungi of Ste‘lenbosch District and immediate yicinity. Trans. Roy.

Soc. S. Afr. 10: 281—288. ‘

Verwoerpb, L. & pu Puessis, S. J., 1931. Descriptions of some new species of South African fungi

and of species not previously recorded in South Africa. UI. S. Afr. J. Sci. 28 : 290—297.

347

Bothalia 10, 2 : 347-349

A New Species of Gonatobotryum from

South Africa

K. T. van Warmelo*

ABSTRACT

An isolate of Gonatobotryum from barley kernels is described as a new species,

Gonatobotryum sclerotigenum.

Early in 1969 a fungus was isolated from partially sterilized barley kernels

received from Stellenbosch. The isolate proved to be an unknown species of

Gonatobotryum Sacc. and is here described as new. Type specimens in the form

of dried-down cultures on agar have been deposited im the National Herbarium

(PRE) (Mycological Herbarium), Department of Agricultural Technical Services,

P.O. Box 994, Pretoria, South Africa.

This appears to be the first record of this genus in South Africa.

Gonatobotryum sclerotigenum van Warmelo, sp. nov.

Fie, il, 2, 3, ch Os

Hyphae fuscae, erectae, ramosae, septatae. Conidiophora ex substrato aut

hyphis vegetatis enata, ramosa, septata, 10u in diametro, spicula hyalina. Conidia

in greges aliquando conferta, acatenulata, septata aut perraro uniseptata, ellip-

soidea vel ovata, brevi pedicellata, pallide olivaceo-brunnea, 11 — 19 x 6 — 8u.

Sclerotia praesentia, pervariabilia, fere globosa, nigra.

Specimen examined: Van Warmelo, PRE 44252, (Holotype), on Hordeum spp.

kernels, from Stellenbosch, Cape, May 1969. Cultures have also been deposited

in the Centraalbureau voor Schimmelcultures, Baarn, Netherlands.

Colonies on corn-meal agar at 25°C covering the plate in approximately

five days, loosely textured and woolly, at first hyaline but eventually darkening

to light olivaceous brown, aerial mycelium abundant, reverse pale red. Mycelium

hyaline at first, darkening eventually but remaining hyaline in parts, branched,

septate, 10u in diameter. Conidiophores arising from the substrate or from aerial

vegetative hyphae, branched, becoming olivaceous brown, septate, 10. in diameter,

giving rise to hyaline spicules on which the conidia are borne. Conidia borne

in clusters at intervals along the conidiophores, non-catenate, aseptate or only

very rarely uniseptate, ellipsoid to ovate, briefly pedicellate, light olivaceous brown,

11 — 19 x 6 — 8u. Sclerotia variable in shape and size, usually globose, black,

irregularly distributed through the colony, up to 2 mm in diameter.

*Plant Protection Research Institute, Private Bag 134, Pretoria. Present address: Dept. of

Botany, Rand Afrikaans University, P.O. Box 524, Johannesburg.

348

1. Conidiophores with spore clusters (46x).

Fig. 2. Branched conidiophore showing spore cluster and growing tip (750x).

3 and 4. Sporogenous spicule showing conidia borne at different heights on the

same spicule (1500x).

. Mature conidium (2000x).

349

This species was compared with the other species of Gonatobotryum and was

found to differ from them as detailed below. The conidia of G. fuscum Sacc.

(Syll. Fung. 4 : 278, 1886), G. maculicolum (Wint.) Sacc. (Saccardo. Syll. Fung.

4 : 278, 1886), G. bahiense Batista (Ann. Soc. Biol. Pernambuco 13 : 154, 1955)

and G. indicum Munjal & Gill (Ind. Phytopath. 16 : 62, 1963) are all consider-

ably smaller than in G. sclerotigenum. The conidia of G. dichotomum Cooke &

Mass. (Saccardo. Syll. Fung. 10 : 579, 1892) are not very different from those

of G. sclerotigenum which, however, differs from G. dichotomum in having non-

dichotomously branched conidiophores. The cenidia of G. apiculatum (Peck)

Hughes (Can. J. Bot. 31 : 594, 1953) are similar in size to those of G. sclerotigenum

but differ by being borne in chains.

An additional important difference is the absence of sclerotia in all the

previously described species of Gonatobotryum.

The sporogenous spicules are extremely delicate structures which degenerate

and disappear after producing a variable number of conidia in different planes,

leaving only a raised pad-like structure on the surface of the conidiophore, around

which the spores remain clustered. The spicules can only be seen on developing

portions of the conidiophore, and then only with some difficulty.

The placing of this species in the genus Gonatobotryum justifies some comment.

The genus Gonatobotryum Sacc. (Saccardo. Syll. Fung. 4: 278, 1886) was de-

scribed as being a dark or pigmented Gonatobotrys Corda (Saccardo, Syll. Fung. 4:

169, 1886). These two genera are morphologically similar and can thus be separa-

ted only on the basis of colour. Barron (Mycologia 56: 313 — 316, 1964) discussed

the value of colour as a distinguishing criterion between the genera Stachybotrys

Corda and Hyalostachybotrys Srinivasan, and concluded that the presence of

colour as the sole difference between these two genera is unacceptable.

On the basis of the pigmented conidia and conidiophores this new fungus

is correctly placed in Gonatobotryum Sacc. It is, however, felt that, following

the same reasoning as Barron (loc. cit.), the genus Gonatobotryum Sacc. should be

reduced to synonymy with Gonatobotrys Corda. As no authentic material of

either genus was examined, however, this change is not made here.

ACKNOWLEDGEMENT

The author wishes to thank Dr. W. F. O. Marasas of this Institute, for his

interest in and assistance with the identification of this fungus.

Bothalia 10, 2 : 351-354

The Acacia Species with Glandular Glutinous Pods

in Southern Africa

lp Jal, Joss

ABSTRACT

The complex of Acacia species with glandular glutinous pods, consisting of seven

species, is endemic to southern Africa. A map showing the distribution of these species is

given. A new key to the identification of these species is provided and characters that enable

members of the complex to be distinguished from A. karroo Hayne are discussed.

The South African species of Acacia with glandular glutinous pods were

dealt with by Verdoorn in Bothalia 6 : 153-160 (1951). Recently Brenan in Kew

Bull. 21 : 480 (1968) has described yet another species with glandular pods from

Mozambique. As there is now an additional species within the complex and, as

more material and information has become available since Verdoorn’s treatment,

it seems worthwhile to re-examine the entire complex in southern Africa.

The complex now consists of seven species, namely, with species enumerated

chronologically: A. nebrownii Burtt Davy, A. borleae Burtt Davy, A. permixta

Burtt Davy, A. swazica Burtt Davy, A. exuvialis Verdoorn, A. tenuispina Ver-

doorn and A. torrei Brenan. The complex is endemic to southern Africa. The

distribution of each species may be seen in Fig. 1.

Fig. 1 reveals that no species within this complex occurs in the Orange

Free State, in Lesotho or in the Cape. The Transvaal has the greatest number

of species; of the seven within the complex only A. torrei is absent. A. exuvialis

and A. permixta are endemic to the Transvaal, A. exuvialis occurring in the

eastern portion and A. permixta in the western. A. torrei is endemic to the

Manica e Sofala region of Mozambique. A. borleae occurs in southern Mozam-

bique, the eastern Transvaal, eastern Swaziland and northern Zululand. A.

swazica occurs in the south-eastern Transvaal, in Swaziland and just south of

Abercorn Drift in northern Zululand. A. tenuispina is found in eastern Bots-

wana and in the western and central Transvaal, whilst A. nebrownii ranges from

South West Africa to Botswana and the northern Transvaal.

This complex of species is taxonomically difficult. The species are all

very closely related to one another and to A. karroo Hayne and appear to

have been derived from A. karroo ‘stock’ during earlier times. It is not clear

whether each species was independently derived from the ancestral stock or

whether some of the species have given rise to others. Each species seems to

have different habitat preferences, the different species being possibly the outcome

of differing environmental conditions acting upon a common or similar gene pool.

Some specimens cannot be referred either to A. karroo or to A. tenuispina with

. borleae

. swazica

. nebrownii

. tenuispina

. permixta ick /

. exuvialis i MyBO RISONN2 190 290 390490 Mites

. torrei Kilometer 100 50 190-200 «390 4 st 690 Kilometres

1 n z ! |

Fic. 1— The known distribution of the Acacia species with glandular glutinous pods in

Southern Africa.

certainty. For example, Codd 7047 (PRE) from north of Pienaars River (Grid

Reference: 2528 AB Pretoria) was described by the collector as ‘possibly a

hybrid between A. karroo and A. tenuispina.’’ Some plants from the Spring-

bok Flats agree vegetatively with A. tenuispina, but lack glandular pods and

are therefore hesitantly referred to A. karroo. The two species may well hybridize

but this requires careful field studies.

Apart from certain specimens referred to above as being possible inter-

mediates between A. karroo and A. tenuispina, each of the species can usually

be distinguished from A. karroo fairly easily. The species differ from A. karroo

in that they are usually small, slender plants and are smaller in all of their

parts. With the exception of A. borleae and A. torrei the remaining species have

typically fewer pinna and fewer leaflet pairs than A. karroo. The young branch-

lets, leaf petioles, rachides and rachillae are slender and not robust as in A.

karroo. Unlike A. karroo, the flowers are not aggregrated into such dense

terminal panicles, but tend often to occur in axillary fascicles along the young

stems. In addition, the pods are smaller, of a different shape and texture. The

species perhaps most likely to be confused with A. karroo are A. swazica and

A. exuvialis. However, the bark in both species is different being pale greyish

to chestnut or reddish-brown and often peeling off in strips in contrast to the

dark brownish-black or reddish-brown rough bark of A. karroo.

353

A new key to the identification of the species within this complex has

been drawn up which makes provision for A. torrei. In contrast to the key given

by Verdoorn (l.c.), where the first dichotomy was based on pod shape and

whether the glands on the pods are conspicuous or not, the emphasis in this

key is initially on vegetative characters. This is because pod shape in some of

the species is more variable than initially recognized and also because it enables

some of the species to be identified when in the flowering or even in the vege-

tative state. However, even in the presence of pods, it is felt that the vegetative

characters used in the key enable certain species to be identified far more

readily than by relying on pod characters. All species usually have flowers and

pods in various stages of maturity contemporaneously. It may be argued that

in the absence of pods it is not possible to establish whether the plant in question

is a member of the complex with glandular glutinous pods. Whilst this may

occasionally be true, it is usually possible and, certainly with a little practice.

relatively simple to establish whether or not a plant belongs to this complex.

Young branchlets densely tomentose with spreading white hairs ee A. permixta

Young branchlets glabrous, subglabrous, or occasionally very sparingly pubescent:

Mature leaves with 7-20 pinna pairs:

Leaflets glandular-punctate with distinct glands on the lower surface and on

margins giving the margin a crenate appearance; pods glabrous or subglabrous,

glands inconspicuous ..... : A. borleae

Leaflets not glandular-punctate on lower surface and on margins; pods distinctly,

although sparingly, pubescent especially on the raised portions over the sceds,

glands conspicuous ; A. torrei

Mature leaves with up to 6 pinna pairs:

Involucel at or near the base of the peduncle; usually only | pinna pair, rarely 3

A. nebrownii

Involucel at or above the middle of the peduncle:

Stoloniferous shrubs; leaflets small, up to 4.3 mm long, 1.5 mm wide; pods

straight or almost so, 4.5—8.0 mm _ wide A. tenuispina

Non-stoloniferous shrubs; leaflets larger, up to 10 mm long, 5.1 mm wide; pods

distinctly torulose or only slightly falcate, occasionally straightish; usually

broader than above:

Spines slender; midrib and veins usually very conspicuous abaxially; pods

slightly to distinctly falcate, occasionally + straight, 0.7—1.2 cm wide,

glands conspicuous; seed circular or subcircular A. swazica

Spines frequently enlarged and swollen basally; midrib and veins inconspicuous

abaxially; pods torulose, 4.5—9.0 mm wide, glands not conspicuous; seed

Cllipticge aaa nue A. exuvialis

A. permixta clearly differs from the remainder of the complex in having

young branchlets that are usually clothed with a dense indumentum of spreading

white hairs, a feature that enables the species to be distinguished even in the

vegetative state. A borleae and A. torrei differ from the remaining species in

having mature leaves with at least seven pinna pairs. A. borleae differs from

A. torrei in having leaflets that are glandular-punctate on the lower surface

and on the margins which gives an irregularly crenate appearance. However,

the margins of leaflets of A. torrei that have been subjected to the ravages of

certain insects and sometimes appear crenate must not be confused with leaflets

of A. borleae. The pods of A. borleae are usually glabrous and the glands

inconspicuous, whereas in A. forrei the pods are distinctly, although sparingly.

pubescent especially on the raised portions over the seeds, and the glands are

conspicuous. In addition, a large geographical discontinuity separates the two

species. A. permixta, A borleae and A. torrei can therefore be distinguished

from the remainder of the complex and from one another on the basis of

vegetative characters alone.

354

The position of the involucel on the peduncle is the important character

in differentiating A. nebrownii from the remaining species. The stoloniferous

habit, slender spines and small leaflets distinguish A. tenuispina from A. swazica

and A. exuvialis. In A. exuvialis the thin bark peels off, but this feature has

also been noted on some specimens of A. swazica. The spines of A. swazica

are slender, whereas in A. exuvialis they are often somewhat enlarged and the

members of each pair united basally. The conspicuous midrib and veins on

the lower surface of the leaflets provide a useful means of identification in

dried specimens of A. swazica, but it must be mentioned that the midrib and

veins are not so conspicuous in living plants. They apparently only become

conspicuous when the tissues have dried out. Pod shape also enables the two

species to be differentiated.

355

Bothalia 10, 2 : 355-358

Two New Ornithogalum Species from

South West Africa

A. A. Obermeyer

Ornithogalum candidum Oberm., sp. nov., O. rautanenii Schinz affinis, sed

plantae robustiores, foliis latioribus, perianthio majore staminum filamentis ex

appendiculatis differt.

Bulbus angustus in collo longo productus. Folia 2 (raro 3) linearia ad late

linearia ad basin vaginata, apice acuminata, 6-15 cm longa, 6-12 mm lata patentia

glauca glabra striata margine saepe undulata. Racemus simplex ad 20-florus;

scapus validus teres; bracteae minorae membranaceae; pedicelli filiformes patentes.

Perianthium candidum circa 2 cm diam., segmentis anguste oblongis. Staminum

filamenta anguste lanceolato-subulata. Ovarium ovoideum sessile, ovulus in

quoque locula 8. Capsula late obovoidea chartacea; semina applanata oblanceo-

lata nigra.

SoutH West Arrica.— 2317 Rehoboth: Biullsporier Flache. Giess 388

(PRE, holo.; M; WIND); Form Vrede. REH 433, Giess 9/10. 2417 Mariental:

between Mariental and Aris, Werdermann & Oberdieck 2268; farm Witvley

near Mariental, Giess 3878; Swartrand, + 30 miles W. of Mariental, Basson 146.

2115 Karibib: farm Nudis (Altenbrunn), Seydel 245.

Bulb narrowly ovoid, 2-4 cm in diam., produced into a neck 8-10 cm

long, outer covering papery, rough, greyish- or light brown. Leaves 2, rarely

3, linear to broadly linear, up to 15 cm long and 1.2 (-2) cm broad, clasping

below, spreading, apex acuminate often laxly curled upwards, the margins usually

wavy and involute, glaucous, glabrous, striate. Raceme simple, up to 15 cm

high and up to 20-flowered: scape terete. firm, bracts small, ovate-acuminate,

clasping, membranous; pedicels filiform, 1-3 cm long, patent, becoming firm in

fruit. Flowers with a white shiny star-shaped perianth about 2 cm in diam.,

segments broadly linear, 1 cm long and 4 mm broad. apex obtuse, with a broad

5-nerved midrib. Stamens somewhat shorter than the perianth-segments; filaments

linear, anthers versatile, introrse. Ovary narrowly ovoid, sessile, with about 8

ovules per locule, style terete, stigma apical, papillate. Capsule broadly obovoid,

3-angled, about 1 cm long, chartaceous; seeds flattened, oblanceolate in outline,

about 6 mm long, tapering to the funiculus, with a blackish-brown shiny loose

skin (uncertain if seeds were mature).

Apparently confined to the Rehoboth-Mariental region although there is

one record from Karibib. Found in open stands and covering extensive areas.

According to collectors the flowers appear in their thousands after the first

rains and cover the ground with a white, shiny carpet and on the ridges the

flower masses resemble snowdrifts. Mr. W. Giess, who collected the species in

“in

b 6 ky

Prate 1. — Ornithogalum candidum, Farm Vrede, Rehoboth, South West Africa

(Giess 9110). (Photos by W. Giess).

357

several localities, noted that they preferred coarse quartzite gravel with under-

lying limestone but were absent from pure limestone or in red sand. They were

eaten by animals which is surprising since species of Ornithogalum are known

to be very poisonous.

var. tubiforme Oberm., var. nov. Perianthii segmentis ad basin connati.

SoutH West Arrica.— 2317 (Rehoboth): farm Buellsport, Strey 2624 (PRE.

holo.).

The fusion of the bases of the perianth-segments to form a short tube about

3 mm long is very unusual for this genus. The stamens are fused to the tube

at the base but exserted above it. The leaves are narrow and folded, c.2 mm

wide.

Ornithogalum recurvum Oberm., sp. nov., O. rautanenii Schinz affinis, sed

plantis majoribus, foliis latioribus, racemis elongatis, perianthii segmentis re-

curvis differt.

Planta ad 55 cm alta. Bulbus oblongo-globosus 10 cm longus, 5 cm diam..

tunicis tenuibus brunneis transverse striatis. Folia c. 7 lineari-lanceolata ad 27 cm

longa, 3 cm lata, ad basin vaginata apice attenuata, glabra glauca nitida. Racemus

elongatus c. 35 cm longus; scapus teretus, bracteis minutis subulatis; pedicelli

filiformes patentes ad 18 mm longi. Perianthii segmenta recurvata late linearia

alba. Staminum filamenta in dimidio inferiore abrupte expansa in squamam

latam, parte superiore erecta stylos cingentia. Ovarium disco basali leviter 6-

lobato, infra in stipite brevi contractum. Capsula late triangularis; semina dis-

coidea nigra.

SoutH West Arrica.— 1712 (Posto Velo): Kaokoveld, banks of the Kunene

River at long. 1226, lat. 1715, among rocks in mountains, Story 5848 (PRE,

holo., bulb cultivated at PRE). 1812 (Sanitatas): flats near waterhole at Orupembe,

gravelly calcareous soil, De Winter & Leistner 5732 (fruiting).

Bulb oblong, 10 cm long and 5 cm in diam. covered by thin shiny brown

transversely striate scales; neck absent. Leaves rosulate, erect, about 7, linear-

lanceolate, up to 27 cm long and 3.5 cm broad, clasping at the base, apex

attenuate, glabrous, dark glaucous green, shiny. Raceme overtopping leaf-rosette

(in the cultivated type-specimen the scape is up to 30 cm long and straight) or

about equal to the scape, which is bent outwards near the middle; with a few

minute sterile subulate bracts above; fertile bracts similar. Flowers up to 40,

on patent filiform pedicels up to 2 cm long, which become wiry in fruit. Perianth

with the segments recurved, their tips touching the apex of the pedicel, broadly

linear, 8 mm long and 1.5 mm broad, white with a broad green 3-5 nerved

central band, apex obtuse cucullate, barbate. Stamens erect, filaments expanded

in lower half to form a square base which folds inwards above around the

ovary from where the filiform upper part then forms a connivent tube around

the style; anthers versatile, introrse. Ovary expanded below in a shallowly 6-

lobed disc from a shortly stipitate base, turbinate above, with about 18 biseriate

ovules in each chamber; style filiform, about as long as the stamens, stigma

apical, shortly 3-lobed, papillate, exserted when stamens shrivel. Capsule broadly

triangular. c. 18 mm long and broad, depressed in the centre, walls thin, greenish;

seeds discoid, 7 mm in diam., black, shiny.

The perianth, with its recurved segments, resembles a Drimia flower but

in all other respects it conforms to Ornithogalum. The segments, after anthesis,

return to the bud position closing over the ovary. In the type-plant the recurving

was very characteristic, as can be seen in the photographs, but in the other

Pirate 2. — Ornithogalum recurvum, Kaokoveld, South West Africa (Story 5848).

(Whole plant at left, x 4; inflorescence at right, x 1).

cited specimen, which was in the fruiting stage (De Winter & Leistner 5732),

this could not be observed as no flowers were present. A few staminal filaments,

that had persisted below the capsule, showed the typical broadened bases and

it matched the type in all other respects. The species is very near O. stapfti

but the plant is larger, with the larger and broader leaves forming a distinct

upright rosette, and the recurved perianth especially distinguishes it from that

species.

359

Bothalia 10, 2 : 359-362

A Note on the Acacia giraffae x A.

haematoxylon Hybrid

J. H. Ross

ABSTRACT

Over twenty years ago the first specimens of a hybrid between Acacia giraffae Willd.

and A. haematoxylon Willd. were collected in the Hay district of the Cape Province.

From an examination of all available herbarium specimens the characteristics of A. giraffae,

A. haematoxylon, and the hybrid are tabulated. Some of the characters displayed by the

hybrid, for example number of pinna pairs, are found to be intermediate between the

values recorded for the parent species, while other characters, for example the degree

of pubescence and the presence of glands, tend to be inherited from one parent species

only. It appears, therefore, that there is a marked tendency for certain characters asso-

ciated together in a parent to be associated in the hybrid.

ne ge

/ i

in = es oe ves ee ee

Q@xa-

‘

®@® A. giraffae

x A. giraffae x A.

haematoxylon

° 100 200 300 400 500

——— EEE

KILOMETRES

Fic. 1.—The known distribution of Acacia giraffae, A. haematoxylon and A. giraffae

< A. haematoxylon, based on an examination of herbarium specimens.

360

Over twenty years ago the first specimens of a remarkable Acacia were

collected in the Hay district of the Cape Province by Mr. J. P. H. Acocks (No.

12689) and Dr. L. E. Codd (No. 1261). The collectors noted that the plant

exhibited characteristics of A. giraffae Willd. and of A. haematoxylon Willd.

and concluded that it was in all probability a hybrid between the two species.

A number of specimens have been collected subsequently from other localities

in the Cape Province and these have all lent support to the idea that the

plants are hybrids between A. giraffae and A. haematoxylon. Leistner in Mem.

Bot. Surv. S.Afr. 38:67 (1967) noted: “In river areas where Acacia giraffae

and A. haematoxylon form mixed stands. especially in the lower Auob and

Nossob, a tree is occasionally encountered (Plate 24), which is regarded as a

hybrid between the two species.’ Evidence suggests that although the hybrid

plants are relatively widespread they are nowhere common. Usually only a

solitary plant is found or at the most five or six individuals.

It is now proposed to examine the relevant characteristics of A. giraffae,

of A. haematoxylon, and of the hybrid to determine the position that the last-

named occupies in relation to the two species and, if possible, to attempt to

establish some of the characters which each parent contributes. The information

given in Table 1 has been compiled from an examination of all available her-

barium specimens. Seventy-six specimens of A. giraffae, 35 of A. haematoxylon,

and 15 of the hybrid were examined.

The hybrid has the spreading, rather weeping habit of A. giraffae and the

bark resembles that of A. giraffae, while the foliage is greyish and resembles

that of A. haematoxylon.

Leaflet size tends to be almost intermediate between that of A. giraffae and

A. haematoxylon. The leaflets are not tightly compressed laterally as in A.

haematoxylon and so do not appear as though simply pinnate. The hybrid

plants are not uniform in regard to leaflet size as some have decidedly larger

leaflets than others. There is, therefore, a small but readily apparent range in

leaflet size within the hybrid complex. For example, Acocks 12689 (PRE) is

typical of the one extreme with large leaflets and Leistner 1340 (KMG, PRE) is

typical of the other extreme with smaller leaflets. Leistner 1728 (KMG, PRE)

is almost intermediate in size between the two extremes.

Leaflets in the hybrid are sparingly to densely puberulous as in A. haema-

toxylon which is in sharp contrast to the glabrous leaflets of A. giraffae. Like-

wise, in contrast to A. giraffae, the petiole, rachides and rachillae are densely

puberulous and have numerous minute scattered reddish glands as in A. haema-

toxylon.

A. giraffae has (1-) 2-3 (-6) pinna pairs, A. haematoxylon (6-) 12-20 (-22)

and the hybrid (3-) 7-11 (-12) pairs. Thus the number of pinna pairs in the

hybrid is intermediate between the values recorded for the parent species. Similar

results were obtained by Moffett in Heredity 20: 621-629 (1965) who found

that the number of pinna pairs in hybrids between A. irrorata Sieb. ex Spreng.

and A. mearnsii De Wild. was “‘approximately midway between the parent

Species values,.... ” This behaviour was also very similar to that found in

hybrids between A. decurrens (Wendl.) Willd. and A. mearnsii by Moffett and

Nixon in Heredity 12: 199-212 (1958).

Inflorescences in the hybrid are grey in bud as in A. haematoxylon owing

to the dense grey pubescence on the corolla lobes, whereas in A. giraffae the

buds are yellow. As opposed to the glabrous, eglandular peduncle of A. giraffae,

361

Taste 1. — Synopsis of differences between A. giraffae Willd., A. haematoxylon Willd., and

A. giraffae x A. haematoxylon.

A. giraffae

Stipular spines usually fused

basally, often inflated into

rounded “ant-galls” up to 2.5

cm in diameter, tapering

apically.

Foliage dark green

Petiole 4-14 mm long, gla-

brous or subglabrous

Rachis 0—S.5 (0.7-2.7) cm

jong, glabrous or subgla-

brous

Pinnae 1—6 (2-3) pairs

Rachillae 1.3—4.2 (1.6-3.2)

cm long, glabrous or sub-

glabrous

Leaflets 7—18 (9-12) pairs,

4—11.5 (8.1+ 2.0) mm long,

0.7—2.4 (1.5+0.3) mm wide,

linear to linear-oblanceo-

late, glabrous

Penduncle 1.8—5.5 (3.1 +

0.8) cm long, glabrous or

subglabrous

Involucel apical

Calyx 1.5—2.8 mm_ long,

apices of lobes glabrous or

sometimes with few glan-

dular hairs

Corolla 2.7—3.6 mm long,

glabrous or apices of lobes

with few glandular hairs

Stamen filaments to 7.5 mm

long, often connate into

groups basally

Ovary sessile

Legume semi-lunate or sub-

orbicular, sometimes curled

into almost a complete cir-

cle, woody, not umbonate

over the seeds, 6—13 cm

long, 1.8—5.0 cm wide, 0.8—

2.0 cm thick

Seeds 10—14 mm x 8—10

Areole 6—9 mm x 3.5 —

5.5 mm, + closed

A. giraffac xX

A. haematoxylon

Stipular spines + straight,

usually slightly stouter than

in A. haematoxylon, not

fused basally

Foliage greyish

Petiole 2—9 mm long,

densely grey tomentose, with

minute scattered glands

Rachis 0.9—4.8 (2.0—4.4) cm

long, densely grey tomentose,

with minute, scattered glands

Pinnae 3—12 (7—11) pairs

Rachillae 0.4—2.2 (1.4—2.0)

cm long, densely grey to-

mentose, with minute scat-

tered glands

Leaflets 11—25 (15—22)

pairs, I1—4 mm long, 0.4—

1.1 mm _ wide, linear to

linear-oblong, often slightly

falcate, sparingly to densely

puberulous

Peduncle 1—3 cm _ long,

densely grey puberulous

Involucel apical

Calyx 1.8—2.4 mm _ long,

apices of lobes sparingly to

densely pubescent

Corolla 2.2—3.0 mm long,

apicces of lobes sparingly to

densely pubescent

Stamen filaments to 4 mm

long, often connate into

groups basally

Ovary sessile

Legume falcate or curl:d

into a complete circle, mar-

gin irregular, often constrict-

ed between the seeds and

+ moniliform, umbonate

over the seeds, 7—14 cm

long. 1.2—2.3 cm wide, up

to 1 cm thick

Seeds 9—12 mm x 6—8 mm

Areole €—8 mm x 2.5 —

3.5 mm, + closed

A. haematoxylon

Stipular spines + straight.

slender, not fused basally

Foliage greyish

Petiole 1—S mm long,

densely grey tomentose, with

minute scatttered glands

Rachis 0.8—5.1 (1.9—3.7)

cm long, densely grey tomen-

tose, with minute scattered

glands

Pinnae 6—22 (12—20) pairs

Rachillae 0.3—1.0 (0.5—0.8)

cm long, densely grey tomen-

tose, with minute scattered

glands

Leaflets 12—24 (14—21)

pairs, up to 0.8 mm long,

0.4 mm wide, very tightly

compressed laterally, super-

ficially appearing simply

pinnate, oblong, densely pu-

berulous

Penduncle 1.0—2.4 cm long,

densely grey puberulous

Involucel apical

Calyx 1.4—1.9 mm _ long,

apices of lobes sparingly to

densely pubescent

Corolla 1.8—2.6 mm long,

apices of lobes sparingly to

densely pubescent

Stamen filaments to 4.5 mm

long, often connate into

groups basally

Ovary sessile

Legume linear, falcate or

curled into a complete circle,

margin entire or irregularly

constricted between the seeds

and += moniliform, umbo-

nate over the seeds, 8—21

cm long, 0.6—1.4 cm wide,

up to 0.9 cm thick

Seeds 8.5—11.5 mm x 6.5—

9 mm

Areole 5—7 mm x 3.5—5

mm, + closed

362

peduncles are sparingly to densely puberulous and glandular. The stamen fila-

ments in A. giraffae and in A. haematoxylon are sometimes connate into groups

basally, whereas in the hybrid the degree of fusion is sometimes greater than

in either parent.

Pods of the hybrid are almost intermediate in width between those of A.

giraffae and those of A. haematoxylon.

Although the parentage of the hybrid is known, it is not known which

Species functions as the male parent and which as the female. Furthermore, it

is nct known whether the same species is always, for example, the male parent

or whether the same species may sometimes serve as the female parent. Con-

sequently there is at present no understanding of differences arising in the progeny

as a result of this. As the hybrid is fertile it should be possible to find all

stages of back-crossing with the parents. Careful field studies are necessary to

resolve these problems. A few plants grown from seed of a hybrid tree have

been established on the National Botanic Garden, Pretoria.

Thus it may be seen that some of the characters displayed by the hybrid,

for example number of pinna pairs and leaflet size, are intermediate between

the values of A. giraffae and those of A. haematoxylon, while other characters,

for example the degree of pubescence and the presence of glands, are those

exhibited by a single parent, namely A. haematoxylon. The hybrids are usually

as densely pubescent and as glandular as A. haematoxylon and are not only

sparingly pubescent as an intermediate state between glabrous A. giraffae and

densely pubescent A. haematoxylon. Recombination of the characters of the

two parent species apparently does not take place at random, but there is a

marked tendency for characters associated together in one species to remain

associated in the hybrid. The multifactorial genes tend, therefore, to be linked

together in the hybrid in the same way as they were in the parents, certain

combinations perhaps being favoured by natural selection. As in the case of

A. haematoxylon and the hybrid, the presence of glands and the degree of

pubescence were also found (Ross in Webbia 22 : 203-223, 1967) to be correlated

in A. caffra (Thunb.) Willd., the greater the degree of pubescence the more

numerous the glands.

I am grateful to Dr. K. D. Gordon-Gray of the Bews Botanical Laboratories,

University of Natal, for reading the manuscript and for advice.

363

Bothalia 10, 2 : 363-371

New and Interesting Records of African Plants

Various Authors

ASCLEPIADACEAE

A New ComMBINATION IN SCHIZOGLOSSUM

Bullock, in Kew Bull. 1952:417 (1952), proposed the upholding of the

generic status of Stenostelma to accommodate S$. capense Schltr. (= Schizoglos-

sum aciculare N.E.Br.) and the closely related species Stenostelma corniculatum

(E. Mey.) Bullock.

Huber, in Prodr. Fl. S.W.Afr. 114 : 52 (1967), transferred Stenostelma capense

Schltr. to Schizoglossum capense (Schltr.) Huber.

The latter classification appears preferable and it thus becomes necessary

to transfer Stenostelma corniculatum (Lagarinthus corniculatus E. Mey.) also to

Schizoglossum with synonymy as follows:

Schizoglossum corniculatum (FE. Mey.) R. A. Dyer, comb. nov.

Lagarinthus corniculatus E. Mey., Comm. 208 (1837).

Gomphocarpus corniculatus (E. Mey.) Dietr., Syn. Pl. 2: 901 (1840).

Krebsia corniculata (E. Mey.) Schltr. in Bot. Jahrb. 20, Beibl. 51 : 41 (1895):

ae in Fl. Cap. 4, 1 : 587 (1907). (Krebsia Harv. 1868, not of Eckl. & Zeyh.,

1 ;

Stenostelma corniculatuin (E. Mey.) Bullock in Kew Bull. 1952 : 417 (1952).

R. A. DyEr.

DISTINCTIONS BETWEEN DUVALIA AND HUERNIA

In an article published earlier in this volume, pp. 45 - 54 (1969), L. C. Leach

transferred Duvalia tanganyikensis Bruce & Bally, D. procumbens R. A. Dyer

and D. andreaeana Rauh to the genus Huernia. In the absence of any published

comment it could be construed that the changes met with general agreement.

For my part this is not so.

It has been pointed out by various authors that distinctions between genera

of the Stapelieae are often arbitrary. A close study makes it fairly clear that

there is no sharp distinction between the two genera Duvalia and Huernia as at

present constituted. On the other hand the respective type species D. elegans

364

(Masson) Haw. and Huernia campanulata (Masson) R.Br. leave no room for

confusion. There is no call for the amalgamation of the two genera and the

problem is to select the most natural line of distinction.

N. E. Brown in Fl. Cap. 4, 1: 526 (1907) used the presence of teeth be-

tween the corolla-lobes as a distinctive character for Huernia. He went on to

describe the outer corona as spreading upon and more or less adnate to the

bottom of the corolla-tube and the inner corona as arising from the upper part

of the staminal column, of 5 simple lobes incumbent upon the backs of the

anthers and equalling or exceeding them, subulate or clavate or thickened at the

apex, often with a slight transverse dorsal ridge at their base but no crest, wing

or dorsal horn.

In the case of Duvalia, N. E. Brown described the corona as arising near

the top of the staminal column and stipitate, and the inner corona-lobes as turgid,

ovoid, more or less pointed at each end, subhorizontal, with the dorsal point

usually somewhat raised and the inner closely incumbent on the backs of the

anthers and sometimes longer than them but not produced into erect points.

Leach tabulates the four main characters on which he relies to distinguish

the two genera, characters present in Duvalia and absent in Huernia: Corona

stipitate; denticles (usually) at base of leaves; corolla (usually) replicate; corolla

lobes (usually) ciliate. He lists known exceptions to the last three of these

characters in Duvalia, and it still remains to be seen whether Duvalia maculata

N. E. Brown var imaculata Luckhoff is an exception in not having the stipitate

corona normally found in Duvalia.

Duvalia tanganyikensis, D. procumbens and D. andreaeana are excluded from

Duvalia and placed by Leach in Huernia, because the coronas are not stipitate,

the stem-teeth (rudimentary leaves) are devoid of denticles at the base and the

corolla-lobes are neither replicate nor ciliate, although he allows other species

without these characters to remain in Duvalia.

Leach omits, as not being diagnostic, (a) intermediate corolla-lobes, as occur-

ring in both genera to some extent, although usually much more prominent in

Huernia; (b) corona shape, since, as he says, the outline of the outer corona

and the form of the inner lobes are closely matched in both genera.

It is in our interpretation of the importance of the structure of the inner

corona-lobes that we differ most. Leach, p. 54, points out that there is consider-

able variation in the inner corona of D. procumbens and, in his words, the lobes

may be either widely spreading with the inner face somewhat channelled, or

strictly erect and more or less triangular in cross-section. What Leach is describ-

ing is not the inner face of the lobe but the upper face of its dorsal prolongation.

His figure, p. 47, shows this very nicely. This dorsal prolongation is quite foreign

to the genus Huernia, as stressed by N. E. Brown. Leach is in further error,

therefore, in saying at the foot of p. 46 that his rearrangement requires no amend-

ment to the existing generic circumscription of Huernia.

In an effort to tidy up the genus Duvalia, Leach has introduced exceptions

into Huernia where none existed before. None of Duvalia tanganyikensis, D.

procumbens or D. andreaeana has teeth between the corolla-lobes; in none is the

inner coronal-lobe limited to the presence of a transverse dorsal ridge and all

have a dorsal prolongation of the inner corona-lobe giving them the charac-

teristic Duvalia appearance. Bruce, Bally and Rauh expressed no doubt in

placing their species in Duvalia where, together with D. procumbens, they should

remain.

365

Huernia verekeri Stent var. pauciflora Leach, the controversial species,

does appear to show a tendency towards Duvalia but does not quite reach the

borderline in the dorsal development of its inner corona-lobes and should remain

in Huernia as concluded by Leach.

ReeAteD YER:

CELASTRACEAE

A New NAME IN CaTHA

Catha transvaalensis Codd, nom. nov.

Lydenburgia cassinoides Robson in Bol. Soc. Brot. 39 : 35 (1965).

Catha cassinoides (Robson) Codd in Bothalia 9: 124 (1966), nom. illegit., non

Webb & Berth.

I am grateful to Dr. G. Kunkel of Las Palmas Herbarium for drawing my

attention to the prior use of the name Catha cassinoides by Webb & Bertholet.

L. E. Copp.

GRAMINEAE

A New Species or ARISTIDA FROM SouTu West AFRICA

Aristida dewinteri Giess, sp. nov., A. stipoidi Lam. affinis, sed gluma _ in-

feriore breviore apice rotundo tenuiter fimbriato, glumis superioribus et interiori-

bus flavis apice macula nigra praedito differt.

Gramen annuum, ad | m altum. Culmi simplices, nodis 2-3, recti vel sub-

geniculati. Panicula diffusa. Spiculae pallidae, sed glumae apice macula nigra

praeditae. Glumae valde inaequales, 1-nervatae, subcoriaceae; inferior 3.5-4.0 mm

longa, late oblonga, apice obtuso rotundato fimbriato; superior 10-11 mm longa,

lanceolata, apice plus minusve acuto membranaceo fimbriato vel sublacerato.

Lemma 3-nervatum, cylindricum, coriaceum, in dimidio inferiore glabrum

superne scabridum, 7-8 mm longum (callo incluso), inter apicem lemmatis et

basin columnae articulatum; columna 1.3-1.4 mm longa, tortilis, scabrida; aristae

3, scabridae; centralis laterales excedens, c. 3.5 cm longa; laterales c. 2.5 cm

longae; callus c. 1 mm longus, manifeste bifidus, dense retrorse setosus setis

basin versus brevioribus. Palea 0.7 mm longa, subcoriacea, late oblonga, enervis

hyalina, rotundata. Lodiculae magnae, 1-1.25 mm. longae, ovatae, tenuiter mem-

branaceae, apice rotundato; nervi 7-9, distincti. Stamina 3; antherae cultratae,

3 mm longae, pallide luteae. Caryopsis linearis, 3-3.5 mm longa; hilum lineare

caryopsidem aequans (embryo indistinctus ob caryopsides immaturas).

Type: South West Africa, Kaokoveld (Grid 1613, Uncocua), 18 miles west

of Enyandi, Giess 9345 (PRE, holo.; K; M; WIND; US).

Annual forming erect tufts up to 1 m high. Culms simple, 2-3-noded, straight

or somewhat geniculate, nodes and lower internodes usually purple, glabrous,

very finely striate. Leaf-sheath shorter than the internodes, rather lax, often striate

particularly upwards, with a flake of wool at the mouth. Collar smooth, usually

purplish. Ligule a dense fringe of long woolly hair. Panicle effuse and open,

pallid, up to 30 cm long and 15 cm wide. Spikelets pallid, glabrous, the glumes

each with a black spot at the apex. Glumes very unequal, I-nerved, firm and

thick in texture; lower 3.5-4.0 mm long, broadly oblong, apex obtuse and rounded,

fimbriate; upper 10-11 mm long, lanceolate, apex more or less acute, thin, mem-

branous. Lemma 3-nerved, cylindrical, coriaceous, scabrid in the upper half,

366

otherwise glabrous, 7-8 mm long (callus included), articulated between the apex

of the lemma and the base of the column; column up to 2.2 cm long, twisted,

scabrid; awns 3, scabrid; central awn exceeding the others in length, about 3.5

cm long; lateral awns about 2.5 cm long; callus about 1 mm long, distinctly

bifid, densely retrorsely bristly, the bristles diminishing in length towards the

base. Palea 0.75 mm long, subcoriaceous, broadly oblong, nerveless, apex hya-

line, rounded. Lodicules large, 1-1.25 mm long, ovate, thinly membranous, apex

rounded; nerves 7-9 in number, distinct. Stamens 3; anthers cultrate, 3 mm long.

pale yellow. Caryopsis linear tapering to an acute apex, 3-3.5 mm long. Hilum

linear, as long as the grain. Embryo (indistinct due to immature grains).

This species has so far been found only along the Kunene River between

the Epupa Falls and Enyandi in the Kaokoveld of north-western South West

Africa. It often occurs in pure stands and is then conspicuous due to the bright

yellow inflorescences. This species is named in honour of Dr Bernard de Winter

for his contributions to the study of South West African plants.

W. Giess.

LILIACEAE

A New Aor From SoutH WEsT AFRICA

Aloe erinacea Hardy, sp. nov. A. melanacanthae Berger affinis, sed foliis

compactioribus brevioribus biconvexis, floribus brevioribus subventricosis non

incurvatis differt.

Plantae solitariae vel glomeratae, acaules vel caulescentes, caulibus decum-

bentibus 60 cm longis. Folio griseo-viridia dense rosulata suberecta deinde

patula vel recurvata deltoideo-lanceolata 8-16 cm longa 3-4 cm base lata, supra

convexa vel ad basim plana, nonnunquam ad apicem spinis nigricantibus paucibus

armata, subtus convexa ad apicem carinata, carinis 5-8 spinis armatis, margine

spinis nigricantibus deltoides 5-9 mm longis 10 mm distantibus armata. /nflores-

centia simplex semper solitaria arcuato-adscendens usque ad | mm _ longa.

Pedunculus laete griseo-viridis vel ochraceus base complanatus circa 1.5 cm

diam., circa 25 bracteis vacuis membranaceis leviter fuscis ovato-acuminatis 2.5-3

cm longis | cm latis 8-10 nervatis instructus. Racemi dense multiflori 24-26 cm

longi 5-6 cm diam., gemmis congestis patulis deinde cernuis carmesinis apicibus

schistaceis. Bracteae membranaceae fuscae 3-5 nervatae infimae 25-27 mm

longae 4.5 mm latae pedicellis longiores. Pedicelli infimi 18-19 mm_ longi.

Perianthium subcylindricum leviter ventricosum 28 mm longum ad medium 7 mm

latum base rotundatum haud stipitatum; segmenta exteriora libera apicibus sub-

acutis; segmenta interiora libera eburnea apicibus viridibus. Antherae 4 mm

exsertae. Ovarium viridi-brunneum 5-6 mm longum 2-2.5 mm diam.; stigmata

denum 7 mm exserta. Capsula non vidi.

Type: South West Africa, Liideritz District, Witputz South, 21.9.1968, Hardy

2619 (PRE, holo.).

Plants succulent, solitary or in compact groups of five or more, sometimes

with a decumbent stem up to 60 cm long. Leaves grey-green, densely rosulate,

suberect becoming spreading or recurved, deltoid-lanceolate, 8-16 cm long, 3-4

cm broad at the base; upper surface convex to flat low down often with a few

sharp, black spines near the apex, grey-green; lower surface convex with a

distinct keel for one-third of its length, the keel armed with 5-8 deltoid, sharp,

black teeth, otherwise same colour as the upper, margins armed with deltoid,

horny, black teeth which are evenly spaced throughout, 5-9 mm long, 10 mm

367

Pretoria, x

inflorescence from the natural habitat,

plant growing in the National Botanic Gardens,

x lg

Il.

Owe x I,

1.— Aloe erinacea.

(approx.); 2, cross section of leaf,

4, bud, x 1;

Fic.

368

apart. /nflorescence simple, usually one from a rosette, arcuate-ascending up

to | m high. Peduncle pale grey-green to yellowish brown, flattened low down,

about 1.5 cm diameter, clothed with about 25 sterile bracts which are 2.5-3 cm

long, | cm broad at the middle, ovate-acuminate, membranous, brownish, 8-10

nerved. Racemes densely flowered, 24-26 cm long; buds crowded, horizontally

spreading to spreading downwards, crimson with slate-grey tips. Bracts thin,

membranous, brownish, 3-5 nerved, longer than the pedicel, the lowest 25-27

mm long, 4.5 mm broad at base. Pedicels about 18-19 mm long on the lower

flowers. Perianth cylindric and slightly narrowed above the ovary, subventricose,

about 28 mm long, 7 mm broad at the middle, rounded at the base, not stipitate;

outer segments free to the base, apices subacute, with 4-5 nerves which meet

near the apex, inner segments free, yellowish-white, greenish at apex. Anthers

exserted by 4 mm. Ovary greenish-brown, 5-6 mm long, 2-2.5 mm diam.; stigma

exserted by 7 mm. Capsule not seen.

This species is nearest Aloe melanacantha Berger from which it can be

distinguished by the more compact rosette, the shorter leaves which are more

biconvex and eventually more spreading and by the shorter, subventricose

perianth which is not incurved.

Aloe erinacea was first collected at Witputz-Suid by Mr. Jan Botha of

Rosh-Pinah, South West Africa, in 1967. In September 1968 he accompanied

the author to this locality where a number of plants were collected, some of

which were planted at Rosh-Pinah and the balance being cultivated at the

National Botanical Gardens, Pretoria.

The accompanying description was based on plants which flowered at Rosh-

Pinah in August 1969.

D. S. Harpy.

MY RSINACEAE

A Note oN MyYRSINE IN SOUTHERN AFRICA

When preparing the text for an illustration of Myrsine africana L. for

publication in Flowering Plants of Africa, the second author noted that the

species is dioecious. This fact is not recorded in the first author’s treatment

of the species in the Flora of Southern Africa 26: 5-6 (1963), although it is

mentioned in the generic description that flowers are often unisexual. Cne

result of the oversight is that the key in the Flora is not entirely accurate.

The first character given in the key to distinguish Myrsine africana from M.

pillansii Adamson is ‘‘Anthers exserted beyond corolla-lobes” versus ‘‘Anthers

included within corolla-lobes.” In M. africana the anthers are exserted only in

the male flowers; in the female they are included. As regards M. pillansii, the

anthers referred to are of female flowers. In other words. the anthers of male

flowers in M. africana are compared with those of female flowers in M. pillansit.

The leaf characters, however, used in the key to distinguish the two species,

hold good. So far, male flowers of M. pillansii do not appear to have been

collected, so we are ignorant of their structure. The female flowers contain

“male” organs resembling fertile stamens, but in all specimens examined by

the authors none has shown pollen, consequently they are merely staminodes.

Adamson in his original description of M. pillansii in J. S. Afr. Bot. 7 : 204

(1941) refers to male flowers as follows: “‘in floribus masculis ovarium abortivum

minutum adest.’’ However, in the type material which was kindly sent on loan

369

from the Bolus Herbarium, there is no trace of functionally male flowers. What

Adamson probably observed were young female flowers with undeveloped

ovaries and staminodes.

After Vol. 26 of the Flora of Southern Africa appeared, a specimen of

M. pillansti (Moss sub TRV 19775) from Witpoortjie, Krugersdorp district, in

the Transvaal, was located in the National Herbarium, Pretoria, and further

specimens from the Transvaal have since been collected by Mr. J. H. Vahrmeijer

(Nos. 1763 and 1819) of the Botanical Research Institute, at Breed’s Nek in

the Magaliesberg and by Mr. G. K. Theron (No. 1954) at Loskop Dam. It is

clear that M. pillansii has a much wider distribution than was at first realized

and it is all the more surprising that no functionally male flowers have yet

been preserved. It would be appreciated, therefore. if collectors would search

for male trees of H. pillansii to fill this serious gap in our knowledge of the

species.

Ine AX, IDAs Ce IDE J, 1B, [amici

ZAMIACEAE

THe ConeES oF ENCEPHALARTOS INOPINUS

When Encephalartos inopinus R. A. Dyer was first described in Bothalia

8: 169 (1964), only a few plants and two damaged young male cones had been

recorded. No additional information came to light for inclusion in the account

for the Flora of Southern Africa 1 : 13 (1966). In January 1969, however, both

male and female cones became available for description: a beautiful full-sized,

but immature, female cone collected by Mr. G. J. Alberts of the Nature Con-

servation Section of the Transvaal Administration, and male cones from wild

and cultivated plants through the help of Mr. C. T. Phillips of Verwoerdburg

Mr. Alberts had undertaken extensive field excursions in the rough terrain

of the Olifants River Valley in the Lydenburg district of the Transvaal to deter-

mine the distribution range of the species and to estimate if possible the degree

to which unauthorized collecting had been taking place. He found the cycad

population to be fairly widely scattered but nowhere was there a colony with

young plants to prove that active regeneration is taking place from seed. Approxi-

mately 50 per cent of the plants recorded by him were on north-facing cliffs,

krantzes or rocky outcrops and the other half on various other aspects including

some in south-facing positions. The tallest perfectly upright stem measured

approximately 1.5 m, whereas the longest stem, now procumbent, measured

nearly three times this length (15 ft).

The male cone from the wild — one of a cluster — was subcylindric, more

or less elliptic-oblong in outline, whereas three cones from a single stem on

a garden plant, were more oblong-lanceolate. The scales of the cone from the

wild were dense and the beak slightly decurved, by comparison with the rela-

tively open, spreading scales of the plant in cultivation. It was found also that

the garden cones must have had a considerably higher moisture content judging

by the shrivelling and weight for size ratio. The following details amplify the

original description.

Cones 1 to several in a head; the scale faces green and densely covered

with small white papillae giving a general light green .,matt’ appearance. Male

cones subcylindric, narrowed more or less equally to both ends or sometimes

more gradually tapered towards the apex, 18-25 cm long, 6.5-8 cm diam.,

370

Piare 1—Encephalartos inopinus R. A. Dyer; left, male cone; right, female

cone, both densely and minutely whitish papillate; female cone-scales

largest toward the apex of the cone.

Sil

pedunculate; peduncle 6-8 cm long, 2.5-3 cm diam. at top, not much thinner

at base; median scales spreading more or less horizontally from axis with de-

flexed beak, 2.5-3 cm long, 7-8 mm thick vertically, with sharp lateral angles,

moderately ridged down upper surface, nearly flat on microsporangial surface,

with pollen-sacs not spreading quite to margin; bulla-face minutely papillate,

projected into a beak 9-10 mm long; upper facet with median receding ridge:

lower facet continuous with microsporangial surface; terminal facet subquadrate

7-8 x 7-8 mm. Female cones broadly subcylindric, slightly narrowing to obtuse

apex, 31 cm tall 15 cm in greatest width near base, and with stout peduncle;

peduncle obconic, 5-6 cm long, 5 cm thick at top and narrowed to base, sub-

tended by numerous linear-filiform tomentose bracts about the same length as

the peduncie; median scales about 5 cm long, 4-4.75 cm broad, 3 cm_ thick

vertically, with lateral ridges extending into incurved lateral lobes; bulla-face

about 10 mm prominent, minutely and densely whitish papillate; upper ‘acct

with a slightly undulate surface and 1 or 2 receding ridges; lower facet similar

to upper but more acutely angled from the terminal facet; terminal facet slightly

below centre of bulla-face, slightly concave. about 2.5 cm broad, 1.5 cm wide

vertically (slight cracking of surface at time of photograph and description

probably due to slight drying out after period of 4-5 weeks since removal from

parent plant); scales in the upper 4 of the cone, broadest. up to 5 cm broad,

with the terminal facet narrower vertically by comparison with those of the lower

scales.

The leaf characters of this species are distinctive among the South African

species of the genus and the minutely papillate nature of the cone-scales recorded

above is one more obvious distinguishing feature.

R. A. Dyer.

oe eee ee ee

7 Ciiny wt © stn! Qadl

p- fre : } ; ; 1 @ © Mia i)

373

Bothalia 10, 2 : 373-378

Notes on the Genus Brachystelma

R. A. Dyer

ABSTRACT

The distribution and synonymy of Brachystelma blepharanthera H. Huber, B. circinatum

E. Mey. and B. dinteri Schltr. are discussed in the light of material recently available. New

names include a new combination: B. stenophyllum (Schltr.) R. A. Dyer (Siphonostelma

stenophyllum Schltr.) and three new species: B. cupulatum R. A. Dyer from South West

Africa, B. tenue R. A. Dyer and B. vahrmeijeri R. A. Dyer, both from Natal.

An effort has been made to clarify some long outstanding nomenclatural

problems in southern Africa. For this purpose specimens collected by K. Dinter

in South West Africa were kindly sent on loan to the National Herbarium

from the S.A. Museum Herbarium by the Director of the National Botanic

Gardens, Kirstenbosch. As an introduction to the conclusions arrived at, it

seems appropriate to recall an observation by Dinter when describing his

Brachystelma grossarthii in 1914; the variation in the length and width of the

leaves of the species were so great that when dealing with extreme forms one

might readily think that one was confronted by two completely distinct species.

He went on to say that the same phenomenon of variability is met with in

Brachystelma dinteri Schltr. and several other related plants growing on the

Rehoboth sand flats in South West Africa.

Brachystelma arnotii Baker in Ref. Bot. | : t.9 (1869); N.E. Br. in Fl. Cap. 4,

1: 845 (1908).

B. grossarthii Dinter, Neue Pfl. S.W. Afr. 16 (1914); H. Huber in Prodr. FI.

S.W. Afr. 114: 12 (1967).

Dinter referred in his description of B. grossarthii to the similarity of B.

arnotii Baker but gave no diagnostic characters to distinguish between them.

He stated, however, that he was unable to find an inner corona in B. grossarthii.

An examination of the syntypes, Grossarth sub Dinter 2698 and Dinter 2790 a,

shows that the inner corona-lobes are inconspicuous as in B. arnotii. They are

represented by inwardly inclined cushion-like swellings between the outer corona-

lobes and are pressed against the base of the filaments. No significant difference

between the two concepts was discerned.

Specimens in PRE which have also been identified with B. arnotii are

Bradfield 350 from the Waterberg of South West Africa and H. Hechter sub

PRE 30681 from between Windhoek and Gobabis (cult. Uitenhage, South

Africa).

374

Brachystelma blepharanthera H. Huber in Mitt. Bot. Miinchen 4 : 33 (1961);

Prodi. Fl. S.W. Afr. 114 (1967), partly, as to Dinter 410, 680 and 1514, excl.

Story 6400.

Belpharanthera dinteri Schltr. in Bot. Jahrb. 51 : 146, Fig. 2, A-F (1913).

Blepharanthera edulis Schltr. 1.c. Fig. 2, G-M (1913).

Since the two epithets dinteri and edule had already been created for other

species in Brachystelina, Huber gave the new name blepharanthera to cover the

combined concept. It was only in his second publication (1967) that he cited

Story 6400 under his blepharanthera, but the Story specimen has a cupular outer

corona and reduced inner corona-lobes, quite unlike typical B. blepharanthera.

Story 6400 is specifically equal to Dinter 2699 in its greater part in the S.A.

Museum Herbarium and is designated as the holotype of the species described

below under the epithet cupulatum.

Brachystelma circinatum /. Mey., Comm. 196 (1836); N.E. Br. in Fl. Cap. 4.

LSD S8e908):

» Orme lskray, Wie, Cay, 358, i 93 (ESO eE WIE. Bye, loc, Go (OCs).

. ovatum Oliver in Refug. Bot. 4: t. 226 (1870); N.E. Br., 1.c. 860 (1908).

. cinereum (Schltr.) N.E. Br.. l.c. 862 (1908).

. galpinii (Cchitr.) N.E. Br., 1.c. 860 (1908).

. pallidum (Schltr.) N.E. Br., 1.c. 861 (1908).

. undulatum (Schitr.) N.E. Br., 1.c. 859 (1908).

. zeyheri (Schlitr.) N.E. Br., 1.c. 855 (1908).

Dichaelia forcipata Schltr. in Bot. Jahrb. 51:145 (1914); Dinter, Neue Pfl.

S.W. Afr. 27, t. 10 (1914).

Many attempts at the National Herbarium to distinguish clearly between

the several species recognised by N. E. Brown l.c. (1908), within the B. circinatum

complex, have failed. Consultation with Mr. A. A. Bullock of Kew in 1966

revealed that he had experienced similar difficulties and, although he favoured

the retention of the generic status of Dichaelia Schltr. for species with united

ccrolla-tips, he held the view that B. pallidum (Schltr.) N.S. Br., B. galpinii

(Schitr.) N.E. Br., B. ovatum Oliver, B. bolusii N.E. Br., and B. commixtum N.E.

Br. are not specifically distinct from B. circinatum E. Mey. In agreeing with

Mr. Bullock, the question arose whether or not to go further and reduce the

specific status of B. undulatum (Schltr.) N.E. Br., B. filiforme Harv., and B.

zeyheri (Schitr.) N.E. Br. A good range of material from the type localities is

not available for further assessment but the characters relied upon for dis-

tinction by the authors are unconving in the light of our knowledge of natural

variability.

BBnnansns

Several specimens from South West Africa conforming to the description

and illustration of Dichaelia forcipata Schltr. have been received at the National

Herbarium, and they too agree with Brachystelma circinatum in essential

characters. The decision to accept Dichaelia forcipata as a synonym of B.

circinatum E. Mey. gives the species a very wide range of distribution from

the eastern Cape Province northwards to the Transvaal and westwards to the

northern parts of S.W. Africa.

It goes without saying that the concept of B. circinatum becomes a very

broad one, and is even broader than the sum total of the above synonyms,

because of the inclusion of yet other forms in the Transvaal, one with longer

S15)

corolla lobes, Galpin 9150, from a ravine near Potgietersrust and Schlieben 7627,

from the Soutpansberg, with flat leaves, thickly pubescent on both surfaces.

Brachystelma cupulatum R. A. Dyer, sp. nov. pulla affinitate manifesta,

corolla 6-9 mm longa viridi, corona exteriore cupulata 1.5-2 mm alta., lobis

interioribus parvis pulvinatis valde insignis.

Herba tuberosa erecta simplex vel parum ramosa, 8-15 cm alta, breviter

pilosa, pilis decurvis; tuber depressum, circiter 10 cm diam. Folia ovata, lan-

ceolata, elliptica vel linearia, 3-8 cm longa, 4-5 mm lata, breviter petiolata, in

superficiebus ambabus breviter pilosa. Flores plures extra axillares, pedicellis

2-3 mm longis; calyx sparse pilosus segmentis ovatis vel lanceolatis 1.5-2 cm

longis. Corolla viridis, 6-9 mm longa, extus sparse pilosa intus glabra, tubo

1.5-2.5 mm rariter 3 mm longo, lobis plus minusve oblongis attenuatis vel

lineari-lanceolatis, marginibus leviter recurvis. Corona exterior cupulata circiter

2 mm alta, margine 5-emarginato, lobi interiores 5 pulvinati antheris oppositi.

Type: S.W. Africa, Grootfontein, about 8 km west of Aha Mts., sandy

grasslands, Story 6400 (PRE, holo.).

Tuber up to about 10 cm in diam., compressed from above and _ below.

Stem single or once-branched near the base and sparsely above, up to 15 cm

tall, sparsely hairy with decurved hairs. Leaves ovate, lanceolate, elliptic or

linear, 3-8 cm long, 0.4-1.5 cm broad, contracted at base into a short petiole,

shortly pubescent on both surfaces. Flowers several together, extra-axillary,

opening successively; pedicels 2-3 mm long. Calyx with sepals ovate to lan-

ceolate, 1.5-2 mm long, sparsely hairy. Corolla green, 6-9 mm long, thinly hairy

on outer surface, glabrous within; tube 1.5-2.5 mm rarely up to 3 mm long;

lobes more or less oblong, slightly narrowed to apex, with slightly recurved

margins. Corona arising about 0.5 mm above base of staminal column; outer-

corona cupular, about 2 mm high and much exceeding the staminal column,

sometimes with 5 slits 0.5 mm deep on rim alternating with the inner corona-

lobes and 5 small notches above (opposite) the inner corono-lobes cushion-like,

arising from near base of inner wall of outer corona and pressing on base of

filaments. Pollinia about 0.25 mm diam. compressed, translucent on upper half

of inner margin, with short connectives from about middle, attached to small

carrier.

S.W. Africa. — Grootfontein: about 8 km (5 miles) west of Aha Mts., sandy

grasslands, Story 6400. Rehoboth: Dinter 2699, for the greater part, cult.

Okahandja. Windhoek: between Windhoek and Gobabis at Witvlei, H. Hechter

sub PRE 30682, cult. Uitenhage.

Dinter may have distributed a mixture of species under his number 2699

but it is more likely that the mixing which took place happened inadvertently

later. All but one twig on the three sheets with that number in the S.A. Museum

collection, although differing in the shape and texture of their leaves, are growth

forms of the one species. The epithet cupulatum is given in reference to the

cup-shaped outer corona. The foreign twig on one of the sheets mentioned is

B. blepharanthera Huber, equal to Dinter 410 & 680.

Brachystelma dinteri ScAltr. in Bot. Jahrb. 51 : 144 (1913); Dinter, Neue

Pfl. S.W. Afr. 15, Fig. 7 (1914); H. Huber in Prodr. Fl. S.W. Afr. 114: 12 (1967).

B. brevipedicellatum Turrill in Kew Bull. 1922: 29 (1922).

B. ringens E. A. Bruce in Flow. Pl. Afr. 28 : t. 1096 B (1951).

The syntype numbers are Dinter 775, 1515, 1890 and 2384, the last two

being represented in the S.A. Museum collection. They are matched very closely

by Hechter sub PRE 30680, cultivated in Uitenhage from tubers collected near

376

Witvlei, between Windhoek and Gobabis. The coronal structure of these and

of the concepts described by Turrill and Bruce as B. brevipedicellatum and B.

ringens respectively, are very similar. Identified with these also are specimens

collected in the Transvaal, Pretoria district, near Hammanskraal, by D. S$. Hardy

2205, 2210; Waterberg district, per W. J. Louw sub PRE 30679; and Letaba

district, east of Tzaneen, by B. van Zyl sub PRE 28904 (cult.). Plants come

into flower at an early age as seen in B. ringens and the leaf formation, their

texture and pubescence show considerable variation according to the conditions

of growth. The present records give the species a considerable extension of

distribution from S.W. Africa to the Northern Cape Province and Transvaal.

Brachystelma stenophyllum (Sch/tr.) R. A. Dyer, comb. nov.

Siphonostelma_ stenophyllum Schltr. in Bot. Jahrb. 51 : 148, Fig. 3 A—E

(1913); Dinter, Neue Pfl. S.W. Afr. 49, Fig. 37 (1914); H. Huber in Prodr. FI.

S.W. Afr. 114: 53 (1967).

The isotype of Dinter 2361 in the S.A. Museum Herbarium shows that the

original figure is inaccurate in the proportion of corolla-tube to corolla-lobes;

the lobes are appreciably longer than the tube and the tube is more campanu-

late than depicted in the figure. It is agreed, as interpreted by Phillips in his

Genera, ed. 2: 607 (1951), that Siphonostelma should be regarded as a synonym

of Brachystelma. Both Dinter and Huber compare the species with Ceropegia

pygmaea, but the affinity is certainly only remote. The corona is comparable

to that in B. cupulatum described above.

The distribution range of B. stenophyllum is extended into the Transvaal

by the identification with it of Galpin M217, from the Waterberg, and Hardy

2206, from north of Pretoria.

Two New Species FROM NATAL

In September 1965 Mr. J. Vahrmeijer, of the Botanical Research Institute,

undertook a collecting expedition to some little explored parts of Zululand.

Among many interesting records, two somewhat diminutive species of Brachy-

stelma proved to be undescribed.

Brachystelma tenue R. A. Dyer, sp. nov., B. circinatae E. Mey. affine sed

pilis spatulis longis, pedicellis longioribus et tenuioribus, sepalis longioribus dif-

fert.

Herba tuberosa humilis hirsuta e basi ramosa, ramis 5-10 cm altis, 0.75-1.25

mm diam. Folia breviter petiolata, ovata, ovato-elliptica vel oblongo-elliptica,

1-2 cm longa, 1.5-5 mm lata, infra sparse hirsuta, supra glabra. Flores extra

axillares singuli vel bini, gracili-pedicellati, pedicellis hirsutis 8-10 mm _ longis,

0.25 mm diam. Calyx sparse hirsutus 5-partitus, segmentis linearibus acuminatis

3 mm longis. Corolla luteo-fusca, plus minusve 1.5 cm longa extus sparse hir-

suta intus glabra, basi in tubum 1 mm longum connata, segmentis e basi ovatis

linearibus erectis, apice cohaerentibus. Coronae lobi exteriores circiter 0.75 mm

longi, profunde bilobati, lobulis filiformibus; lobi interiores ovato-oblongi,

antheris incumbentibus aequilongis.

Type: Natal, Ubombo-Ingwavuma border, near Lala Nek, about 3 km from

sea, open veld, Vahrmeijer 1049 (PRE, holo.).

Tuber red, up to about 4 cm in diam., slightly compressed. Stem sparsely

branched from the base and sometimes above; branches erect or somewhat

straggling, 5-10 cm long, slender 0.75-1.25 mm diam., with spreading transparent

SPHONAL

RASS ag HEMBAIEM, PRETUAES.

me eek a ihe

It Fg é 2 Go hen LAAN

: . Se be

é BA & pas eo KREG

SS ss ips ~ £4

Fe 5100.

a if T } T ‘3 PRE. Ns Lexat AOL ALF i

RS oe « sme me 2 |

Dopmanscr ora sen Un eae NAN i aL ARPA

Pate 1.—A. Brachystelma tenue. B, B. vahrmeijeri.

378

hairs 0.5-1 mm long. Leaves opposite, shortly petiolate or subsessile, ovate.

ovate-elliptic, or oblong-elliptic, 1-2 cm long, 1.5-5 mm broad, thinly hirsute on

the back and margins, glabrous or with very few scattered hairs on upper surface.

Flowers 1 or 2 produced laterally at the nodes, opening together, subtended by

filiform bracts about 1.5 mm long; pedicels slender, 8-10 mm long, 0.25 mm

diam., soon enlarging up to 1 mm diam. after fertilization. Sepals linear, 3 mm

long, sparsely hairy. Corolla yellowish-brown, about 1.5 cm long, cage-like.

sparsely hairy on outer surface; tube more or less saucer-shaped, 1 mm deep,

slightly recurved at the sinuses; lobes ovate at base, linear above and united

at the tips. Corona arising from about the middle of the staminal column;

outer-lobes about 0.75 mm long, deeply divided into 2 filiform horns, slightly

overtopping the staminal column; inner corona-lobes ovate-oblong, incumbent

on the backs of the filaments and about equal in length. Pollinia somewhat

pyriform 0.25 mm long, slightly compressed, with short translucent wing on

upper 4+ of the inner margin, with short connectives about the middle attached

to the small wingless carrier.

Brachystelma vahrmeijeri R. A. Dyer, sp. nov., B. flavido Schltr. affine sed

corollae tubo longiore, lobis latioribus, corona exteriore cupulata differt.

Herba tuberosa basi ramosa usque 10 cm alta; tuber rubrum leviter com-

pressum, usque 5 cm diam.; rami recti minute pubescentes vel glabrescentes.

Folia plus minusve elliptico-lanceolata, usque 3 cm longa, 7 mm lata, glabra

vel nonnunquam minute et sparse ciliate, breviter petiolata vel subsessilia. Flores

2-3 extra axillares, pedicellis 5-10 mm longis; calyx glaber segmentis lineari-

lanceolatis vel lanceolatis, 2.5 mm longis. Corolla plerumque flavo-virens plus

minusve 8 mm longa, glabra, tubo 3 mm longo infundibuliformi, lobis ovato-

triangularibus 5 mm longis, basi 2.5 mm latis. Corona exterior cupulata circiter

1 mm alta, 1.5 mm diam., margine minute 10-emarginato, lobi interiores ovato-

oblongi, antheris leviter incumbentibus.

Type: Natal, Ubombo-Ingwavuma border, near Lala Nek, about 3 km from

sea, open veld, Vahrimeijer 1050 (PRE, holo.).

Tuber red, up to about 5 cm in diam., slightly compressed from above and

below. Stems several from the centre of the upper surface, up to about 10 cm

tall, rarely rebranched above, minutely pubescent or glabrescent. Leaves more

or less elliptic-lanceolate, up to 3 cm long, 7 mm broad, the lowest smaller,

glabrous on upper and lower surfaces, sometimes remotely and minutely ciliate

on margin, contracted into a short petiole or subsessile. Flowers 2-3 together,

extra axillary, the oldest uppermost and opening in succession; pedicels 5-10 mm

long, subtended by short linear-lanceolate bracts. Calyx with sepals linear-

lanceolate to lanceolate, 2.5 mm long, glabrous. Corolla yellowish-green or

cream, rarely white or somewhat maroon, glabrous, 8 mm long; tube 3 mm long,

funnel-shaped, 4 mm diam. at mouth; lobes ovate-triangular, 5 mm long, 2.5 mm

broad at base. somewhat spreading and somewhat fleshy. Corona arising very

slightly above base of staminal column; outer corona cupular, 1 mm high, 1.5 mm

diam., twice as high as the staminal column, shallowly 10-notched or subentire,

with few long hairs within; inner corona-lobes arising about halfway down inner

wall of outer corona, linear-oblong, incumbent on back of filaments but not as

long as them. Pollinia about 0.25 mm diam., somewhat pyriform, compressed,

with narrow translucent margin 3 length of inner margin, with short connectives

from near base to small wingless carrier.

379

Bothalia 10, 2 : 379-383

A Further New Species of Cycad from the

Transvaal

R. A. Dyer

Encephalartos cupidus R. A. Dyer, sp. nov., planta pygmaea foliolis plus

minusve glaucis, marginibus utrinque 2-5-spinosis, megasporophyllorum vulticulo

terminali 2.5-3 cm lato, 1.5-2 cm alto distinguitur.

Planta solitaria vel basi ramosa, truncis usque 75 cm altis cylindricis 20-30

cm diam. Folia plus minusve glauca lineari-oblonga prope basin attenuata, circa

1 m longa, foliolis infirmis 1-3 spinosis reductis, petiolorum basibus lanatis gla-

brescentibus sed pulvinis extus permanente lanatis, foliola mediana_lineari-

lanceolata acuminata 10-15 cm longa, rariter usque 19 cm, 10-13 cm lata, pun-

gentia, marginibus utrinque 2-5-spinosis. Strobilus masculinus subcylindricus

usque 18 cm longus, 5 cm latus, pedunculo 5-10 cm longo, microsporophylla

congesta, mediana 2.2-2.5 cm longa, 1.7-1.9 cm lata; bulla glabra circa 5 mm

prominens; vulticulus terminalis, circa 4.5 « 4.5 mm. Strobilus femineus im-

maturus glaber ovoideus, 18-20 cm longus, 10 cm_ latus; megasporophylla

mediana circa 3 cm longa; bulla 3.5-4 cm lata, circa 2.5 crassa, 5-8 mm longa;

lobi 2 laterales incurvi circa 5 mm longi; vulticulus termiralis 2.5-3 cm latus,

1.5-2 cm altus, margine paulo prominenti.

Type: Transvaal, Pilgrim’s Rest Division, cultivated at White River, D.

van Heerden sub PRE 30546 (PRE, holo.).

Plants unbranched or branched from the base, up to about 75 cm tall,

20-30 cm in diam., covered by alternate series of leaf-bases and bracts; bracts

about 5 cm long, lanceolate, acuminate, whitish woolly on outer surface. Leaves

about 1 m long, linear to narrowly oblong in general outline, slightly narrowed

to apex and to base; rhachis more or less straight or spreading and curved,

with the tip up-turned rather than recurved, more or less glaucous-green; pul-

vinus about 3.5 cm broad and long, woolly on outer surface; petiole-like base

4-8 mm long, up to 10 mm diam., tomentose when young, glabrescent; leaflets

more or less glaucous-green, inserted up to 3 cm apart towards base, usually

closer and more crowded in the upper half and often slightly overlapping, in

V disposition from rhachis, reduced to 2-3 prickles at the base; median leaflets

linear-lanceolate, more or less equal-sided. curved slightly upwards, 10-15 cm

long, rarely up to 19 cm, 10-13 mm broad, about 17-nerved, usually with both

upper and lower margins with 2-5, rarely more, forward-directed sharp prickles

about 4-7 mm long and terminating in a single prickle or rarely with 2-3 sub-

equal apical prickles. Cones: male and female apparently usually solitary from

apex of stems, dissimilar, glabrous. Males cones (nearly mature, slightly dried

out) with peduncle 5-10 cm long, subcylindric, about 18 cm long, 5 cm diam.

about the middle, narrowed to apex and base; median scales spreading at right

angles to axis, congested, 2.2 cm long, 1.7 cm broad; cuneate at base, with

Piare 1.—Encephalartos cupidus: left, male stem with cone (PRE 30546a); right, female

stem with cone (PRE 30546b).

microsporangia extending to margins and 3-5 mm from terminal facet; upper

surface more or less Sat with indication of keel, becoming minutely honey-

combed on drying; bulla glabrous, projecting about 5 mm, up to 1.7 cm broad,

6-7 mm thick vertically; upper facet humped in centre with narrow shoulders;

lower facet is a continuation of the sporangial surface; terminal facet 4-5 mm

broad and 4-5 mm wide vertically, slightly concave. Female cones (immature)

18-20 cm long, about 10 cm in their greatest diam., slightly narrowed to apex,

glabrous, bright apple-green, with short stout peduncle, subtended by slender

woolly bracts; median scales about 3 cm long; bulla protruding 5-8 mm, 3.5-4

cm broad, about 2.5 cm thick vertically, with lateral ridges extending into in-

curved lobes about 5 mm long clasping the seeds, with the median clasping

lobe broadly rounded (extension of upper facet); upper facet irregularly shaped,

sometimes with 1 or 2 indistinct ridges, sometimes more or less rounded, fairly

381

PLate 2.— Encephalartos cupidus, young plant producing suckers from base (PRE 30546c).

smooth; lower facet receding more or less at right angles from the terminal

facet; terminal facet irregular in shape; 2.5-3 cm broad, 1.5-2 cm wide ver-

tically, fairly smooth, with slightly raised rim; seeds immature.

TransvaaL. — 2430 DA (Pilgrim’s Rest Division): cultivated at White

River, D. van Heerden, sub PRE 30546; PRE 30546a (male cone); b (female

cone); c (small plant); Blyde River Nature Reserve (farm Steenveld). Alberts,

Nature Conservation Officer, sub PRE 30545a (male plant); b (female plant);

c (sterile plant).

The first evidence received of the present new species was a leaf sent to

the Botanical Research Institute, Pretoria, by Mr. J. S. Oliver from a plant in

cultivation in the garden of Mr. David van Heerden of White River. The

382

parent plant had originated some time back from the farm Steenveld on the

border of the Pilgrim’s Rest and Lydenburg districts in the eastern Transvaal.

The area has since been incorporated in the Blyde River Nature Reserve.

Between Mr. van Heerden and Nature Conservation Officers of the Trans-

vaal Province, and Dr. Leistner of the Botanical Research Institute, sufficient

material has been made available for a fairly full description to be prepared.

So far, however, only 2 immature female cones have been seen. They agree

closely in essential characters and have a conspicuously large terminal facet to

the scales. These, together with its dwarf habit, somewhat glaucous leaves and

strongly prickly leaves, distinguish the species from among its southern neigh-

bours. On the characters mentioned, its nearest relative appears to be E. munchii

Dyer & Verdoorn, described from south of Vila Pery in Mozambique. The

latter species, however, has far larger cones, the males cf which are more than

one on a plant, tall and lax.

Prare 3.—Encephalartos cupidus, plant in natural habitat in Blyde Rivet

Nature Reserve, Eastern Transvaal.

383

The full distribution range of FE. cupidus has not yet been surveyed but it

is likely to be of strictly limited extent. The colony which was investigated by

Mr. G. Alberts on the farm Steenveld covers a few square miles and consists

of approximately 120 specimens. The locality is still rather inaccessible and

the terrain very rough, as indicated by the name of the farm meaning stony

ground.

Astonishment was expressed in 1964, when first describing Encephalartos

inopinus R. A. Dyer, that an undescribed species of this remarkable genus

should be discovered at this comparatively late stage in the botanical exploration

of southern Africa. And now, five years later, another undescribed species has

been brought to light.

So great is the interest in specimens of Encephalartos to adorn gardens,

both public and private, that the specific epithet given here, meaning ‘desirous’

could be applied in its good sense to any species of the genus. However, the

strict protective laws and penalties designed to ensure the sanctitiy of plants

in their native habitats are, in fact, insufficient to safeguard plants or to deter

some collectors. Thus unfortunately the tarnished meaning of the specific epithet

implying a passionate desire, to the extent of greed or lust, would also be

appropriate on occasions. At the same time it is well to remember that pro-

tective laws should not impede the bringing to light of new scientific information

and this should be borne in mind by those empowered to administer regulations

for Nature Conservation.

385

Bothalia 10, 2 : 385-401

Acacia karroo in Southern Africa

J, H. Ross

ABSTRACT

Reasons for adopting the name Acacia karroo Hayne are considered and the nomen-

clature and synonymy are dealt with. A broad description of the species is provided. The

relationship of A. karroo to the closely related A. seyal Del., A. hockii De Willd. and

the glandular podded Acacia species is mentioned. Acacia karroo Hayne, which is the most

widespread Acacia in southern Africa, is an extremely variable species which occupies a

diverse range of habitats. The range of variation within A. karroo, and in particular in

Natal, is considered. At least six entities are recognizable within A. karroo in Natal. The

nature and range of variation within these entities is considered. No infraspecific categories

are recognized within A. karroo.

The southern African Acacia species for which the correct name is now

known to be Acacia karroo Hayne often forms a conspicuous feature of the

landscape in the western Cape where it is the only Acacia to be seen for miles.

On account of its occurrence in proximity to Cape Town it was the first Acacia

encountered by early travellers in the interior of southern Africa. A number

of these travellers (Simon van der Stel, Barrow, Lichtenstein, Sparrman, Thun-

berg and later Burchell) mentioned the plants in the accounts of their journeys

and collected specimens which were later sent to Europe. This was ultimately

responsible for some of the confusion concerning the correct name for this

species.

Mimosa _ nilotica Burm. f., Prodr. Fl. Cap. 27 (1768), is apparently the

earliest name applied to this southern African Acacia. For this species Burman

quoted a figure published by Plukenet in his Phytographia, t. 123, Fig. 1 (1691)

and mentioned that there were dried specimens preserved. Examination of

Plukenet’s figure, which consists only of a small vegetative shoot, shows that

the leaflets are far too large for Mimosa nilotica L. (Sp. Pl. 521, 1753) whilst

it is known that M. nilotica, or to give it its correct name, Acacia nilotica (L).

Willd. ex Del., does not occur in the Cape. Verdoorn in Bothalia 6 : 409 (1954)

mentions having received photographs of the only two Acacia specimens in

Burman’s collection from Prof. Baehni, Director of the Conservatoire et Jardin

Botaniques, Geneva. One specimen is a seedling that cannot be identified with

certainty. The other specimen has written on it “Mimosa nilotica Linn.” and

“Plukn. Tab. 123. Fig. 1°. The name Mimosa capensis was later written over

Mimosa nilotica. This specimen of Burman’s is certainly referable to what is

now known as Acacia karroo so it is evident therefore that Mimosa_nilotica

was a wrong identification by Burman.

Mimosa nilotica Thunb., Prodr. Pl. Cap. 92 (1800) was also an incorrect

identification. This is clarified by Barrow in his Travels: 89 (1801) where he

pointed out that the Swedish travellers (Sparrman and Thunberg) had erro-

neously called the Karroo Mimosa the nilotica, or that which produces the

gum Arabic.

386

Mimosa capensis Burm. f., Prodr. Fl. Cap. 27 (1768), was based on a

figure published by Plukenet in his Phytographia, t. 123, Fig. 2 (1691). However,

Linnaeus referred to this same figure for his M. reticulata L., Mant. 1 : 129

(1767), and added to it a description of a plant from the Upsala garden with

a flat reticulate pod. Linnaeus’s description of the pod is quite at variance

with the pod figured by Plukenet. Unfortunately Plukenet’s illustration cannot

be identified with any South African species of Acacia with certainty. Both

names based on Plukenet t. 123, Fig. 2, that is Mimosa capensis and M. reticu-

lata, must therefore be rejected as there is no preserved dried specimen of either

species and as the figure cannot be identified.

Burchell in his Travels, | : 114 (1822), used the name Acacia capensis B.

for the Karroo-thorn-tree but did not give a Latin diagnosis as was his custom

with a new species. This suggests that Burcheli was making a new combination

but nowhere does he confirm this. He provided an unmistakable description

of the plant on page 189 and there is a vignette of it on the same page. How-

ever, Bentham, who worked with Burchell on his Leguminosae, wrote in Trans.

Linn. Soc. 30:507 (1875) “Burchell adopted the name A. capensis for this

species, supposing it to be the Mimosa capensis Burm., Fl. Cap. Prodr. 27,

which it probably is.” Burchell’s name, which is the combination in Acacia

based on Mimosa capensis Burm. f., must therefore also be rejected.

Jacquin’s plate of Mimosa leucacantha, Hort. Schoenbr. 3 : 75, t. 393 (1798),

provided the first readily identified illustration with a definite name but the

combination of the epithet with Acacia is rendered inadmissable by A. leuca-

cantha Vatke in Oesterr. Bot. Zeitsch. 30: 276 (1880) for an entirely different

species.

For many years A. karroo was incorrectly called A. horrida (L.) Willd.

Hillcoat and Brenan in Kew Bull. 13: 39-40 (1958), in establishing the true

identity of A. horrida (L.) Willd., explain how the name “Acacia horrida’,

which was originally based on a quite distinct Indian species (Mimosa horrida

L., Sp. Pl. 521, 1753), was gradually accepted as the correct name for this

South African species.

Acacia karroo was described by Hayne in Arzneyk. Gebr. Gewachse 10 : t.

33 (1827). Hayne stated that his illustration was made from a specimen named

Acacia vera in the Willdenow Herbarium (No. 19184 fol. 2) and from another

specimen which he received from the Cape. The Director of the Botanischer

Garten und Museum, Berlin-Dahlem, to whom I am very grateful, sent me a

photograph of this specimen from Willdenow’s Herbarium. I have not suc-

ceeded in tracing the other specimen mentioned by Hayne. Examination of

the labels on this sheet from the Willdenow Herbarium (No. 19184 fol. 2) reveals

that the specimen was initially called Mimosa nilotica. Mimosa _ nilotica was

subsequently erased on the one label and “‘Acacia vera’’ was written over it

in dark ink. A. vera Willd. (Sp. Pl. 4: 1085—1806) is given as a synonym of

A. arabica (now A. nilotica (L.) Willd. ex Del.) by Bentham in Trans. Linn.

Soc. 30: 506 (1875). From Willdenow’s description of A. vera it is clear that

Bentham’s decision was correct. It is quite certain that this specimen from

Willdenow’s Herbarium (No. 19184 fol. 2) is not referable to A. nilotica for

the leaflets are far too large for that species. The name “‘Vieweg”’ also appears

on one of the labels. Wagenitz in Willdenowia Bd. 3, Heft 1 : 109-136 (1962)

mentions that the name ‘‘Vieweg” occurs frequently on specimens and on covers

in the Willdenow Herbarium. Because of this frequent appearance of the name

“Vieweg”’ it might be concluded that Vieweg was the collector of these plants.

Specimens from as far afield as North America, Jamaica, Europe and particu-

larly from the Mediterranean area, plus 100 specimens from South Africa bear

387

the name “Vieweg’’. Despite this, nothing much is apparently known about

Vieweg and there is no evidence that he was a collector. Nobody could have

travelled so extensively before 1800 and yet have remained so unknown. Vieweg

was apparently the owner of an Herbarium hence the appearance of his name

on so many specimens. It is thought that Vieweg handed the specimens on to

Willdenow who then described them. Unfortunately Willdenow seldom gave

the name of the collector when describing species. There is no record of where

the specimen No. 19184 in the Willdenow Herbarium was collected. Although

one of the labels reads “Habitat in Aegypto, Arabia fh” I know of no species

from North Africa to which the specimen could be referred. However, the

specimen is a very good match of a number of specimens that I have examined

from the Cape. This close resemblance of the specimen in the Willdenow Her-

barium (No. 19184) to many other Cape specimens suggests strongly that the

specimen was collected in the Cape and that the phrase “‘Habitat in Aegypto.

Arabia fh’? does not refer specifically to this specimen but rather to the species

A. nilotica and is taken either from Linnaeus’ description of Mimosa_nilotica

or from Willdenow’s Acacia vera. Presumably either specimen No. 19184 or the

other species mentioned by Hayne was collected in the Karoo, whence the

specific epithet for this species. Acacia karroo Hayne is therefore the earliest

valid name for this common southern African Acacia.

A. karroo was introduced into Mauritius and became naturalized there. It

was erroneously identified as Mimosa eburnea L.f. by Bojer and was listed

under the name in his Hort. Maurit. 115 (1837).

A. karroo Hayne, Arzneyk. Gebr. Gewachse 10: t. 33 (1827); Glover in Ann.

Bolus Herb. 1 : 150 (1915); Burtt Davy in Kew Bull. 1922 : 328 (1922); Marloth,

RES heAticen 25-2 (1925)2Bakent. eos irops Atri 843) 7(1930). Burtt. Davy, El:

Transv. 2: 346 (1932); Henkel, Woody Pl. Natal 229 (1934): Gerstner in J.S.

Afr. Bot. 14: 19-27 (1948); Codd, Trees & Shrubs Kruger N. Park : 44, fig. 38h

& i (1951); Miller in J.S. Afr. Bot. 18 : 22 (1952); Verdoorn in Bothalia 6 : 409

(1954); Fl. Pl. Afr. 31 : t. 1220 (1956); Palmer & Pitman, Trees of S. Afr.

[Sal SSeS 6nes70 (1961) sawihitess Kor He Ne Rhods (855 figs TsDN (962) sv.

Breitenbach, Indig. Trees S. Afr. 2 : 298 (1965); de Winter et al., Sixty-Six Tvl.

Trees 50-51 (1966); Schreiber in Prod. Fl. S.W. Afr. 58:9 (1967). Type: Herb.

Willdenow 19184 (B, lecto.: PRE, photo.).

Mimosa nilotica sensu Burm. f., Prodr. Fl. Cap. : 27 (1768), non L.

M. capensis Burm. f., Fl. Cap. : 27 (1768). pro parte.

M. leucacantha Jacq., Hort. Schoenbr. 3 : 75, t. 393 (1798), non Acacia leuca-

cantha Vatke (1880).

Acacia horrida sensu Willd., Sp. Pl. 4, 1082 (1806), pro parte, quoad Jacq.

fig.; sensu auct. mult. : *E. Mey., Comm. | : 166 (1835); Harv. in Fl. Cap. 2 : 281

(1865); Benth. in Trans. Linn. Soc. Lond. 30: 507 (1875); Engl. in Bot. Jahrb.

10:23 (1888): Marloth in Trans. S.A. Phil. Soc. 5:270 (1889); Fourcade,

Report on Natal Forests : 106 (1889); Schinz in Mem. Herb. Boiss. | : 113 (1900)

quoad Marloth 1334, excl. specim. Liideritz 122 [The identity of Ltideritz 122.

which I have not seen, is in some doubt as Schinz cited it as A. horrida on

* As mentioned earlier the name “A horrida (L.) Willd.” was widely misapplied to this

common southern African Acacia, the correct name of which is now known to be

A. karroo. A horrida (L.) Willd: is a distinct species that is found in East Africa and

Asia.

388

p. 113, as A. trispinosa Marl. & Engl. on p. 115 and as A. aff. trispinosa on

p. 116]; Sim, For. Fl. Cape Col. 211, Pi. LXI (1907); Burtt Davy in Kew Bull.

1908 = 158 (11908)3 Sims Fors FE PAE? Air 257 (1909):

A. capensis sensu Burch., Trav. 1: 114, 189 (1822); Sw., Hort. Britt. 1: 103

(1826); Colla in Mem. Acad. Torin 35: 175 (1831); Eckl. & Zeyh., Enum. 260

(1835).

Mimosa nilotica Thunb., Fl. Cap. ed Schult. 432 (1823), non L.

Acacia hirtella B. Mey., Comm. 1: 167 (1835); Harv. in EI Cap: 23281

(1865): Benth. in Trans. Linn. Soc. Lond. 30: 513 (1875). 2Glover in Ann. Bolus

Herb. 1: 150 (1915). Type: Natal, between Umkomaas and Umlaas, Drege

(whereabouts unknown).

A. natalitia E. Mey., Comm. 1: 167 (1835); Harv. in Fl. Cap. 2 : 281 (1865);

Benth. in Trans. Linn. Soc. Lond. 30: 508 (1875); Burtt Davy in Kew Bull.

1908 : 159 (1908); Glover in Ann. Bolus Herb. 1: 150 (1915); Burtt Davy in

Kew Bull. 1922 : 329 (1922); Fl. Transv. 2 : 347 (1932); Gerstner in J.S. Afr. Bot.

14:22 (1948). Syntypes: Natal, Port Natal (Durban) and Umgeni, alt. 300 ft,

Drege (? K, iso.); Pondoland, between Umgazana and Umzimvubu, alt. 600-

1000 ft, Drege (whereabouts unknown).

Mimosa eburnea sensu Bojer, Hort. Maurit. 115 (1837), non L.

Acacia horrida Willd. var. transvaalensis Burtt Davy in Kew Bull. 1908 : 158

(1908). Syntypes: Transvaal, Pretoria district, Groenkloof, near Pretoria, Burtt

Davy 2468 (BOL!; PRE!); Arcadia, Pretoria, Burtt Davy 2807 (PRE!).

A. karroo Hayne var. transvaalensis (Burtt Davy) Burtt Davy in Kew Bull.

NOQQTIS 23 (1922) Srl eiransvaee eS 47 93D).

A. inconflagrabilis Gerstner in J.S. Afr. Bot. 14 : 24-26 (1948). Syntypes: Natal,

Nongoma district, Nongoma township Gerstner 4562 (NBG!; NH!; PRE!);

Gerstner 4635 (NBG!; NH!; PRE!); Gerstner 4637 (NBG!).

Shrub, often many stemmed, or a tree to 22 m, sometimes very slender

and sparsely branched; crown rounded, often irregularly so, or flattened; trunk

to 0.75 m in diameter. Bark dark brown, reddish-brown, brownish-black to

black, rough, often fissured, or white to pale greyish-white or greyish-brown and

smooth, the latter green when young and with conspicuous transversely elon-

gated lenticels. Young branchlets reddish- to purplish- or blackish-brown or

white to yellowish- or greyish-brown, flaking minutely or smooth, conspicuously

or inconspicuously lenticellate, glabrous or sometimes sparingly pubescent,

especially when young. Stipules spinescent, glabrous, in pairs below the nodes,

2non Sim in Agric. Jour. 19 (1900); non Sim in For. and For. Fl. Cape Col.: 211,

t. LIX (1907). Sim was apparently referring to A. sieberana DC. var. woodii (Burtt Davy)

Keay & Brenan as evidenced by his description of the bark “yellowish white flaky” and

of the pod as “4 inches long, 3-1 inch wide, solid, indehiscent, tomentose.” However,

A. sieberana var. woodii seldom has only 4 pinna pairs as described by Sim and the

involucel iis in the upper half of the peduncle or apical and not in the lower third

as illustrated in t. LIX. Sim described the inflorescence as “light yellow or nearly white”

which is in contrast to the deep yellow inflorescence of A. karroo.

A. robusta Burch., however. which often has only 4 pinna pairs, has a_ whitish

inflorescence and has the involucel in the lower third so it appears as though Sim’s

description of A. hirtella might possibly be taken from A. sieberana var. woodii and

from A. robusta Burch.; non Sim in For. Fl. P.E. Afr.: 57, t. XXXV A (1909). In this

instance Sim was clearly referring to A. robusta Burch.

389

straight or slightly curved, white or the same colour as the stem, usually 0.4-4.5

(-10.0) cm long, sometimes swollen and greatly elongated to 25.0 cm long, latter

elongated spines usually united basally, entire plant frequently exceedingly

spinescent. Leaf: petiole 0.5-1.8 cm long, glabrous or sometimes sparingly

pubescent, adaxial gland usually present, variable in position, usually rounded

or oval, at times slightly stalked, to 1.5 x 1.5 mm; rachis (0-) 1.0-4.6 (-9.0) cm

long, glabrous, sometimes sparingly pubescent, abaxial surface without recurved

prickles, sulcate adaxially; glands rounded or sometimes stalked, yellowish- to

reddish-brown or black, between top 1-3 pinna pairs, between all pinna pairs

or absent from some; pinnae (1—) 2—6 (—13) pairs; rachillae (1.0-) 1.5—3.8 (—7.2)

cm long, glabrous or sometimes sparingly pubescent; leaflets 6-15 (—24) pairs,

(2.8-) 3.5-8.0 (-12.5) mm long, 1.0-2.5 (—5.0) mm wide, linear, linear-oblong

to obovate-oblong, base oblique, apex rounded to sub-acute or acute, margin

entire, glabrous or occasionally sparingly puberulous. /nflorescence capitate, on

axillary peduncles, fascicled or sometimes solitary, forming terminal racemes,

sometimes on lateral axillary branchlets the entire inflorescence producing an

irregular terminal panicle; flowers sessile, bright yellow; peduncle 0.7—2.4 (—4.0)

cm long, terete, olive- or reddish-brown, glabrous, occasionally sparingly pubes-

cent, sometimes glandular; involucel at, slightly above or below middle (down

to one third) of peduncle (when the flowers are young the involucel appears

to be at the apex of the peduncle, however, as the peduncle lengthens the

involucel soon assumes its true position). Calyx deep yellow, campanulate,

glabrous throughout or apices of lobes sometimes sparingly pubescent, tube

1.2-1.8 mm long, lobes up to 0.5 mm long. Corolla deep yellow, glabrous,

tube 1.5-2.3 mm long, lobes to 0.8 mm long, reflexed, alternating with calyx

lobes; stamen filaments free. up to 5 mm long, yellow: anthers with deciduous

apical gland; ovary glabrous, shortly stipitate, up to 1.5 mm long; style glabrous,

up to 5 mm long. Legume dark ycllowish- or reddish-brown to brown, straightish

or slightly to strongly falcate. irregularly constricted between the seeds, often

distinctly moniliform, (4.4-) 5.0-10.5 (21.0) cm long, 0.5—0.7 (-1.1) cm wide.

apex rounded to acute or acuminate, sometimes attenuate at both ends, dehiscing

longitudinally, subcoriaceous, venose, usually longitudinally so, often very con-

spicuously, glabrous, sometimes glandular, umbonate over the seeds. Seeds olive-

brown to reddish-brown, elliptic or lenticular, sometimes + quadrate or sub-

orbicular, compressed, (3.5-) 4.5-6.5 (-9.0) mm x (2-) 3-4 (-7) mm_ wide:

areole elliptic or lenticular, sometimes subcircular, 3.0-5.5 (-7.5) mm x 2.0-3.5

(4.5) mm.

A. karroo Hayne is the most widespread Acacia in southern Africa (see

Fig. 1). Not only is A. karroo widespread, but it is numerically well-represented

throughout most of its range. A. karroo, being so widespread, has exploited

many diverse habitats and is consequently an exceedingly variable species. Story

in Mem ss Bot. Sunve (S) Afr, 27-28 (11952), noted; ~ Ihe fact that Acacia (A:

karroo) is South Africa’s most widely distributed tree suggests strongly that it

is also the least exacting in its demands, and that it would often be the first to

migrate into an unfavourable area. One could also expect to find pure and

permanent stands more and more strongly marked according as the areas were

progressively less suited to trees — obviously as far as these areas were not

too unfavourable to preclude its growth altogether.” A. karroo has the ability

to encroach rapidly into grassland grazing areas and is consequently considered

a serious menace in parts of its range. Attempts to eradicate plants by chop-

ping often result in a vigorous coppice growth. Of all the indigenous Acacia

species, A. karroo appears to be subjected to the severest attacks by the wattle

bagworm, Kotochalia junodii (Heyl.). The degree of infestation is often sufficient

to kill fairly large trees.

Fic. 1— The known distribution of

specimens, field observations, on an unpublished map prepared by Acocks in 1965S,

and on De Winter et al. in Sixty-Six Transvaal Trees : 50-51 (1966).

Plants tend often to have a different “look”? in various parts of the species

geographical range. In the arid regions of the northern Cape and in South West

Africa plants are confined to the banks of dry watercourses or other areas where

underground water is available. In parts of the Transvaal plants are often exceed-

ingly robust and are vegetatively easily confused with A. robusta Burch. In the

reproductive phase, A. karroo is, however, readily distinguishable from A. robusta

in having bright yellow flowers and much narrower, less woody pods.

A tendency of A. karroo, shared also by A. seyal Del., A. hockii De Willd.,

A. nilotica (L.) Willd. ex Del. and sometimes also by A. davyi N.E. Br., is for

a few flowers to cevelop in the involucel on the peduncle, sometimes giving the

appearance of a smaller secondary capitulum below the main one. ‘The flowers

in this secondary capitulum in A. karroo often develop before those in the main

capitulum. Most of these flowers are male only although in a few flowers examined

the ovary was present, but the style absent. The flowers are apparently sterile, but

this needs further investigation.

The relationship of A. karroo to the very closely related A. hockii and to

A. seyal needs careful investigation. A. seyal is widespread in northern tropical

Africa, extending to Egypt and southwards to Zambia, Malawi and Mozambique.

A. hockii occurs from French Guinea in west Africa to the Sudan in the north

39]

and southwards to Angola, Zambia and Mozambique. The northern limit of

distribution of A. karroo, therefore, corresponds roughly to the southern limits

of distribution of A. seyal and A. hockii. Although all three species are found

in Zambia and Mozambique their ranges scarcely overlap. Brenan in Fl. Trop.

E. Afr. Mimos. : 103-105 (1959) enumerates the differences between A. seyal and

A. hockii whilst White in For. Fl. N. Rhod. : 85 (1962) provides descriptions of

all three species.

The glandular podded Acacia species (A. borleae Burtt Davy, A. exuvialis

Verdoorn, A. nebrownii Burtt Davy, A. permixta Burtt Davy, A. swazica Burtt

Davy, A. tenuispina Verdoorn and A. torrei Brenan) in southern Africa all have

a close affinity to A. karroo and appear to have been derived from the latter during

earlier times. It is not clear whether each of the above species was independently

derived from A. karroo or whether some of the species have given rise to others.

Most of the species are now fairly distinct from A. karroo, but certain specimens

cannot be referred either to A. karroo or to A. tenuispina with certainty. Codd

7040 (PRE) from north of Pienaars River (Grid. Reference: 2528 AB PRE-

TORIA) was described by the collector of “possibly a hybrid between A.

karroo and A. tenuispina.’ Some plants have the growth form of A.

fenuispina and agree with it vegetatively but lack the glandular pods. The two

species may well hybridize but this requires careful field investigations. Burtt

Davy 4075, 4077 (PRE) from the Springbok Flats are difficult to place. They are

vegetatively very close to A. tenuispina, but lack glandular pods and are therefore

referred to A. karroo.

A. karroo, therefore, is very closely related to a number of other species.

It is apparently one of the least exacting in regard to habitat preference and has

consequently been able to inhabit a wide range of habitats.

Before proceeding to examine the variation within A. karroo it is necessary

to consider the value of recognizing infraspecific categories within such a widespread

and variable species. Two alternatives are available: to fragment the species and

accord each recognizable entity formal taxonomic status, or to recognize only one

variable species with no infraspecific categories. Clearly a decision must be taken

as to which of the two alternatives to follow. For ecological purposes it is useful

for variants to have names for ease of reference. The significance of infraspecific

“labels”’ is obvious in that if two variants occur in the same community the use

of the same name for each will suggest a degree of similarity which may be mis-

leading. However, the problem of identifying each entity clearly and thereby

facilitating identification by other workers is often extremely difficult.

In Natal, A. karroo occurs from sea level to the top of the Low Berg at

Van Reenen’s Pass (1524 m). It is perhaps in Natal that A. karroo exhibits its

greatest range of morphological variation. At least six entities may be recognized,

namely:

1. shrubs or small trees with dark, rough bark (see Pl. 1) growing in dry thornveld

or dry valley scrub;

2. large trees with dark, rough bark forming a narrow riverine fringe along the

banks of streams in dry thornveld or dry valley scrub:

3. white barked trees or shrubs with short spines (A. natalitia E. Mey.):

4. “fire-resistant” shrubs found in northern Zululand (A. inconflagrabilis

Gerstner);

5. slender, sparingly branched trees in Zululand (pcpularly termed “‘spindle

A. karroo’’);

6. trees with whitish bark, long white spines and long, moniliform pods found

along the Zuiuianad coast.

392

Piare 1—The rough, dark brown to brownish-black bark of A. karroo. Bisley, near

Pietermaritzburg (2930 CB Pietermaritzburg), April, 1967.

393

The first two entities are comparable with specimens of A. karroo found in

other areas of distribution and present no difficulty. The riverine plants are larger

than those in the surrounding dry thornveld or valley scrub and consequently often

enable the course of a stream to be detected from afar.

A. natalitia E. Mey.

E. Meyer, Comm. | : 167 (1835), described A. natalitia from specimens collected

at ““Port Natal et Omgeni . . .; inter Omgaziana et Omsamwubo . . .” by Drege.

Meyer held that A. natalitia differed from A. karroo in its whitish and not blackish

bark, in its spines being short or nearly wanting (‘aculeis saepe brevissimis et vix

ullis, numquam 9 lineas longis’’), in its more numerous pinnae (4-7) and leaflet

(12-18) pairs, and in its smaller and narrower leaflets. Mr. J. P. M. Brenan,

Keeper of the Herbarium and Library, Royal Botanic Gardens, Kew, to whom

I am extremely grateful, informed me that there is a specimen at Kew which may

possibly be an isotype of A. natalitia. This specimen was originally in Bentham’s

herbarium and bears a label reading “‘Acacia natalitia E. M. a.” plus a biblio-

graphical reference in Bentham’s hand. There is a pencil note on the sheet by

Dr. N. E. Brown reading “*Port Natal Umgeni 300 ft. alt.” It is thought that this

specimen may be part of the first Drege gathering mentioned by Meyer (l.c.).

However, the label of this specimen does not bear any collector’s name and it

is only inference that it is part of a Drege specimen although it must have been

accepted by N. E. Brown who was probably very familiar with the early collectors.

The sheet bears a type specimen label. I have not succeeded in tracing the

whereabouts of the other specimen mentioned by Meyer. There are, however,

specimens from Port Natal (Durban) collected by Gueinzius (K, PRE) and

Krauss 66 (K) which agree with the description. A further selection of

specimens, for example Ross 802, 803 (NU) from Uvongo (3030 CD PORT

SHEPSTONE) and Ross 806 (K, NU) from Port Edward (3130 AA PORT

EDWARD), serves to establish the identity of the entity referred to A. natalitia

by Meyer.

Meyer (l.c.) also described A. hirtella from a specimen collected between

Umkomaas and Umlaas. The description of A. hirtella differed from that of

A. natalitia in a few minor points, namely the hairiness of the leaflets in A. hirtella,

the presence of a gland between the first and last pinna pairs as opposed to a

gland between each pinna pair in A. natalitia, and the somewhat acute leaflet

apices in A. hirtella in contrast to obtuse leaflet apices in A. natalitia. Unfortunately

I have not been able to establish the whereabouts of the type specimen of A. hirtella.

However, the specimen Pole Evans in H. Medley Wood 12014 (BOL, NH, SAM)

from Winklespruit (3030 BB PORT SHEPSTONE) serves to establish the identity

of the entity referred to A. hirtella. Gerstner in J. S. Afr. Bot. 14 : 19-27 (1948)

considered A. hirtella to be ‘‘only a young and local variety or modification of

A. natalitia’ and consequently regarded A. hirtella as a synonym of A. natalitia.

The name A. natalitia has been loosely applied to the variant of A. karroo

with white bark not only in Natal but also in the eastern Cape, the Transvaal,

Swaziland and Mozambique. Burtt Davy, Fl. Transv. 2 : 347 (1932) cited several

specimens of A. natalitia from the Transvaal, for example Pott 5304 (PRE) from

Barberton (2531 CC KOMATIPOORT). Gerstner (l.c.), although maintaining

A. natalitia as a distinct species, also mentioned certain specimens which he

regarded as hybrids between A. karroo and A. natalitia. For example, Gerstner

6225 (PRE) from Chipese in the northern Transvaal (2230 CA MESSINA) which

“has bark of A. karroo, leaves of natalitia .. .”

394

From an examination of herbarium specimens it is obvious that Gerstner

studied the variation within A. karroo and A. natalitia in some detail, and over

a period of years. Apparently he initially considered the variant of A. karroo with

long spines and long moniliform pods that is found along the Zululand coast as

a mew species which he proposed calling ‘“A. psammophila.”’ However, on dis-

covering that this name had been used for an Australian species, A. psammophila

Pritz in Engl. Bot. Jahrb. 35 : 294 (1904), he adopted the name ‘‘A. zululandensis.”’

He evidently then considered some specimens from the northern Transvaal, for

example Gerstner 5800, 5846 (PRE) to represent a new species for which he

proposed the name “‘A. karrooidea MS (= forma suluensis Ms)”. Subsequently

Gerstner united his “A. zululandensis’’ and “‘A. karrooidea’’ with the short-spined

variants referred to A. natalitia by Meyer under one species which he called A.

natalitia. A. natalitia in Gerstner’s view in J.S. Afr. Bot. 14 : 19-27 (1948) incor-

porated all of the white-barked variants irrespective of their locality. The wide-

spread application of the name A. natalitia for the variant of A. karroo with long

spines and long, moniliform pods seems to originate from Gerstner’s publication.

Although the pod of A. natalitia was unknown to Meyer, it seems quite clear

to which entity he intended his “A. natalitia’’ to be applied. Meyer made special

mention of the spines being very short or nearly wanting in A. natalitia as opposed

to those of A. karroo as evidenced by his description. The application of the

name ‘‘A. natalitia’’ for the long-spined variant with long monilitorm pods seems

therefore in error.

In the western and northern Cape and in the Karoo itself, A. karroo has

usually 2-3 pinna pairs, although the range is 1-5 pairs, and 6-12 leaflets pairs.

Consequently, the presence of 4-7 (up to 13 are recorded) pinna pairs and 12-18

(—24) leaflet pairs in parts of the eastern Cape, Natal, Swaziland and the eastern

Transvaal suggests at first sight these characters are of importance in distinguishing

the entity from A. karroo. However, despite this tendency of A. natalitia to have

a greater number of pinna and leaflet pairs, when the entire range of morphological

variation of A. karroo throughout its distributional range is examined, the differ-

ences provide no discontinuity. Consequently, A natalitia is not regarded as specifi-

cally distinct from A. karroo nor is it maintained at infraspecific rank within A.

karroo.

A. inconflagrabilis Gerstner

Gerstner (l.c.) described A. inconflagrabilis from the Nongoma district of

Zululand (2731 DC LOUWSBERG). A. inconflagrabilis was said to be always a

shrub “‘in the mistbelt area and transition from mistbelt to grassveld and bushveld”’

in contrast to its nearest relations A. karroo and A. natalitia which “grow into

trees and inhabit the dry bushveld.”’ The leaflets of A. inconflagrabilis were said

to be shiny and narrower than in the other two species although Gerstner conceded

that “‘Purely vegetative Herbarium specimens of these two (A. inconflagrabilis

and A. natalitia), if already dried, are impossible to distinguish.” The type locality

is an area that is usually heavily overgrazed. Consequently, the grass cover is kept

very short and at times is very scant. Grass fires, therefore, do not generate so

much heat and it would be interesting to ascertain whether A. inconflagrabilis 1s

indeed more fire-resistant than A. karroo, or whether this impression is gained

because plants of A. inconflagrabilis are never subjected to such intense heat

as are plants of A. karroo growing in tall grassland.

A. inconflagrabilis is not considered sufficiently distinct from A. karroo for

retention at specific rank nor at infraspecific rank within A. karroo.

“Spindle A. karroo”’

Henkel in his report on the Plant and Animal Ecology of the Hluhluwe Game

Reserve : 18 (1937) referred to ‘‘a dwarf or spindly form” of A. karroo. Henkel

wrote “This (Dichrostachys glomerata and dwarf Acacia karroo association) is

the most important of the lowlands associations and ccvers a large area, chiefly

the eastern part of the lowlands.’ This variant of A. karroo, which has subse-

quently been widely known as “‘spindle A. karroo,” is also found in the Umfolozi

Game Reserve, in the corridor linking both reserves and northward to Rooirand.

Plants grow typically as slender, relatively unbranched trees up to 6 m high

(see Pl. 2). Typically the bark is bright reddish-brown and flaking minutely, the

foliage is glaucous, the petiolar gland is large, flattened and discoid, and there is

a large gland between each, or almost every pinna pair. The paired spines are

usually very short although often they are completely absent. A few specimens

will serve to establish the identity of this variant: Downing 451, 452, 453 (NH,

NU) from Umfolozi Game Reserve; Bourquin H60307, H60308 (NH) from Hluh-

luwe Game Reserve.

Piate 2.— Slender, relatively unbranched specimens of “Spindle A. karroo” up to 6 m high.

A. caffra (Thunb.) Willd. in foreground, Maytenus senegalensis (Lam.) Exell left

foreground and Ceratotherium simuni simum Burch. centre. Hluhluwe Game Reserve

(2832 AA Mtubatuba), April, 1963.

“Spindle A. karroo’’ is not as common within the Hluhluwe Reserve as

“typical” A. karroo. However, there is a complication because “‘typical” A. karroo

also tends to be slender and often only sparingly branched (see Pl. 3), especially

when young, but plants ultimately become fairly well branched with a fairly

396

Ee ~

Prater 3.—Slencer, relatively unbranched young specimens of “‘tyrical” A. karroo up to 7 m

high. A. carira ii foregrouna. Corridor between Hluhluwe and Umfolozi Game Reserves

(2831 BD Nkané@la), March, 1964.

dense crown (see Pl. 4). The bark on these plants, although sometimes reddish-

brown, is often greyish-black with a reddish-brown inner bark and the foliage

only slightly glaucous as opposed to the bright reddish-brown bark and distinctly

glaucous foliage in ‘spindle A. karroo.’’ However, these plants possess the large

petiolar gland and the large glands between the pinnae exhibited by “spindle

A. karroo”’ (Ross & Moll 1773).

On the badly overgrazed areas outside the southern entrance to the Hluhluwe

Game Reserve plants grow as small, slender, often much-branched trees or shrubs

up to 2 m high. These plants have bright reddish-brown bark that flakes minutely

and glaucous foliage. However, the large flattened petiolar gland and the large

glands between the pinnae that are usually associated with the glaucous foliage

are absent whilst the peduncle and young pods are distinctly glandular and the

latter somewhat viscid (Ross & Moll 1770).

A “spindle”? growth form is also recorded (Codd 8435 in PRE) from near

the Loskop Dam in the Transvaal (2529 AD WITBANK). This specimen differs

from the Natal “‘spindle A. karroo”’ in leaf and in pod characters.

Growth form alone does not distinguish this variant, because nearly all of the

A. karroo in the Hluhluwe Reserve has the slender relatively unbranched habit,

especially when young. However, whereas “‘typical’’ A. karroo continues to grow

and branch until it is a fairly large tree up to 10 m high “spindle A. karroo”

Ber acd

Prare 4.—Fairly well branched, more mature specimens of “typical” A. karroo up to 8 m

high growing together with more slender specimens. Hluhluwe Game Reserve (2832

AA Mtubatuba), March, 1964.

remains slender and seldom appears to attain a height of over 6 m. These slender

plants are often subjected to fairly severe mechanical breakage during strong winds.

A. karroo often grows in extremely dense, pure stands within the Reserve,

individuals being apparently of similar age as if germination was stimulated

simultaneously by some environmental factor such as an unseasonal fire.

Emphasis on the growth form of “spindle A. karroo’’ has masked what is

probably a more important taxonomic character in attempting to distinguish the

variant from “‘typical’’ A. karroo, namely the glaucous foliage. Leaflet shape

is perhaps also important because in “spindle A. karroo’’ the leaflets are often

broader in relation to their length than in “typical” A. karroo. However, this

character provides no clear distinction when leaflet shape of A. karroo from the

entire distributional range is examined.

Field observations within the Hluhluwe and Umfolozi Reserves indicate that

“typical” A. karroo and “spindle” A. karroo are linked by a number of inter-

mediates. In its typical form ‘“‘spindle A. karroo”’ is readily recognizable, yet

when an attempt is made to delimit it from “typical” A. karroo, great difficulty

is encountered. This difficulty is especially apparent from an examination

of herbarium specimens. It is appreciated that this inability to distinguish specimens

in the herbarium does not, of course, imply that the field differences are of no

consequence. There are undoubtedly differences, but the characters appear to vary

Pirate 5.— Tall, slender specimens of the variant of A. karroo with

whitish bark, long spines and long moniliform pods, up to 20 m

high and forming a dense community in dune forest. Undergrowth

mainly Isoglossa woodii C. B. Cl. Mapelane (2832 AD Mtubatuba),

Nov. 1965. Photo: E. J. Moll.

399

PLate 6.—Well branched specimen of the variant with long spines and long moniliform pods,

up to 5 m high and with a rounded crown. Growing on the bank of the Amatikulu

river estuary in an area subject to tidal inundation. The variant is dominant on the

lower slopes of the hillside in the left foreground. (2931 BA Stanger), March, 1967.

independently as inconsistent tendencies. For example. the glaucous foliage, the

large petiolar gland and glands between each pinna pair appear to typify “spindle

A. karroo.”” However, in some plants the glands are present but tne foliage is green

whilst in others the foliage is glaucous but the glands are absent. There is 3

gradation in leaf colour from glaucous to semi-glaucous to green. The “spindle

A. karroo”’ recorded from the Transvaal has neither markedly glaucous foliage

nor large glands.

It is not intended to accord “spindle A. karroo’’ formal infraspecific taxonomic

status. This decision must not be taken as an indication of uniformity with

A. karroo. The variant is considered as a local expression of an extremely variable

species. In an ecological account the variant may be distinguished by reference to

it as “the spindle form” and thus convey the lack of uniformity within the com-

munities. The term ‘spindle’ is perhaps an unfortunate one, but the term has

been so widely adopted for this variant that an attempt to substitute the term

with a new one would probably merely create confusion.

The variant with long spines and long moniliform pods

This variant extends northwards along the coast of Zululand from about the

mouth of the Tugela River to central Mozambique, including the offshore islands

of Inhaca and Bazaruto. Plants are confined to a fairly narrow belt along the

coast which is often narrower than one kilometre. They grow on the coastal plain,

400

amongst the coast dunes, in the mouths of many river estuaries, for example,

the Amatikulu, and around the shores of the fresh water Lake Sibayi. The plants,

which usually form very dense, pure stands and are dominant to the exclusion

of other trees, often act as pioneers in stabilising loose sand dunes, especially in

disturbed areas and in patches of regenerating coast dune forest. When growing

in dense communities the plants are tall, fairly slender and relatively unbranched

(see Pl. 5). In the open they are well branched with rounded crowns (see Pl. 6).

The bark is typically greyish-white or whitish, fairly smooth, often lenticellate

and encrusted with crustose lichens (see Pl. 7). However, on exposed plants the

bark becomes at times quite dark greyish-brown and rough. On young stems the

bark is typically green with numerous white, transversely elongated lenticels whilst

on the very young branches it is usually smooth and whitish although it may at

times be purplish. Plants are armed with white spines that are frequently large

and slightly swollen and which may attain a length of 25 cm. Many plants display

persistent paired spines on the trunk, a feature not observed in any of the other

variants. Some plants are exceedingly spinescent, a feature which renders them

conspicuous. The view has been expressed, although not in print, that the large

spines are a characteristic of this variant alone and that in other areas of distribution

large spines are confined to young plants, mature plants bearing small spines. This

is not true, for large spines (over 10 cm long) are found on mature plants in

other areas of distribution, although largest spines are admittedly found in this

variant. For example, Ross 640 (K, NH, NU), from near Muden (2830 CD

DUNDEE) has spines up to 15 cm long. The foliage is often a dark green

similar to that of A. robusta and there is a tendency for the glands between the

pinna pairs to be slightly stalked.

None of the characters mentioned is sufficient to warrant the separation of

this variant as an infraspecific entity of A. karroo. The smooth, whitish bark

is shared by the entity referred to A. natalitia. Indeed, it will be recalled that

Gerstner united both variants under A. natalitia.

This variant does, however, tend to differ from “typical” A. karroo in having

longer and broader peds that are typically almost moniliform, larger seeds, larzcr

areoles and longer peduncles. However, in no instance does a single character

provide a clear discontinuity, the characters tending rather to occupy one extreme

of the range of variation of A. karroo.

Although the smooth bark on the young branchlets, coupled with the above

tendencies may be considered sufficiently distinctive it is not intended to accord the

variant formal taxonomic status. This variant, which grows on the geologically

recent sands of the Zululand coast, is in its extreme form perhaps the most

distinctive of all the variants within A. karroo. The plants are adapted to the

prevailing range of environmental conditions and are probably best regarded as

an ecotypic response to this habitat. Some of the characters enabling the plants

to flourish are no doubt physiological and genetical and are not primarily mor-

phological. Consequently the differences do not manifest themselves morphologic-

ally in a manner that is sufficiently distinctive to facilitate taxonomic recognition.

Further investigation is necessary and sufficient grounds may ultimately be found

to accord the variant formal taxonomic status.

To date only one infraspecific category has been formally recognized. Burtt

Davy in Kew Bull. Misc. Inf. 1908: 158 (1908) recognized var. transvaalensis

within A. horrida Willd., the variety differing from typical A. horrida in being

“pubescent on the younger parts.’ Burtt Davy failed to nominate a type specimen

for var. transvaalensis in this paper. Subsequently, in Kew Bull. 1922 : 328 (1922),

401

Pratt 7 Pale greyish-white, smooth, lenticellate bark of the variant

with long spines and long moniliform pods. Lake Sibayi (2732

BC Ubombo), Feb. 1968.

402

after learning that the correct name for the South African plants previously

referred to A. horrida was A. karroo, Burtt Davy transferred his var. transvaalensis

to A. karroo. He maintained his variety in Fl. Transy. 2 : 347 (1932) and it is

here that he mentioned the syntypes Burtt Davy 2468, 2807 for the first time. The

type specimens are only sparingly pubescent and this sparse development of the

indumentum is not considered sufficiently distinctive to warrant recognition at

varietal rank.

In this treatment A. karroo has been regarded as an exceptionally variable

species in which no infraspecific categories have been formally recognized. Within

the species numerous biotypes are recognizable, each of which varies independently

but always within certain limits, the limits of each falling within the range of

variation that is accepted as A. karroo. Some of these biotypes, for example the

variant with long spines and long moniliform pods, are more distinctive than

others.

The extremes of each of the variants are usually quite distinctive and naturally

it is these extremes that attract immediate attention. However, it has been found

that the extremes of each variant are linked to the “central A. karroo gene-pool”

by numerous and varied intermediate stages that become progressively less and

less distinct until a stage is reached where it becomes extremely difficult to assign

specimens to any particular entity with any degree of certainty. It has consequently

been considered of dubious value to fragment such an inherently variable species

into a number of taxonomic entities. Examination of A. karroo suggests that the

A. karroo gene-pool is an ancient one, and one that has continually been able to

adapt itself to new habitats. A. karroo is apparently one of the least exacting

species in regard to habitat preference, a feature that enabled the species to inhabit

a diverse range of habitats.

Acknowledgements

I am very grateful to Dr. L. E. Codd, Chief, Botanical Research Institute,

for some valuable suggestions, for advice and for comments; Dr. H. Schulz of the

Botanischer Garten und Museum, Berlin-Dahlem, for information relating to

Vieweg, and to Dr. K. D. Gordon-Gray of the Bews Botanical Laboratories,

University of Natal, Pietermaritzburg, for reading and commenting on the original

manuscript.

403

Book Review

FuNDAMENTALS OF Myco.ocy, by J. H. Burnett, 546 pp., numerous text figures, Frontispiece and

8 plates. Edward Arnold (Publishers) Ltd. London, W1, 1968. Price £6.10.0 net (65/- net

paper).

The title is slightly misleading for, in a book on fundamental mycology, one would

expect due attention to be given to fungal taxonomy. This branch of mycology, which has

undergone important changes of approach in recent years, deserves more than the cursory

treatment it receives in this book, even if one takes into account that the author’s aim

has been to elucidate what all fungi have in common and what makes a fungus tick.

Fundamentals of mycological biology would therefore have been a more appropriate title.

The author has, however, succeeded admirably in bringing together a wealth of informa-

tion which has never before been available in one text book. Thus the structure of fungi,

their growth and other physiological aspects, genetical phenomena, their speciation and

dispersal are all dealt with at length. This comprehensiveness is a valuable feature of the

book, though the review style adopted to condense as much information as possible in

a book of this size, makes for exhausting reading.

The contents are divided into four broad sections, Structure and Growth, Function,

Recombination, Speciation and Evolution. The first gives a detailed account of our present

knowledge of hyphae and all components as elucidated by, for instance, electron microscopy.

Hyphal differentiation and growth, the production of reproductive structures and spore

liberation, dispersal and germination are the other main topics discussed in this section.

The second section, Function, contains chapters on the following subjects: general

aspects of fungal nutrition and metabolism, transport processes in fungi, translocation and

transpiration, carbohydrate catabolism, accumulated and synthetised products and_ their

metabolism, reaction and interaction. Under the heading “reaction to environmental

factors” in the last chapter I looked in vain for a reference to adaptation of fungi. An

essay on this subject rather than the unnecessary detailed conflicting evidence on bio-

chemical processes should have found a place in the book. Adaptation in fungi is of

paramount importance to applied mycology which has repercussions in such diverse fields

as, among others, plant protection, the soft drink industry and microbiological oceanology.

In Section III attention is drawn to the great diversity of types of somatic division and

to the uniformity of meiotic divisions. The reasons for asexual variation are clearly eluci-

dated. This section is of a very high standard as is Section IV in which an admirable

expostion is given of what forces are at work which lead to speciation in fungi. The

result is a genetical isolation which in fungi is of a quite different nature than in other

organisms. The book closes with a chapter on phylogenetic considerations, in which weak-

nesses in phylogeny based on comparative morphology without considering function, are

exposed.

For a book of this size it has surprisingly few typographical errors. I noted however

that in Plate 1, Fig. 3 (p. 8) microfibrils are shown connecting particles with wall at left

and not top of picture, and also that the publication of Emerson & Weston, 1967 (pp.

197, 198) is missing from the list of references. The illustrations are well chosen and the

microphotographs are of high quality.

The outstanding feature of the book is, however, that the author repeatedly points

out the gaps in our knowledge. This fills a need for leadership in mycological investigations

and thus the book should find a very useful place on the shelves of all research mycologists.

G. J. M. A. Gorter.

i Bil

i dy aah Aad 4 7 ite

Oo” te, an

i are en _

; i}

oan S|

- a ¥

- :

7 : we) i gat ae ‘Th at

No \ ON ee Ro

' , i

Tes 39 Sali Bea ee

ae oz a2

Lea:

bee Rte

é ee

Johan Graham Anderson.

Bothalia, 10, 3: 405—406

Johan Graham Anderson (1926-1970)

B. de Winter

Johan Graham Anderson, Senior Professional Officer in the Botanical

Research Institute and, for many years, in charge of the Grass Section, died

after a short illness brought on by a coronary thrombosis on 10th March, 1970,

at the age of 43 years. Born at Douglas, Cape Province, on 17th October, 1926,

he was the son of Hugh Graham Anderson, an Irish medical doctor who settled

in South Africa, and Hailie Johanna Anderson (née Steenkamp). When eight

years old he contracted poliomyelitis and meningitis simultaneously. Miracu-

lously, he survived but was totally paralysed for a year. After months of

constant nursing by his mother, who specially for this purpose took a course in

MESSER, he started regaining the use of his arms, but his legs remained very

weak.

In Cape Town he had several operations on his legs and was fitted with

calipers, but it was only after his school years that he eventually learnt to walk

with crutches. After private tuition as well as several years of regular schooling

he passed the matriculation examination. He was then seventeen years of age

and his progress must be seen as a remarkable achievement, in view of all the

setbacks he had received. During all these years he had the constant help and

support of his loving mother and younger brother, Wallace, who stood by him

to the end of his life.

Although regarded as being physically unfit for a normal occupation, he

was very active mentally and, in 1944, registered at the Potchefstroom University,

entering for a B.Sc. course in Botany and Zoology. He left before fully com-

pleting the course and, on the 12th of February, 1949, was appointed as Technical

Assistant in the then Division of Botany and Plant Pathology. Shortly after

this appointment, he registered at the University of South Africa and received

his B.Sc. degree in 1951, during which year he was also promoted to Assistant

Professional Officer.

On the 14th of January 1956 he married Maria Bezuidenhout and from

this marriage four children, two sons and two daughters, were born.

In 1963 he was promoted to Senior Professional Officer. In this rank he

served the Botanical Research Institute until his untimely death. From the start

of his career at the Institute he was associated with the Grass Section, from the

end of 1950 as the officer in charge. The majority of his papers deal with this

family, and are listed at the end of this appreciation. During the several years

in which he was personally responsible for the general identification of grasses

he amassed a deep knowledge of this group. which was to stand him in good

stead in his research in later years.

The quality of his work was high and in addition to the listed publications

he left a number of important manuscripts (such as the grasses for the revision

of the Genera of South African Flowering Plants, the grasses for the Flora of

Pretoria and a check-list of the South African grasses) in various stages of

completion. His contributions to botany in South Africa should not be seen in

406

the light of his research output alone. During his career he personally identified

more than 25,000 specimens and in addition supplied a wide range of informa-

tion to numerous fellow botanists as well as scientists in other fields. Few if

any of the authors of major works dealing with S.A. grasses — published in

recent years — were not to some extent, indebted to Johan Anderson and his

assistants at the Botanical Research Institute. Numerous technicians of the

Department of Agricultural Technical Services received their basic courses in

grass taxonomy as well as training in the identification of grasses from him.

In other spheres he also made significant contributions. He was one of

the staff members who worked on the as yet unpublished list of botanical terms

compiled jointly by the late Professor A. P. Goossens and the staff of the

Botanical Research Institute. He was extremely competent in both official

languages and gave invaluable service as a translator of botanical articles. In

administrative matters he was meticulous and assisted his seniors by keeping

records of various activities of the Institute. For many years he acted as

secretary to the regular staff meetings.

It is clear that the constant and usually successful battle against his dis-

abilities and the pain he had to endure deeply influenced his attitude to life.

Very few problems seemed to him unsurmountable and he had a cheerful

optimism which stimulated all those in contact with him. Perhaps his most out-

standing characteristic was his extremely warm and human nature. He drew

people to him in a remarkable way, and he was genuinely loved by his colleagues.

He had as many friends in the lower ranks as he had among the seniors, thus

acting as a link between the young and older staff members. Because of his

warmth and inherent fairness he contributed greatly to the friendly atmosphere

and the efficiency of the Institute. With his passing many have lost a good

friend and colleague.

List oF PusLiications — J. G. ANDERSON

1959. Cenchrus brownii, Burr-grass — Knopklitsgras. Department of Agriculture

Pamphlet.

1960. Notes and New Records of African Flowering Plants (Andropogon ravus;

Danthonia stereophylla; Panicum volutans) — Bothalia 7: 417-422.

1961. New and Interesting Taxa from Southern Africa (Andropogon platybasis;

Tricholaena monachne var. annua; Brachiaria dura var. pilosa) — Kirkia

1: 102-104.

1962. Notes and New Records of African Plants (Andropogon lacunosus) —

Bothalia 8: 113-114.

1963. Man and Beast Live on Grass. — Fmg in S. Afr. 39.

1964. Notes and New Records of African Plants (Eriochrysis brachypogon subsp.

australis; Danthonia aureocephala) — Bothalia 8: 170-172.

1966. Typhaceae. Fl. S. Afr. 1: 53-56.

1966. The genus Andropogon in Southern Africa. Bothalia 9: 5-30.

1966. A new combination in Hyparrhenia. Bothalia 9: 130.

1966. Common Weeds in South Africa] Algemene Onkruide in Suid-Afrika. Mem.

Bot. Surv. S. Afr. No. 37, 440 pp. (With M. Henderson).

1967. A new species of Panicum. Bothalia 9: 341.

1967. A new variety of Festuca costata. Bothalia 9: 341.

1967. Trachypogon spicatus. Flow. Pl. Afr. 38, t.1512.

Bothalia, 10, 3: 407—410

Cotton Staining caused by Crebrothecium ashbyi

(= Eremothecium ashbyi) in South Africa

W. FE. O. Marasas*

ABSTRACT

A description is given of Crebrothecium ashbyi (Guill.) Routien (= Eremothecium ashbyi

Guill.) which was isolated from yellow stained cotton fibres. This fungus, in association

with cotton stainer insects, caused extensive cotton staining in the Vaalharts area of the

Cape Province during 1969-197).

Cotton staining, also known as internal boll rot or stigmatomycosis, is a

serious disease which is caused by yeast-like fungi in the genera Ashbya, Eremo-

thecium, Nematospora and Spermophthora and spread by sucking insects.

Although the taxonomy of these fungi has been a source of some controversy

(Pridham and Raper. 1950; Lodder and Kreger-Van Rij, 1952), these four genera,

characterised by needle-shaped spores, are currently placed in the family Spermoph-

thoraceae of the Endomycetales by most authors (Gaumann, 1964; Von Arx,

1967).

Three species of Spermophthoraceae, Ashbya gossypii (Ashby and Nowell)

Guill. (= Nematospora gossypii Ashby and Nowell), Nematospora coryli Pegl.

and Crebothecium ashbyi (Guill.) Routien (— Eremothecium ashbyi Guill.),

have been reported as causing cotton staining in association with sucking insects,

primarily Dysdercus spp. in South Africa (Moore, 1930; Ullyett. 1930; Pearson,

1934, 1947; Wallace. 1939; Wickens. 1940; Rainey. 1948; Pridham and Raper,

1950; Doidge, 1950; Doidge, Bottomley. Van der Plank and Pauer. 1953; Empire

Cotton Growers Corporation Progress [eports, 1933-1948; C.M.I. Distribution

Maps of Plant Diseases No. 153 and 163). A. gossypii is common in the more

humid parts of the South African cotton belt while N. coryli has been found

only occasionally in the northern Transvaal (Moore, 1930; Wickens, 1940).

Crebrothecium ashbyi has been recorded only once from a single piece of lint

at Barberton, Transvaal (Wickens, 1940).

During April 1970, a sample of lemon yellow stained cotton was received

from Vaalharts Irrigation Seitlkement, Cape Province. According to Dr Gillham

of the J. L. Clark Cotton Corp., this type of lemon yellow staining had not

been observed in the area before and was widespread in the Vaalharts region

during the 1969-1970 season although the cotton stainer insect population

was low. Microscopic examination of these yellow fibres revealed that they

were packed with needle-shaped spores. A lemon yellow, yeast-like fungus,

Crebrothecium ashbyi (Guill.) Routien, was isolated from these fibres by incubating

them on potato dextrose agar at 25°C.

C. ashbyi was first isolated from cotton in the Sudan and described as

Eremothecium ashbyi Guill. (Guilliermond, 1935; 1936; Tarr, 1955). The new

genus Crebothecium was erected by Routien (1949) because F. ashbyi differs

from the type species of Eremothecium, E. cymbalariae Borzi, in the presence

* Plant Protection Research Institute, Private Bag 134, Pretoria.

408

of intercalary chains of “‘asci’” and in the indefinite arrangement of spores in the

spore sacs.

According to Krneta-Jordi (1962), the only culture of C. ashbyi which had

previously been isolated is Guilltermond’s type strain from the Sudan. The ten

isolates of C. ashbyi at the Centraalbureau voor Schimmelcultures, Baarn, Nether-

lands, are apparently all subcultures of the type strain, except possibly CBS

185.6/, B.I.G.. which was received from Dr Mach, Bot. Inst. d. Ernst Moritz

Arndt Universitat. Greifswald, without further particulars (Dr M. A. A. Schipper,

personal communication). A culture of C. ashbyi which was obtained from the

Fic. 1—3. — Crebrothecium ashbyi (Nomarski interference con-

trast, X 1000). Fig. 1, spore sac filled with spores. Fig. 2.

dichotomous branching of hypha and deliquescent spore sac.

Fig. 3, spores.

409

Northern Utilization Research and Development Division, United States Depart-

ment of Agrculture. Peoria, Illinois, was deposited by A. C. Thaysen without

any further information on its source (Dr L. J. Wickerham, personal communi-

cation).

Crebrothecium ashbyi (Guill.) Rceutien in Mycologia 41: 184 (1949).

Eremothecium ashbyi Guill. in Compt. Rend. Acad. Sci. Paris 200: 1556 (1935);

CMI Descriptions of pathogenic Fungi and Bacteria, No. 181 (1968).

eT Is A, Be

Colonies grow rapidly on potato dextrose agar and produce spores within

2 days at 25°C. Colonies are Lemon Yellow (Ridgway, 1912, Plate IV), probably

due to the production of riboflavin which also diffuses out and stains the agar

yellow, flatly appressed. have the appearance of melted snow and produce no

aerial mycelium. Vegetative hyphae are dichotomously branched. filled with

yellow protoplasma, non-septate except in old cultures. 5—7.5u wide. Usually

an entire hypha or a large portion of it becomes fertile, giving rise to chains

of intercalary, truncate-ellipsoid spore sacs (asci?). Spore sacs are separated

by constricted areas in which the hyphal wall is thickened (Callosepfonfen of

Krneta-Jordi, 1962), 55—100u (mostly 70—80u) long, 7—l0u wide and com-

pletely filled with spores (ascospores?). Spores are irregularly disposed without

any arrangement in fascicles. entangled and overlapping one another. difficult

to count but apparently ranging from 8 to 32 (mostly 16) per sac. Spores are

released by deliquescence of the spore sac walls, hyaline, non-septate, multi-

globulate, needle-shaped. curved and sharply attenuated to a pointed projection

at one end, 24—28 x 2—2.5u (mostly 25 x 2u).

Assimilation reactions: non-fermentative; glucose, sucrose, trehalose. raffinose,

glycerol and succinate are assimilated (kindly determined by Dr. J. P. van der

Walt).

Specimen examined: PRE 44335 (Mycological Herbarium), cultures on

potato dextrose agar isolated from yellow stained cotton (Gossypium sp.) bolls,

Vaalharts, Cape Province, April 1970. Cultures have also been deposited in

the Centraalbureau voor Schimmelcultures, Baarn, Netherlands (CBS 741.70) and

the Northern Utilization Research and Development Division, United States

Department of Agriculture, Peoria, Illinois (NRRL Y-7249).

The South African isolate was compared with the following isolates of

C. ashbyi and found to be identical in all respects: CBS 204.36, Guilliermond,

Type Culture; NRRL Y-1363, A. C. Thaysen. The assimilation reactions of

the South African isolate also agree well with the data obtained for NRRL Y-1363

(Dr. L. J. Wickerham, personal communication).

ACKNOWLEDGEMENTS

The author is grateful to Dr. J. P. van der Walt. Microbiology Research

Group. Council for Scientific and Industrial Research, Pretoria, for determining

the assimilation reactions and for valuable advice; to Dr. L. J. Wickerham,

Northern Utilization Research and Development Division, United States Depart-

ment of Agriculture, Peoria, Illinois, and Dr. M. A. A. Schipper, Centraalbureau

voor Schimmelcultures, Baarn, Netherlands, for cultures and helpful information.

REFERENCES

Arx, J. A. Von, 1967. Pilzkunde. Cramer, Lehre.

Doince, ErHet M., 1950. The South African fungi and lichens. Bothalia 5: 1-1094.

410

Doince, EtHer M., Borromuey, A. M., VAN per PLANK, J. E. & Paver, G. D., 1953. A revised

list of plant diseases in South Africa. Union S.Afr. Dept. Agr. Sci. Bull. 346: 1-125.

GAuUMANN, E., 1964. Die Pilze. Birkhauser, Basel.

GuILLierMonpD, A., 1935. Sur un champignon nouveau, parasite des capsules du cotonnier,

VEremothecium ashbyii et ses re‘ations possibles avec le Spermophthora gossypii et les

ascomycetes. Compt. Rend. Acad. Sci., Paris 200: 1556-1558.

GuILiiermonp, A., 1936. L’Eremothecium ashbyii, nouveau champignon parasite des capsules

du cotonnier. Rev. Mycol. (N.S.) 1: 115-156.

Krneta-Jorpi, M., 1962. Cytologische und physiologische Untersuchungen an Eremothecium

ashbyi. Arch. Mikrobiol. 43: 76-108.

Lopper, J. & Krecer-van Rig, N. J. W., 1952. The Yeasts. North-Holland, Amsterdam.

Moore, E. S., 1930. Internal boll disease of cotton in South Africa. Union of S. Afr. Dept. Agr.

Sci. Bull. 94: 11-18.

Pearson, E. O., 1934. Preliminary observations on cotton stainers and internal boll disease of

cotton in South Africa. Bull. Ent. Res. 25: 383-414.

Pearson, E. O., 1947. The development of internal boll disease of cotton in relation to time

of infection. Ann. Appl. Biol. 34: 527-545.

Pripuam, T. G. & Raper, K. B., 1950. Ashbya gossypii — its significance in nature and in the

laboratory. Mycologia 42: 603-623.

Rainey, R. C., 1948. Observations on the development of the cotton boll, with particular

reference to changes in the susceptibility to pests and diseases. Ann. Appl. Biol. 35: 64-83.

Ripeway, R., 1912. Color standards and color nomenclature. Hoen, Baltimore.

Rourten, J. B., 1949. Crebrothecium ashbyi. Mycologia 41: 183-185.

Tarr, S. A. J., 1955. The Fungi and plant diseases of the Sudan. Commonwealth Mycological

Institute, Kew.

Uttyert, G. C., 1930. The life-history, bionomics and control of cotton stainers in South Africa.

Union S. Afr. Dept. Agr. Sci. Bull. 94: 1-10.

Watiace, G. B., 1939. Plant diseases spread by bugs. FE. Afr. Agr. J. 4: 268-271.

Wickens, G. M., 1940. Plant pathology. Emp. Cotton Grow. Corp. Prog. Rep. 1938-1939:

43-64

Bothalia, 10, 3: 411—416

New and Interesting Records of South African Fungi,

Part VII

W. EF. O. Marasas* and G. C. A. van der Westhuizen*

ABSTRACT

Four species of fungi recorded for the first time in South Africa, are described and

illustrated. These are: Acremoniella verrucosa Togn. from roots of Medicago sativa;

Coniella pulchella Hohn. from roots of pine-apple; Periconia igniaria Mason & Ellis from

seed of Medicago sativa; and Stachybotrys subsimplex Cooke from the cocoon of Para-

stizopus armaticeps.

Four species are described and discussed below. Dried down cultures of

all these have been deposited in the mycological collection of the National Her-

barium (PRE) at 590 Vermeulen Street, Pretoria.

Acremoniella verrucosa Togn., Rend. Inst. Lombardo Sci. Lett., 2 ser., 29:

864 (1896); Horne & Williamson in Ann. Bot. 37: 393 (1923); Horne & Jones

in Ann. Bot. 38: 354 (1924); Mason, C.M.I. Mycol. Papers 3: 34 (1933); Groves

& Skolko in Can. J. Res. C. 24: 77 (1946).

Eidamia tuberculata Horne & Jones in Ann. Bot. 38: 334 (1924).

Figures: 1, 2.

On corn meal, potato carrot and 14% malt extract agars at 25°C colonies

grow rapidly and reach a diameter of 85 mm in 7 days. On potato dextrose

agar the growth is slower (25 mm) while no visible growth occurs on Czapek-

Dox agar. Vegetative mycelium is thinly effused, hyaline and consists of repeatedly

branched, septate, hyaline hyphae 5—7.5u in diameter. Colonies appear brown

because of the macroconidia. Mdacroconidia are aleuriospores which are borne

singly either on straight to procumbent conidiophores which develop laterally on

the main hyphae, or, on conidiophores which proliferate sympodially from a point

behind the apex to produce one or more secondary sporogenous cells. Conidio-

phores are hyaline, simple or sparingly branched but the sporogenous cells may

branch repeatedly in a sympodial manner to give rise to complex masses, non-

septate or up to 8-septate, 10—110u long, 5—7.5u in diameter at the base, ter-

minating in sharply tapered sporogenous cells, 2—S in diameter at the rounded

apex. Macroconidia solitary, terminal one-celled, brown, globose, thick-walled,

tuberculate, 19—27.5u in diameter (mostly 22.5u). Microconidia not seen.

Specimen examined: PRE 44334 (Mycological Herbarium), dried culture,

isolated from Medicago sativa L. roots, Grootrivier, Knysna District, Cape Pro-

vince, October 1969.

The South African isolate of A. verrucosa described here agrees well with

the descriptions by Mason (l.c.) and Groves and Skolko (l.c.), except that asper-

gilliform phialophores and microconidia were not produced by cultures on corn

meal agar incubated at 25°C for 4 weeks. This isolate also failed to grow on

Czapek-Dox agar in pure culture but grew normally in mixed culture with other

fungi such as Fusarium oxysporum (Schlecht.) Snyd. et Hans. This confirms

* Plant Protection Research Institute, Private Bag 134, Pretoria.

412

the findings by Mason (l.c.) and Groves & Skolko (l.c.) that A. verrucosa also

exhibits the so-called Heald-Pool reaction on Czapek-Dox agar.

This is the first record of the occurrence of this species in South Africa. —

W.F.O.M.

Fic. 1—2. — Acremoniella verrucosa. Fig. 1, septate conidiophores bearing terminal macro-

conidia (x 1,000). Fig. 2, verrucose macroconidia (x 1,000).

Fic. 3—6. — Periconia igniaria. Fig. 3, macronematous conidiophore showing branched

chains of conidia developing from a sporogenous cell (x 1,000). Fig. 4. micronematous

conidiophores (x 1,000). Fig. 5, intercalary spore (x 1,000). Fig. 6, conidia (x 1,000).

413

Coniella pulchella Héhnel in Ber. Dt. Bot. Ges. 36: 316 (1918): Sydow &

Petrak in Repert. Nov. Spec. Regn. Veg. Beih. 42: 459 (1927) (repr. 1969);

Sutton in Can. J. Bot. 47: 603-608 (1969).

Figures: 7, 8 9.

On potato dextrose agar, mycelium thin, white, floccose or cobwebby, or

submerged, covering the plate after 8—1l/ days at 25°C. Pycnidia scattered.

single or gregarious. subglobose to more or less rounded conical, blabrous, seated

on a thin white subiculum, 0.05—0.5 mm in diameter, at first hyaline, translucent

then darkening with maturation of conidia; pycnidial wall hyaline to sub-hyaline

internally, 12—20u thick and composed of several layers of pseudoparenchymatous

cells, ostiolate at maturity. Conidiophores hyaline, short. unbranched or branched

at the base, tapering apically, arizing from a basal sporodochium-like dome of

tissue, 15—20 x 2—3u, tapering to Iu in diameter apically. Conidia olivaceous-

brown. flattened on one side, with a paler, raphe-like longitudinal mark, apex

conical or subapiculate and with the base truncate, 8—11 x 5.5—6.5u.

Specimen examined: PRE 44310, dried down culture. isolated from roots

of pine-apple, Ananas sp., East London, C.P., February 1968.

The fungus described here has many features in common with Cyclodomella

nigra Mathur. Bhatt & Thirumalachar (Sydowia 13, 143-147, 1959), a species

regarded by Sutton (l.c.) as a synonym of Coniella diplodiella (Speg.) Petrak &

Sydow (l.c.). The conidia of the specimen described here have their apices

conical rather than obtuse, and, because this feature distinguishes C. pulchella

from C. diplodiella (Sutton, |.c.) this fungus is assigned to the former species.

This specimen was isolated from decaying roots of pine-apple plants and

was observed in culture only. Although freshly formed conidia were examined

even from pycnidia which had not developed to the ostiolate stage, neither the

gelatinous “‘raphe-like structures’’ nor the ‘‘small stalk cell’? of the conidia. as

described in Cyclodomella nigra by Mathur et al (l.c.) were ever seen. Also, the

pycnidial wall in this isolate remained hyaline with the innermost cells becoming

pale olivaceous with age. The mature pycnidia. however. remain essentially

hyaline. The dark coloured inner layers reported in pycnidia of Cyclodomella

nigra by Mathur ef al (l.c.), were not seen in pycnidia of this isolate. This

difference may be of generic importance or may be a variable character. This

isolate agrees so well in other respects with the genus Coniella Hohnel, however,

that it’s segregation from that genus does not appear to be justifiable.

This is the first record of this genus in South Africa. — G. C. A. v.d. W.

Periconia igniaria Mason & M. B. Ellis, C.M.I. Mycol. Papers 56: 104

(1953); Booth in Brit. Mycol. Soc. Trans. 51: 803 (1968).

Figures: 3, 4, 5, 6.

Considerable variation in cultural appearance occurs on different culture

media (Table 1), but colonies are generally woolly and always produce a typical

rose-madder or vinaceous pigment. Vegetative mycelium is composed of hya-

line, thin-walled, smooth or verruculose, branched, septate hyphae approximately

3 in diameter. At the base of macronematous conidiophores the hyphae are

swollen, brownish and very coarsely warted or encrusted. Conidia (blastospores)

are borne on micronematous and macronematous conidiophores but on some

media only micronematously (Table 1). Micronematous conidiophores are formed

by enlargement of the vegetative hyphal cells to become cylindrical, brown,

thick-walled, smooth or verruculose sporogenous cells, 9—20 x 4—6u. The

sporogenous cells yive rise to straight or branched chains of conidia which

developed in acropetal succession but mature basipetally (Fig. 4). The cells of

414

micronematous conidiophores apparently also have the ability to round off and

form intercalary chlamydospores which are morphologically indistinguisable from

the blastospores (Fig. 5). Macronematous conidiophores arise singly, in small

groups or in dense clusters on the vegetative mycelium (Fig. 3). Stipes are erect,

stout, unbranched, brown, smooth or verruculose, 4—8-septate, 225—540wu long,

5—7.5u in diameter at the base and 5u in diameter at the obtuse apex. The

cells of macronematous conidiophores give rise to conidia directly or to distinguish-

able sporogenous cells which are light brown, ovoid, smooth or verruculose,

8—l0 x 6—7u (Fig. 3). Branching chains of conidia which form loose heads

are borne apically and laterally on the stipes. Coxidia are blastospores which

develop acropetally but mature basipetally, spherical, dark-brown, one-celled,

7—l1 (mostly 8u) in diameter, thick-walled, spinose with spines approximately

Iu long (Fig. 6).

Specimen examined:

PRE 43738 (Mycological Herbarium), dried culture,

isolated from Medicago sativa L. seed, Upington, Cape Province, July 1969.

The isolate from Medicago sativa seed described here agrees very well with

the description by Mason & Ellis (J.c.).

No reference could be found in the

literature to the apparent “‘intercalary chlamydospores” frequently observed in

These structures may develop in two different ways: (1) A cell

of the micronematous conidiophore rounds up to form an intercalary chlamydo-

our culture.

spore.

(2) A sporogenous cell of the micronematous conidiophore gives rise to

TasL_eE 1. — Cultural characteristics of Periconia igniaria*

Culture Growth Sporula- Colony colour after 21 days

media” rate tion

(mm/day) | (21 days) Surface Reverse

CMA 39) Micro- and| Centre Rose-madder | Brown tinged with

Macrone- Margin Buff Rose-madder

matous

OMA 4.6 Micro- and| Centre Rose-madder | Greenish-grey tinged

Macrone- Margin Buff with Rose-madder

matous

1, MA 4.8 Micro- and} Centre Smoky-grey Brown tinged with

Macrone- Margin white tinged Rose-madder

matous with Rose-madder

PCA 4.7 Microne- Centre Rose-madder | White tinged with

matous Margin white Rose-madder

PDA 4.7 None Centre Smoky-grey Brown tinged with

Margin white tinged Rose-madder

with Rose-madder

a Based on five single spore isolates of PRE 43738 on each of five culture media incubated

at 25°C for 21 days.

b CMA = corn meal agar.

OMA = Oat meal agar.

1} MA = 13% Malt extract agar.

PCA = Potato carrot agar.

PDA = Potato dextrose agar.

415

a blastospore which in turn gives rise to another sporogenous cell in stead of a

successive blastospore with the result that the blastospore appears to be an inter-

calary chlamydospore. Somewhat analogous heavy-walled cells giving rise to a

further conidiophore have been described in Acremoniclla velata by Onions &

Jones (Brit. Mycol. Soc. Trans. 51: 151-152. 1968). They referred to these cells

as “rudimentary conidia”. This question will have to be resolved by a develop-

mental study in slide culture to determine the exact method of formation of

these spores.

Booth (l.c.) described the perithecial state of P. igniaria as Didymosphaeria

igniaria. He reported that the species is homothallic and that uniloculate asco-

stromata are produced when cultures are grown on potato dextrose agar with

pieces of wheat straw and subjected to near ultraviolet light.

Mason & Ellis (/.c.) found P. igniaria on plants which have been scorched

or prematurely killed by burning. They recorded this species on eight different

host plants in England and on Borassus flabellifer var. aethiopica in Ghana.

P. igniaria has also been isolated from soil by Stenton at Wicken Fen, Cambridge-

shire (Mason & Ellis, /.c.) and from the surface layer of a sand dune at Sandwich,

Kent by Brown (J. Ecol. 46: 641-664, 1958). The South African isolate was

obtained from lucerne seed surface sterilized with Nance solution (1 g HgCl,

10 ml 0.1IN HCl, 12 ml Teepol, diluted to 1 litre) for 90 seconds, washed

five times with distilled water and plated on potato dextrose agar. One hundred

seeds were treated in this way and only two yielded fungal colonies, both of

which proved to be P. igniaria. These findings suggest that the spores of P. igniaria

are very resistant to heat and chemical treatment, probably because of the thick

epispore.

This is a new host record for P. igniaria, the first record of the occurrence

of this species on seed and the first record of the occurrence of this species in

South. Africa. — W.F.O.M.

Stachybotrys subsimplex Cooke in Grevillea 12:33 (1883); Bisby in Trans.

Brit. Mycol. Soc. 26: 133-143 (1943).

Figures: 10, 11, 12.

Colonies on potato-malt-filter paper agar, slow growing, reaching a diameter

of 30 mm in two weeks, black, woolly-funiculose with trailing ropes of hyphae

bearing conidiophores which terminate in black, glistening, slimy spore masses.

Hyphae hyaline at first becoming fuliginous, branching, septate, 2—4u in

diameter. Conidiophores arizing from funicles of aerial hyphae, hyaline at first

later fuliginous to dark olivaceous brown in age, simple or occasionally branched,

tapering gradually from the base to the tip, smooth or finely roughened, 0—3

septate and bearing a whorl of 3—7 sporogenous cells terminally, 30—55 x 2.5—

5.0u. Sporogenous cells hyaline at first, later fuliginous oblong-ovoid often

somewhat flattened on one side and curving, 7.0—12.0 x 3.0—S5.0u. Conidia dark

olivaceous, thick-walled, finely verrucose, subglobose 5.0—7.0u in diameter or

ellipsoidal and somewhat pointed, 5.5—7.0 x 5.0—6.0u, borne in globules of

slime.

Specimen examined: PRE 44311 (Mycological Herbarium) on potato-malt-

filter paper agar, isolated from cocoon of Parastizopus armaticeps, Twee Rivieren,

Kalahari Gemsbok Park, May 1969.

The fungus described here agrees very well with the description of this

species in culture by Bisby (J.c.). He thought that this species is a saprophyte

of the warmer regions and that the genera Gliobotrys Hohnel and Memnoniella

Hohnel are based on this species. In the South African isolate it was noticed

that the young conidia were somewhat ellipsoidal smooth-walled and pale coloured

416

but that older conidia are sub-globose, dark and roughened. The conidia were

always produced in slime balls, however. No chains of conidia which may

suggest a ““Memnoniella stage’ were ever seen in this isolate.

This is the first record of this species in South Africa. — G.C.A. y.d. W.

Fic. 7—9. — Coniella pulchella. Fig. 7, conidia (xX 1,000). Fig. 8, conidiophores with

young conidia and part of sporodochium-like dome (xX 1,000). Fig. 9, part of pycnidium

wall (x 1.000).

Fic. 10—12. — Stachybotrys subsimplex. Fig. 10, conidiophores arising from hyphal strand

(< 400 phase contrast). Fig. 11, sporogenous cells on conidiophores (xX 1,000 phase

contrast). Fig. 12, conidia at various stages of maturity (x 1,000).

Bothalia, 10, 3: 417—418

New Combinations in the Genera Brettanomyces,

Kluyveromyces, Lodderomyces and Wingea

J. P. van der Walt*

Van der Walt and Van Kerken (1959) and Van der Walt (1965, 1966, 1967)

introduced several non-valid combinations in the genera Brettanomyces, Kluyvero-

myces, Lodderomyces and Wingea by omitting to list relevant basionyms according

to the requirements of Article 33 of the International Code of Botanical Nomen-

clature. As these combinations, however. have been adopted in the new taxo-

nomic treatise edited by Lodder (1970), they are now validated in conformance

with the requirements of the Code.

Brettanomyces intermedius (Krumbholz & Tauschanoff) v.d. Walt & vy. Ker-

ken, comb. nov.

Basionym: Mycotorula intermedia Krumbholz & Tauschanoff in Zentr. Bak-

teriol. Parasitenkunde Abt. II, 88: 367 (1933).

Kluyveromyces aestuarii (Fell) v.d. Walt, comb. nov.

Basionym: Saccharomyces aestuarii Fell in Antonie van Leeuwenhoek 27: 29

(1961).

Kluyveromyces bulgaricus (Santa Maria) v.d. Walt, comb. et stat. nov.

Basionym: Saccharomyces fragilis Jorgensen var. bulgaricus Santa Maria in

Anales Inst. Nac. Invest. Agric. 8: 165 (1956).

Kluyveromyces delphensis (v.d. Walt & Tscheuschner) v.d. Walt, comb. nov.

Basionym: Saccharomyces delphensis v.d. Walt & Tscheuschner in Antonie

van Leeuwenhoek 22: 165 (1956).

Kluyveromyces dsbzhanskii (Shehata, Phaff & Mrak) v.d. Walt, comb. nov.

Basionym: Saccharomyces dobzhanskii Shehata, Phaff & Mrak in Mycologia

47: 805 (1955).

Kluyveromyces drosophilarum (Shehata, Phaff & Mrak) yv.d. Walt, comb.

nov.

Basionym: Saccharomyces drosophilarum Shehata, Phaff & Mrak in Mycologia

47: 804 (1955).

Kluyveromyces fragilis (Jdrgensen) v.d. Walt, comb. nov.

Basionym: Saccharomyces fragilis Jorgensen in Die Mikro-organismen der

Girungsindustrie S5te Auflage. Berlin, p. 377 (1909).

Kluyveromyces lactis (Dombrowski) v.d. Walt, comb. nov.

Basionym: Saccharomyces lactis Dombrowski in Zentr. Bakteriol. Parasiten-

kunde Abt. II, 28: 366 (1910).

* Microbiology Research Group, Council for Scientific and Industrial Research, Pretoria.

418

Kluyveromyces lodderi (v.d. Walt & Tscheuschner) y.d. Walt, comb nov.

Basionym: Saccharomyces lodderi v.d. Walt & Tscheuschner in Antonie van

Leeuwenhoek 23: 188 (1957).

Kluyveromyces marxianus (Hansen) v.d. Walt, comb. nov.

Basionym: Saccharomyces marxianus Hansen in Meddelelser Carlsberg Labo-

ratoriet 2: 222, 1883-88 (1888).

Kluyveromyces phaffii (v.d. Walt) v.d. Walt, comb. nov.

Basionym: Fabospora phaffii v.d. Walt in Antonie van Leeuwenhoek 29: 320

(1963).

Kluyveromyces phaseolosporus (Shehata, Phaff & Mrak) v.d. Walt, comb.

nov.

Basionym: Saccharomyces phaseolosporus Shehata, Phaff & Mrak in Mycologia

47: 806 (1955).

Kluyveromyces vanudenii (v.d. Walt & Nel) v.d. Walt, comb. nov.

Basionym: Saccharomyces vanudenii v.d. Walt & Nel in Mycopathol. Mycol.

Appl. 20: 73 (1963).

Kluyveromyces veronae (Lodder & Kreger-v. Rij) v.d. Walt, comb. nov.

Basionym: Saccharomyces veronae Lodder & Kreger-v. Rij in The Yeasts,

Ist Edition, Amsterdam, p. 669 (1952).

Kluyveromyces wickerhamii (Phaff, Miller & Shifrine) v.d. Walt, comb. nov.

Basionym: Saccharomyces wickerhami Phaff, Miller & Shifrine in Antonie

van Leeuwenhoek 22: 150 (1956).

Lodderomyces elongisporus (Recca & Mrak) v.d. Walt, comb. nov.

Basionym: Saccharomyces elongisporus Recca & Mrak in Food Technology

6: 451 (1952).

Wingea robertsii (v.d. Walt) v.d. Walt, comb. nov.

Basionym: Pichia robertsii v.d. Walt in Antonie van Leeuwenhoek 25: 342

(1959).

Saccharomyces kloeckerianus y.d. Walt, nom. nov.

Synonym: Debaryomyces globosus Klécker in Compt. rend. trav. lab. Carls-

ens 2 2/3, WO7eOD (190),

REFERENCES

Lopper, J (Ed.), 1970. The Yeasts 2nd Ed., Amsterdam: North Holland Publ. C.

Van vER Watt, J. P., 1965. The emendation of the genus Kluyveromyces v.d. Walt, Antonie

van Leeuwenhoek 31: 341- 348.

VAN DER WALT, J. P., 1966. Lodderomyces, a new genus of Saccharomycetaceae. Antonie van

Leeuwenhoek 32: 1-5.

Van pdER Watt, J. P., 1967. Wingea, a new genus of Saccharomycetaceae. Antonie van Leeuwen-

hoek 33: 97-99.

Van per Watt, J. P. & VAN Kerken, A. E., 1959. The wine yeasts of the Cape. Part V. The

occurrence of Brettanomyces intermedius and Brettanomyces schanderlii in South African

table wines. Antonie van Leeuwenhoek 25: 145-151.

Bothalia, 10, 3: 419—426

Acacia brevispica and A. schweinfurthii

jo Ink, IRess*

ABSTRACT

Recently Brenan, in Kew Bull. 21 : 477-480 (1968). upheld Acacia brevispica Harms and

A. schweinfurthii Brenan & Exell as distinct species and recognized two subspecies within

A. brevispica. These conclusions differ from earlier results published by Ross & Gordon-

Gray in Brittonia 18: 44-63 (1966). Consequently it was considered necessary to re-examine

these species in preparation for the account of Acacia for the Flora of Southern Africa.

Distribution maps of the two species are provided and the differences between the species

tabulated and discussed. Despite certain difficulties in Natal, Brenan’s taxonomic conclusions

are adopted. A map showing the distribution of A. schweinfurthii and A. brevispica subsp.

dregeana in Natal is provided and a selection of Natal specimens cited.

Recently Brenan, in Kew Bull. 21 (3): 477-480 (1968), discussed Acacia

brevispica Harms and A. schweinfurthii Brenan & Exell in preparation for his

account of the Mimosoideae for Flora Zambesiaca. Brenan’s taxonomic con-

clusions are, briefly, that A. brevispica and A. schweinfurthii are specifically

distinct, and that two subspecies are recognizable within A. brevispica namely:

(a) subsp. brevispica;

(b) subsp. dregeana (Benth.) Brenan.

The two varieties within A. schweinfurthii namely var. schweinfurthii and

var. sericea recognized by Brenan and Exell in Bol. Soc. Brot. sér. 2, 31: 114-5

(1957) were upheld.

Brenan’s conclusions differ from those reached by Ross and Gordon-Gray

in Brittonia 18: 44-63 (1966) after a study of these two species, with particular

reference to Natal. The taxonomic conclusions reached by Ross and Gordon-

Gray were, briefly, that A. brevispica and A. schweinfurthii are not specifically

distinct and that three varieties are recognizable within A. brevispica, namely:

(a) var. brevispica;

(b) var. dregeana (Benth.) Ross & Gordon-Gray;

The two varieties within A. schweinfurthii (in Brenan and Exell’s sense)

namely, var. schweinfurthii and var. sericea were not upheld.

Therefore, whilst there is general agreement that three main entities are

present, namely, brevispica, dregeana and schweinfurthii there is disagreement

about the taxonomic status of each entity. I have also had to come to a

decision for the account of the Mimosoideae that is being prepared for the

Flora of Southern Africa. It seems therefore necessary to examine the cause

of this disagreement. I have been fortunate in having had the opportunity of

discussing this matter at some length with Mr. J. P. M. Brenan whilst visiting

Kew some years ago.

* Botanical Research Institute.

420

The distribution of A. brevispica subsp. brevispica (the three entities are

referred to in accordance with Brenan’s conclusions) and of A. brevispica subsp.

dregeana may be seen in Fig. 1, whilst that of A. schweinfurthii is given in

Fig. 2. A. brevispica subsp. brevispica occurs in tropical north-east Africa to

as far south as central Tanzania and then again in Angola. Subsp. dregeana

occurs in southern Mozambique, Natal and Pondoland. There is therefore a

large geographical discontinuity between the populations of subsp. brevispica in

central Tanzania and those in Angola. In addition there is a large discontinuity

between the populations of subsp. brevispica in these two territories and the

populations of subsp. dregeana in southern Africa. A. schweinfurthii occurs in

north-east tropical Africa and extends southwards to Mozambique, the Transvaal

.and Natal. The species is not recorded from Kenya nor Uganda.

subsp. brevispica

Subsp. dregeana

400 800 1200 1600

KILOMETRES

Fic. 1. — The known distribution of A. brevispica subsp. brevispica and A. brevispica

subsp. dregeana in Africa based upon an examination of herbarium specimens and on

information supplied by Brenan & Exell in Bol. Soc. Brot., sér. 2, 31: 99-140 (1957).

750 1000 800 1200 1600

a =.)

KILOMETRES

Fic. ey — The known distribution of A. schweinfurthii in Africa based upon an examination

of herbarium specimens and on information supplied by Brenan & Exell (l.c.) 1957.

A. brevispica and A. schweinfurthii form part of a complex of very closely

related species (see Brenan & Exell in Bol. Soc. Brot.. sér. 2: 99-140, 1957, and

Brenan. Fl. Trop. E. Afr. Legum. — Mimos., 1959). In tropical Africa, brevispica

and schweinfurthii are readily distinguishable on the basis of petiole length: in

brevispica petiole length varies from 0.4—1.3 (—1.5) cm and in schweinfurthii

from 2.6—3 (—5.5) cm. No difficulty is experienced in referring specimens

either to one species or to the other on the basis of petiole length alone. How-

ever, petiole length in dregeana is exceedingly variable. even on a single plant.

and ranges from 1—3.5 cm. In southern Mozambique and in Natal, where

dregeana and schweinfurthii both occur. no distinction can be drawn between

these two entities on the basis of petiole length.

422

Similarly rachis length, rachilla length, leaflet length, leaflet width, leaflet

spacing, pod length and pod width reflect no discontinuity between dregeana

and schweinfurthii in southern Mozambique and in Natal (see Ross & Gordon-

Gray l.c.). It is in this area, particularly in Natal, that difficulty is sometimes

encountered when attempting to distinguish specimens of dregeana and schwein-

furthii. This difficulty is not experienced in tropical Africa as dregeana does

not occur there and the difficulty scarcely makes itself felt in southern Mozam-

bique.

There are differences between dregeana and schweinfurthii and there is

general agreement on the characters by which the two entities may be differen-

tiated. The problem essentially is how much significance is to be placed on these

differences in reaching a decision concerning the taxonomic status of these two

entities.

What then are the differences between dregeana and schweinfurthii?

TaBLE 1. — Synopsis of the differences between brevispica, dregeana and

schweinfurthii.

brevispica dregeana schweinfurthii

Young branchlets grey-

brown

Indumentum on young

branchlets and rachides

spreading and exceed-

ing the glands

Petiole 0.4—1.5 cm long

Petiolar gland 1.5—3.0

mm long, sometimes

absent

Leaflets 0.5—1.25 mm

wide; with conspicuous

whitish often spreading

marginal cilia, glabrous

abaxially apart from the

cilia

Pods puberulous,

glandular

Young branchlets grey-

brown

Indumentum on young

branchlets and rachides

shortly appressed-

pubescent, pubescence

shorter than the glands

Petiole 0.5—3.5 cm long

Petiolar gland 0.5—1.5

mm long, sometimes

absent

Leaflets 0.6—1.2 mm

wide; usually with short

inconspicious marginal

cilia, abaxial surface

usually densely appres-

sed-pubescent, some-

times only portion of

the leaflet sparingly

appressed-pubescent or

occasionally entire leaf-

let glabrous

Pods puberulous,

glandular

Young branchlets olive-

green to olive-brown

Young branchlets

puberulous when young

Petiole 2.6—5.5 cm long

| Petiolar gland humped,

1—1.8 mm long, some-

times 2 glands present

or gland absent

Leaflets 0.83—2 mm wide;

with conspicuous whit-

ish appressed marginal

cilia, invariably glab-

rous abaxially apart

from the cilia (very few

specimens with appres-

sed-pubescence are re-

ferrable to var. sericea)

| Pods glabrous or almost

so, glandular

423

Ross & Gordon-Gray (/.c.: 59) stated that: ‘“‘The only observed differences

(between dregeana and schweinfurthii) are in leaflet and pod pubescence. number,

position, and to some extent, shape of petiolar glands and colouration of young

twigs’. Brenan in Kew Bull. 21: 479 (1968) uses these same characters to dis-

tinguish dregeana and schweinfurthii. We-examination of all available specimens

has not revealed any additional useful! characters (see Table 1).

Brenan in Kew Bull. 21: 479 (1968) enumerates in some detail the differences

between brevispica and dregeana. These two entities are quite readily distinguished

and it is unlikely that any difficulty will be experienced in referring specimens

to either entity, particularly in view of the large geographical discontinuity

between the two.

Whilst working on dregeana and schweinfurthii in Natal prior to the publi-

cation of the paper in Brittonia, it was felt that dregeana bridged the gap

between brevispica and schweinfurthii in respect of petiole, rachis and rachilla

length, leaflet length and width, leaflet spacing, pod length and width, and that

the characters enumerated in Table 1 were not sufficiently distinctive to enable

brevispica and schweinfurthii to be maintained as distinct species. Brenan

(Kew Bull. 21: 479, 1968) contends that “this wide range of petiole-lengih (and

presumably the other characters) is characteristic of dregeana, and that dregeana

is a taxon more distinct from typical A. brevispica and A. schweinfurthii than

implied by Ross & Gordon-Gray ...”. Brenan (/.c.) continued ‘ta case may

be made out for considering dregeana as a species distinct both from A. schwein-

furthii and A. brevispica, but this does not seem to be really justified since, in

general appearance and characters, dregeana is so very close to A. brevispica’’.

Little would be achieved by giving dregeana specific status since the problem

of satisfactorily differentiating dregeana and schweinfurthii remains irrespective

of the taxonomic rank held by dregeana.

The distinguishing criteria between dregeana and schweinfurthii are essen-

tially differences in the colour of the young branchlets, differences in the degree

of pubescence of the young branchlets. leaflets and pods and, to a lesser extent,

in leaflet size. Differences in the degree of pubescence are often difficult to

evaluate. For example, it may be difficult to establish whether a young branchlet

is puberulous or whether it is shortly appressed-pubescent. Although all of the

above characters must be used in conjunction, the pubescence of the lower leaflet

surface seems to offer the most promising character in distinguishing dregeana

from schweinfurthii. In schweinfurthii the leaflets are invariably glabrous ab-

axially apart from the appressed marginal cilia. Very few specimens with

appressed-pubescence on the abaxial leaflet surface (referred to var. sericea

Brenan & Exell) have been recorded throughout the species range. In dregeana

the abaxial leaflet surfaces are either densely appressed-pubescent throughout or

else the pubescence is confined to a portion of the surface. Sometimes, and

particularly in the middle reaches of the Tugela River, for example Ross 184

(NU) from near Keats Drift. and on the Lebombo mountains in Zululand, for

example Ross 305, 1175 (NH, NU), the leaflets are entirely glabrous or have

only a few inconspicucus marginal cilia. Specimens of var. sericea with their

appressed-pubescent abaxial leaflet surfaces may prove difficult to distinguish

from dregeana on this character but identification should be possible by using

the other characters.

On the basis of the characters enumerated in Table | most specimens from

southern Mozambique and from Natal can be fairly readily sorted. However.

some specimens do create difficulties and it is sometimes debatable whether they

are robust specimens of dregeana or depauperate specimens of schweinfurthii.

424

One specimen in particular, namely Ross 874 (NU) from just south of Mandini

near the Tugela river, is difficult to place with certainty. The specimen, which

was growing in a thicket of schweinfurthii, has densely appressed-pubescent

abaxial leaflet surfaces and densely puberulous, glandular pods similar to those

of dregeana. In general facies the specimen resembles schweinfurthii and although

not typical of schweinfurthii is probably best placed in schweinfurthii.

In deciding to sink A. schweinfurthii under A. brevispica (Ross & Gordon-

Gray, /.c.) a certain amount of reliance was placed on Brenan and Exell’s comment

(/.c.: 115) that three specimens from central and southern Tanzania (two of which

were examined) “‘are perhaps crosses between A. brevispica and A. schweinfurthii,

having longer petioles and rather smaller heads than the former and narrower

leaflets than the latter’. It was felt that these specimens pointed to the

occurrence of intermediates in east Africa similar to the situation prevailing in

Natal. Now Brenan in Kew Bull., /.c.: 478 has stated: “‘The three possible

hybrids mentioned by Brenan & Exell in Bol. Soc. Brot., sér. 2, 31: 115 (1957)

are very dubious. Two of them, from Lindi District in Tanzania, are outside

the range of typical A. brevispica and are more likely to be poor specimens of

A. taylorti Brenan & Exell, at that time imperfectly known’. Whilst these

specimens may well be A. faylorii this prevailing uncertainty does nevertheless

illustrate how difficult it is to identify some of the species within this complex

with certainty. Examination of specimens and of the keys to the identification

of the species within this complex provided by Brenan & Exell (/.c.) and by

Brenan (l.c., 1959) reveals how closely related many of the species are and on

what slender grounds some of the species are recognized.

A decision concerning the taxonomic status of dregeana and schweinfurthii

must be seen against the background of all the African species within this

complex. For almost its entire range of distribution schweinfurthii is an easily

recognizable species. It is only near its southern limit of distribution in Natal

that schweinfurthii is sometimes confused with dregeana. Natal, a small province

comprising only 8% of the area of the Republic of South Africa, forms a very

small part of the distributional range of schweinfurthii. The occurrence of the

characteristically variable dregeana at the southern limit of distribution of this

complex of species need not therefore prevent brevispica and schweinfurthii from

being maintained as distinct species, even although they are very closely related.

Consequently I am prepared, albeit somewhat hesitantly, to follow Brenan’s

decision to regard dregeana and schweinfurthii as belonging to different species

even although the differences between them are at times rather ill-defined, and

although difficulties will sometimes be experienced when attempting to identify

specimens. With A. brevispica and A. schweinfurthii now established as distinct

species subspecific status is the correct taxonomic rank for dregeana within

A. brevispica.

The distribution of A. schweinfurthii and of A. brevispica subsp. dregeana

in Natal is seen in Fig. 3. A. schweinfurthii is typically coastal in distribution

and plants usually grow on the margin of riverine forest or in riverine fringing

vegetation. A. brevispica subsp. dregeana is found more commonly in the

interior and plants thrive in dry thornveld and in the dry scrub of the interior

river valleys. There is apparently a fairly large geographical discontinuity between

the population in the Umkomaas valley near Richmond and the plants in Pondo-

land, the latter being the type locality of subsp. dregeana.

As a consequence of its preference for moister situations, plants of A. schwein-

furthii appear more robust and verdant in the field. The flowers of A. schwein-

furthii are frequently in larger and more open terminal panicles than in the

425

@ dregeana

X schweinfurthii

20 3 40 50

MILES

100

KILOMETRES

Fic. 3. — The known distribution of A. brevispica subsp. dregeana and A. schweinfurthii

in Natal based upon field studies and an examination of herbarium specimens.

case of A. brevispica subsp. dregeana. The pods of A. brevispica subsp. dregeana

dehisce quite readily whilst those of A. schweinfurthii are usually only tardily

dehiscent or at times apparently indehiscent.

_ In view of the past confusion within this complex in Natal the opportunity

is taken of citing a selection of the Natal material.

A. brevispica Harms subsp. dregeana (Benth.) Brenan

Nata. — 2731 (Louwsburg): 1 km from Pongola bridge on Magudu road (—BC),

Edwards 3187 (NU. PRE). 2732 (Ubombo): Ingwavuma (—AA), Moll & Strey 4021 (NH);

Majozini, approx. 5 km S. of Pongola Poort (—AC), Ward 3917 (NH, NU); Mkuze Game

Reserve (—CB), Ward 3504 (NH. NU). 2830 (Dundee): Mngwenya valley, river crossing

on Weenen-Middelrest road (—CC), Edwards 2811 (NU); Muden (—CD). Sim 19082 (NU):

426

24 km from Greytown on Keats Drift road (—DC), Ross 184 (K, NH, NU); Jamesons

Drift (—DD), Ross 834 (NH, NU). 2831 (Nkandla): Umfolozi Game Reserve, bank of

White Umfolozi river (—BC), Ross 2029 (NH, PRE). 2929 (Underberg): Estcourt Pasture

Research Station (—BB). Acocks 9893 (NH). 2930 (Pietermaritzburg): Ashburton (—CB),

Ross 443 (K, NH, NU); 16 km from Richmond on Ixopo road, Umkomaas valley (—CD),

Ross 116 (NH, NU); 2.5 km S.E. of pumping station on Nagle Dam road, Umgeni valley

(—DA), Ross 246 (K, NH, NU). 2931 (Stanger): opposite Nembe river confluence, Lower

Tugela valley (—AB), Edwards 3045 (NH, PRE).

A. schweinfurthii Brenan & Exell var. schweinfurthii

NaTtaL. — 2632 (Bela Vista): Ndumu Game Reserve. near Banzi Pan (—AB). Ross 697

(K, NH, NU). 2732 (Ubombo): Mkuze bridge, 3 km N. of Mkuze on Candover road

(—CA), Ross 1022 (K. NH, NU); Mkuze Game Reserve, Fig Park (—CB), Ross 310

(K, NH. NU); False Bay Park (—CD), Ross 2326 (NH, PRE). 2831 (Nkandla): Umfolozi

Game Reserve, Matshamshlope (—BD), Downing 561 (NH, NU); Heatonville (—DB),

De Waal 39203 in NH. 2832 (Mtubatuba): Hluhluwe Game Reserve (—AA), Ward 1835

(NH, NU). 2931 (Stanger): 45 km from Kranskop on Mapumulo valley road (—AA),

Moll 924 (NU); 3 km S. of Mandini on old main road (—AB), Ross 872 (NH, NU);

Verulam (—CA), Ross 167 (K, NH, NU); 1.5 km N. of Virginia Airport (—CC),

Ross 484 (NU).

Bothalia, 10, 3: 427—430

A Variant of Acacia karroo from Sekukuniland

Io Jal, Joss”

ABSTRACT

The relationship of some densely pubescent specimens from Sekukuniland in the

eastern Transvaal to Acacia gerrardii Benth. and to A. karroo Hayne is discussed. The

specimens were found to represent part of the range of variation of A. karroo. A map

showing the known distribution of the densely pubescent plants is provided.

Nearly twenty years ago specimens of a most interesting Acacia were

collected by Dr. L. E. Codd near Steelpoort in the Lydenburg district of the

Transvaal. A few specimens have been collected subsequently and these have

all lain unnamed in the National Herbarium, Pretoria.

Initially these specimens were thought to represent a new species some-

what intermediate in position between A. karroo Hayne and A. gerrardii Benth.

The densely pubescent young branchlets, leaves, leaflets and pods are similar to

those of A. gerrardii whilst the bright yellow flowers with reflexed corolla lobes

resemble those of A. karroo. However, a closer examination of the specimens

suggested that the relationship to A. gerrardii is superficial and that the true

relationship is with A. karroo. This suggestion was subsequently strengthened

by field observations.

A. karroo, which is an extremely widespread and variable species, typically

has glabrous young branchlets, leaf petioles, rachides, rachillae, leaflets, peduncles

and pods. However, specimens are found with sparingly pubescent young branch-

lets, petioles, rachides, rachillae, leaflets and peduncles. These specimens from

Steelpoort differ from ‘“‘typical’” A. karroo in the dense development of the

indumentum on these organs, particularly the conspicuous spreading marginal

cilia on the leaflets and the densely pubescent, glandular pods.

Although superficially resembling A. gerrardii, the specimens have bright

yellow flowers in contrast to the white flowers in A. gerrardii. In A. gerrardii

the flowers are in fascicles on axillary peduncles along the branchlets, often on

the previous season’s growth whereas in these specimens the fascicled axillary

peduncles tend to form a terminal raceme or sometimes the flowers occur on

lateral axillary branchlets, the entire inflorescence forming an irregular terminal

panicle. The specimens differ vegetatively from A. gerrardii in that they lack

the large cushion-like abbreviated shoots between each pair of spines from which

the leaves arise. These ‘“‘cushions’ persist on older branches of A. gerrardii

and provide a very useful means of identification. Although the pods of these

specimens resemble those of <A. gerrardii the seeds are elliptic and not

+ quadrate as in A. gerrardii. Similarly areole shape also differs. These

specimens differ therefore quite significantly from A. gerrardii and are readily

distinguished from that species. Specimens of this Acacia and A. gerrardii were

Observed growing together near the Mapochs River in Sekukuniland and both

were readily distinguishable.

* Botanical Research Institute.

428

Last year Mr. J. W. Morris, to whom I am very grateful, collected an

excellent set of specimens in the vicinity of Roossenekal (see Fig. 1). Although

these specimens had densely pubescent young branchlets, leaves and pods they

nevertheless differed somewhat from the Steelpoort specimens and had a distinct

“A. karroo look” about them. The Roossenekal specimens had larger leaves,

more pinna pairs and, in some instances, a much sparser development of the

indumentum than the Steelpoort specimens.

In the field in Sekukuniland from a distance of a few metres the densely

pubescent plants of this Acacia (hereafter referred to as Acacia) are indisting-

uishable from plants of A. karroo. It is only on approaching the plants and on

seeing the indumentum on the young branchlets, leaves and pods that it is

possible to establish with certainty that the plants differ from “‘typical’’ A. karroo.

29° 30° 30° 00’ 30° 30’

TO PIETERSBURG

BURGERSFORT

™ AGNEETSHOOGT

LYDENBURG

10 KM

—— 10 MILES

DULLSTROOM

Fic. 1. — Map showing portion of Sekukuniland in the eastern Transvaal. The shaded

areas indicate the known distribution of the densely pubescent Acacia specimens.

429

Growth form as in A. karroo is either a small spreading shrub, a tree with a

somewhat rounded crown, or a sparingly branched, spindle-like tree. Acacia

is often dominant and forms dense stands.

Now to consider a few further specimens of Acacia from Sekukuniland.

Ross 2089 and 2090 were growing next to each other at Laersdrif. Ross 2089

has the typical dense indumentum on the pods, young branchlets, leaf petioles,

rachides and rachillae and the leaflets have very dense, spreading marginal cilia.

Ross 2090, which was initially mistaken in the field for a specimen of “‘typical”

A. karroo, has very sparingly pubescent leaf petioles, rachides and rachillae and

only some leaflets have few marginal cilia. The pods are sparingly, but never-

theless quite distinctly, pubescent. Ross 2095, which was growing close to 2089

and 2090, resembles 2090 in having pods with a sparse, but nevertheless distinct,

indumentum. Ross 2094 is from Steelpoort. Here on what is apparently an

old abandoned field the plants exhibit a spindle-like growth form and it is here

that the greatest development of indumentum is evident. Ross 2094 together

with Codd 6702, Codd & Dyer 7713, 7716 fror the same locality differ from

the remaining specimens in having smaller leaves and fewer pinna pairs. How-

ever, these Steelpoort specimens agree well with the other pubescent specimens

from Laersdrif and Roossenekal in all other characters and no grounds can be

found for separating them.

Field observations and an examination of specimens indicated that specimens

of Acacia could not be differentiated from A. karroo at specific level. Never-

theless, because of the dense indumentum on the young branchlets and especially

on the pods of some specimens of Acacia, I initially felt somewhat hesitani to

include Acacia in the already very variable A. karroo. In a previous paper

dealing with the variation within A. karroo [Ross in Bothalia 10(2): 385-401, 1971 |

where a sparse indumentum on the young branchlets and leaves of some specimens

in other areas of distribution in southern Africa was reported, in no instance

were densely pubescent pods recorded. However, the very sparingly pubescent

pods in Ross 2090, 2095 bridge the apparent discontinuity between the glabrous

podded ‘typical’ A. karroo and the specimens of Acacia with densely pubescent

pods from Steelpoort. Sparingly pubescent pods do occur quite frequently in

specimens from the Transvaal highveld.

It is interesting to recall that Burtt Davy in Kew Bull 1908: 158 (1908)

recognized var. transvaalensis within A. horrida Willd., the variety being created

to accommodate “‘the form met with in moist soils at the foot of kopjes, or near

fonteins and streams, around Pretoria and on the high veld which is pubescent

on the younger parts... ”’. Subsequently in Kew Bull. 1922: 328 (1922) after

learning that the correct name for the South African plants previously referred

to A. horrida was A. karroo, Burtt Davy transferred his var. transvaalensis to

A. karroo. Burtt Davy’s statement on p. 328 that var. transvaalensis “approaches

A. natalitia E. Mey. but the rachis and rachillae are less densely pubescent... ”

is difficult to comprehend as many specimens named A. natalitia by Burtt Davy

himself are essentially glabrous. Burtt Davy maintained his var. transvaalensis

in Fl. Transv. 2: 347 (1932) and it is here that he mentioned the syntypes Burtt

Davy 2468, 2807 for the first time. Burtt Davy 2807 is a flowering specimen

and 2468 is sterile. The syntypes have only very sparingly pubescent branchlets

and leaves. Burtt Davy never cited any other specimens of var. transvaalensis

and never mentioned whether plants referred to his variety had pubescent pods

or not. The degree of pubescence of the branchlets, leaves and pods exhibited

by Ross 2090 from Laersdrif is similar to the degree of pubescence exhibited

by other specimens of A. karroo in the vicinity of Pretoria and, with the excep-

tion of the pods, to the syntypes of var. transvaalensis.

430

Field observations and an examination of herbarium specimens indicate that

there is continuous variation in the degree of pubescence from the glabrous and

very sparingly pubescent plants in the vicinity of Pretoria and on the Transvaal

highveld to the more densely pubescent plants in Sekukuniland. Variety trans-

vaalensis was distinguished from typical A. karroo solely on the presence of

pubescence and there does not appear to be any other distinguishing character.

Pubescent specimens are found in many parts of the species range so that a

rather heterogeneous assemblage of plants from all areas of distribution could

be referred to var. transvaalensis. For example, A. hirtella (a synonym of A.

karroo), which was described by E. Meyer (Comm. Pl. Afr. Austr. 1: 167, 1835)

from the south coast of Natal, was distinguished from A. natalitia and from

A. karroo in having pubescent leaflets and branchlets and yet specimens differ in

many respects from those of var. transvaalensis. It is felt that var. transvaalensis

would make a rather poor variety and consequently it is not intended to uphold

the variety. The densely pubescent specimens from Sekukuniland which cannot

be satisfactorily differentiated from var. transvaalensis, and which represent an

extreme form of it, will therefore receive no formal taxonomic recognition.

It appears that on the Transvaal highveld there is a local tendency to the

production of a sparse pubescence on the young branchlets, leaves and pods.

However, occasionally this tendency is so extreme, for example at Steelpoort,

as to alter the general appearance of the plants completely. The area between

the two apparently distinct populations in Sekukuniland (see Fig. 1) is rather

mountainous and consequently has not been thoroughly investigated. It is quite

likely that the range of distribution is greater than reflected and that the two

populations may in fact be one continuous population. The soils in this area

are derived from magnetite, a fact which may be of great significance. The

possibility exists that this edaphic factor plus perhaps some other evironmental

influences have resulted in the development of the dense indumentum. The

growth of seedlings under experimental conditions may shed light on this matter.

This decision not to uphold var. transvaalensis was taken after consideration

of the range of variation within A. karroo throughout its distributional range.

As discussed previously (Ross, l.c.) A. karroo is an extremely variable species

in which numerous variants are recognized. The extremes of each variant, for

example the densely pubescent plants from Steelpoort, are usually quite distinctive

and naturally it is these extremes that attract attention. However, the extremes

of each variant are linked to the “central A. karroo gene-pool” by numerous

varied and intermediate stages that become progressively less and less distinct

until a stage is reached where it is difficult to assign a specimen to a particular

entity with any degree of certainty. It becomes extremely difficult to identify

each entity clearly and thereby facilitate identification by other workers. It

seems preferable therefore to regard A. karroo as an inherently variable species

in which no infraspecific categories are recognized rather than to fragment the

species into a number of somewhat arbitrary infraspecific entities.

ACKNOWLEDGEMENTS

I am grateful to Mr. J. P. M. Brenan, Keeper of the Herbarium and Library,

Royal Botanic Gardens, Kew, for examining and commenting on a set of the

densely pubescent specimens and for much valuable information.

Bothalia, 10, 3: 431—436

New and Interesting Records of African Plants

Various Authers

ASCLEPIADACEAE

Brachystelma catheartense 2. A. Dyer, sp. nov., B. tuberoso R. Br. et

B. campanulato N.E. Br. affine, ab ambobus coronae lobis exterioribus et in-

terioribus, corollae colore et pilis differt.

Herba tuberosa humilis basi sparse ramosa; tuber leviter compressum, supra

concavum, 4—5 cm diam., 2 cm crassum; rami adscendentes, 5—15 cm longi,

minute scabridi. Folia lanceolata breviter petiolata vel supera lineari-lanceolata

usque 2.5 cm longa, sparse et minute scabrida. Flores 1—2, extra axillares, pedi-

cellis plus minusve 8 mm longis, sepalis lineari-lanceolatis circiter 5 mm longis.

Corolla 2—2.2 cm longa, extra glabra vel sparse pilosa; tubus campanulatus

9—9.5 mm longus, 1.8—2 cm latus, extra rubro-punctatus, intra infra medium

glaber, rubro-lineatus et maculatus, supra medium rubiginosus, plus minusve

hirsutus; lobi basi rubiginosi, plus minusve hirsuti, apicem versus virides, pilosi,

marginibus sparse ciliatis, leviter recurvatis, apice leviter incurvato. Corona basi

breviter tubulata; lobi exteriores 5, subquadrati, erecti 1—1.5 mm longi, bifidi;

lobi interiores spathulato-oblongi leviter crassi incumbenti-erecti. Pollinia sub-

globosa compressa, circiter 0.25 mm longa.

Type: Cape, Cathcart District, on Viviandale farm, Nov. 1967, P. F. du Toit

in PRE 31309 (PRE, holo.).

Herb with tuberous root; tuber 4—5S cm diam., about 2 cm thick vertically,

rounded on under surface and slightly concave on upper surface. Stem 5 cm

tall when first flowering, elongating to 15 cm under cultivation, sparsely branched,

minutely scabrid. Leaves towards base of stem subsessile, lanceolate, less than

1 cm long; upper leaves longer and narrower, linear-lanceolate, up to 2.5 cm long.

thinly and minutely scabrid-pubescent on both surfaces; the margin slightly folded

upwards. Flowers 1 or 2 at a node, extra axillary, sometimes on opposite sides

of the node, each subtended by a short bract, with carion odour; pedicel about

8 mm long, minutely pubescent; sepals linear-lanceolate, about 5 mm _ long.

Corolla 2—2.2 cm long, 10-ribbed, 5 ribs extending to the sinuses, the others

to the tips of the lobes, with minute recurved teeth at the sinuses (Huernia-wise),

glabrous or with few scattered hairs on outer surface; tube subcampanulate

9—9.5 mm long, spreading somewhat abruptly above and 1.8—2.0 cm across

tim; lobes 1.1—1.3 cm long, 9—10 mm broad across base, triangular, ciliate with

slightly swollen hairs towards sinuses; finely maroon-spotted on outside of tube

and base of lobes: basal 4 of tube within yellow, marked with small radial

maroon stripes and spots, becoming solid maroon and with long hairs above

and on base of lobes, with the tips green and shortly hairy, but somewhat

variable; margins somewhat recurved, apex incurved. Corona arising from above

base of staminal column, with saucer-shaped tube 0.5 mm high; outer lobes

subquadrate 1—1.5 mm long, bifid to about } way with slightly spreading lobules;

inner lobes arising from within basal tube, incumbent-erect. slightly thickened,

spathulate-oblong, 2.5—3 mm long. Pollinia about 0.25 mm long, subglobose,

432

compressed, with narrow translucent inner margin; caudicles slender; carrier

narrowly winged.

Cape.— 3227 (Stutterheim): Viviandale farm near Cathcart, Nov. 1967. (—AC)

P. F. du Toit in PRE 31309; GRA without No., 1970 (—AC), P. F. du Toit in

PRE 31310.

Brachystelma cathcartense was discovered by Mr. P. F. du Toit, Pasture

Research Officer, in November 1967 in the eastern Cape, Cathcart district, on

the farm Viviandale. The species was noted to occur in several of the farm

camps, although nowhere plentiful. Mr. Du Toit took tubers to the office of

the Botanical Research Institute in Grahamstown in 1967, where Miss Grace

Britten induced one tuber to flower quite freely each year. Further collections

were made by Mr. Du Toit in February and November 1970. The main flowering

period is during November and December. Fine colour photographs were taken

by Col. Roy Bayliss and Messrs. Hepburn & Jeanes in 1969. There was a

certain amount of variation in size, pubescence and colouration of the flowers

of different tubers but nothing to warrant taxonomic recognition. One has to

bear in mind the variation which results from the preservation of flowers under

different conditions and at different stages of maturity; for instance the more

the corolla shrinks in drying, and it may be to less than a quarter of its original

size, the denser will appear the pubescence. Miss Britten records separate plants

with the corolla marked with dark purple transverse lines in the tube, maroon

above and with scattered purple hairs tipped with yellowish-green; others were

yellowish-green with dark purple transverse lines in the tube and mulberry-

coloured above with numerous purple hairs giving it a velvety appearance; and

yet others were green with darker green and purple spots in the tube and

purplish-black above with purple hairs.

At first it was thought that B. cathcartense might be a form of B. meyerianum

Schltr. from the same geographic region, but Miss Britten noted that the latter

is of sprawling habit and favours rock crevices for its growth. It was noted also

that the corolla-tube is considerably narrower. Further examination showed that

the outer corona is dissimilar, although Brown’s description in Flora Capensis 4,

1: 842 (1908), based on dried material, is somewhat misleading. He describes

5 transversely rectangular lobes (outer corona) with a central linear obtuse point

(inner corona-lobes). The outer corona in B. meyerianum is actually in the

form of V-shaped pockets between the inner corona-lobes and it is their united

adjacent shoulders which form the rectangular shaped lobes behind the inner

corona-lobes to which Brown refers.

In the shape of the corolla it seems that B. tuberosum R. Br. and B. cam-

panulatum N.E. Br. are the nearest affinities to our species. But besides marked

differences in colouration and pubescence, the difference in coronal structure is

very marked between B. campanulatum and B. cathcartense. It probably is with

B. tuberosum also, but in this case the original description is inadequate, merely

stating that the corona is 5-cleft with the segments triangular and conniving at

the points. This could indicate that the outer corona consists of V-shaped

pockets similar to those of B. campanulatum and B. meyerianum. B. cathcartense

is no exception to the rule that the flowers give off a carrion-like odour.

In applying the epithet cathcartense to this species the name commemorates

the type locality, in fact the only locality so far established, and at the same

time Sir George Cathcart, Governor at the Cape of Good Hope, 1852-1854,

and who was killed in the Crimean War at the battle of Inkerman in 1854.

433

Xysmalobium trauseldii R. A. Dyer, sp. nov., X. orbiculari (E. Mey.)

D. Dietr. affine, sed habitu minore, capite gynostegii peltato concavo exserto

differt.

Asclepias sp., Trauseld, Wild Flows. Natal Drakensberg 154 (1969).

Herba perennis, radice tuberosa elongata. Rami 1—2 erecti, caudicis apice

editi, simplices, 20—35 cm alti, 5—10 mm crassi, puberuli. Folia 6—12, paribus

oppositis, breviter petiolata; lamina late elliptico-oblonga, oblonga, vel oblongo-

lanceolata usque 15 cm longa 8 cm lata, basi rotundata vel cordata, glabra,

margine minute scabra. Umbellae 1—4, pedunculatae multiflorae; pedunculi

5—6 cm longi, puberuli, plus minusve extra-axillares; pedicelli 1—1.3 cm longi.

Sepala lineari-lanceolata, 5—7 mm longa, pilis curvatis pubescentia. Corolla

prope basin divisa; petala 9—10 mm longa, circiter 6 mm lata, expansa, apice

leviter reflexa, extra glabra, intra minute papillosa, cremea demum erubescens.

Coronae lobi carnosi contigui, 3 mm alti, 2.5 mm lati, 1.5 mm crassi, columnae

adpressi, intra margines interiores concavi, medio lobis 2 minutis praediti.

Pollinia cylindrica, 0.6 mm longa. Gynostegii caput exsertum, peltatum, con-

cavum, 4.5—5 mm diam., margine plus minusve undulato vel leviter 5-angulato.

Folliculus 1, puberulus, 10—11 cm longus, 1.5—2 cm diam., pedunculo 13 cm

longo; semina 7—8 mm longa, 4—4.5 mm lata, rugosa.

Type: Natal, Estcourt District, Giants Castle Game Reserve, Nov. 1969,

Trauseld 1107, in PRE 30955 (PRE, holo.).

Perennial herb; rootstock tuberous up to 15 cm long and 25 cm diam.,

hard, corky, fissured and pitted. Stems 1—2 per annum, erect, unbranched,

20—25 cm tall, 5—10 mm thick, puberulous, with 3 to 6 pairs of leaves. Leaves

variable; petoile 0—7 mm long; blade broadly elliptic-oblong, oblong or oblong-

lanceolate, lowest pair 2.5—5 cm long, 2—4 cm broad, upper pairs 7—15 cm

long, 3—8 cm broad, rounded or cordate at base, sometimes stem-clasping,

glabrous except for minutely scabrid margin; median vein prominent below

and slightly sunken on upper surface. Umbels 1—4 per stem from upper nodes,

pedunculate, many-flowered. Peduncles more or less extra-axillary at the nodes,

5—6 cm long, stout, erect, thinly puberulous; pedicels 10—13 mm long. Calyx

deep rose-coloured, divided to base; sepals linear-lanceolate, 5—7 mm _ long,

pubescent on back with minute curved haris. Corolla divided nearly to base,

cream turning deep rose; petals 9—10 mm. long, about 6 mm broad, spreading,

with slightly recurved apex, glabrous on outer surface, minutely papillate within.

Staminal-column arising from base of corolla. Corona-lobes cream-coloured,

arising about 0.5 mm from base of staminal column, fleshy, subquadrate, con-

tiguous 3 mm tall, 2.5 mm broad and 1.5 mm thick, obtuse with depressions on

inner surface on either side of the median thickening and 2 minute lobules slightly

above the point of attachment; anther-thecae hard, filaments adpressed to style

base; stigma exserted above stammal column, 4.5—5 mm diam., peltate, concave,

more or less crenate on margin and 5-angled, pale yellow turning red. Peollinia

cylindric about 0.6 mm long with slender caudicles laterally attached. Follicle

single by abortion, on stout peduncle 13 cm long, puberulous 10—I11 cm long,

1.5—2 cm diam., seeds 7—8 mm long, 4—4.5 mm broad, rugose, concave on one

surface, with apical tuft of hairs.

NataL. — 2828 (Bethlehem): Mont-aux-Sources National Park, grassveld,

1,500 m alt., very rare, Nov. 1963 (—DB), Trauseld 117 (PRE). 2929 (Under-

berg): Giants Castle Game Reserve, on bank of Bushmans River in black turf,

about 1,800 m alt., Oct. 1966 (—AD), Trauseld 1107, in PRE 30955.

Mr. W. R. Trauseld, field officer on the staff of the Natal Parks, Game

and Fish Preservation Board, after whom this species is named, found the first

specimen in November 1963 in Themeda triandra grassveld at Mont-aux-Sources.

434

The second record was in October 1966 on the bank of the Bushmans River at

Giants Castle. In spite of diligent searching in these areas, only a very few

other plants of the species were located, the most recent being a robust specimen

from Giants Castle in November, 1969, Trauseld 1107.

Among southern African species, X ysmalobium orbiculare (E. Mey.) D. Dietr.

shows the closest affinity to X. trauseldii which, however, is readily distinguished

by the unusual feature of a protruding stigma from the apex of the staminal

column. In this character X. trauseldii has an affinity with X. angolense Scott-

Elliot from distant Angola.

Mr. Trauseld took pains to preserve flowers in solution in different stages

of maturity. Dissections from the dried specimens failed to reproduce the

original shape of the fleshy corona-lobes, which once more highlights the value

of a good spirit collection as an adjunct to herbarium specimens. Illustrations

of the habit of the species and of an inflorescence, with a central open flower,

are shown in the recently published book by Mr. Trauseld, entitled Wild Flowers

of the Natal Drakensberg, p. 154 (1969).

R. A. Dyer.

* *

FLACOURTIACEAE

A New Species oF DovyAtis

Dovyalis revoluta Thom, sp. noy., foliorum forma et nervatura D. zeyheri

similis sed praecipue lobis calycibus revolutis, baccis subrotundis papillosis, semi-

nibus lanatis differt.

Arbor vel frutex 5.4 m altus, saepe armatus, dioecius interdum floribus

polygamis. Folia alterna, petiolata; lamina coriacea obovata, 1.5—5S cm longa,

1—3 cm lata, glabra, e basi triplinervia. Flores feminei solitarii, lobis calycibus

revolutis, persistentibus. Fructus suborbiculatus, papillosus. Semina lanata.

Type: Natal. — 2832 (Mtubatuba): False Bay Park, Moll 5112 (PRE, holo.).

Tree or shrub up to 5.4 m tall, deciduous; dioecious or rarely polygamous,

often armed with glabrous spines up to 6 cm long, bark light grey to black,

lenticellate. Leaves exstipulate, alternate, petiolate; blade coriaceous, obovate,

1.5—S cm long, 1.3 cm broad, glabrous, 3-veined from the base; apex obtuse,

base cuneate, margin light green, entire or faintly serrate; petiole 3—5 cm long,

glabrous. Male flowers light green, pedicellate, in fascicles of 2—6; pedicels

surrounded at the base by small hairy scales, 3—4 cm long. Calyx 4—5 lobed;

lobes 3—5 mm long, ovate-elliptic, pubescent. Coralla 0. Stamens 15—25,

3 mm long; filaments surrounded at the base by glabrous nectaries which form

a honeycomb structure; anthers bilocular, dehiscing by means of longitudinal

slits. Female flowers yellow-green, pedicellate, solitary; pedicels surrounded at

the base by small pubescent scale-like bracts, 4—7 mm long. Calyx 5—7 lobed:

lobes 4 mm long, linear-elliptic, pubescent, revolute. Ovary unilocular, surrounded

at the base by a glutinous, sparsely hairy, lobed annulus, with 2 parietal placentas,

each placenta with one ovule; styles 2, channeled. Fruit a subglobose berry,

2 cm in diam., minutely papillose, orange when ripe. Seeds 2, embedded in

fleshy pulp, 13 mm long, densely woolly.

Recorded from Zululand in sand forests.

NataL. — 2732 (Mtubatuba): False Bay, Edwards 3199; Gerstner 4735;

Moll 2823; 5112; Ward 4781.

The Dovyalis species in South Africa can be divided into two distinct

sections. D. caffra is the only representative of the one section and all the

other species, including D. revoluta, form the other section. D. caffra is charac-

435

terised by: (1) the leaves being fascicled on cushion-like abbreviated shoots;

(2) the ovary having 5—7 placentas, each placenta bearing two ovules; and (3)

the fruit being 3—4 cm in diameter with many seeds. All the other species

have the following characteristics: (1) the leaves are alternate, not fascicled;

(2) the ovary has 2—3 placentas, each placenta bearing one ovule; and (3) the

fruit is 0.6—2 cm in diameter with 1—3 seeds.

Although D. revoluta and D. caffra are placed in different sections, they

have a character in common, namely, their suborbicular fruits, whereas the

remaining species have oblong fruits.

D. revoluta and D. lucida are the only species in which the exocarp of the

fruit is papillose. The exocarp in the other species is either glabrous or hairy.

The leaves of these two species are. however, very distinct. D. revoluta has

obovate leaves with an obtuse apex, while the tertiary veins are not as prominent

as the primary and secondary veins. D. lucida, on the other hand, has rhomboid,

acuminate leaves and the tertiary veins are as prominent as the primary and

secondary veins.

The leaf shape and venation of D. revoluta and D. zeyheri are very similar

and the two species might be confused in the vegative state. D. revoluta, how-

ever, posesses one unique character which distinguishes it from all the other

species: the calyx lobes of the female flower are revolute before and at the

fruiting stage. In all the other species they are either erect or bent outwards,

not revolute. The seed testa in D. revoluta is densely woolly while in all the

other species the testa is glabrous or hairy.

Fruiting specimens of D. revoluta have been collected by various collectors

near False Bay in Zululand over several years. It was only in October 1970

that Mr. Moll of the Botanical Research Institute, who is engaged on a botanical

survey of the area, succeeded in finding male and female flowers.

A fruiting specimen, Galpin 8074, collected in 1911 near the Zwart Kei River

in the Eastern Cape, may belong in D. revoluta, though the fruits are somewhat

smaller. Further investigation is necessary to establish with certainty whether

D. revoluta does in fact also occur in the Eastern Cape Province. ee

da 1a HOM.

* * *

THYMELAEACEAE

A New ComMBINATION IN GNIDIA

C. H. Wright in Fl. Cap. 5, 2: 583 (1925) based his description of Arthro-

solen compactus on a specimen collected by Wylie at Nigunya, 1,980 m altitude,

in Natal (Wylie sub Wood 10531). Examination of an isotype in the Natal

Herbarium, Durban, revealed that the species would be better placed in the

genus Gnidia as the inflorescences are not in dense heads and lack the surrounding

coloured membranous bracts typical of Arthrosolen. Phillips in his “Notes on

some Genera of the Thymelaeaceae” in J. S. Afr. Bot. 10: 61-67 (1944) trans-

ferred eight of the eleven species of Arthrosolen enumerated in FI. Cap. 5, 2:

6-9 to the genus Gnidia but he apparently overlooked Arthrosolen compactus

which was described in the addenda on p. 583.

The combination of this specific epithet with the genus Gnidia has apparently

never been validly published. It is proposed therefore to remedy this now by

effecting the necessary combination.

Gnidia compacta (C. H. Wr.) J. H. Ross, comb. nov.

Arthrosolen compactus C. H. Wr. in FI. Cap. 5, 2: 583 (1925). Type: Natal,

Nigunya, Wylie sub Wood 10531 (NH, iso.!).

436

Apart from the type specimen from Nigunya (the exact locality of which

is unknown) a few other specimens from Natal have been examined. These

include: Killick 1187 (NH) from the Cathedral Peak Forest Research Station

(Bergville district); Trauseld 464; 639 (NU) from the Giants Castle Game Reserve

(Estcourt district); Hilliard & Burtt 5676 (NU) from the Highmoor Forest Reserve

(Estcourt district) and W. F. Wright 234 (NU) from Storm Heights (Impendhle

district).

J. H. Ross.

* * *

VITACEAE

New ComBINATIONS IN CYPHOSTEMMA

In Not. Syst. 16: 113-125 (1960), Descoings published a number of new

combinations in Cyphostemma but, because he omitted to cite the publications

in which the basionyms were published, the combinations were not valid. In

Kirkia 2: 139 (1961) and Kirkia 3: 20 (1963), Wild & Drummond validly

published additional combinations and, in Flora Zambesiaca 2: 459 (1966), they

met the requirements for certain of the combinations attributed to Descoings.

Descoings corrected the remainder of his combinations in Nat. Monsp. 18: 227

(1968), but there still remain a few South African species for which combinations

have not been made. These are now effected.

Cyphostemma natalitium (Szyszyl.) J. v. d. Merwe, comb. nov.

Vitis natalitia Szyszyl., Polypet. Rehmann 2: 45 (1889).

Cissus natalitia (Szyszyl.) L. E. Codd in Bothalia 6: 545 (1956).

Cissus connivens Lam. var. meyeriana Planch. in DC. Monag. Phan. 5: 584

(1887), partly; Siissenguth in Pflanzenfam. 20d: 250 (1953), partly.

Cissus glabra E. Mey. in Drege, Zwei Pfl. Docum. 173 (1843), nomen nudum;

Planch. I.c. 584 (1887), in synonomy.

Cissus orientalis sensu Hary. in Fl. Cap. 1: 253 (1860), non Lam.

Cyphostemma suicatum (C. A. Sm.) J. v. d. Merwe, comb. nov.

Cissus sulcatus C. A. Sm. in Burtt Davy, Flow. Pl. Ferns Transv. Swaziland 2:

xx, 477 (1932).

Cyphostemma segmentatum (C. A. Sm.) J. v. d. Merwe, comb. nov.

Cissus segmentatum C. A. Sm. in Burtt Davy, Flow. Pl. Ferns Transv. Swazi-

land 2: xx, 476 (1932).

Cyphostemma oleraceum (Bolus) J. v. d. Merwe, comb. nov.

Cissus oleraceus Bolus in J. Bot. Lond. 47: 55 (1909).

Cyphostemma dasypleurum (C. A. Sm.) J. vy. d. Merwe, comb. nov.

Cissus dasypleurus C. A. Sm. in Burtt Davy, Flow. Pl. Ferns Transv. Swaziland

2: xx, 476 (1932).

Cyphostemma hisp‘diflorum (C. A. Sm.) J. v. d. Merwe, comb. nov. :

Cissus hispidiflorus C. A. Sm. in Burtt Davy, Flow. Pl. Ferns Transy. Swazi-

land 2: xx, 476 (1932).

A name which has not yet been transferred to Cyphostemma is Cissus

sandersonii Harv. After a study of a wide range of modern material, it is

concluded that this cannot be satisfactorily separated from Cyphostemma cirrhosum

(Thunb.) Descoings ex Wild & Drummond, nor can a satisfactory distinction

be found between the typical form of the latter and C. cirrhosum subsp. trans-

vaalense (Szyszyl.) Wild & Drummond.

J. J. M. vAN DER MERWE.

Bothalia, 10, 3: 437—450

Principal Components Analysis of Acacia burkei

and A. nigrescens in Natal

Jel Ross and |, We) Morris:

ABSTRACT

Four principal component analyses were carried out to study the perplexing relations

within Acacia burkei Benth. and between it and A. nigrescens Oliv. Sampling methods are

described in detail. Ten morphological parameters were noted from 163 plants of 21

populations. The results confirmed conclusions of earlier non-multivariate studies. The

two species can be distinguished on the basis of the ten parameters and it is of doubtful

value to recognize infraspecific categories within A. burkei as the variation within the

species is continuous. The technique of principal components analysis was most useful

in this study.

INTRODUCTION

Acacia burkei Benth. and A. nigrescens Oliv. form part of a complex of

closely related species which are taxonomically most perplexing. Within this

complex the degree of pubescence of the calyx is the character of prime import-

ance in distinguishing two main groups. In their typical forms A. nigrescens

and A. burkei are readily distinguishable: the former with its large leaflets and

glabrous calyces and the latter with smaller leaflets and pubescent calyces. How-

ever, there are numerous plants with leaflets intermediate in shape and in size

between those of A. nigrescens and those of A. burkei. Leaflet size varies con-

siderably and an entire range from those the size of A. burkei to those the size

of A. nigrescens may be found on a single plant. However, as these plants

have pubescent calyces their relationship seems to be with A. burkei rather than

with A. nigrescens.

This range of morphological variation within A. nigrescens and _ within

A. burkei has been considered in some detail (Ross 1968a, 1968b). It had

been customary to distinguish loosely between “‘small leaflet” A. burkei and

“big leaflet’ A. burkei, the former typically having leaflets less than 3 mm wide

and the latter leaflets more than 3 mm wide. However, it was found (Ross 1968b)

that the characters typifying ‘small leaflet’ and “‘big leaflet’ A. burkei were

not necessarily correlated but varied independently, certain combinations of

characters being commoner than others. Thus, although specimens at either

extreme of the range of morphological variation could be readily sorted into

two groups, there remained numerous specimens that could not be referred to

either group with certainty. Consequently no infraspecific categories were recog-

nized within A. burkei.

An examination of the means of the morphological parameters (see below)

obtained for A. nigrescens (Ross, 1968a) and for A. burkei (Ross, 1968b)

indicated that rachilla length, number of pinna pairs, number of leaflet pairs,

leaflet length and leaflet width provided discontinuities between the two species.

When the extremes of the morphological parameters were examined, however,

* Botanical Research Institute.

438

these differences were not so readily apparent. Consequently it was decided to

subject the morphological parameters to a principal components analysis in an

attempt either to confirm or contradict earlier findings that A. nigrescens and

A. burkei are quite readily separated, and that it is of doubtful value to recognize

infraspecific categories within such an inherently variable species as A. burkei.

Kendall (1957) and Seal (1964) described principal components analysis in

detail and one introduction to the subject, in a taxonomic setting, is given by

Jeffers (1965). The results of many taxonomic applications have been published

recently but as this is one of the first in South Africa the method is described

in detail. In the context of this paper the object of the technique is to extract

a set of components from the populations x parameters matrix which account

for as much as possible of the parameter variation between the Acacia popula-

tions and which are mathematically independent of one another.

SAMPLING TECHNIQUE

To assess the morphological variation within and among trees, and within

and among populations, some statistical procedure was essential. The application

of statistical methods brought with it the need for reliable, yet practical sampling

techniques. The average herbarium collection is unsuitable, consisting often of

isolated specimens selected as being “typical”, either of a single plant, or of a

population, or of aberrants sufficiently atypical to have attracted attention.

The prime requirement for a statistical study is that samples be representa-

tive. This proved difficult since populations were not always clearly defined

and often occupied rugged terrain. Individual plants because of their large,

woody, much-branched growth form and abundant foliage presented yet other

sampling problems. All such problems had to be met by employing techniques

that yielded representative samples, yet were essentially practicable.

Twenty leaves, twenty pods and twenty inflorescences from each plant

were regarded as a satisfactory number for a sample. To obtain such samples

from individual plants, a sampling method devised for and tested out on A.

robusta Burch. (Gordon-Gray, 1965) was employed.

The distal one to two feet of not less than ten branches representative of

the crown of a plant were collected.!_ In no instance were coppice shoots included

since preliminary work showed that the leaves of such shoots differ, either in

size or in pubescence. The branches collected from any one plant constituted a

sample.

Each sample was treated separately. All mature leaves were stripped from

the branches, heaped together and thoroughly mixed. Immature leaves were

ignored. From this heap twenty leaves were taken by an operator with eyes

closed. The same procedure was followed to obtain a sample of twenty pods

and twenty inflorescences.

As many populations as possible of each species, which occur scattered

through Natal (almost entirely in Zululand), were visited and sampled (see

Fig. 1). Most populations visited covered large areas. Because of the rugged

terrain, plants growing on, or near, the roadside were sampled. Availability

alone governed the haphazard intervals at which plants were sampled. As far

1. It was appreciated at the outset that a truly random sampling method, such as ‘Randomised

Branch Sampling’ (Jessen, 1955), was not practicable in this study. Consequently the

word ‘random’ has been omitted throughout, lest its use infringe mathematical require-

ments. In all sampling procedures followed, however, care was exercised to ensure that

samples were representative and without bias.

439

30 40 50

MILES

100

KILOMETRES

Fic. 1. — Localities of sampled populations in Natal.

as possible, ten plants were sampled from each population, but where popula-

tions were small fewer were sampled. A rough sketch map of the population

was made on which the positions of the sampled plants were plotted. Each

plant sampled was identified by means of a numeral painted on the bark. This

was important since it was usually necessary to re-visit the plants as few had

flowers and ripe pods contemporaneously.

The morphological parameters for each of the twenty leaves and pods which

constituted the sample from each tree were:-

Petiole length (mm).

Rachis length (mm).

Leaf length (mm).

Rachilla length (mm) (the length of the right-hand member of the rachilla

pair central on the leaf, abaxial surface uppermost).

Pwnr

440

5. Number of pinna pairs (mean).

6. Number of leaflet pairs (mean).

7. Leaflet length (mm) (length of the leaflet midway along the right-hand

member of the rachilla pair central on the leaf, abaxial surface uppermost).

8. Leaflet width (mm) (as for leaflet length).

9. Pod length (mm).

10. Pod width (mm).

Means for each tree, referred to below as tree means, and for each population,

referred to as population means, were calculated. Parameter means for the 15

populations of A. burkei (1-15) and the six populations of A. nigrescens (16-21)

are given in Tabie 1.

Tasie |. — Population means for each morphological parameter. The parameters

are enumerated in the text and the localities of the populations are

indicated in Fig. 1.

Popula- Parameter

tion

Number I] 2 3 4 5 6 7 8 9 10

1 DEO G42 66 Bel 3.0 4.0 52 G52 Wat

D 20.8 386 59.4 20.6 2.0 2.0 6.5 101.1 19.1

3 | cea a 10 P20 PS) 4.4 4.9 6.2 108.2 18.0

4 17.8 6 64. 23.4 6.4 9.7 7.6 DA BID W5.7/

5 eld Onl OS 2 Onl 6.3 9.4 7.9 Ri IBIS = 7A

6 16.0 47.0 63.0 24.1 5.6 6.2 8.3 38 173 Bly

7 | 15.8 38.0 588 G4 4.5 4.1 8.4 44 814 20.2

8 15.6 45.0 60.6 25.8 V8 Wil 5.2 Dil Fell 16.5

9 153 360 51.0 20.8 4.3 40 10.3 59 72.0 18.6

@ | 146 477 G23 25.3 7.6 10.9 6.6 2.6 101.2 16.2

11 13.2 346 47.8 18.0 8.0 12.2 4.6 lg Wa Weal

12 12.8 284 412 192 5.6 8.8 5.5 2.3 78.3 #4217.9

13 12.8 31.8 446 20.4 4.0 3.0 9.5 Sil = S443} 13.7/

14 11.0 343 45.3 19.5 7.1 10.7 5.1 I WS 16

15 10.2 314 41.6 20.0 8.6 11.1 4.1 1.4 104.1 18.3

16 ¢ M3 GDA 9.5 2.9 10 22.9 18.7 109.4 20.8

17 17.1 40.8 57.9 8.0 3.1 IO Bl 73 W051 IS52

18 38.2 53.0 6.6 3.1 10 209 15.6 106.0 417.1

1G | Waal Bye Sil 8.3 3.2 10 B07 Ii O67 Ive

mm) | Waa DOS Aare loss 3.0 1@ D7 17 l@83 73

21 37 DBS = 4S 7.5 3.0 10 18.0 146 115.0 16.6

44]

Data ANALYSIS

Four principal component analyses were performed on the available data.

Firstly, population means for both A. burkei and A. nigrescens were used and

secondly, tree means for both species were used. The third and fourth analyses

were carried out on, respectively, population means and tree means for A. burkei

alone. Thus the raw data matrix (population x parameters) for the first analysis

contained 21 population means, the second 163 tree means, the third 15 popula-

tion means and the fourth 118 tree means. In each analysis all ten morphological

parameters were used. The raw data for the first and third analysis are given

in Table 1. Shortage of space precludes inclusion of the raw data for the second

and fourth analysis but it is available from the authors on request.

For each analysis the first step was the computation of correlation co-efficients

between each parameter and each other one over all population or tree means,

resulting in a symmetrical 10 x 10 matrix. The principal components were

extracted from this matrix.

An eigenvalue and eigenvector are associated with each principal component.

The value indicates the proportion of the total variation accounted for by the

component and thus the “importance” of the component, and the vector gives

the weighting of each parameter. Components are extracted in descending order

of eigenvalues, hence the name principal components. The vector is scaled so

that the highest value is unity. In practice it has been found that parameters

having weightings of over 0.7 and under —0.7 are important, the importance

being proportional to the absolute value.

Two-dimensional scatter diagrams were constructed from the analyses. The

position of a population along an axis is found by summing the products of the

eigenvector and parameter vector for the population.

RESULTS AND DrscussION

First Analysis

Eigenvalues and eigenvectors resulting from the first analysis are given in

Tables 2 and 3, respectively. Inspection of Table 2 shows that almost half the

variation within the correlation matrix is extracted by the first component, that

over 90 per cent is extracted by the first four components and virtually all is

extracted by the first six components. Further discussion will be limited to the

first three components which account for 88 per cent of the variation.

TaBLe 2. — Eigenvalues of the first seven components extracted by the first analysis.

Component Eigenvalue Percentage of variability

Component Cumulative

1 4.846 48.467 48.467

2 DSB 29319. 77.847

3 1.057 10.577 88.424

4 0.557 She)//s) 94.000

5) 0.358 3.583 97.583

6 0.184 1.840 99.424

7 0.044 0.449 99.873

442

TaBLeE 3. — Eigenvectors of the first three components extracted by the first

analysis.

Parameter Eigenvectors corresponding to component:

Number 2 3

1 0.295 0.848 —0.833

2 —0.176 0.926 0.149

3 —0.030 1.000 0.052

4 —0.836 0.493 —0.045

5 —0.930 —0.161 0.274

6 —0.968 —0.037 0.141

7 1.000 0.008 —0.012

8 0.992 —0.102 0.042

9 0.699 0.383 0.282

10 0.206 —().074 1.000

The morphological parameters contributing most to the first component’s

variation are 4, 5, 6, 7 and 8 (values given in italics in Table 3). The first three

parameters and the last two are positively correlated between themselves, but

the two groups are negatively correlated. Parameter 9 also has a high weighting

on the first component. Parameters 1, 2 and 3 contribute most to the second

component and parameter 10 is the only important one on the third component.

Positions of the populations along the first and second and first and third

components are given in Fig. 2. A clear discontinuity between A. burkei and

A. nigrescens is shown along the first component. There is also a discontinuity

LNINOdWOD 351

2nd COMPONENT 3rd COMPONENT

Fic. 2. — Positions of A. burkei (1-15) and A. nigrescens (16-21) populations plotted against

the first and second components (a) and the first and third components (b) of the first

analysis.

443

on the second component within A. burkei. However, on the third component

there are no distinct discontinuities within A. burkei although population 13 is

somewhat separate from the remaining populations. Within the A. nigrescens

populations a cluster is formed by all the populations except 16 which is some-

what separate, particularly along the third component. Very little variation is

evident within A. nigrescens along the first component.

Figure 2 indicates a definite distinction between A. burkei and A. nigrescens

based on the sampled populations and on the morphological parameters used,

and suggests that A. burkei is a more variable species than A. nigrescens. As

almost three times more A. burkei than A. nigrescens populations were sampled,

it is not possible to conclude with certainty that the former is the more variable

species, but a trend which supports findings of previous, non-multivariate studies

(Ross 1968a, 1968b) is evident. The reason for the greater variation within

A. burkei has not been studied.

Parameters responsible for the separation of A. burkei and A. nigrescens

are those mentioned above, with high absolute values within the first eigenvector.

Likewise, parameters responsible for the spread amongst A. burkei populations

along the second and third components are those with high absolute values within

the second and third eigenvectors respectively. The values for three parameters

which have high absolute values within the first eigenvector are plotted in Figure 3

14.6

71 156 17-3

Fic. 3. — (a) Rachilla length (parameter 4), (b) number of leaflet pairs (parameter 6) and

components of the first analysis.

against the first and second components of Figure 2a. The clear difference

between the two species with respect to these parameters can be seen. It will

be noticed that better gradients along the first component are shown by number

of leaflet pairs and leaflet width than by rachilla length. As rachilla length has

an eigenvector value of only 0.636 a very good fit is not expected. The good

gradient along the second component is, however, expected as rachilla length has

the highest eigenvector value on this component.

In Figure 3 the first two parameters are positively correlated because in both

cases the higher values are found amongst the A. nigrescens populations. As the

higher values for leaflet width are found amongst the A. burkei populations,

leaflet width is negatively correlated with rachilla length and number of leaflet

pairs.

444

The positive and negative correlations discussed here and earlier had been

discovered before the multivariate analysis was undertaken. The negative correla-

tion of parameters is because the longer leaves have relatively fewer pinna pairs

and, similarly, long rachillae have relatively fewer pairs of larger leaflets. Con-

versely, short rachillae carry a larger number of smaller leaflets. The agreement

between what was known and the results of the analysis add to one’s confidence

in the technique.

Table 3 reveals that petiole length, rachis length and leaf length are the

most important characters on the 2nd component and that pod width is the most

important character on the 3rd component. As these characters mainly affect

the distribution of A. burkei populations within the ordination, they will be

discussed later where analyses without the presence of A. nigrescens populations

are presented. The first two analyses were undertaken to study the relationship

between the two species and not within each.

163.cq 153

147 144 148'34 16

1320141

149141 138

162 149, 151

(2)

2nd COMPONENT

Fic. 4. — Positions of A. burkei (1-118) and A. nigrescens (119-163) tree means plotted

against the first and second components of the second analysis. Means for populations

11-15 are circled and means of population 5 are boxed. A dotted line separates

A. burkei from A. nigrescens populations.

445

Second Analysis

Tree means for A. burkei and A. nigrescens are plotted against the Ist and

2nd components in Figure 4. Although there is not a clear discontinuity, the

two species are completely separated by the first component.

It is known that A. burkei and A. nigrescens are distinct species and that

they are readily distinguished on the degree of pubescence of the calyx. This

character, being of the presence/absence type, was not mixed with the other

characters which are approximately continuous. The object of using the data,

even though this taxonomically significant character had been omitted, was to

establish whether or not the two species could still be separated by multivariate

analysis. Certain trees of A. nigrescens (e.g. 121 and 127) are similar to certain

trees of A. burkei (e.g. 4, 6, 68) with regard to the 10 characters sampled but

the usual clear distinction between the species is indicated in Figure 4 by the

two distinct clusters formed alorg the first component.

As in Figure 2, there is a tencency for the A. burkei trees to be more

spread along the 2nd component than are the A. nigrescens trees. Populations

11—15, represented in Figure 4 by trees numbered 84—118 (circled), are again

in proximity although in this instance there is no discontinuity between these

trees and the remainder as in Figure 2. It was appreciated at the outset that

population means were of limited value but they give a useful summary of the

situation. Comparison of Figure 2 with Figure 4 shows how erroneous a picture

can be obtained from the use of population means alone. Furthermore, in

Figure 4, where the means of each tree were used, it is seen that there is con-

siderable variation within each population. For example, population 5 of Figure

2 is represented by trees numbered 29—38 in Figure 4.

Third Analysis

For the third analysis population means ‘or A. burkei alone were used.

Inspection of the eigenvalues showed that over 54 per cent of the variability

within the correlation matrix was extracted by the Ist component and over 90

Taste 4. — Eigenvectors of the first three components extracted by the third

analysis.

Parameter | Eigenvectors corresponding to component:

Number l 2 3)

1 0.945 0.000 —0.141

2 0.728 0.962 0.057

3 0.893 0.724 —().008

4 0.636 1.000 —0.063

5 —0.859 0.791 0.093

6 —0.780 0.907 —0.073

7 1.000 —0.406 —0.030

8 0.892 —0.698 0.087

9 0.683 0.580 0.410

10 —0.164 —0.183 1.000

446

per cent by the first three components together. Eigenvectors for the first three

components are given in Table 4 and positions of populations plotted against

the first and second and first and third components are given in Figure 5S.

ANINOdWOD 351

2nd COMPONENT 3rd COMPONENT

Fic. 5. — Positions of A. burkei populations plotted against the first and second components

(a) and the first and third components (b) of the third analysis. A. burkei ‘small’

populations are circled and mixed “big” and “small” populations are marked by a

dotted circle.

Fic. 6. — (a) Rachilla length (parameter 4), (b) number of leaflet pairs (parameter 6) and

components of the third analysis.

447

With the exception of pod width all parameters have relatively high weightings

along the first component, while parameters 2 through 6 have high weightings

along the second component and pod width is the only important parameter along

the third component. Many high weightings are often found on the first com-

ponent of an analysis and can usually be attributed to overall size differences.

However, the first component does not separate “big leaflet’? and ‘‘small leaflet’

populations as one would then expect. Instead, “big” and “‘small’’ populations

occur scattered along the first axis. This is also shown in Figure 7 (see below).

The second component spreads the populations in such a way that “big”

and ‘‘small’” populations can be separated by a diagonal line extending from

between populations 12 and 13 to between | and 4. This is, however, the axis

of maximum variation along the first two components. This means that the

“big” to “small” difference is secondary to another, more important, gradient

which separates populations 11 to 15 from the rest.

In Figure 6 the three morphological characters used in Figure 3 were plotted

against the first and second components of the third analysis. Similar positive

and negative correlations as in Figure 3 are shown. There is an indistinct gradient

along the first component in leaflet width and number of pinna pairs. Leaflet

width is, however, the character on which “‘big’’ plants are separated from “‘small’’,

once again suggesting that there is some other character, or characters, which

are more important than leaflet width in drawing out the populations and splitting

populations 11 to 15 from the rest. Table 4 reveals that petiole length, rachis

length, leaf length, number of pinna pairs, number of leaflet pairs, leaflet length

and leaflet width are all important in creating variation between populations. All

of these characters contribute either positively or negatively to the split between

populations.

Fourth Analysis

The eigenvalues and eigenvectors of the fourth analysis are very similar to

those of the third. The similarity is to be expected as the data for the third

analysis are derived directly from those of the fourth. Slightly less variability

Taste 5. — Eigenvectors of the first three components extracted by the fourth

analysis.

ae a Ln ne

Parameter Eigenvectors corresponding to component:

Number 1 2 3

1 0.896 0.287 —0.072

2 0.690 1.000 0.145

3 0.835 0.871 0.088

4 0.544 0.941 0.128

5) —0.882 0.691 0.206

6 —0.809 0.883 0.097

7 1.000 —0.345 —0.055

8 0.943 —0(.606 —0.111

9 0.290 —0.130 1.000

10 —0.051 —0.538 0.980

448

ININOdWOD

2nd COMPONENT

Fic. 7. — Positions of A. burkei tree means plotted against the first and second components

of the fourth analysis. “Small” populations are circled.

449

(47 per cent) is extracted by the first component and a total of 81 per cent is

extracted by the first three components together. Comparison of Tables 4 and

5 shows only differences of degree within each eigenvector.

The gradients and apparent division discussed under the third analysis are

caused by the use of means of means (cf. first and second analyses) as no such

discontinuities are obvious from Figure 7 where tree means were used. The

“small” populations tend to be clustered along the lower half of the elongated

scatter but merge completely with trees from the ‘‘big’’ populations. The scatter

in Figure 7 shows that there is a greater difference between trees number 11

and 91 than between trees 6 and 78. Thus there is much more variation

within both the “big” and the ‘small’? populations than between them.

CONCLUSIONS

A. burkei and A. nigrescens are closely related, but nevertheless distinct

species, that are readily distinguished from each other on the degree of pubescence

of the calyx. In the absence of this taxonomically significant character A. burkei

and A. nigrescens still separate (see Figure 4) although there is no absolute dis-

continuity. The analysis indicates that rachilla length, number of pinna pairs,

number of leaflet pairs, leaflet length and leaflet width are additional characters

that enable A. burkei and A. nigrescens to be differentiated.

There is a rather ill-defined tendency for the appearance of a discontinuity

within the A. burkei populations. However, contrary to expectations, this dis-

continuity does not differentiate the “‘big leaflet’ trees from the “‘small leaflet’

trees for both “‘big’” and “‘small’’ leaflet plants occur on either side of the dis-

continuity. Leaflet width is, therefore, not the most important character in

creating this discontinuity between the A. burkei populations. Petiole length,

rachis length, leaf length, number of pinna pairs, number of leaflet pairs, leaflet

length and leaflet width are all important characters in creating this discontinuity

when considered collectively. Past emphasis on leaflet width alone, a character

that provides a rapid visual assessment, as a means of loosely distinguishing

between “‘big leaflet’? and ‘‘small leaflet’? A. burkei has tended to obscure the

many characters that do contribute to the range of variability within the species.

A. burkei is an extremely variable species and although the specimens at

either extreme of the range of morphological variation appear distinctive it is

not possible to divide this range of variation satisfactorily and thereby facilitate

the recognition of infraspecific categories. As concluded previously (Ross 1968b)

it is therefore of doubtful value to recognize infraspecific categories within

A. burkei.

As a technique for studying taxonomic and ecological problems, principal

components analysis is gaining in popularity overseas. After its convincing

performance in the present study we hope its popularity will spread to South

Africa. One of its attributes is its ability to stimulate further investigation. The

reason for the apparently greater variation within A. burkei is an example of

this stimulus.

ACKNOWLEDGEMENTS

We would like to thank Dr. D. Edwards and Mr. J. N. R. Jeffers for

reading and commenting on this paper prior to its publication.

450

REFERENCES

Gorpon-Gray, K. D., 1965. Acacia robusta Burch. and Acacia clavigera E. Mey. in Natal,

South Africa. Brittonia 17: 202-213.

Jerrers, J. N. R., 1965. Principal component analysis in taxonomic research. Statistics Sect.

Pap. For. Commn, 83.

Jessen, R J., 1955. Determining the fruit count on a tree by randomized branch sampling.

Biometrics 11: 99-109.

KenpaLt, M. G., 1957. A course in multivariate analysis. London: Griffin.

Ross, J H, 1968a. Acacia nigrescens Oliv. in Africa with particular reference to Natal. Bol.

Soc. Brot., Sér. 2, 42: 181-205.

Ross, J. H., 1968b. Acacia burkei Benth. in Southern Africa, with particular reference to Natal.

Bol. Soc. Brot., Sér. 2, 42: 275-304.

Seat, H., 1964. Multivariate statistical analysis for biologists. London: Methuen.

Bothalia, 10, 3: 451—460

The Rate of Forest Tree Growth and a Forest

Ordination at Xumeni, Natal

E. J, Moll’ and D. B. Woods?

ABSTRACT

The rate of increment in circumference at breast height of canopy tree species in the

southern temperate, Mist-belt forest at Xumeni, Natal, South Africa, is very slow. The

mean for all trees calculated from measurements in 1929 and 1966 is 0.201 + 0.015 inches

per year.

Results from an_ ordination analysis of 39 plots were interpreted in terms of a

successional gradient from seral sites on steep ground, characterized by Kiggelaria africana,

Xymalos monospora and Fagara davyi, to climax sites on flatter ground with Podocarpus

spp. Two climax types are indicated, with P. henkelli on moist soil and P. falcatus

on drier soils.

INTRODUCTION

The Xumeni forest lies 110 km south-west of Pietermaritzburg in the Polela

magisterial district of the Natal midlands. The indigenous forest is under the

control of the Government Forestry Department.

The forst occurs on the south-facing slopes of a dolerite ridge. The topo-

graphy is rugged and the soils are shallow. Boulder-strewn areas are common.

The most important canopy tree at Xumeni is Podocarpus henkelii but

generally the canopy, which has an average height of about 30 m, is of mixed

composition (Moll and Haigh, 1966). The area is classified as southern temperate

(K6ppen’s Cwb division; Schulze, 1947). According to Acocks (1953), Xumeni

is Mist-belt forest, and is rich in epiphytic bryophytes, ferns and angiosperms

which festoon tree boles and rocks.

In 1929 the resident forester laid out a line through the forest and on it

recorded all canopy trees in 40 one chain square, systematically placed, plots

(Cook, 1929). He took various measurements including circumference at breast

height (CBH). In 1966 the same trees were re-measured for this parameter.

RESULTS

1. TREE GrowTH

In Fig. 1 the tree CBH measurements for 1929 have been plotted against

their CBH 1966. A regression line fitted for all species, shows that the average

CBH increment rate over 37 years is very small. Regression lines were also

calculated for the six most common species, and with the exception of Fagara

davyi these were almost identical to the regression line for all species.

* Botanical Research Institute.

+ Department of Botany, University of Natal, Pietermaritzburg; present address: 464 West

Broadway, New York, N.Y. 10003.

452

200

150

CBH 1966 ( inches )

ra)

[o}

y= 13-07 + 0:92x (All species °¢ x )

y=: 35°33 + 0°53x (Fagara ~x )

0 50 100 150 200

CBH 1929 ( inches )

Fic. 1. — CBH increment 1929—1966 for all species in all plots, and Fagara davyi alone.

The regression line for F. davyi probably reflects that this species has the

fastest rate of CBH increment of all species when young, but that the older

individuals approximate to the behaviour of all species. This conclusion though

tentative, there being only seven trees encountered, is in keeping with field

observations on F. dayyi.

P. henkelii exhibits a very slow rate of CBH increment (all the individuals

being mature), while F. davyi shows the most rapid rate of CBH increment (all

the individuals being comparatively young).

The rate of CBH increment of forest trees at Xumeni is very slow averaging

0.201 + 0.015 inches* (n= 160) per year. Phillips (1931) quotes average incre-

ment of CBH per year in the Knysna Forest, Cape Province (Table 1). Com-

paring the data shown for the two species in common, the rate of increment

* As the original measurements were in feet and inches, these units have been retained

rather than adopt the metric system for the recent measurements.

of P. henkelii in both Knysna and Xumeni is similar.

453

1 The other common species,

Kiggelaria africana, has twice the increment rate of this species in Knysna.

TaBLE 1. — Mean annual CBH increment rate (in inches) of selected forest tree

species at Knysna and Xumeni.

Species

Observed Rate

Knysna (1924-5)

Estimated Rate + S.E.M.

Xumeni (1929-66)

Podocarpus_ henkelii

Xymalos monospora

Podocarpus falcatus .. .. ..

Kiggelaria africana ..

Fagara davyi ..

0.139

no data

0.131

no data

0.132

0.191

0.231

0.267

0.459

hr tae Ine ihe Wie

0.017

0.064

0.049

0.036

0.113

(n = 68)

(n — 13)

(n — 15)

(n = 26)

(== 7/)

In Fig. 2 the mean plot increment 1929—1966 was plotted against the mean

plot CBH 1929, and a regression line fitted. The results indicate that the plots

with the smallest mean CBH 1929 exhibit maximum increment, reflecting that

CBH increment is inversely proportional to age of canopy tree species.

—d

Mean Plot Increment 1929-1966 (inches)

y = 12:25-0:06434x

US ee Pe ee ee Ee ee

50 100

Mean Plot CBH 1929 (inches)

200

Fic. 2. — The relation of mean plot increment 1929—1966 to mean plot CBH 1929.

454

The CBH data were separated into 20 inch size classes and their increments

drawn as a histogram (Fig. 3). As reflected in Fig. 2 the trees with the smallest

CBH have increased the most, with one exception, CBH size class 141-160

inches. Upon further investigation of the four individuals in this size class, it

was found that the unexpected high rate of increment was due entirely to two

individuals, which were extremely fluted.

—

Mean Total Increment CBH 1929-1966 (inches)

oF eS) te Sars qureene nae

Set OS Se)

CBH Size Classes (inches)

Fic. 3. — Histogram of CBH increment for 20 inch size classes (the number of individuals

per size class are given above each bar).

2. ORDINATION

The floristic similarity of Cook’s 1929 data for 39 plots (one plot was

excluded because it contained no canopy tree species), was investigated using

the ordination procedure developed by Bray and Curtis (1957). Two axes, X

and Y, were sufficient to account for the floristic variability and the resultant

two dimensional scatter diagram of points (representing plots) is shown in Figs.

4—16 (the orientation of the X and Y axes is given in each figure). Isolines

are used to draw attention to (i.e. separate) high and low ratings, unless otherwise

explained.

455

Canopy tree species (Figs 4—8)

Podocarpus henkelii and P. falcatus (Figs 4 and 5) separate off with centres

of maximum density at the two extremes of the X axis, while K. africana and

X. monospora (Figs 6 and 7) separate off at the upper end of the Y axis, the

latter species being apparently unrelated to the X axis. F. davyi (Fig. 8) is

intermediate between K. africana and X. monospora.

Canopy (Fig. 9)

Subjective estimates of canopy were made in 1966. Comparison of Figs 4

and 9, reveals that P. henkelii is the most important species contributing to plots

with closed canopy.

Density and CBH (Figs 10 and 11)

Highest densities tend to occur in the K. africana and P. henkelii region of

the ordination, while plots with trees of the largest CBH are found mainly in

the P. henkelii and P. falcatus region.

Increment (Fig. 12)

Minimum plot increment 1929—1966 is found in the P. henkelii region

and maximum increment is limited to K. afracana, X. monospora and F. davyi

at the upper region of the ordination.

Slope (Fig. 13)

At each site a subjective estimate cf slope was made. Although the pattern

is not altogether clear, the Y axis appears related to a gradient from steep plots

with I’. davyi and other species, through gentle and flat plots with predominantly

P. henkelit.

Rocks (Fig. 14)

A subjective estimate of the amount of surface rocks in each plot was

made. The most rocky plots are found in three groups: flat plots with P. henkelii,

flat and geatle plots with, predominently, P. falcatus and three plots with K.

africana.

Mist (Fig. 15)

Nine plots are stuated, approximately 100 m above the rest, on a ridge

exposed to winds from the south. These plots receive most orographic mist.

The soils are generally rocky and well drained. and the majority of these plots

contain the highest densities encountered of P. falcatus (compare Figs. 5 and

15). Three plots isolated from the rest (Fig. 15), fall in the K. africana and

X. monospora region of the ordination.

Selaginella kraussiana (Fig. 16)

The presence of this herb, recorded in 1966, reflects a high soil-moisture

content and the majority of plots which contain this species are in close proximity

to small flowing streams or noticeably dank. The distribution of this species

in the ordination coincides with P. henkelii and the closed canopy region (Figs.

4 and 9).

456

Fics. 4—16. — Distribution on the ordination of:-

4—8: Plot density of selected canopy tree species.

9: Canopy (c = closed, o = open and * = no canopy; isoline encloses region of

closed canopy).

10: Plot density (absolute values plotted).

11: Mean plot CBH 1929 (CBH size classes in inches: 1 = < 60, 2 = 60—90, 3 =

91—110, 4 = > 110).

12: Mean plot increment 1929—1966 (CBH increment classes in inches: 1 = < 5,

2 = 5S—10, 3 = > 10; isolines enclose regions of maximum and minimum incre-

ment).

13: Slope (S=steep, G = gentle, F = flat).

14: Rocks (R= rocky, S =semi-rocky, O =no rocks).

15: Mist (M = plots with most orographic mist).

16: Selaginella kraussiana present in the herbaceous field layer.

Fig.14. Rocks Fig.15.Mist

Fig.16. Selaginella kraussiana

458

CONCLUSIONS

Tree GrowTH

(a) Rate of forest tree growth at Xumeni is very slow, mean CBH incre-

ment for all species being 0.201 + 0.015 inches per year.

(b) P. henkelii exhibits the minimum increment, X. monospora a_ higher

increment approximating to the average of all trees, and P. falcatus,

K. africana and F. davyi the highest. F. davyi shows the maximum

CBH increment of all canopy species.

known to foresters, is clearly demonstrated for this area of indigenous

forest.

ORDINATION

(1) Specific conclusions

(a)

(b)

(ec)

(1)

Two axes, X and Y, were sufficient to account for the floristic

variability within the 39 plots.

Comparing Figs 4 and 13—16 it appears that P. henkelii occurs

on moist, flat, rocky terrain situated below the level of most

orographic mist.

Comparing Fig. 5 and Figs 13—i6, it appears that P. falcatus is

associated with gently sloping, dry, rocky areas with most orographic

mist.

From b and c, and with reference to Figs 10, 15 and 16, it is

suggested that the X axis reflects a density gradient with P. henkelii

and P. falcatus at the two extremes from respectively flat, moist,

rocky plots exposed to damper atmosphere at a higher altitude.

This gradient is mirrored in decrease in canopy and the occurrence

of S. Kraussiana, the presence of S$. kraussiana showing high soil-

moisture and low light intensities.

The Y axis separaes the two Podocarpus spp. from K. africana,

X. moncespora and F. davyi (Figs 4—8), the latter three species

growing on the steeper, least rocky slopes, and appears to show

an environmental gradient from steeper, drier sites to gentle and

flat, wetter sites (rigs 13 and 16).

Figs 11 and 12 snow the inverse relationship, stated previously,

between mean plot CBH and increment, and the Y axis is related

to these two measures of productivity.

With reference to Figs 4—9, 11—13, 15 and 16 it is suggested

that the Y axis reflects a successional trend from seral sites with

K. «africana, X. monospora and F. davyi to climax sites with

P. henkelii and P. falcatus, the X axis showing the separation

between the two climax species as described in (d).

(h)

459

Fig. 17 summarizes the suggested site-succession relationships.

Stoop

SERAL STANDS

MAXIMUM INCREMENT OPEN CANOPY

2 MINIMUM CBH STEEP TOPOGRAPHY

Ke africana

EB any!

XY. mondspora

“ CLIMAX STANDS ™

FA ANI AUMINGREMENT MEDIUM INCREMENT

ye MAXIMUM CBH MAXIMUM CBH %&

my CLOSED CANOPY OPEN CANOPY

FLAT TOPOGRAPHY GENTLE TOPOGRAPHY

“aw

P henkelii ike P ‘faleatus

Soil Wet Soil Dry

Least Mist Most Mist

Fic. 17. — Diagrammatic representation of suggested environmental gradients and successional

trend on the ordination.

(ii) General conclusions

(a) The plotting of information collected in 1966 onto an ordination

(b)

scatter derived from data collected in 1929, and the resultant correla-

tions, indicate that the forest environment has remained stable over

this period of years. The value of plotting information onto an

ordination based on floristic data collected in previous seasons has

been shown to have considerable use (cf. Woods, 1964).

Consider two hypothetical plots A and B, containing species a, b,

c, d and e in the following densities:

Plot Species a b c d e

A Ll — ] ] yD}

B —- — 1 ] 1

C (coefficient of similarity) between A and B = 75.0 per cent.

Suppose species c were absent from plot B, then C would have

a value of 57.1 per cent.

The plot density values at Xumeni for all species were low (Fig. 10),

and it was initially suspected that such low values might have

yielded little meaningful information in the matrix of similarity

coefficients. However, the ordination technique employed proved

460

reasonably robust in view of the nature of the data used. The

value of a correlation coefficient between 151 randomly selected

inter-point distances and the respective dissimilarity (100 —C)

values was 0.678.

compared with such a technique as principal components analysis

(Orloci, 1966), it is realized that the Wisconsin ordination technique

does not yield high correlations between inter-point distances and

similarity coefficients. However, the actual distances between points

are not necessarily of paramount importance unless, for example,

it is desired to classify the samples. Numerous workers have

shown that the Wisconsin technique yields a meaningful overall

pattern of variation in vegetation samples analysed. It has the

considerable advantages of not necessarily requiring electronic com-

putation facilities and unlike such classificatory techniques as Asso-

ciation Analysis, ordination is primarily useful in the study of the

autecology of species when overall trends, rather than inter-point

differences, are studied.

ACKNOWLEDGEMENTS

We are grateful to the Department of Forestry and in particular Mr. H. Haigh

for stimulating our interest and for valuable assistance in the field. Our thanks

too, to Mr. E. L. Abraham for drawing the figures. Finally we thank the

Director, Botanical Research Institute, for facilities and for permission to publish

our findings.

REFERENCES

Acocks, J. P. H., 1953. Veld types of South Africa. Mem. Bot. Surv. S. Afr,, No. 28.

Bray, J. R. & Curtis, J. T., 1957. An ordination of the upland forest communities of southern

Wisconsin. Ecol. Monogr. 27: 325-349.

Cook, G., 1929. Sample Area Records: Xumeni Forest. Unpublished data, records in Natal

Region of Forestry Department.

Mott, E. J. & Haicu, H., 1967. A report on the Xumeni forest. Natal. Forestry in S. Afr. 7:

99-108.

Ortoct, L., 1966. Geometric models in ecology. I. Theory and application of some ordination

methods. J. Ecol. 54: 193-215.

Puiuires, J. F. V., 1931. Forest-succession and ecology in the Knysna region. Mem. Bot. Surv.

S. Afr. No. 14.

Scuuize, B. R., 1947. The climates of South Africa according to the classification of Koppen

and Thornthwaite. S. Afr. geogr. J., 29: 32-42.

Woops, D. B., 1964. A study of the ecology of selected annual and bienniel species on sand-

dunes. Unpublished Ph.D. thesis. University of Wales.

Bothalia, 10, 3: 461—500

The Flora of the Mariepskop Complex’

H. P. van der Schijfff and E. Schoonraadt

ABSTRACT

The physiography, soils, climate and the main plant communities found on Mariepskop,

Eastern Transvaal, are briefly described; a checklist of the plants recorded is appended.

CONTENTS

Page

INSEROD WCW Nam aR em onl cat cae PGS. 9 Py eee tee SR grate ii Ca 462

CEOnOGNas ee: Se eng em ema en ear aie See Or Te ne ere 463

SON, “has Ee i SM Are Ee ko a ae ee 464.

CiTIVPAT Emer ean ee ah ct hgnen pe A ieee 7h RS ey cata, ae. tun 465

THE PLANT Counce Petcee SMP) gt eel, Nee, Be GR i Geet ge 465

Pee owsAltitudemwoodl andi yuscea ee eae 465

(a)) Lowveld (Gallery Forest 20. 0. cen cen em Re Pe eee 466

(b) Savanna on mountain slopes. ..... LS TD CRO ie ster Pine Bs 466

(d) Submontane forest in moist and sheltered kloofs 20... cee one 468

Zee hiiohe Mountains Grasslandy 2 (ee ef ee4 Gee) ee ee ee Ge 469

35 Mountains lateauaCommunities: 22. soc) os ee ee ee eee ee 470

AmaVvaddringtonva, Communities!) ae ee eee ee) 472

Semel ViontanesHonestwecpiets et eet as Neer rs | Sen ee 474

(GUY (CENANOy Oy eat eae ee eae a a ee oe 475

) Shade-tolerant small-tree stratum oi, ec, cece cece eee sete senee 476

()eShrubystratum 3. a ees, Ges cee Mn wae. ere. 2 476

Ca) menbmstraterrigane es seers rete ee a aera ricy EE va petenn Wane 476

J) ESET.) age an OAR ae aoe aco a IDO Ne tay Rc J ES 476

(GE) MBE on Wiyy tC Sta te peste ei tree ee a eer Se aN | cnt AT7

6. Indigenous Communities in Plantations ..... ..... Pitt pies seve eee 477

HESAREINIGIES OGRE PEPORA a ok eo ee ks, en ee, eS 478

HACTORSPENDANGERING SRIIE (INDIGENOUS) SLORAU fs Sa) a. eons urea nae 478

Te Silivicul tire Mamet etre cng ok Pode I Gt Mi eM pd. ot ae 478

ee 1 ee ae MM Ray | ete pee Chey Seed aie Ml plea tke 479

SUIMINGAH Vane PE whe SN ee, eee re al in Oech Gd.» Tbe. Cats 479

AICKNOWIEDGEMENESH i Maison. nick (feb eee “ake, aan) Jue wean Sek seed ee 480

INEBEREN CES Gigi aimee en ets Sen ate Hg A a Oa aia tema aes 480

PRELIMINARY CHECK LIST OF THE VASCULAR PLANTS i, cess cunt, sense nee 481

* This work was made possible by grants from the Research and Publications Committee of

the University of Pretoria and from the Council for Scientific and Industrial Research.

+ Department of General Botany, University of Pretoria.

462

INTRODUCTION

Mariepskop (altitude 1946 m) is situated at 30°52’ E and 24° 30’ S in

the district of Pilgrim’s Rest, in the Transvaal Drakensberg Mountains. Access

to the area is via Lydenburg and through the Abel Erasmus Pass towards Acorn-

hoek, or via Nelspruit and thence northwards to Acornhoek and _ Klaserie.

Mariepskop is approximately 32 km west of Klaserie station and is not yet

accessible by road through the Blyde River Valley.

The name of the mountain is derived from the Bantu chief Maripi who with

his tribe found sanctuary from raiding Swazi warriors on this high mountain

known by the Bantu as Mohuluhulu, “the great one”. The flat summit formed

a natural fortress of great strength which they were able to defend by rolling

boulders down the cliffs. Although the Swazis attacked the fortress on numerous

occasions they did so without success. In the words of Bulpin (1965) “The

bones of their dead are still to be seen as mementoes of a tough siege and

bitter defence”’.

(2)

SOUTHERN

Fic. 1. — Location of Mariepskop, the study area.

463

The Mariepskop-Magalieskop complex is separated from the rest of the

Drakensberg Range by a tributary of the Klaserie River and by the Blyde River,

the latter forming a deep canyon. The Klaserie River, which rises in the rain

forests on the southern slopes of Mariepskop, is a perennial stream that provides

irrigation water for the riparian farmers in the Lowveld. Here, the Mariepskop

complex marks the position where the Drakensberg escarpment changes direc-

tion, from a north-south direction to the south of Mariepskop to a _south-

easterly to north-westeriy direction towards the Wolkberg. The valleys of the

Klaserie and Blyde Rivers are important because they form migration routes

for the Lowveld flora westwards into the mountains and for the montane flora

eastwards to the Lowveld.

As at 3lst March 1969, 2954 ha of the Mariepskop Forest Reserve were

covered by indigenous forest and 1560 ha by plantations, while 121 ha are

suitable for future plantations and 378 ha could not be utilized for silviculture.

GEOLOGY

According to Hall (1910) and Du Toit (1954), Mariepskop is partly formed

of the erosion-resisting quartzites of the Black Reef Series that extends from

Duiwelskantoor near Kaapsehoop to the Wolkberg. This formation outcrops

along a narrow belt running roughly north and south, parallel to the Dolomite

area on the west, and gradually turning towards the north-west after crossing

the Blyde River. etween Belvedere and Mariepskop the Series attains its

maximum surface width of 8.5 km.

SECTION OF MARIEPS MOUNTAIN

BLYDE RIVER, MARIEPS LOW

MOUNTAIN. COUNTRY.

eT hata

Fic. 2. — Geological section of Mariepskop Complex, (after Hall, 1910).

464

Northwards and north-westwards from the source of the Treur River, the

Black Reef plateau undergoes a marked change, owing mainly to the increased

erosion by the Treur, Blyde and Ohrigstad Rivers. These have cut deeply into

the succession of quartzites and sandstones and have carved out a series of

long, deep kloofs.

“The quartzites weather in the peculiar and fantastic manner of the Table

Mountain Sandstone; are vertically jointed and give rise to scenery of consider-

able grandeur, the rugged grey crags of the Escarpment contrasting with the

rounded, grassy, and often forested granitic slopes of the ‘Low Veld’ beneath”

(Du Toit, 1954).

The Black Reef Series consists of a succession of sandstones, quartzites,

conglomerates and sandy shales. This division of the Black Reef Series is clearly

recognizable around Mariepskop. The major stratigraphical sub-divisions are:

Shaly sandstones

Quartzite

Shales and shaly sandstones

Quartzite with conglomerate bands at base.

Rassage-bedSise ar

780 m Upper quartzite group .

Middle shale group ... ..

Lower quartzite group ..

SoILs

Most soils at Mariepskop belong to the Lateritic Red Earths of Van der

Merwe (1940). The mature soils are strongly weathered and deep with the

mineral content generally low. Horizon development is poor, but the soil is

well drained. The soils of the higher areas with a higher rainfall are more

leached and laterized than those in the Blyde River Canyon and at the foot-

hills of the mountain. The humus content in the topsoil is fairly high.

No direct correlation could be found between the vegetation and soil type,

except that both are influenced by the climate. For details of the soil conditions

of Mariepskop Complex, see Table 1.

TaBLE 1. — Soil conditions of Mariepskop Complex.

Mountain Lowveld

Soil Summit Montane Sour

depth Grassveld Forest Bushveld

2.5-5 cm pH Spl 4.9 6.3

35 cm 5.0 4.4 5.4

2.5-5 cm Texture Coarse Sand Loamy Sand Coarse sandy

loam

35 cm Sand Sandy loam Sandy clay

2.5-5 cm Structure Structureless Structureless Poorly devel-

oped

coarse

block

35 cm Structureless Poorly de- Moderately

veloped developed

coarse coarse

block block

465

CLIMATE

The only data available on the climate of the area are those recorded at

an intermediate altitude on the Mariepskop and Salique Forestry Stations. There

is little doubt that the mean annual precipitation on the higher slopes is much

higher than at these two stations, possibly up to 2500 mm in the southern and

south-eastern kloofs at higher altitudes, while it is probably much lower than

the recorded rainfall along the foot-hills and in the Blyde River Valley. This

presumption is borne out by the foresters at Mariepskop Forestry Station who

have often witnessed rainy days and rainy periods higher up in the kloofs when

it has been dry at the Forestry Station and near the Blyde River in the valley.

At the Mariepskop Forestry Station the mean annual rainfall over a period

of 25 years is 13.7 mm and the average number of days per year on which

rain falls is 96.2. The wettest months are December, January and February

with monthly means of 205 mm, 244.6 mm and 286 mm, respectively, and an

average of 12.7, 14.2 and 13.5 rainy days per month. The driest months are

June, July and August with 2.4, 2.2 and 3.4 rainy days, respectively, and average

precipitations of 23.4 mm, 19.1 mm and 20.8 mm per month. Heavy down-

pours of 50 mm and more within 24 hours are not unusual, and mist is frequent.

The Lowveld, nearly 1220 m lower than the highest point on the moun-

tain, is much warmer and the humidity much lower.

No temperature data are available.

Mist is an important source of moisture for most Bryophytic flora and the

great number of epiphytes growing in the montane forests. The lichen Usnea,

which is mainly dependent on mist for its moisture, is found on most of the

forest trees, especially Podocarpus latifolius, P. falcatus and Widdringtonia cupres-

soides.

THE PLant COMMUNITIES

On the basis of observations made during the past ten years, the vegetation

may be divided into several communities. No quantitative surveys have been

carried out, however, and the subdivisions are based mainly upon physiognomic

features.

1. Low Altitude Woodland Communities (Plate 1).

These plant communities are a continuation of Acocks’ (1952) Lowveld

Sour Bushveld and Van der Schijff’s (1958) Large-leaved Deciduous Bushveld

with Tall Grass. They occur in the Blyde River Valley, on the eastern foot-

hills of the Drakensberg below the Forestry Station and on the western and

northern slopes of Mariepskop.

At these sites the rainfall is relatively low, being from 635—760 mm per

annum. The area is low-lying (altitude 760 m) and on the northern slopes,

where the insolation is greater, the temperatures and the rate of evaporation are

probably much higher than on the southern and south-eastern slopes.

The area is broken and the vegetation varies from relatively open bushveld

with tall grass on the undulating hills and slopes, through dense brushwood and

scrub in the dongas and dry ravines, to patches of submontane forest in sheltered

kloofs. This vegetation type can be subdivided into four main communities.

Pirate 1. — The Mariepskop Complex as seen from the Lowveld near Acorn Hoek.

(a) Lowveld Gallery Forest.

This type of forest is found on the banks of the Blyde and the Klaserie

Rivers, although it is not well developed along the Blyde River on the north-

western side of Mariepskop. Prominent species include:-

Adina microcephala var. galpinii Ekebergia capensis

Ficus capensis Combretum imberbe

F. sycomorus Trichilia emetica

Schotia brachypetala Catha edulis

Spirostachys africanus Acacia robusta

Syzygium cordatum A. ataxacantha

S. guineense Bauhinia galpinii

Combretum erythrophyllum Phoenix reclinata

Acacia albida Bridelia micrantha

Xanthocercis zambesiaca Acacia karroo

Rauvolfia caffra Antidesma yvenosum

Diospyros mespiliformis

(b) Savanna on mountain slopes (Plate 2).

At higher altitudes, the slopes are wetter and more mesic tree and shrub

species are found, e.g. Faurea speciosa, F. saligna, Acacia ataxacantha, Greyia

sutherlandii, Trema orientalis, Iboza riparia, Antidesma venosum, Ficus petersit,

Heteropyxis natalensis, Fagara capensis, Catha edulis, Rhoicissus tridentata and

Acacia caffra.

467

On the drier lower slopes, trees and shrubs found, among others, are:-

Annona senegalensis

Pterocarpus angolensis

P. rotundifolius

Acacia caffra

Dichrostachys cinerea ssp. glomerata

Parinari curatellifolia ssp. mobola

Sclerocarya caffra

Grewia flavescens

G. monticola

Acacia gerrardit

A. sieberana var. woodii

A. davyi

Cussonia natalensis

Euclea divinorum

Commiphora mollis

Chaetacme aristata

Strychnos innocua

Aloe mariothii

Dombeya rotundifolia

Flacourtia indica

Ficus ingens

Lannea discolor

EL. edulis

Pappea capensis

Peltophorum africanum

Ximenia caffra var. natalensis

Vangueria infausta

Ziziphus mucronata

Combretum zeyheri

C. apiculatum

C. imberbe

Holmskioldia tettensis

Ficus smutsi

F. soldanella

F. sonderi

Terminalia sericea

Trichilia emetica

Prate 2. — Woodland on mountain slope with a north-eastern aspect, grading into sub-

montane forest in a sheltered kloof in the foreground (right).

Typical trees and shrubs found in this type of habitat include the following:-

Kirkia wilmsii

Ptaeroxylon obliquum

Terminalia phanerophlebia

Steganotaenia araliacea

Urera tenax

Cussonia natalensis

Aloe marlothii

Combretum apiculatum

Grewia monticola

Ficus sonderi

Sarcostemma viminale

Euphorbia triangularis

E. tirucalli

Schotia brachypetala

Pappea capensis

Ficus ingens

(d) Submontane forest in moist and sheltered kloofs.

In moist kloofs higher up the mountain slopes and in the Blyde River

Canyon itself, the vegetation changes into submontane forest. The following

trees and shrubs are typical:-

Celtis africana

Ficus capensis

Ptaeroxylon obliquum

Ekebergia capensis

Mimusops zeyheri

Homalium dentatum

Syzygium cordatum

Anthocleista grandiflora

Halleria lucida

Pittosporum viridiflorum

Croton sylvaticus

Trimeria grandiflora

Cussonia spicata

Warburgia ugandensis

Protorhus longifolia

Acacia ataxacantha

Piare 3. — High Mountain Grassland on plateau on mountain summit.

469

Lianas that can be expected in these forest patches include Dalbergia armata,

Rhoicissus tomentosa, R. tridentata, R. revoilii, Grewia occidentalis, Smilax kraus-

siana, Pterolobium exosum, Entada spicata and Acacia ataxacantha.

In the kloofs of the eastern foot-hills these communities have mostly been

replaced by plantations, but on the northern and western slopes the natural

vegetation is still relatively intact.

2. High Mountain Grassland (Plate 3).

This type of vegetation is characteristic of the less moist and exposed ridges

between wooded kloofs on the southern and south-eastern slopes, and also of

the mountain summits of the Drakensberg. There are also isolated patches of

mountain grassland scattered in the forest communities, a good example being

in the vicinity of Colonel Reitz’s grave. Frequently these patches of grassland

erve as sponges for the mountain rivulets that constitute the head-waters of

the Klaserie River.

Where the habitat is favourable, a fynbos type of vegetation replaces the

grassveld. Hygrophilous species of shrubs and herbs associated with the grass-

veld in these areas include Cliffortia linearifolia, Erica leucopelta var. luxurians,

Hemizygia albiflora, Hypericum revolutum, Gunnera perpensa and Thelypteris

palustris var. Squamigera.

The typical mesophytic mountain grassveld consists of a mixture of short,

“sour” grasses of which some of the dominant species, such as Danthonia drakens-

bergensis, form tussocks. ‘Characteristic grasses of this community are:-

Eragrostis sclerantha Sporobolus eylesii

E. caesia Themeda triandra

E. capensis Stiburus alopecuroides

Agrostis barbuligera Trichopteryx dregeana

var. longipilosa Andropogon distachyos

Harpechloa falx Eulalia villosa

Loudetia simplex Tristachya hispida

Panicum spp.

Geophytic species associated with the grassveld include:

Brunsvigia natalensis Hypoxis angustifolia

Watsonia densiflora H. argentea

Moraea spathulata H. membranacea

Dierama robustum H. rigidula

Other shrubs and herbs associated with the grassveld are:

Cliffortia linearifolia Lopholaena disticha

Helichrysum lepidissimum Mohria caffrorum

A. latifolium Smithia thymodora

H. wilmsii Cheilanthes multifida

H. odoratissimum Vaccinium exul

H. platypterum Kniphofia triangularis ssp. obtusiloba

H. appendiculatum Restio sieberi var. schoenoides

Erica leucopelta var. luxurians Tetraria cuspidata

E. drakensbergensis Scirpus macer

Protea rouppelliae Vernonia corymbosa

P. rhodantha var. rhodantha Crotalaria doidgeae

P. gaguedi Teedia lucida

470

A feature in this area is that some of the largest patches of mountain

grassland have been replaced by pine plantations, while the small patches that

remain higher up the mountain in the natural forest belt are gradually being

invaded by Pinus patula. Every effort should be made to eradicate these self-

sown trees from the sponges that provide the headwaters of the Klaserie River.

3. Mountain Plateau Communities.

The plateau (altitude 1920 m) is bordered by vertical cliffs with a sheer

drop of nearly 1220 m. The surface of the plateau is extremely rocky with

shallow soil and a very high, though fluctuating water table. The climate is

moist and cool, although the effective rainfall is lower than in the kloofs. For

most of the year strong winds, which impair tree growth, blow over the summit.

Pare 4. — Sheltered rocky basin on summit of mountain with sclerophyllous scrub in the

foreground and Widéringtonia scrub in the background.

The following three habitats can be distinguished on the plateau:

(a) Level, soil-covered areas cccupied by Danthonia Grassland with associated

geophytes. These areas are exposed to winds, the soil layer is approxi-

mately 0.5 m deep and there are no rock outcrops.

(b) Flat, exposed rock habitats without soil and covered by lichens.

occur (Plate 4).

471

Piate 5. — Flat exposed lichen-covered rocks with fissures hundreds of feet deep on moun-

tain summit. Wind-stunted Podecarpus Iatifelius scrub on right and stray Pinus patula

in centre background.

The flat exposed rocks are covered with a variety of chasmophytic crustose

and foliose lichens (Plate 5). Small hollows and depressions in the rock surface

form semi-permanent pools of water where, under certain conditions, a very rich

annual flora, with species such as Ilysanthes conferta and perennials such as

Isoétes natalensis, occurs.

Although wind is a determining factor for species composition and growth

form, huge boulders surrounding shallow basins provide adequate shelter for

local islands of luxuriant vegetation. In these shallow sheltered basins, heath-

like shrubs and their associates are found, the following being either dominant

or conspicuous:

Passerina montana Cliffortia serpyllifolia

Erica leucopelta var. ephebioides C. nitidula ssp. pilosa

E. leucopelta var luxurians Phylica paniculata

E. woodii Psoraled pinnata

Vaccinium exul Muraltia flanaganii

In the same kind of habitat, and more often than not together with the

fynbos, scattered communities of Protea rouppelliaz, P. rhodantha var. rhodantha

and P. gaguedi are found.

472

Other characteristic plants of the mountain plateau include:-

Anisopappus junodii

Otiophora cupheoides

Aloe nubigena

A. arborescens

Cyrtanthus huttonii

Agapanthus inapertus

Zaluzianskya katherinae

Cineraria fruticetorum

Crassula browniana

C. setulosa

Crocosmia _ pottsti

Clivia caulescens

Anthospermum hispidulum

Eumorphia davyi

Hemizygia albiflora

Silene burchellii

Plectranthus fruticosus

Albuca fastigiata

Craterostigma wilmsii

C. plantagineum

Helichrysum spp.

Sebaea erosa

Hebenstreitia comosa

C. parvisepala

Kniphofia triangularis ssp. obtusiloba

K. linearifolia

Watsonia densiflora

Euryops rogersii

Selago villosa

S. nelsonii

Nemesia melissaefolia

Macowania tenuifolia

Carex zuluensis

Monopsis kowynensis

Hesperantha baurii

Lopholaena disticha

4. Widdringtonia Communities (Plate 6).

These communities occur in fissures and kloofs on the southern and south-

western cliffs just below the summit of the mountain, and between huge boulders

on top of the mountain.

In the kloof south of the trigonometrical beacon on top of the mountain,

where the habitat is probably particularly favourable for this community, prac-

tically pure stands of Widdringtonia cupressoides with trees up to 15 m were

found. During 1967 this forest was, however, destroyed by a fire which was

most probably caused by lightning. Fortunately the fire did not spread to the

plateau vegetation or into the montane forest, probably because of the rocky

surface of the areas adjoining this particular kloof. There are still few signs

of life in the black scorched tree trunks of the Widdringtonia community and

it appears that very few of the trees have survived the fire. The area is at

present occupied by small herbs and shrubs, mainly Selago spp., Psoralea spp..

Helichrysum spp., Senecio spp., Hypericum spp., Dioscorea spp., Cyperaceae and

grasses. Cyperaceae occur throughout the area and, near the small stream that

runs through the kloof, some species are approximately 1.25 m tall. Young

Widdringtonia trees are appearing among the grass and, at present, average

approximately 70 cm in _ height.

Between boulders on the mountain summit, where the vegetation is exposed

to strong winds, the trees are often stunted. Here the Widdringtonia shrubs are

associated with other shrubs and small trees such as Rhus dura, Cliffortia nitidula,

Podocarpus latifolius, Phylica paniculata, Cassine eucleaeformis, Pterocelastrus

echinatus, and with Lycopodium clavatum and L. cernuum.

On the slopes below the summit, Widdringtonia cupressoides occurs in

association with various typical forest trees such as Olea woodiana, Podocarpus

latifolius, Xymalos monospora, Schrebera alata and Pterocelastrus echinatus.

473

Pirate 6. — Destruction by fire of Widdringtonia cupressoides community in sheltered

south-west facing kloof near summit of mountain.

5. Montane Forest (Plate 7).

A very large part of the Mariepskop-Magalieskop complex is covered with

evergreen montane forest. In the deep kloofs and on the southern or south-

eastern slopes, these forests attain their maximum development. In these kloofs

the forest extends far down to merge eventually with the submontane forest of

the Drakensberg foot-hills and this in turn into the Gallery Forest bordering

the perennial rivers of the Lowveld.

Common tree species are:-

Podocar pus latifolius Halleria lucida

P. falcatus Apodytes dimidiata

Diospyros whyteana Curtisia dentata

Ficus capensis Celodendrum capense

Kiggelaria africana Fagara davyi

Faurea macnaughtonii Rhus chirindensis forma legatii

Pittosporum viridijlorum Olea capensis ssp. macrocarpa

Ilex mitis Nuxia floribunda

Rapanea melanophloeos Prunus africana

Celtis africana

Prare 7. — Transitional Lowveld \WWoodland—Montane Forest on sheltered cast facing slopes

of Mariepskop.

475

Where the forest attains its maximum development four strata can be dis-

tinguished:

(a) Canopy

The most common species of this stratum are:-

Podocarpus tatifolius

P. falcatus (can attain a height

of 18.5 m or more)

Cussonia umbellifera

Ochna oconnorii

Curtisia dentata

Kiggelaria africana

Nuxia floribunda

N. congesta

Apodytes dimidiata

Brachylaena discolor

Prunus africana

Combretum kraussii

Diospyros whyteana

Syzygium gerrardii

Schrebera alata

Piate 8. — Scrubby undergrowth in open

footpath.

Faurea macnaughtonit

Brachylaena transvaalensis

Ficus petersit

Xymalos monospora

Celtis africana

Flalleria lucida

Olea capensis ssp. macrocarpa

O. woodiana

Ilex mitis

Pterocelastrus echinatus

Bersama tysoniana

Trichocladus grandiflorus

Protorhus longifolia

Cussonia spicata

Cryptocarya liebertiana

C. woodii

patch of typical montane forest along Bedford

476

(b) Shade-tolerant small-tree stratum

In this stratum, trees and shrubs such as Rawsonia lucida, Fagara davyi,

Trimeria grandifolia, Rhus chirindensis forma legatii, Rothmannia capensis, Tri-

calysia capensis, Oxyanthus gerrardii, Grumilia capensis, G. kirkii, Calpurnia

aurea and Croton sylvaticus, are found.

This is a heterogeneous stratum that varies with the aspect and the amount

of light that penetrates the two higher strata. It is best developed towards the

forest margins, where it sometimes forms dense thickets. The most characteristic

components are Bowkeria cymosa, Sclerochiton haveyanus, Plectranthus spp..

Piper capense, Mackaya bella, Pavetta lanceolata, Vangueria cyanescens, Duver-

noia adhatodioides and Canthium obovatum.

Along streams and in other moist places in dense shade, abundant growths

of Pteridophyta are found, e.g. Marattia fraxinea var. salicifolia, Blechnum capense

and Cyathea capensis.

In moist areas with higher light intensity, Hypolepis sparsisora is found, as

well as Adiantum poiretii, Asplenium friesiorum, A. gemmiferum and Blechnum

attenuatum var. giganteum.

(d) Herb stratum

Like the shrub stratum, the herb stratum on the forest floor varies appre-

ciably with the light intensity and consists exclusively of herbs and herb-like

species. Where the tree and shrub strata are very well developed and the light

intensity is consequently low, the floor vegetation is in many cases totally wanting

or limited to species such as Carex spicato-paniculata, Galopina circaeoides, Clivia

caulescens, Selaginella kraussiana and Oplismenus hirtellus. The latter grass

is able to grow in dense shade. Other species characteristic of this stratum

include:

Impatiens sylvicola Thalictrum rhynchocarpum

I. duthieae Alchemilla rehmannii

Cardamine africana Asplenium inaequaelaterale

Pseudobromus africanus A. aethiopicum

Begonia spp. A. erectum

Laportea peduncularis A. rutaefolium

Peperomia retusa Dryopteris inaequalis

Hypoestes verticillaris Thelypteris pozot

Streptocarpus micrantha

Where there is more light the floor vegetation is fairly dense with Plectranthus

spp.. Piper capense, Laportea alatipes and representatives of the Acanthaceae

like Barleria gueinzit.

(e) Lianas

A conspicuous feature of the montane forest is the variety of lianas and

epiphytes. The most important lianas are:

Rhoicissus rhomboideus Jasminum abyssinicum

R. revoilii J. fluminense

R. tomentosa Landolphia capensis

477

Canthium gueinzii Cnestis natalensis

Secamone alpinii Smilax kraussiana

S. gerrardii Behnia reticulata

Riocreuxia torulosa Stephania abyssinica

Quisqualis parviflora (very robust) Helinus integrifolius

In some localities along the edge of the forest, Rubus rigidus, Entada spicata,

Sphedamnocarpus galphimiifolius, Dioscorea retusa, Senecio tamoides, S$. quin-

quelobus and Mikania cordata form an impenetrable thicket or scrub together

with shrubs like Hypericum revolutum, Pteridium aquilinum, Buddleia salvifolia

and Plectranthus spp. Lower down in the kloofs, where elements of the Low-

veld flora become more prominent, the most conspicuous lianas are Rhoicissus

tridentata, Adenia gummifera, Dalbergia armata, Acacia ataxacantha and Ptero-

lobium exosum.

(f) Epiphytes

Most of the epiphytes recorded belong to the Pteridophyta or to the families

Orchidaceae and Gesneriaceae of the higher plants. Epiphytic and epiphyllous

mosses and lichens are also plentiful. The most important epiphytic Pterido-

phyta are:

Vittaria isoetifolia Loxogramme lanceolata

Lycopodium verticillatum Asplenium sandersonii

L. dacrydioides A. anisophyllum

Polypodium polypodioides ssp. ecklonii A. splendens

Pleopeltis macrocarpa Elaphoglossum acrostichoides

The most common epiphytic orchids include:

Tridactyle tricuspus Mystacidium capense

Bulbophyllum sandersonii Angraecum sacciferum

Of the Gesneriaceae, Streptocarpus haygarthii, S. micrantha and S. wilmsii

are common epiphytes, but they also grow, together with Peperomia retusa, on

rocks in the forest.

Clivia caulescens (Amaryllidaceae), usually found on the forest floor, is

often one of the most conspicuous epiphytes. Even Aloe arborescens is some-

times found as an epiphyte on trees at the edge of the forest.

6. Indigenous Communities in Plantations

In the Pinus plantations, especially the older ones, very few or no shrubs

or herbs are found.

A better developed undergrowth is found in the Eucalyptus plantations. It

would be interesting to observe the secondary succession in plantations left

undisturbed for the next 20 or 30 years. In the higher areas with a southern

or south-eastern aspect, it appears as though the succession will eventually result

in a montane forest climax. Hitherto, the weeding that is done, is not selective,

with the result that pioneer seedlings of the montane forest and of true forest

trees are eliminated with normal ruderals.

478

Shrubs that have been found in the plantations include:

Clerodendrum glabrum Smilax kraussiana

Vernonia shirensis Dombeya pulchra

Hibiscus meeusei Heteropyxis natalensis

Pycnostachys urticifolia Antidesma venosum

Rhus intermedia Adenia gummifera

Maesa lanceolata Cassia petersiana

Trema orientalis Rhoicissus tridentata

Lippia javanica Dalbergia armata

The grass, Setaria chevclieri, is common, and the exotic liana, Caesalpinia

decapetala, is encroaching on natural vegetation on the eastern foot-hills of

Magalieskop.

THE AFFINITIES OF THE FLORA

Of the considerable number of interesting species collected in the area, the

following deserve special mention: the ferns, Pyrrosia schimperiana and Hymeno-

phyllum capillare; the lichen, Cladonia rangiformis; and the fire lily, Cyrtanthus

Auttonii.

As far as can be ascertained, Pyrrosia schimperiana and Hymenophyllum

capillare have not previously been collected in South Africa, and the only

specimen of Cladonia rangiformis in the National Herbarium, Pretoria, is from

the Congo. The distribution of Cyrtanthus huttonii, which is abundant under-

neath cliffs that face south and south-east on top of Mariepskop, requires

further study. The nearest locality of this species to Mariepskop is the Katberg

in the Eastern Cape Province. Phytogeographically, Mariepskop is of special

importance, because it is clear that it is an unusually interesting floral area.

The presence of predominantly southern genera like Protea, Erica, Otiophora,

Phylica and Restio, and species such as Cliffortia nitidula, Passerina montana,

Widdringtonia cupresscides, Myrsine africana, Vaccinium exul and Smithia thymo-

dora in the communities on the mountain plateau is of special interest as they

also show affinities with the fynbos of the Chimanimani Mountains (Phipps and

Goodier, 1962). This indicates that the eastern mountain ranges could possibly

form a migration route for the sclerophyll flora of the South-western Cape. On

the other hand, the montane forests have strong affinities with the forests of

tropical eastern Africa, as found at Mount Salinda and the Chimanimani Moun-

tains. Hence the Drakensberg Range forms a migration route for both the

southern and the tropical elements of our indigenous flora.

Factors ENDANGERING THE INDIGENOUS FLORA

1. Silviculture

A very large area, 1 560 ha of the 5013 ha incorporated in the Mariepskop

Forest Reserve, is covered with plantations, of which an area of 1152 ha is

planted to conifers. Apart from the fact that very little undergrowth can survive

under the coniferous trees, they are also very well adapted under these climatic

conditions to encroach on the natural vegetation. Should these trees spread from

the plantations in any great number, it would present a reel danger to the

Mountain Grassland. Pinus patula, in particular, establishes itself very easily

at the forest edge and in grass patches. It is strongly urged that these trees

should be destroyed as soon as they are noticed.

479

Encroachment on the indigenous flora is, however, not the only adverse

influence of silviculture on the Mariepskop vegetation. The catchment area of

the Klaserie River, which rises on the southern and south-eastern slopes of the

Drakensberg, include 3720 ha of plantations, comprising 794 ha at Mariepskop,

2620 ha at Salique and 312 ha at Hebron.

It is already noticeable how springs, streams and rivulets, which flowed

quite strongly about ten to fifteen years ago, have slowly become weaker, and

how many of these have disappeared and completely dried up. It is of course

possible that this phenomenon is due to purely natural causes.

Stream-flow measurements on the Klaserie River, where it is crossed by

the main Tzaneen-Nelspruit road, are given in Table 2.

TasLe 2. — Mean annual rainfall for Mariepskop Forestry Station in millimetres

compared with the mean annual run off for the Klaserie River in cubic

metres between 1935 and 1960.

Period Rainfall (mm) Run off

(cubic metres)

1935-400 Bye es 1 729 143.07 x 10°

V94T AS oy oe il 22 48.72 x 10°

94625 0m esi. ho x 16332 38.43 x 10°

NOS5-600 Me 2 060 Paxcle4 I AKO?

L966 1 308 GSS exes

It is interesting to see how the flow of the river has steadily decreased

even though there has been no permanent drop in rainfall.

If the plantations have a detrimental effect on the subterranean water, the

vegetation on the foot-hills of Mariepskop, as well as the riparian flora of the

Lowveld, will be affected.

2. Fire

Occasional fires, detrimental to certain plant species, occur on the mountain,

but fortunately it seems these are usually the result of natural causes. The last

serious fire was that which destroyed the large pure community of Widdringtonia

cupressoides in the gulley between the road and the trigonometrical beacon.

SUMMARY

The Mariepskop Complex lies in the Eastern Transvaal Drakensberg and

forms part of the Eastern Escarpment which extends northwards from the Cape

Province.

At the forestry station the mean annual rainfall is 1 369 mm, falling mainly

in the summer months. In the sheltered kloofs facing east and south-east,

precipitation, as judged by the vegetation, is probably much higher.

Mariepskop is geologically partly formed of the Black Reef Series. Soils

are predominantly acid, lateritic, strongly weathered, and with generally low

mineral content.

The natural vegetation is divided into five main communities, whereas

indigenous shrubs and trees found in the pine plantations are dealt with separately.

480

Lowveld Sour Bushveld occurs in the Blyde River Valley on the eastern

foot-hills of the Drakensberg below the Forestry Station and on the western

and northern slopes of Mariepskop. The most conspicuous trees and shrubs

are Faurea saligna, F. speciosa, Peltophorum africanum. Sclerocarya caffra and

Parinari curatellifolia ssp. mobola, with typical Lowveld Gallery Forest on the

river banks. Prominent riparian species include Adina microcephala var. galpinii,

Ficus capensis and Syzygium cordatum.

High Mountain Grassland is characteristic of the less moist and exposed

ridges between wooded kloofs on the southern and south-eastern slopes, and also

of the plateau on the mountain summit. It is a short grassveld, the most con-

spicuous species being Danthonia drakensbergensis, Loudetia simplex, Tristachya

hispida and Andropogon distachyos.

The communities of the mountain plateau show affinities with both the

Cape fynbos and the scrub of the ‘Chimanimani Mountains.

On the southern, south-eastern and south-western cliffs just below the summit

of the mountain, in fissures and kloofs and between huge boulders on top of

the mountain. pure Widdringtonia cupressoides scrub is found. Depending on

the habitat, these communities vary from stunted shrubs to trees of appreciable

size. One forest in a kloof south of the trigometrical beacon was destroyed by

fire in 1967.

Evergreen montane forest with strong tropical affinities occurs in deep kloofs

and in slopes facing south and south-east and is the most conspicuous component

of the vegetation of this area. Typical species are Podocarpus latifolius, P. fal-

catus, Diospyros whyteana, Kiggelaria africana, Nuxia floribunda, Faurea mac-

naughtoni and Ilex mitis.

Factors endangering the indigenous flora are silviculture, fire and tourism.

ACKNOWLEDGEMENTS

The authors extend their sincere gratitude to the Forestry Department for

facilities provided, to the different foresters for their kind assistance, to the

Defence Department for their co-operation, to Mrs. F. Wilkens for help with

the compilation of the check list, and to the National Herbarium, Pretoria, for

identifying the specimens.

REFERENCES

Acocks, J. P. H., 1953. Veld Types of South Africa. Mem. Bot. Survey S. Afr. No. 28.

Butpin, T. V., 1965. Lost Trails of the Transvaal. Cape Town: Nelson.

Du Torr, A. L., 1954. The Geology of South Africa. Edinburgh and London: Oliver and Boyd.

Gooptrr, R. & Putpps, J. B., 1962. A revised check-list of the vascular plants of the Chimanimani

Mountains. Kirkia 1: 44-66.

Hatt, A. L., 1910. The Geology of the Pilgrim’s Rest Gold Mining District. Geol. Surv. Mem.

No. 5, Transvaal Mines Department, Pretoria.

Puipps, J. B. & Gooptrr, R., 1962. A Preliminary account of the plant ecology of the Chimani-

mani Mountains. J. Ecol. 50: 291-319.

Scuetpe, E. A. C. L. E., 1969. A revised check-list of the Pteridophyta of Southern Africa.

J. S. Afr. Bot. 35: 127-140.

VAN DER MERWE, C. R., 1940. Soil groups and subgroups of South Africa. Sci. Bull. Dept. Agric.

For. S. Afr., No. 231 (Chem. Ser. No. 165).

Van veR Scuijer, H. P., 1958. Inleidende verslag oor veldbrandnavorsing in die Nasionale

Kruger-wildtuin. Koedoe 1: 60-93.

Van ver Scuiyrr, H. P., 1963. A Preliminary account of the vegetation of the Mariepskop

Complex. Fauna & Flora 14: 42-53.

481

PRELIMINARY CHECK LIST OF THE VASCULAR PLANTS OF THE

MARIEPSKOP AREA, SOUTH AFRICA

The Pteridophyta are arranged into families and genera according to

Schelpe (1969) and the Angiospermae according to Dalla Torre and Harms

(1900-1907) and Phillips (1951). The numbers following plant names represent

the collector’s numbers of the senior author.

PTERIDOPHYTA

LYCOPODIACEAE

Lycopodium carolinianum L. Moist habitat in sum io. oi. cc ec cee sees wees ce cee 6450

Peicernuum ples Moist soillbetweens rocks! =) an oy ee ee ee eee

L. dacrydioides Bak. Epiphyte .0.0 00 0. ccc ce ce sing fe ieee ov Ge fe I are 6253

ES onidioidesplest nm OnUrOCKSie. rte hd. eeagracalath ea le Shady leaned vere ees A ee “5586, ou

D Ee Gunioelossbides Lam. Epiphyte Peeks Bee iain oe Negra esi 613

L. saururus Lam. Moist edge of Blyde_ ‘River footpath . peng, ney Are Aime eee b SONA

L. verticillatum L.f. Epiphyte on forest trees 0.00 oo. ccc cee tee se ee 4684, 6269

SELAGINELLACEAE

Selaginella caffrorum (Mile) Hieron. On plateau near trigonemetrical beacon. Epi-

phy temo nm IRE CSM ren ern ee ee emennn subst iy enrs = Uierltane SUM nsS Ae ee ea 6350

S. dregei (Presl) Hieron. ~ Xerophyte on rocky outcrops. ..... ee ee ee 6349, 4819

S. kraussiana (Kunze) A.Br. Abundant on forest floor... Alera ce 4532, 4696

S. mittenti Bak. On rocky outcrop, Thalene-kop 20 00. 200 uu. sow te alieamare ties 6502

EQUISETACEAE

Equisetum ramosissimum Desf. Sand on Blyde River bank ........ : = 6434

OPHIOGLOSSACEAE

Ophioglossum reticulatum L. Pine plantation on Blyde River road ..... ..... ...... 4515, 4302

MARATTIACEAE

Marattia fraxinea Sm. ex Gmel. var. salicifolia (Schrad.) C. Chr. Dense shade 5577, 4719

OSMUNDACEAE

Osmunda regalis L. In water and on bank of river . ooh ete aes 5137

Todea barbara (L.) Moore. Dense shade along streams oo ee ceo 4924

SCHIZAEACEAE

Anemia dregeana Kunze. Terrestrial, in dry bush . Sith Sass satel iste fines Bi 4513, 5049

Mohria caffrorum (L.) Desv. Grassveld on mountain .... oR eo aks 5129, 4447

Schizaea pectinata (L.) Sw. Summit of mountain, between rocks oe ae senna 2478

GLEICHENIACEAE

Gleichenia polypodioides (L.) Sm. Shade of boulders 2.000 ee ee 4490, 4538

Dicranopteris linearis (Burm.) Underw. Moist open areas 0 ee 4649, 5088

HYMENOPHYLLACEAE

Hymenophyllum capense Schrad. Very moist places 00 00 ce ck ee cee cee 6264, 6314

Hi ‘capillare Desve \ On) overhanging ‘cliffs, 2 Bc ae, Go eee) ee ee ee es 4872, 4301

H. polyanthos Sw. var kuhnii (C. Chr.) Schelpe. Shady places, on stems of trees 4300

H. tunbridgense (L.) Sm. Epiphyte, in dense shade 0... cee 4624, 5825

Trichomanes pyxidiferum L. var. melanotrichum (Schlechtd.) Schelpe. Epiphyte on

stems of forest trees ..... ...... . SA Ata oes eters Pe baat) te attest eh nee hg Same 4722, 4216

T. rigidum Sw. Edge of stream ..... .... Sy Se ee Ree nes RARE oe eer Mee trier err 1604, 1606

CYATHEACEAE

CHEE CAG? VWI, WOE CN EES om comm ce ao Gers Ge Ga oom Gp ai oo como 5900

C. capensis (L.f.) Smith. Moist deep shade, southern kloofs 0 0. un 4306, 4720

482

DENNSTAEDTIACEAE

Hypolepis sparsisora (Schrad.) Kuhn. Moist sunny places in forest .... ..... 4718, 5063

Pteridium aquilinum (L.) Kuhn. Widespread weed on forest margins and moist

BTR EAS a eet eae lee eer ay bce ee fol sateng nal ge age te Bs Gib), Sey) ceed eae oes 5487

Blotiella glabra (Bory) Tryon. Moist open BEDS HM SOE oie eam wea cee vom 5574, 4717

Histiopteris incisa (Thunb.) J.Sm. Terrestrial on damp soil 0 00 0. —

VITTARIACEAE

Vittaria isoetifolia Bory. Epiphyte in mountain forest 2.0 200 ou. ce ee ee 4939, 6024

ADIANTACEAE

Pityrogramma argentea (Willd.) Domin. Between rocks on damp soil ..... ..... 5595, 4489

Adiantum aethiopicum L. Shaded areas near water 0 ie cee ee ie ee ce ne —

A.capillus-veneris L. Shaded areas near water .... ... .... Se oe SRAM Ait a 5486

Aly ROUGE NKMsbr, OFS BWGAG WD (ORBIT oes ccexe ooo com coco ees arm emt comm comin comm doa 4981

Pteris buchananii Bak. ex Sim. Klaserie Drift on edge of forest .... ...... BT Lod! 4936

P. catoptera Kunze. Moist places in forest margin 2.0 ce ee ee 4467, 4996

P. dentata Forsk. ssp. flabellata (Thunb.) Runem. Kloof on western bank of

Kalasenie; VRAvie rye ae ae er Ug eee ashe eee Pea yiee wie ida aucune et 5488

P. vittata L. Kloof on western bank of Klaserie River 00 0 0 ce we wee vn ue 5484

Cheilanthes eckloniana Kunze. Open grassy areas, NE facing .. 0. W.. .... 4960, 4973

C. hirta Sw. North-eastern slopes in grassveld 00 0. i. cee ce ee ce 5560

C. multifida Sw. Shade of boulders 0.0 0. ck sce cies cece vets vente nets une 5621, 4623

Doryopteris concolor (Langsd. & Fisch.) Kuhn var. kirkii (Hook.) Fries. Dry banks

Of COM ay ee ics eR ns: Uitte eee 2th, ea aaa ayia enlace peace es ei hoiafienn Neeah Sena 5048

Pellaea calomelanos (Sw.) Link. Between rocks, relatively dry areas ... ae 5152, 4932

P. goudotii (Kunze) C. Chr. Rocky outcrops o.oo e ck e ee cee ane ek. 6053, 5135

P. quadripinnata (Forsk.) Prantl. Grass patches 0.0 0 ce ce ce ee cee cee 5108

P. viridis (Forsk.) Prantl. Grassveld on north-eastern facing slopes ..... ..... 5094, 4934

P. viridis (Forsk.) Prantl var. glauca Sim. Open areas, between rocks ...... ..... 6455, 5514

POLYPODIACEAE

Pyrrosia schimperiana (Mett.) Alston. Rock crevices on cliffs 0.0. ee 5505

Loxogramme lanceolata (Sw.) Pres]. Epiphyte 0.0 2.0 ce ce te es tee tee tee 5545

Polypodium poly podioides (L) Hitch. ssp. ecklonii (Kunze) Schelpe. Epiphyte 5547, 4949

Pleopeltis excavata (Bory ex Willd.) Sledge. Between rocks at high altitudes 4879, 4762

P. macrocarpa (Willd.) Kaulf. Epiphyte in mountain forest 00 2. 0. 5546, 5580

P. schraderi (Mett.) Tard. On boles of large forest trees 00 20.0 oe eet rt ne $903

DAVALLIACEAE

Oleandra distenta Kunze. Rocky outcrops 0 0. ee Be eae siete eke aia 5901

ASPLENIACEAE

Asplenium aethiopicum (Burm.) Becherer. Forest floor 00 0. 0 ee. un w. 4327A, 5159

A. anisophyllum Kunze. Epiphyte in forest ... . PA era yates "ten eee ae 5107, 5543

A. erectum Bory ex Willd. Rocks on forest floOr 0. ce eee cee cette tee tee 4700

Al, jgstorninn (CiClie, IMIOISE BIREAS 1 TORESE con axes com oem coro om mom comm ed 4789, 5160

A. gemmiferum Schrad. Dense shade in forest 00 0. ee eee A (oath bates 5575

Apsinaequaclateralemrlicron ROLES ttlOOTm ere nr rom ane ee 4724

A. rutaefolium (Berg.) Kunze. Rocky places of forest ktloony)scu\ . ee eee 4682

Arm splenderis meu Ze mae) D1 lytic ee et 4955, 5047

A. theciferum (H.B.K.) Mett. var. concinnum (Schrad.) Schelpe. ae es 1627

Ceterach cordatum (Thunb.) Desv. Rock crevices in shade ... ... ... evel ets 5339

THELYPTERIDACEAE

Thely pteris bergiana (Schlechtd.) Ching. Terrestrial in forest 00. ce ee cae —

T. dentata (Forsk.) E. St. John. Kloof on western bank of Blyde River 2.0 00. 5485

T. palustris Schott var. squamigera Seigentt) Tard. Waterlogged grassveld ‘near

Reitz’s grave Pp tne ic eeeay GLAM Lea) eT Be OT MA at) ts rik 5064

T. gueinziana (Mett.) Schelpe. Small stream on “way to Klaserie Waterfall ...._... 5046

ATHYRIACEAE

Cystopteris fragilis (L.) Bernh. In thick bush near running water... 0. ue ue 1615

Athyrium scandicinum (Willd.) Presl. Moist ground, on forest margin. ..... ..... 4706, 4937

483

LOMARIOPSIDACEAE

Elaphoglossum acrostichoides (Hook.) Schelpe. Near waterfall, rocks in forest

E. angustatum (Schrad.) Hieron. Rocks near waterfall

E. aubertii (Desv.) Moore. Epiphyte on rocks

E. macropodium (Fée) Moore. Among rocks at waterfall

E. lastii (Bak.) C.Chr. Rocky outcrops

ASPIDIACEAE

Dryopteris inaequalis (Schlechtd.) O. Kuntze. Forest floor

Rumohra adiantiformis (Forst.) Ching. Rocky outcrops in open areas in forest

BLECHNACEAE

5902,

6619,

4449,

Blechnum attenuatum (Sw.) Mett. var. giganteum pear Pace of footpath in forest

B. capense (L.) Schlechtd. Moist areas

B. Coan De Sw. Moist banks of river

B. punctulatum Sw. var. atherstonei (Papne & Rawson) Sim. Moist places

B. tabulare (Thunb.) Kuhn. Edge of river ne

GYMNOSPE RM AE

PODOCARPACEAE

Podocarpus falcatus (Thunb.) R. Br. ex Mirb. Canopy tree in montane forest

P. latifolius (Thunb.) R. Br. ex Mirb. Canopy tree in montane forest

CUPRESSACEAE

Widdringtonia cupressoides (L.) Endl. Tree or large shrub on mountain

ANGIOSPERMAE

MONOCOTYLEDONEAE

TYPHACEAE

Typhe latifolia L. ssp. capensis Rohrb. Swampy environment and streams

GRAMINEAE

Imperata cylindrica (L.) Beauv. Moist areas

Eulalia villosa (Thunb.) Nees. Eastern slopes of Magalieskop : se

Microstegium capense (Hochst.) A.Camus. Forest floor

Ischaemum arcuatum (Nees) Stapf. Blyde picnic spot

Urelytrum squarrosum Hack. Eastern slopes...

Ilemarthria altissima (Poir.) Stapf & C. E. Hubb. Localized in moist panes

Rottboellia exaltata Lf. En slopes of mountain : ee

Trachypogon spicatus (L.£.) Kuntze. Mountain grassveld

Andropogon amplectens Nees. Infrequent in grassveld

A. distachyos L. North-eastern slopes of mountain

A. eucomus Nees. Side of road in pine plantation

A. lacunosus J.G. Anders. Mountain grassveld

4790,

6015,

5841,

5842,

5889,

A. schirensis Hochst. var angustifolius Stapf. Yall grassveld on foothills of mountain

Bothriochloa insculpta (Hochst.) A. Camus. Near Blyde River picnic spot

Schizachyrium semiberbe Nees. Eastern slones of mountain

Cymbopogon excavatus (Hochst.) Stapf. Tall grassveld on foothills of mountain

C. plurinodis Stapf ex Burtt Davy. Tall grassveld on foothills of mountain

C. validus Stavf ex Burtt Davy. Foothills of mountain, tall grassveld

Hyparrhenia dichroa Stapf.. Banks of Blyde River

H. hirta (L.) Stapf. Tall grassveld on foothills of mountain

H. tamba Anderss. Bedford footpath, tall grassveld, north-eastern slopes

Monocymbium ceresiiforme (Nees) Stapf. Mountain grassveld

Heteropogon contortus (L.) Beauv. Mountain grassveld

Hyperthelia dissoluta (Nees) Clayton. Tall grassveld on foothills of mountain

Cleistachne sorghoides Benth. Lower slones of mountain in pine plantation

Themeda triandra Forsk. Mountain grassveld

Tragus berteronianus Schult. Small annual. Very localized on fallows

Perotis putens Gand. Small grass, common in disturbed areas

Paspalum commersonii Lam. Banks of Blyde River

P. dilatatum Poir. Banks of Blyde River . i

6144.

5989,

6424,

5904

4928

1618

1630

4927

5558

4779

5092

4500

5 olsy2

4929

5109

4404

4688

4477

6150

6426

5562

4571

4786

6071

5994

6410

484

P. urvillei Steud. Locally along road on lower slopes _..... pee Peet eee te ey aes =

Panicum deustum Thunb. Blyde River picnic spot 2.0 20 ee es 6102, 6464

EeckloniiiNeesamMountainyerassveldie esses in sul nen 6498, 4830

P. filiculme Hack. Roadside between forest station and Klaserie Waterfall ..... 5467

P. maximum Jacq. Pine plantations on lower slopes 0.0 00 ce cn 6148, 6406

Eoumonticolum) Hooking Wlaserics Drift simesmoistplacessse lr ee 6301

P. natalense Hochst. Scarce in mountain grassveld _.... eae Lee hatee anklet acs 6487, 6047

Rarepens# eas sandbanksyatyBlydespicnicyspO team nn nn nnn enn 6430

Alloteropsis semialata (R. Br.) Hitche. Tall grassveld on foothills of mountain _. 6152

Brachiaria brizantha (Hochst. ex A. Rich.) ee Lower slopes of mountain... _

B. serrata (Thunb.) Stapf. Along roadside 0 0.0 ee serrate Wee art deny --

Digitaria eriantha Steud. Mountain grassveld 2.0. cc, ee cess te as

os ternata (Hochst.) Stapf. In pine plantations on lower slopes Lang ce ene ee ee 4568

D. zeyheri (Nees) Henr. Lower slopes of mountain, on roadside ...._ .... 6062

Rhynchelytrum repens (Willd.) C. E. Hubb. Scattered on rocky places in mounta‘n

STASSVEl Ae aiken ee castes gees atery- ie erty ete eee De Oe Re Ac 6468, 4384

R. rhodesianum (Rendle) Stapf & C.E. Hubb. Bedford fooay on north-eastern

slope), of smOuntainies eee ea en pce Rn eet een, OE Cy (ee mE ee arora 5570

R. setifolium (Stapf) Chiov. Mountain grassveld eats ees ageness ia ee Peel td 6457, 5985

Oplismenus hirtellus (L.) Beauv. Forest floor .. ... 2. eat wy 6153, 4325A

Setaria chevalieri Stapf ex Stapf & C.E. Hubb. Banks of Blyde River... wu. 6407, ey

Seu labellatag Staple inunvlcig near Reitzispectia vCal ems nnn nnn ane Ee 598

S. lindenbergiana (Nees) Stapf. Shade of trees near Blyde picnic spot Sep ene Gee e108

S. sphacelata (Schum.) Spe & C. E. Hubb. ssp. nodosa de Wit. Bee patch

IN LORE heats: ee pe Mae tae a hae Rota ten, Gre ne ee aa 4915

Pennisetum macrourum Trin. Mountain grassveld oa Ps, ear ee nie -saiai ace a es 6

Prosphytochloa_ prehensilis (Nees) Schweick. Forest margin .... 0.0.0. 6289, 6270A

Leersia hexandra Swartz. Moist habitat on lower slopes ..... _..... See Se =

Ehrharta erecta Lam. Small grass in shade of trees 20.0 0. 6030

Aristida congesta Roem. & Schult. ssp. barbicollis (Trin. & RUPE) de Wint. Rare

pioneer grass of disturbed grassveld_.... er eaene —

Pseudobromus africanus (Hack.) Stapf. Fringe « on high vforesto On ae see 4406

Sporobolus africanus (Poir.) Robyns & Tournay. Common in disturbed areas _..... —

S. centrifugus Nees. Mountain grassveld _ ..... Pe Serra re CR ees ee 5986

S. eylesii Stent & Rattray. Mountain grassveld ee ee Efe boa Co Pepe eM e 5060

S. pyramidalis Beauv. Blyde River Canyon .... he 6118, 5864

Agrostis barbuligera Stapf var. longipilosa Goossens & Papendorf. Mountain grass-

Viel Pee gees ieee ae ee yer eee gD ele tons) ee BA Sees Bee Det 4776

A. lachnantha Nees. Mountain grassveld Ee A SLE, ERA Reed 4868, 6270

Tristachya hispida (L.f.) K. Schum. Foothills of mountain, tall grassveld, south-

Caster. SlOpess 22) 2, Pca ees Pee Lee eas 5874

Trichopteryx dregeana Nees. Mountain grassveld near Reitz’s grave aed Manne 6029, 4471

Loudetia flavida (Stapf) C.E. Hubb. Mountain grassveld 2. 0 0. eee 6486

L. simplex (Nees) C.E.Hubb. Mountain grassveld 0. 0 2. tee 5981, 5062

Danthonia drakensbergensis Schweick. Common grass on mountain summit .... 5832, 4844

D. macowanii Stapf. Between boulders on summit of mountain o.oo ee 6003

Pentaschistis natalensis Stapf. Mountain grassveld .... be a 5077

Danthoniopsis pruinosa C.E. Hubb. Rock crevices near Blyde River picnic spot. 6092, 6392

Styppeiochloa gynoglossa (Goossens) de Wint. Small grass on plateau... ... 6344

Rendlia altera (Rendle) Chiov. Mountain sourveld 2... 0 0.00 cn seh aie ties = 6329

Cynodon dactylon (L.) Pers. Pioneer on the bare disturbed areas ..._..... PS ee —

Har pechloa falx (L.) Kuntze. Mountain sour grassveld Be Ss 5989A, 5990

Chloris virgata Swartz. Disturbed areas and sandbanks of dry riverbeds .. .... —

Eustachys paspaloides (Vahl) Lanza & Mattei. Locally abundant in Combretum

veldiiniowveldie sae rs Rae Namen: Ul ely Ek, te. 5 OE LARA) tka a, tee Rees =

Eleusine africana Kennedy-O’Byrne. Widespread ruderal and weed of cultivation

andudisturbances mln nnn nee Ea iar —

Phragmites australis (Cav.) Trin. ex Steud. - Along ‘Klaserie River rg ee ee —

Pogonarthria squarrosa (Licht.) Pilg. Tufted grass on disturbed areas .... rae

Eragrostis acraea de Wint. High mountain sour grassveld 2.0. un. Bis tee 6343

E. atrovirens (Desf.) Trin. Blyde River canyon, small grass on riverbanks .... x. 6117

E. caesia Stapf. Mountain grassveld, tufted grass on mountain summit ...._..... ae 6004

E. capensis (Thunb.) Trin. Short grass, mountain sour grassveld 6469

E. curvula (Schrad.) Nees. Variable grass, in sand in dry river bed, Buje River

Canyon .. eS ; 6090

E. racemosa (Thunb.) Steud. Short grass mountain sour grassveld .. a Fe 6438

E. ‘sclerantha Nees. Edge of forest (7 "0. 0 ca at ae ee rise 4470

E. superba Peyr. Rare in grassveld nen ate Be dear ss

485

Koeleria cristata (L.) Pers. Small tufted grass on rocky outcrops in grassveld

Stiburus alopecuroides (Hack.) Stapf. Mountain grassveld .... a RE WE ton 5468,

Brachy podium flexum Nees. Along forest margin

CYPERACEAE

Lipocarpha senegalensis (Lam.) T. & H. Dur. Moist sandbanks ....

Cyperus albostriatus Schrad. Forest floor a Le

C. distans L.£. Lower slopes of mountain ....

C. fastigiatus Rottb. Lower slopes of mountain ‘on riverbank

C. longus L. At roadside on mountain plateau .... ....

C. obtusiflorus Vahl. Common in grassveld on lower eastern ‘slopes

C. rupestris Kunth. Grassveld on mountain plateau

C. textilis Thunb. Riverbank at Blyde picnic spot

Pycreus polystachyus Beauv. Riverbank at Blyde picnic “spot | ee oe

Mariscus congestus C.B. Cl. Mountain grassveld, moist habitat .... won CUPID,

M. sp. cf. M. sieberianus Nees. Summit of mountain ar

Kyllinga alba Nees var. alata C.B. Cl. Grassveld 2 Pe) ee pee: ie

K. cylindrica Nees. Mountain grassveld .. ... ... ..... ap oe ee ah OS I,

KenSp ae Mountaingerassveld Qe ye SEE ee

Ficinia filiformis Schrad. Mountain plateau, between rocks .. ae 5607,

F. gracilis Schrad. Moist places on mountain plateau . : a. 4842,

F. stolonifera Boeck. Between rocks on mountain plateau =

Fuirena chlorocarpha Ridl. Riverbank at Blyde picnic spot

Scirpus falsus C.B. Cl. Between rocks on mountain plateau a wee Fim

S. fluitans L. Hydrophyte in pool in mountain sour grassveld _..... 4774,

S. inclinatus (Del.) Aschers. & Schweinf. Moist riverbank at Dee picnic “spot ae

S. wager OCs Miowminin gracseGl ., con nen aon oop aa ame am am cam

S. 9a Goatly iim mus WALES

Fimbristylis hispidula (Vahl) Kunth. In waterlogged vlei in mountain forest

Tetraria cuspidata C.B. Cl. Mountain grassveld Peo ee em

Scleria natalensis C.B. Cl. In vlei near picnic spot ... eee

Schoenoxiphium schweickerdtii Merxm. Between large boulders on summit of

mountains .... eputy liam ere oe : Renae. Wee

S. sp. On banks of Klaserie River at waterfall nage, bene: Sead eee ae :

Kobresia lancea (Thunb.) Kiik. Mountain sour grassveld 20 200 a

Carex spicato-paniculata C.B. Cl. Herb on forest floor ... ... 4345,

C. sp. cf. C. zuluensis C.B. Cl. Sedge common on summit of mountain

PALMAE

Phoenix reclinata Jacg. Widely distributed on riverbanks. Blyde picnic spot

ARACEAE

Zantedeschia tropicalis (N.E. Br.) CRY: Scattered in eee scrub forest and

Plantations meee ae wise eee) Clb

Stylochiton natalensis Schott. “Common in plantations Eft: a for enn me ie

S. sp. In forest, disturbed area... soe We ee eS Per Ah, Ae See ee

RESTIONACEAE

Restio sieberi Kunth var. schoenoides Pillans. Grassveld on summit of mountain 5610,

XYRIDACEAE

Xyris umbilonis A. Nilss. Klaserie River on road to waterfall

COMMELINACEAE

Commelina africana L. Herb of grassveld and earlier seral and subseral stages 4356,

C. benghalensis L. Herb of grassveld and earlier seral and subseral stages ... 4582.

C. ekloniana Kunth. Mountain forest . : aa petra

Aneilema aequinoctiale Kunth. Trailing to scrambling herbaceous plant of under-

growth, in plantation _.... ww 4583,

Cyanotis lapidosus Phill. Between rocks and stones on “slopes . SAS as

C. speciosa (L.f.) Hank. Infrequently on undisturbed areas

JUNCACEAE

Juncus lomatophyllus Spreng. Riverbed near Klaserie Waterfall : 6266;

J. punctorius L.f. Riverbank, Blyde picnic pore

J. sp. In vlei on way to waterfall

486

LILIACEAE

Gloriosa superba L. Widespread in well-protected areas of scrub forest ... ... ... —

Littonia modesta Hook. Small herbaceous scrambler in scrub and scrub forest 4552

Bulbine trichophylla Bak. Herb with yellow flowers in mountain grassveld ... 5849, 6335

Anthericum angulicaule Bak. Herb on summit of mountain RAM Mia.” 2.5, 5838

A. fasciculatum Bak. Between rocks on mountain slope ... 2... 0. 0. nt ee 6334

Chlorophytum bokweri Bak. Scattered to locally gregarious... ...... ath Wee 5920

C. comosum (Thunb.) Jacques. Geophytic herb on summit of mountain ie 4906

Eriospermum cooperi Bak. Widespread geophytic herb, often in disturbed areas... 6493

Kniphofia triangularis Kunth ssp. obtusiloba (Diels) Codd. Summit of moun-

1260 nner ern. (Re Jhon ea gine mle ee Sel 4518, 6524

K. linearifolia Bak. In pine plantation and mountain. grassveld. . S111, 4618

Aloe affinis Berger. Eastern slope of mountain between stones . oe 5559, 6215

A. arborescens Mill. Widespread shrubby species on rocky areas tae i —

A. marlothii Berger. Widespread in warmer Lowveld aves .. 6261A

A. nubigena Groenewald. Mountain summit, common in roc k crevices . 4890, 4930

A. sessiliflora Pole Evans. Rocky hills and summit wi cei ene Sie

A. sp. Disturbed area, northern slopes of mountain grassveld Et peat ee 6257

Agapanthus inapertus Beauy. Perennial herb on summit of mountain om 4520

Albuca_ fastigiata (L.f.) Dryand. Rocky areas on mountain summit . 4519

A. melleri Bak. Summit of mountain rae race Pa ct ot ee —

Drimia sp. Between boulders on summit of TOLUCA ae mee ud on ae 5829

Dipcadi viride (L.) Moench. Open grassveld near forest station... eae 4944

Scilla cooperi Hook.f. Rocky situations in open spaces in mountain forest... 6363

S. glaucescens F.Z.v.d. Merwe. Lower western slope of mountain 0. 0... 3681

S. natalensis Planch. On cliffs on eastern highland Oe es 6172

S. sp. cf. S. cooperi Hook.f. Summit of mountain ... ... Pe ae 6174

S. sp. Near mountain summit between rocks .. 6332

Eucomis pole -evansii N.E. Br. Edge of forest in peer! grassveld _ on northern

slopes... He a Pees SA en : 4689

E. sp. Grassveld on northern slopes of mountain me Nel Cea 6254

Ornithogalum inandense Bak. Marshy grassveld _.... ee ea ss ate 6244

OMS etifolivrnmieunth ay ROCKMCTCVilCCS mtn nn omen ene me Sere 6331

Dracaena hookeriana K. Koch. In kloof below forest station ..._ ... boise —

Sansevieria sp. Locally common in Lowveld . Mes : —

Asparagus asparagoides (L.) Wight. Woody liane on forest margin = ers : 4784

A. crispus Lam. Between rocks on mountain summit _.... ft Meee wee seg es nel 4854

A. falcatus L. In scrubby and gallery forest... .... be. ules Skates Sa, Genk Me

A. scandens Thunb. In ravines... fy ee ee 6333

A. setaceus (Kunth) Jessop. Undergrowth ‘of scrubby forest seer a a 4698, 6296

A. subulatus Thunb. Between rocks on mountain summit 2.0. ce es 4622

Behnia reticulata (Thunb.) Didr. In forest, scrub forest and gallery forest 4548, 4276A

Smilax kraussiana Meisn. In scrub forest, gallery forest and savanna woodland 4507

AMARYLLIDACEAE

Haemanthus magnificus Herb. Moist shady places, widespread mn 6256

Clivia caulescens Dyer. Epiphyte or on forest floor... _.... SI Asta), Veg 4862

Brunsvigia sp. cf. B. natalensis. Plateau in grassveld 00 ee EN, sae eee 6355

Crinum macowanii Bak. Mountain slope at forest station SAG Mey 5071, 5855

Cyrtanthus huttonii Bak. Eastern slope of mountain 2.0. = 4564

Hypoxis angustifolia Lam. Small grasslike plant, scattered through seral “grass-

veld Ee ER eel Re RAT cote gee tnd een cama Pexe as 4535, 6321

H. argentea Harv. ex Bak. Scattered in disturbed areas .... . Ss rons Se 4679

H. membranacea Bak. Moist places. ae ; any ae 4818

H. rigidula Bak. Herb of seral grassveld .... Roles ee ee rind eager 5891, 6379

H. sp. Near Klaserie Waterfall on side of road... .... Beene Mees Vie 5908

VELLOZIACEAE

Vellozia talbotii Balf. Rocks at waterfall ........ CON Seal Migece he 4363, 4505

DIOSCOREACEAE

Dioscorea cotinifolia Kunth. Forest margin on lower slopes of mountain _.. 4748, 5932

D. dregeana Bak. var. hutchinsonii Burkill. Drier slopes of mountain and low

D. sylvatica Eckl. Scrub forest and forest margins em ome om Geo, AS

D. retusa Mast. Scrub forest and forest margins... OTe ae a ca es 4502, 4897

487

IRIDACEAE

Moraea spathulata (L-f.) Klatt. Grassveld on mountain slopes... . . 4553.

Dietes vegeta (L.) N.E. Br. Undergrowth on side of roads and forest margins 4878,

Aristea ecklonii Bak. Moist open places, undergrowth in forest ... ........ 4701

Als Sos Cis Zl, Goldtonin Ws lsOMest WYN fe eek) es es me 5847

A. woodii N.E. Br. Mountain grassveld 2.00. eee 4972

Hesperantha baurti Bak. Moist places on mountain summit

Dierama robustum N.E. Br. Grassveld on summit of mountain _...

Crocosmia aurea Planch. Geophytic herb of lightly shaded moist places

C. pottsii (Bak.) N.E. Br. Herb on riverbank near waterfall 20 00

Gladiolus papilio Hook.f. Geophytic herb of marshy places .... . 4508,

G. psittacinus Hook., sens. lat. Infrequent to localized geophytic herb of “grassveld

Lapeirousia grandiflora Bak. Small perennial herb of grassveld and scrubby vegeta-

tion .... cnt hes es rae Ne .

L. laxa (Thunb.) N.E. Br. Herb on fringe of forest

Watsonia densiflora Bak. Grassveld, on mountain summit

ZINGIBERACEAE

Kaempferia aethiopica (Schweinf.) Solms-Laub. Widespread in shade of large trees

on lower slopes of mountains

ORCHIDACEAE

Stenoglottis fimbriata Lindl. On rocky outcrops on south-western slopes

Habenaria sp. Ground orchid in sheltered areas on lower slopes ae ae

Satyrium cristatum Sond. Ground orchid between rocks a Ao ate 6323.

Disa saxicola Schltr. In humic pocket on summit of mountain) 0 4824

Polystachya ottoniana Reichb.f. Epiphyte on forest trees ee en ASS

P. transvaalensis Schltr. Epiphyte on forest trees... a. 6364,

Ansellia gigantea Reichb.f. Large tufted epiphyte on Lowveld trees in 1 Blyde Canyon

Eulophia angolensis (Reichb.f.) Summerh. Ground orchid in pine plantations in

LOW Crag ARCAS Mime a erie city cr pemenm eet ie Ween aoe Ga) ahaa pa

E. ensata Lindl. Ground orchid in grassveld o on lower slopes

E. petersii Reichb.f. In tall mountain-gassveld ee

E. speciosa (R. Br. ex Lindl.) H. Bol. Ground orchid in pine plantation

Bulbophyllum sandersonii Reichb.f. Epiphyte on forest trees

Angraecum conchiferum Lindl. Epiphyte on forest trees

A. sacciferum Lindl. Epiphyte in indigenous forest _..... ;

Tridactyle tricuspis (H. Bol.) Schltr. Epiphyte on large (ROO es es ec a

Cyrtorchis praetermissa Summerh. Epiphyte in forest trees 0 0

Mystacidium capense (L.f.) Schltr. Epiphyte on stems of trees 00... 4387,

DIcoTYLEDONEAE

PIPERACEAE

Piper capense Lf. Soft shrub of moist shady kloofs in high forest

Peperomia arabica Decne. In shade of large trees on mountain slopes . on 5483,

P. retusa (L.f.) A. Dietr. Herb on forest floor _.....

SALICACEAE

Salix woodii Seemen. Small tree in streams on lower mountain slopes

MyYRICACEAE

Myrica pilulifera Rendle. Small tree on forest margin in open areas 5566,

M. serrata Lam. Small tree on riverbank in Blyde River Canyon

ULMACEAE

Celtis africana Burm.f. Tall tree, infrequently scattered in scrub forest, high forest

and forest margins .. Sy Ne PE Gree eras De ge deere Mergers eer sede ee

Trema orientalis (L.) Blume. Small to medium-sized tree in ~ gallery ‘forest on

lower slopes _..... PER ease Netee, crae ea ee, eee ai Mer) EAD ONT

MoRACEAE

Ficus capensis Thunb. Gallery forest in Lowveld ...._.....

F. craterostoma Warb. ex Mildbr. & Burret. Strangler in high forest i ae, CIES,

F. ingens (Miq.) Mig. Small to fair-sized tree on Go) outcrops, on lower slopes

F. petersii Warb. Strangler, fairly common . es me a 4549,

F. sonderi Mig. Tree, common on hills ... Shbsserics ae

F. sycomorus L._ Gallery forest, GOATS eee ey Ge ce Meme Nee hea ees cee

Camas samiua il, AE IIEGSTOC WVACSEN oe ceo oe es ec

488

URTICACEAE

Urera tenax N.E. Br. Rocky places on north-eastern slopes _.....

Laportea alatipes Hook.f. Common in open areas on forest floor ... ... ....

L. peduncularis (Wedd.) Chew. Rocky areas in kloofs on lower slopes aie

Pouzolzia hypoleuca Wedd. Common on rocky places on lower slopes

PROTEACEAE

Faurea galpinii Phill. On Bedford footpath on northern slopes of mountain

F. macnaughtonii Phill. Large tree in rain forest near waterfall

6248,

F. saligna Harv. Small to medium sized tree. In scrub and climax savanna wood-

F. speciosa Welw. Small tree on edgewotieionestiige ite mar oan te meee erro en

Protea gaguedi J. F. Gmel. Small tree in grass patches on maintain ..... .....

P. parvula Beard. Mountain grassveld, near waterfall

4795,

P. rhodantha Hook.f. var. falcata Beard. Small tree in mountain grassveld near water

P. rhodantha Hook.f. var. rhodantha. Small tree, in mountain grassveld

P. rouppelliae Meisn. Tree on mountain cliff ....... Laeeai ton Se ienny coe

LORANTHACEAE

Loranthus dregei Eckl. & Zeyh. In Blyde River Lowveld 2.0 0 0 ke

L. minor (Harv.) Sprague. Parasite on bushveld trees ..... _..... ee, OUAaig sae ans

L. zeyheri Harv. Parasite on western slopes of mountain

Viscum nervosum Hochst. ex A. Rich. Parasite on trees on fringe of forest

V. obscurum Thunb. Parasite on forest trees

OLACACEAE

Ximenia caffra Sond. var. natalensis Sond. Shrub in pine plantation ....._.....

POLYGONACEAE

Rumex crispus L. Near nursery on road to picnic spot 2. oc

R. sagittatus Thunb. On side of road to Blyde picnic spot 2.0.

Polygonum pulchrum Blume. Water-logged area on road to Blyde picnic ope ihe

P. salicifolilum Brouss. Near Klaserie Waterfall in marsh 00 200 0 oe

AMARANTHACEAE

Cyathula cylindrica Moq. Herb in grassveld on western slopes _..... Ain, Boke

Pupalia lappacea (L.) Juss. In shade of scrub on western slopes _.....

Achyranthes sicula (L.) All. On Blyde River bank in scrub

NYCTAGINACEAE

Commicarpus pentandrus (Burch.) Heim. Herb on banks of Blyde River

PHYTOLACCACEAE

Psammotropha myriantha Sond. In grassveld on summit of mountain ...._....

Phytolacca octandra L. Exotic small shrub in disturbed areas on lower slopes

CARYOPHYLLACEAE

Polycarpon tetraphyllum L.f. Weed in vicinity of forest station

Silene burchellii Ott. On side of road on summit of mountain _....

RANUNCULACEAE

AOA ORNO Sop WN FAASSVONG! ooo com umn com comm) cam ao omen coo

Clematis brachiata Thunb. Widespread liane in seral and climax communities

Thalictrum rhynchocarpum Dill. & Rich. Wet shady cool sites in high forest

MENISPERMACEAE

Cocculus hirsutus (L.) Diels. Kloofs in high forest...

Stephania abyssinica (Dill & Rich.) Walp. var. tomentella (Oliv.) Diels. Early

scrubby seral stages and high forest margins

Cissampelos torulosa E. Mey. ex Harv. Twiner on forest margins

ANNONACEAE

Annona senegalensis Pers. Small tree in scrub and young pine plantations

MONIMIACEAE

Xymalos monospora (Harv.) Baill. Tall tree in mountain forest

4534,

4533,

4832,

4586,

4671.

6261,

4497.

4635

4672

6401

4631

489

LAURACEAE

Ocotea viridis Kosterm. Very scarce in high forest _..... OP ey ee bene —_

Cryptocarya liebertiana Engl. High forest tree of canopy and margin .. .. ... 6107A

GevoodisEne aa Malletreeminetonest nts en ees Diese 6107

Cassytha filiformis L. Parasite on shrubs iMPniveLOed suerte any wie Ge i hae ee 5113

CRUCIFERAE

Heliophila rigidiuscula Sond. Herb between grass 0.0. ce cette

Cardamine africana L. Herb on bare patches and rocky areas on forest floor 6267, 4716

CAPPARIDACEAE

Capparis tomentosa Lam. Shrub, common on western lower slopes. ..... toe 4744

Maerua cafra (DC.) Pax. Pine plantations on lower slopes 00. ee ee tee tees 5871

M. rosmarinoides (Sond.) Gilg & Ben. Blyde River picnic spot... ........ 6099, 6259

DROSERACEAE

Drosera dielsiana Exell & Laundon. Wet areas in grassveld 2.0 0. un 4544

CRASSULACEAE

Kalanchoé rotundifolia Haw. Widespread succulent herb, in scrub and on banks

OfM CONG ast ener ents mee, Set et eee te teen Geo, bene cod ree aes 5154

Crassula browniana Burtt Davy. Between rocks on summit of mountain .... 6500, 4393

C. filamentosa Schonl. Small pulpy herb on summit of mountain 2.0 0. uu... 6458

C. lineolata Dryand. Prostrate herb on Blyde footpath 0 00.0 wu... 5526

C. parvisepala Schénl. Succulent shrub on summit of mountain .. 0... 5085, 5072

C. rubicunda E. Mey. ex Harv. Moist areas in rocky situations ... ..._ .... 4371, 4847

C. setulosa Harv. Herb in small annual pools on summit 200 2 tes 4625

C. thorncroftii Burtt Davy. Herb on forest floor 20.0 20 4995

SAXIFRAGACEAE

Choristylis rhamnoides Hary. Shrub on northern slopes of mountain _....._..... 4999, 5571

PITTOSPORACEAE

Pittosporum viridiflorum Sims. Rocky outcrops and forest margins .... . 4690, 5974

HAMAMELIDACEAE

Trichocladus grandiflorus Oliv. Large forest tree _.... es es eee I RS)

ROSACEAE

Rubus pinnatus Willd. Widespread scrambler on edge of forest 2.00 20 ee 5014

Reprigiduss sms SE Tingel Of SfOrest)) fe ee mie cep ee 4562, 6249

Alchemilla cryptantha Steud. Herb on sideloferoad memset 4694

A. rehmannii Engl. Herb on forest floor 0... ... 4992, 6442

Agrimonia odorata Mill. Small shrub in scrub on forest margin. ae 4605, 6369

Cliffortia linearifolia Eckl. & Zeyh. Shrub on open area in indigenous forest 4781, 5090

C. nitidula (Engl.) R.E. Fr. & T. C. E. Fr. ssp. pilosa H. Weim. tis Weidespieas

shrub along streams and on mountain summit ee 5588

C. repens Schltr. Small shrub on mountain summit and slopes 2a ce ee ee 4475

C. serpyllifolia Cham. & Schlechtd. Small shrub common between rocks on mountain 4828

GRstrobilifera Murry Oni mountainy slopes 22) ea) ee eee ce ee eee a 6132B

Prunus africana (Hook.f.) Kalkm. Tall tree scattered in high forests es 4978, 5146

Parinari curatellifolia Planch. ex Benth. ssp. mobola (Oliv.) R. Grah. Ee

tree, tall grassveld, on lower slopes of mountain 2.00. ce —

CONNARACEAE

Cnestis natalensis (Hochst.) Planch. & Sond. Liane in mountain forest ... 4733, 5058

LEGUMINOSAE

Mimosoideae

Albizia versicolor Welw. ex Oliv. Small to medium-sized tree in Lowveld .. ;

Acacia ataxacantha DC. In scrub and scrub forests in kloofs and lower areas 4556, 5005

caffra (Thunb.) Willd. Small trees, in lower areas ek eS elle the aie NS --

. davyi N.E. Br. In Blyde River poort ; ae fe eee: 5016, 6068

. gerrardii Benth. Lowveld tree _.... —

. karroo Hayne. Small to medium-sized tree. Abundant in seral scrub and savanna

of lower foothills y ee Ue os ee ee ae es ae an ee ee —_—

490

A. robusta Burch. Common tree on banks of Blyde River 0.0. eee

A. schweinfurthti Brenan & Exell var. schweinfurthii. Untidy woody. “scrambler.

common in thick bush of lower foothills 2... 0. ce cue cane tee cee ane

A. sieberana DC. var. woodii (Burrt Davy) Keay & Brenan. Medium-sized tree on

foothills of mountain Sep CS mee a CE EIA pe nue beh en Biineiac® loans SEB, had

Dichrostachys cinerea (L.) Wight & Arn. ssp. nyassana (Taub.) Brenan. Infrequent

in seral grassveld and scrub on loWeraSlOPES gee eee

Entada spicata (E. Mey.) Druce. Robust liane of scrub forest and high forest 4732,

Caesalpinioideae

Schotia brachypetala Sond. Gallery forest (2. 0. ee

Bauhinia galpinii N.E. Br. In scrub forest of Lowveld 0.0. ee tes

Paden thonningii (Schum.) Milne-Redhead. Shrub or small tree, locally on foot-

TSH ee ee ee ee

Cassia laevigata Willd. Banks of Blyde Riverat ieee ASE oe eeu pee 4601,

C. mimosoides L. Along footpath on western mountain slopes 00. ees

C. occidentalis L. Small shrub in plantation and disturbed areas .... .... 5856,

C. petersiana Bolle. Small shrub in tall grassveld of lower foothills 20. 0 nn

Pterolobium exosum (Gmel.) Bak.f. Robust, thorny scrambler in foothills scrub _ .....

Caesalpinia decanctala Roth) 4 Alston. eee spiny exotic scrambler in foothill

scrub al | POR Mace ects Uses RAST ots eee The ee eee

Papilionatae

Calpurnia aurea (Ait.) Benth. Small tree, fairly widespread in high forest

C. Sod. Sioalll wee iin MOWAT HOMESE nen cee on ems em ese oncom

Pleiospora cajanifolia Harv. Small shrub with yellow flowers me ee ee ee

Crotalaria capensis Jacq. Small shrub in pine plantation and forest. ‘margin 4555,

C. doidgeae Verdoorn. Small shrub on lower slopes 20. 0. eee 5473,

CerectaySteudysSmalllitshrub Mon tones temas inigge ieee me une nee

Argyrolobium collinum Eckl. & Zeyh. Small herb on summit of mountain ... ....

A. tomentosum (Andr.) Druce. Summit of mountain een .. 4476

Indigofera garckeana Vatke. Forest margins 2.0. eect eee tees en ante

I. hedyantha Eckl. & Zeyh. Herb on mountain summit ... 0... am OIG,

I. swaziensis Bol. North-eastern slopes, small shrub, with red flowers... ate ae

I. tristoides N.E. Br. Small shrub, on north-eartern slopes in grassveld ........ .....

I. sp. cf. I. malacostachys Benth. Open space in forest _..... RAT et canes ee on

Psoralea pinnata L. Shrub with mauve flowers on moist areas, on summit of mous

1420 5c la POO cate | PIR Al Seema ORC On| ey aaa tera Murer Wess | ale ates 4379,

Mundulea_ sericea ‘(Willd.) ‘A. Chev. Common in tall grassveld of foothills and

Blydex (Camo: see ae arcsec ree eet) ern reer metre ere

Sesbania cinerascens Welw. ex Bak. Small shrub on slopes of mountain

Ormocarpum trichocarpum (Taub.) Harms ex Burtt Davy. Small shrub on water-

logged areas on slopes of mountain 2.0 0 es pay thes

Smithia erubescens (E. Mey.) Bak.f. Small shrub. in grassy patches enue 6243,

§. thymodora Bak.f. Small shrub in mountain grassveld 2.0 00 ee ee «= 4547,

Zornia capensis Pers. Small herb in mountain grassveld esa tas Mee

Desmodium hirtum Guill. & Perr. Herb in mountain grassveld es com (AST,

D. repandum (Vahl) DC. In scrub and grassveld of forest margins ..... 4941,

Pseudarthria hookeri Wight & Arn. Shrubby forb in plantations

Dalbergia armata E. Mey. Robust liane of gallery and riverine scrub, scrub fores:

Aid LORESt) eae oe eee! rea eee aU ee econ gece me ones wo Gat,

Pterocarpus angolensis DC. Tall tree on foothills of Drakensberg

P. rotundifolius (Sond.) Druce. Blyde River Canyon _....

Lonchocarpus capassa Rolfe. Stunted tree in Blyde River Canyon ae

Abrus fruticulosus Wall. ex Wight & Arn. Slender subwoody twiner. Widespread

in scrub, savanna and woodland eee ope tata bates

A. precatorius L. Small twiner in Lowveld scrub

Erythrina lysistemon Hutch. Tree in scrub and gallery forest of Lowveld

Mucuna coriacea Baker. Robust soft perennial twiner in pine plantations

Canavalia virosa (Roxb.) Wight & Arn. Herbaceous twiner near streams, low country

Rixmnchosia Carlneaa IDXC, IFORES! WHER AIMNS oc oon oa ome ame ame ame 4592,

R. clivorum S. Moore. Small shrub with yellow flowers. in pine plantations oc

R. hirsuta Eckl. & Zeyh. Blyde picnic spot .

R. sordida (E. Mey.) Schinz. Slender fruticose forb—

Flemingia grahamiana Wight & Arn. In grassveld on mountain summit...

Vigna vexillata (L.) Benth. Small soft twiner. Widespread, but restricted to gra»

veld and early scrub stages 0.0 ue Fees ee see eee

49]

V. sp. Herb in grassveld _..... Se) (A Me PR Ec Se ae ry 5516

Lablab niger Medik. Herb on forest margin eR anes Be cane ete eee 4595

Dolichos taubertii Bak.f. Occurs frequently in grassveld OE A Mg AIG AEM te : —

OXALIDACEAE

Oxalis corniculata L. On side of road to Blyde picnic spot .... . See test ete 6378

O. obliquifolia Steud. ex Rich. In grassveld on NW mountain slopes ae a eee 5003

O. smithiana Eckl. & Zeyh. Small herb, common in grassveld 2.0 00. ees =

O. sp. cf. O. semiloba Sond. Small herb in grassveld 2. 00... woe ee 9004, 6494

LINACEAE

Linum thunbergii Eckl. & Zeyh. NE slopes of mountain, open bushveid _..... _..... 5000

RUTACEAE

Fagara capensis Thunb. Small tree in scrub forest 2.0 0. ue see 4746, 6396

F. davyi Verdoorn. Common tree of high forest and forest margin... 4632, 4674

Calodendrum capense (L.f.) Thunb. Fair-sized tree of high forest and forest margin 5928

Vepris reflexa Verdoorn. Bane ok SBlydem@Rivers -sopete) feet frome) Revere 6105

V. undulata (Thunb.) Verdoorn. Small tree or shrub on fringe of mountain forest 5850

Teclea natalensis Engl. Small understory tree on banks of Blyde River ..... ..... _ .... 5495

Clausena anisata (Willd.) Hook.f. ex Benth. Shrub in Blyde Canyon ....._.... 4749, 4792A

SIMARUBACEAE

Kirkia wilmsii Engl. Tree near Blyde picnic spot... _..... 611036260

BURSERACEAE

Commiphora mollis (Oliv.) Engl. Small tree in scrub of donga .... ..... se bapeeds Wy eee —

C. harveyi Engl. Small tree on banks of Blyde River 2.0 0 ee ae es 6095

MELIACEAE

Ptaeroxylon obliquum (Thunb.) Radlk. Shrub in vicinity of eae picnic aso! - 6098

Ekebergia capensis Sparrm. Tree in forest 2.0 ccc ec ctee 4650, 4714

E. pterophylla (C.DC.) Hofmeyr. Shrub on fringe of foresee ee a. 4364, 4396

Trichilia emetica Vahl. Tall tree in gallery forest on foothills of mountain... 5858

MALPIGHIACEAE

S phedamnocar pus SEOs Coss) Paes Perennial slender subwoody liane

of forest margins _.... ioe nee ape tart neo pen eee Eagar fd ke ae . 4509, 4969

POLYGALACEAE

Polygala sphenoptera Fresen. On sandbank at Blyde River picnic spot 0... ... 6431

P. rehmanni Chod. Small shrub in grassveld on mountain summit Bes eect oe 6454

P. virgata Thunb. Small shrub on forest margin 2.00 on. te FENN ten ana 4656

Muraltia flanaganii H. Bol. Small shrub on summit of mountain cue ve 6187, 4408

EUPHORBIACEAE

Securinega virosa (Roxb. ex Willd.) Pax & Hoffm. On road to Blyde picnic spot,

Shrubgeinidnyaeklo ofa Monit. sr nee aca heey ce soyeuey veto Lene tat 6129, 6474

Phyllanthus reticulatus Poir. Untidy shrub or woody twiner common in thick bush

Ot WOWEP OIEAS cio mms atm om Gen on or gem ome ae cet cm mom oom om =

Antidesma venosum E. Mey. ex Tul. Low spreading tree on foothills of mountain

TANOUMINS SNONSS SUUENNOINS: en one oe oo one oop ome oop Gms aap om amp 5051

Bridelia micrantha (Hochst.) Baill. Fair-sized ‘tree on lower banks of Klaserie

RUN C Tae ene cme es eat ea O ete eh eatiad ts Ue eo Sa ee ce mee 4799, 6419

B. mollis Hutch. Common in rocky areas in Lowveld .... ee 6389

Croton sylvaticus Hochst. Fair-sized tree, in high forest and forest margins ... 4662, 4609

Alchornea hirtella Benth. var. glabrata (Prain) Pax & Hoffm. Small tree on fringe

OMB LOTeS mre et re ek ee eS rio atc Suey canto Jeers setinl ere 6156

Acalypha punctata Meisn. “Subwoody forb in grassveld and forest margins on western BR

SIO PES Wee tee eee ee ahr le ee cee anmte hits chr Anse marten corneas 3

Tragia okanyua Pax. Slender twiner in underscrub of the Lowveld areas... ... ae 6114

Ricinus communis L. Weed on Blyde River bank 2.0 nes =

Jatropha hirsuta Hochst. Eastern foothills of mountain 2.0.0 we ee ee 5882

Clutia affinis Sond. Near Klaserie Waterfall 2.0 0.0... Aker te nee ee Pear e es 4352

C. monticola S. Moore. Western slopes of mountain .. 0 2 ee ee eae eae 5565

C. natalensis Bernh. In water-logged area near Klaserie Waterfall... eae 5021

492

Spirostachys africana Sond. Common tree in Lowveld areas ea pe es

Euphorbia cooperi N.E. Br. ex Berg. Blyde River Lowveld ...... ...... .....

E. ingens E. Mey. ex Boiss. Blyde River Lowveld .

E. lydenburgensis Schweick & Letty. Blyde River ow velde eee, eee ae aS

E. kraussiana Bernh. Small shrub in forest margin 2... 0 cc ee es

E. tirucalli L. In Blyde River Bushveld ..... a ie econ Pecctu raGy O a as) are

ANACARDIACEAE

Sclerocarya caffra Sond. Tall deciduous tree in Lowveld near Klaserie River _.....

Lannea discolor (Sond.) Engl. Small tree in rocky places in Lowveld 0 uu. uuu.

L. edulis (Sond.) ee Herb in mountain grassveld, between forest station and

waterfall Salt g Geet RAO Geo Ja ee RENO S Macti, ce a ee ane oe a ae

Protorhus longifolia -(Bernh.) Engl. Large forest’ tree’ ee 4559,

Ozoroa reticulata (Bak.f.) R. & A. Fernandes. Small tree on foothills of Drakensberg

Rhus chirindensis Bak.f. forma legatii (Schonl.) R. & A. Fernandes. Small tree on

PORES( OMMAT OU Mes. at, “eae er oe nh eee eee Wa nD Ne Paes ds Ma 2c 6167,

R. dura Schonl. Small tree or shrub on summit of mountain... .... .... 4893,

R. ernesti Schonl. Small tree among boulders on summit of TmOUn tac nae

R. gueinzii Sond. Small shrub in Blyde River scrub _..... eae pale le oe sea

R. intermedia Schon]. Small shrub in pine plantation . ee FokG eee ees

JR, Unrerelo WL, Min WOTESE WOGISTEATOWTUN cee care comms ween Gum ome omm camo anm amo omm oom “913,

R. pentheri Zahlbr. Small shrub in Blyde River Poort 2. 00. wu. woo Soil3,

R. pyroides Burch. Small tree on banks of Blyde River ou. ccc cee coc ce we cs 6111,

R. rehmanniana Engl. Small tree on lower western slopes of mountain

R. transvaalensis Engl. Small tree in Blyde River scrub .... ... . Systk Te OUl24;

R. sp. Small shrub in undergrowth and fringe of forest 0. 00. 0 cu 5171,

AQUIFOLIACEAE

Ilex mitis (L.) Radlk. Widespread waterside tree 0.0 ic ccc cee nee cee na vi ve ee 4558,

CELASTRACEAE

Maytenus ocuminata (L.f.) Loes. Tree on frange of mountain forest ..... x

M. heterophylla (Eckl. & Zeyh.) N. Robson. Variable small tree. In scrubby

Vegetation =n." yo otis nlite, Si eee geen ety aria Pes vam eee eae oe

M. mossambicensis (Klotzsch) Blakelock var. rubra (Harv.) Blakelock. Small tree

at altitudes upwards of about 1,300 m. 56

M. peduncularis (Sond.) Loes. Small to fair- sized tree of ‘undergrowth. and margin

POTEStS Fie Seek Ae ese ee) Ee DSO gee ARP pie ag me NOL eee,

M. undata (Thunb.) Blakelock. Small tree of margin and undergrowth eee

Catha edulis (Vahl) Forsk. ex Endl. Widespread small tree in Lowveld regions _

Pterocelastrus echinatus N.E. Br. Large tree common in forest 0... ... 4640A,

P. galpinii Loes. Large tree in indigenous forest .... ...._ .... eet om cron ATOILAN,

12, Fo, MOY ILAMYS (WHO sia TOGNESMOMS WORE cin com ams) om am, ans ree

Cassine eucleaeformis (Eckl. & Zeyh.) Kuntze. Small tree of rain forest .. 6223,

ICACINACEAE

Apodytes dimidiata E. Mey. ex Arn. Small to medium-sized tree of forest and

SCL) FONESt ei Wey heed tices Bae tere enon Ce Ere ey renee 6221,

Pyrenacantha grandiflora Baill. Scrambling shrub in forest... .... MP Seebeck

SAPINDACEAE

Allophylus melanocarpus (Sond.) Radlk. Tree on mountain forest margin .... ....

A. transvaalensis Burtt. Davy. Small tree on forest margin .... ... ate 5940,

Pappea capensis Eckl. & Zeyh. Small tree or shrub in forest margin Ral iia

Hippobromus pauciflorus (L.f.) Radlk. Common in scrub forest and forest margin

MELIANTHACEAE

Bersama_ transvaalensis Turrill. Fair-sized tree of forest and forest margin 5579,

B. tysoniana Oliv. Large tree in indigenous FOLESE. Pei Pat ey ee ae 4780,

J, Goh NGS 0 MAOWIENIN TOTES rus camo cues am «co ome _ cameo eae gm | am ans 6219,

Greyia radlkoferi Szyszyl. Large shrub or small tree on NE foothills 0 cc. son

G. sutherlandii Hook. & Harv. Shrub on NE foothills _.... ede) Ree eee 5553,

BALSAMINACEAE

Impatiens duthieae L. Bol. Soft herb near water... ee Se me 4502A,

I. sylvicola Burtt Davy & Greenway. Soft herb, near water on forest floor 4349,

493

RHAMNACEAE

Ziziphus mucronata Willd. Small tree, in scrubby vegetation _.... Hee Oe.

Phyllogeiton zeyheri (Sond.) Suesseng. Small tree or shrub, widespread oe tee

Scutia myrtina (Burm.f.) Kurz. Robust woody scrambler or liane... 5019,

Rhamnus prinoides L’Herit. Widespread shrub of scrub and scrub forest ..... 4964,

Phylica paniculata Willd. Small shrub on summit of mountain 4481,

Helinus integrifolius (Lam.) Kuntze. Small woody liane or scrambler. Scrubby vege-

tablonyaee nee ee es baer ee el aoe wee ce 460

HETEROPYXIDACEAE

Heteropyxis natalensis Harv. Small tree in seral scrub or savanna

VITACEAE

Rhoicissus revoillii Planch. Robust woody liane of high and scrub forest. ..... 4979,

R. rhomboidea (E. Mey. ex Harv.) Planch. Robust woody liane of high and gallery

LOLES tear Mee me nee mes eee meena eked eee Ent, Soathe Aae 4561

R. tomentosa (Lam.) Wild & Drummond. “Very robust woody liane of high. and

PallenymlOLes tmmey tee coe tee emcee Mit weet cis! Naw end eae 5 UE 46

R. tridentata (L.f.) Wild & Drummond. ~ Shrubby plant, on disturbed retarded grass-

SUGIIGLS > aan eaten ak ahaa eee Ce Gauri ures Car on Av USN WO CaM ET Ee

Cissus sp. [eianeencatirivcr men Hiern Certain te

Cyphostemma anatomicum &: A. Sm.) Wild & Drummond. Common liane of high

forest and kloof 0... et coe wens: eee sass elt Bo eee ed 2

TILIACEAE

Grewia flavescens Juss. Untidy climbing shrub with square stems, widespread

G. monticola Sond. Shrub common on rocky hills and sandy soil . 3

G. occidentalis L. Widespread straggly shrub or small tree in scrub forest... 4745,

iiriumijetia,annua ley sWeedvoniftorest: margin’ (0) 2) ye eee ee eee

T. pilosa Roth. var. tomentosa SESE ex Sprague & Hutch. Small to large

annual ..... . ee ee Oar Ee ee ied (MRM PER cere tie Meek setutiy Vaaeandle a 5480,

MALVACEAE

Abutilon angulatum (Guill. & Perr.) Mast. Small shrub on riverbank thickets _.....

A. fruticosum Guill. & Perr. Shrub in grassveld _ .... een Sr Mert re

Pavonia columella Cav. Small shrub on riverbank es > am CSS,

Hibiscus cannabinus L. Herb in pine plantations 3 me

H. meeusei Exell. Small herb in pine plantations re Beene tae ee hae

H. pedunculatus L.fé. Small shrub on riverbank ... Uae 6055,

H. praeteritus R. A. Dyer. Small shrub with red flowers in Blyde River Poort 0...

BOMBACACEAE

Adansonia digitata L. Large tree with thick stem in Blyde River Lowveld

STERCULIACEAE

Melhania prostrata DC. Herb with yellow flowers in fallows at Blyde River Poort 5954,

Dombeya burgessiae Gerr. ex Harv. Robust shrub in scrub on mountain summit .

D. cymosa Harv. Robust shrub on banks of Blyde River silted Renta gett

D. rotundifolia (Hochst.) Planch. Small Lowveld tree _.... Bo RS ae ek els MSN ral Ber

D. pulchra N.E. Br. Robust shrub in pine plantations _....

Sterculia murex Hemsl. Tree in kloof ana on rocky outcrops of foothills of

newb Choma mn nin

OCHNACEAE

Ochna holstii Engl. In scrub on banks of Blyde River aes ; 5654,

O. natalitia (Meisn.) Walp. Small shrub in scrub forest 00 ee 6237,

O. oconnorii Phillips. Fair-sized understory and canopy tree in high forest 4713,

GUTTIFERAE

Hypericum lalandii Choisy. Widespread pioneer of forest margins and moist

ATCA STM ery Pte ee cas rascn, Tassie, gate Hee We oe: 6021, $

H. revolutum Vahl. Spreading shrub along forest margin Pan viene, Weeeeal ein 4367,

H. roeperianum Schimper. Shrub with yellow flowers on side of road .... . 4975,

CANELLACEAE

Warburgia ugandensis Sprague. Tree in Lowveld kloofs _....._..... 5503,

494

FLACOURTIACEAE

Rawsonia lucida Harv. & Sond. Wnderstonystreesofshighetonesta ee 4528,

Kiggelaria africana L. Variable tree in upper montane forest .... .... ... .. . 4667,

Gerrardina foliosa Oliv. Shrub between boulders on summit of mountain

Homalium dentatum (Harv.) Warb. Infrequent in riparian forest... 0.0... 6082,

Trimeria rotundifolia (Hochst.) Gilg. Lower parts in bush .... ..... ..... ... 6065,

Aphloia theiformis (Vahl) Benn. Shrub near Klaserie picnic spot ..... ... .... 4911,

Dovyalis caffra (Hook.f. & Harv.) Warb. Spiny shrub in thick bush 2. 0. ue

D. rhamnoides (Burch.) Harv. Shrub with spines in wooded kloof 2.0 2.0 en

PASSIFLORACEAE

Adenia digitata (Harv.) Engl. Widespread herbaceous scrambler with large woody

TOOL STOCK IM cee eet PAN Dee oleae see) tee ane een oot eR L

A. gummifera (Harv.) Harms. Widespread liane on forest margins .... 0.0. 0... 4736,

BEGONIACEAE

Begonia caffra Meisn. Against moist cliffs in Inds ENOUSHTOLeSt ie ar a= re

B. sp. Shrub with orange flowers, against moist rocks in indigenous forest ..... 4365,

THYMELAEACEAE

Peddiea africana Harv. Shrub or small tree of high forest 0 0 ccc ee es 5147A,

Lasiosiphon polyanthus Gilg. Small shrub with yellow flowers in forest. margin 6481,

L. splendens Endl. Mountain grassveld shrub with yellow flowers 2... 0... ee

Passerina montana Thod. Ericoid shrub on mountain summit between rocks ..... 4478,

Dais cotinifolia L. Small tree on lower slopes of mountain 0.0 2. eee

RHIZOPHORACEAE

Cassipourea gerrardii (Schinz) Alston. Understory or canopy tree of high forest 4707,

COMBRETACEAE

Combretum apiculatum Sond. In Lowveld ..0 0. ck ce es Bae

» collie Isroxen, Saoelll Gree iim Wennyelle! enn em ben cen ence «me «camo aes

. erythrophyllum (Burch.) Sond. Large tree on lower slopes of mountain _....

. imberbe Wawra. Large tree in Blyde River Lowveld 00. nn Peabo

. kraussii Hochst. Tree in forest 0.00. ce ee ei) tae oe an 4676,

. zeyheri Sond. Small tree in Lowveld 2.0. ee cies 1h A La eae

» Ses Smell tise on Ihyels TRUE LATS ces aes ems scan

Quisqualis parviflora Gerr. Robust woody liane in forest 2. 0c. een 5585,

Terminalia phanerophlebia Engl. & Diels. Small tree in thick bush on lower slopes

OL MAO MUTA ITY, | te elena asst cea ees en cero eee ts ame Ceara ae

D@IGIDIGI®

MYRTACEAE

Eugenia natalitia Sond. Small tree in mountain forest 2.00. tes 4630,

E. paniculata Banks. ex Gaertn. var. australis Baill. Exotic shrub in disturbed areas

E. sp. nov. (Renny 266 + 181). Small shrub on banks of Blyde River ..... _.....

Syzygium cordatum Hochst. ex Harv. & Sond. Lowveld riparian tree ..... ..... 4496,

S. gerrardii (Harv. ex Hook.f.) Burtt Davy. Large tree on mountain forest ..... 4636,

S. guineense (Willd.) DC. Lowveld riparian tree _..... Need) Grates g ORa |, pene Gong MRE s 4801,

MELASTOMATACEAE

Antherotoma naudinii Hook.f. Small herb in disturbed areas) 0. 0 eee

Dissotis canescens (E. Mey. ex Grah.) Hook.f. Suffrutex of earlier stages of hydro-

SETS tag eke Re leat. artheeha tu bch eee ames esi ns gM de erg oo

HALORRHAGIDACEAE

Gunnera perpensa . Perennial rhizomatous herb in water-logged areas ..... 5065,

ARALIACEAE

Cussonia natalensis Sond. Fairly large tree, scattered on rocky outcrops between

patches im forest i cis, ce ie nee ne site) ate Gnesi | ane Basar at

C. spicata Thunb. Tree of fountain forest. “no ukee ken ee eens Re ay ee

C. umbellifera Sond. Large tree of mountain forest 0 0 ce ee ten 4399,

Seemannaralia gerrardii (Seeman) Vig. Small tree in scrub forest of western slopes

495

UMBELLIFERAE

Sanicula elata Ham. ex D. Don. Low herb on forest floor . 4501,

Alepidea amatymbica Eckl. & Zeyh. Herb in mountain grassveld -

Heteromorpha pubescens Burrt Davy. Shrub on riverbank and forest margin om Bowls

H. trifoliata (Wendl.) Eckl. & Zeyh. Small tree on forest margin . 4617,

Steganotaenia araliacea Hochst. Tree on rocky outcrops on banks of Blyde River

CORNACEAE

Curtisia dentata Cunt) C.A. Sm. Fair-sized tree of high forest and forest

OR GIN oom ceo oe zh ae ae ae A O42,

ERICACEAE

Vaccinium exul H. Bol. Small shrub in mountain grassveld we 4712,

Erica caffrorum H. Bol. Scrub on summit of mountain _.... S843.

E. drakensbergensis Guth. & H. Bol. Small shrub on side of mountain road... ...

E. leucopelta Tausch var. opie nieigess | H. Bol. Small shrub on summit of moun-

tain sn Nn, eg ee pene we ua 4395,

MYRSINACEAE

Maesa lanceolata Forsk. Widespread small tree in forest margin and mountain

OTassvel damm reeeh ee ye ed a ak ee By a ears Se ae Me 4551,

Myrsine africana L. Small shrub on forest margin me oo sIS),

Rapanea melanophloeos (L.) Mez. Medium-sized tree of subclimax forest 4454,

PLUMBAGINACEAE

Plumbago auriculata Lam. Shrub on rocky outcrop... .. ... See ee eee Gas

P. zeylanica L. Shrub in scrub forest and common grassveld ..... _.....

SAPOTACEAE

Bequaertiodendron magalismontanum (Sond.) Heine & J. H. Hemsley. Shrub or

small tree on lower foothills Be on 4886,

B. natalense (Sond.) Heine & J. H. Hemsley. Small tree in shrub forest... .... es

Mimusops zeyheri Sond. Fair-sized to large tree in foothill kloofs oo. 0... 5935,

M. obovata Sond. Small tree on fringe of mountain forest... .... ae =

EBENACEAE

Euclea crispa (Thunb.) Guerke. Large tree of mountain forest ... .. ... 4959,

E. divinorum Hiern. Small tree or shrub on mountain slope : 5012,

Diospyros lycioides Desf. subsp. sericea Bern) De Wint. Low shrub, widespread

mn Gamlior GAGES OF GUESS eno au amb op cam ceo omy Gem gm ;

D. mespiliformis Hochst. ex A. DC. Tree on riverbanks, outcrops and hills _.... ee

D. whyteana (Hiern) F. White. Understory tree of high forest _..... ee ee O86"

OLEACEAE

Schrebera alata (Hochst.) Welw. Large tree common in rain forest 4371.85

S. argyrotricha Gilg. Shrub, common in scrub forest of Blyde River Poort

Linociera foveolata (E. Mey.) Knobl. Large tree in mountain forest 20.00. cies

L. foveolata (E. Mey.) Knobl. ssp. major Verdoorn. Large tree in mountain forest

Olea africana Mill. Blyde River scrub forest 22.00 20. cee ee eee cee nee teen

O. capensis L. ssp. macrocarpa (C. H. Wr.) Verdoorn. Fair- sized canopy tree of

high forest ..... Ee a Peay ee a Spence sat ea

O. woodiana Knobl. “Large tree in indigenous forest.

Jasminum abyssinicum Hochst. ex DC. Twining shrub in forest margin enh vceae

J. fluminense Vell. Woody liane in mountain forest (0.00.

J. streptopus E. Mey. var. transvaalensis (S. Moore) Verdoorn. Small shrub on

forest margin Peay FaSiaun ees aatrertes eisserigh gleetasis ais Ma eee eae 4737,

LOGANIACEAE

Strychnos decussata (Pappe) Gilg. Medium-sized to small tree in Lowveld scrub

S. spinosa Lam. Small tree in Lowveld bushveld 2... oe. cee see steee seeee eteee nee

Anthocleista grandiflora Gilg. Tree with large leaves on banks of Klaserie River

Om JOMISr TOONS, aie cee ao ee ES cea EL cee ce eee Re cee Pom te

Nuxia congesta R. Br. ex Fresen. Small to fair-sized tree of mountain forest 6205,

N. floribunda Benth. Small to fair-sized tree, high forest and forest margins ..... _ .....

N. oppositifolia (Hochst.) Benth. Shrub widespread on riverbanks and river

COULSES eee es ce nal Se eee! es eee S eeEaene kos 6081,

Buddleia salviifolia (L.) Lam. Shrub in forest margins om oom Geb, Sb

496

GENTIANACEAE

Sebaea erosa Schinz. Small herb with yellow flowers in grassveld on mountain

SUMIMIt e eee oe treet ceca eee te) gh Une ae ee ee nese wees 4849, 5840

S. macrophylla Gilg. ‘Herb with yellow flowers on top ‘of mountain in moist

places... pega RAs ee ee tet Wee as de ee ; . 4380, 4483

ne NACEAE

Carissa bis pinosa (L.) Desf. ex Brenan var. acuminata (E. Mey.) Codd. Widespread

shrub in undergrowth and along margin of forest... ... wor ee 4768, 5853

. edulis Vahl. Large subscandent shrub or small tree in scrub... ... ... 6512, 6078

eee aia capensis @Olivanlbianes withwlatexcminietOneStie anne nnn 4677

Rauvolfia caffra Sond. Large tree on banks of Klaserie River _ 5053

Strophanthus speciosus (Ward & Harv.) Reber. Liane in mountain forest 6274

ASCLEPIADACEAE

Xysmalobium acerateoides N.E. Br. Perennial forb in mountain grassveld ...._..... 5022

X. confusum Scott-Elliott. Robust perennial forb in mountain grassveld be See 4966

X. undulatum (L.) Ait.f. Herb in mountain grassveld .... oar ... 6387, 6127

Asclepias decipiens N.E. Br. Herb in mountain grassveld.. apenas os ite eee oa 6121

A. gibba Schltr. Herb in pine plantation, on side of road... 2... Se pence arate 5865

A. physocarpa Schltr. Tall weed of disturbed veld 0 0. wo.) ee ek 5892

A. stellifera Schltr. Herb on slopes of mountain . ne area ; 5944

Sarcostemma viminale R. Br. Leafless succulent perennial iwiner ee =

Secamone alpinii Schultes. Slender liane in high forest . heres mgs pee 6211, 4980

Sp fnutescensm@ Decne, Wiane om mockysnidgesy an ye ec ences ecennneee 5520

S. gerrardii Harv. ex Benth. & Hook.f. Liane in high and marginal forest ... 5018, 4638

Brachystelma pulchellum Schltr. Bulbous shrub in grassveld on top of mountain ..... 6322

Riocreuxia picta Schltr. Subwoody twiner in high forest ; om on oem = OHO, CUB

R. torulosa Decne. Subwoody twiner in scrub forest ...0 4670

Marsdenia dregea Schltr. Slender twiner in Lowveld scrub on . banks of Blyde River 6390

CONVOLVULACEAE

Convolvulus ulosepalus Hall.f. Herb with white flowers on foothills of mountain 6322

Ipomoea albivenia (Lindl.) Sweet. Woody twiner on rocky places _..... ee core Ree 6116

E GHOCHP NG IR, lit IsISHOSSOUS GRSSTSP um coe oe eS mek 4579

T. wightii (Wall.) Choisy. Widespread and perennial twiner in grassveld and scrub

OLCS tame ene eae ee ; ands ee Soe Gavieal « Rete Me ee 5890

ROAACINAGIND

EhretiavamocnammMotzschy ss InyWowveldisciubige sr oe eee ae nee tn ce

Trichodesma zeylanicum R. Br. Annual weed in fallows in . Blyde Canyon ...... .... 5952

Cynoglossum lanceolatum Forsk. Weed in pine plantation 2... 6060

Myosotis afropalustris C.H. Wr. Small herb in grassveld on foothills of mountain 5893

VERBENACEAE

lemon Conner IL, “Weel iin roo wav! Sew) an ae em 40m Sen am aro 4603

L. mearnsii Moldenke. Weed in foothill scrub 20 000 0. LA ere Seren ty eee cere 4574

L. montevidensis Brig. Herb along roads . a seg Ue eee ane 4659

Lippia javanica (Burm.f.) Spreng. Small shrub on lower foothills... 6514, 5040

Chascanum sp. Small shrub on lower foothills ... 0 0. eo bs Seat 4597

Duranta repens L. Garden escape of disturbed scrub and scrub forest... es

Gmelina arborea Roxb. Foreign tree on foot of mountain 0.0... 6049

Clerodendrum glabrum E.Mey. Shrub or small tree of early scrubby seral

StQIPeS| ka Misses (lee) Ieee ce ee eee : DS re eS OS/ena 50)

LABIATAE

Leonotis leonurus (L.) Ait. Sonsexueds herb of forest margins and grassveld

ATCAS) cs cae teh, seca) ee ee meee Si ONC a 4588, 5166

Leucas glabrata R.Br. Shrub in scrub forest in grassveld : 5508

Stachys aethiopica L. Creeping shrub with mauve flowers on side of road... .. 6337

S. galpinii Brig. In mountain grassveld near waterfall 0 0 ih amelie tae 6039

S. cf. S. rehmannii Skan. Laxly spreading low forb 00 oc ee a ceed, Mates 5139

Endostemon obtusifolius (E. Mey.) N.E. Br. Herb in forest margin _... . 4921, 6056

Pycnostachys reticulata (E. Mey.) Benth. Soft open shrub of moist places 4611

P. urticifolia Hook. Widespread soft shrub of open places 00. a 4572, 5524

Plectranthus arthropodus Brig. Robust herb on forest margin Pe ee tee ciety, icin 4375

P. fruticosus L’Hérit. Robust herb on forest margins 0 uu. 4460, 4458

P. laxiflorus Benth. Robust herb on forest margins ae wee ~— 4448, 4345

P. myrianthus Brig. Herb in mountain grassveld bce a ons ES eee 5100

497

P. nummularius Brig. Prostrate herb in rock areas

Coleus tysonii (Guerke). Herb on forest margin Sets Ete ons mek Leen, On ee

Hoslundia opposita Vahl. Herb in mountain grassveld ; rears Ee

Iboza riparia N.E. Br. On banks of Blyde River 2.0 0 ce ee

Syncolostemon eriocephalus Verdoorn. Shrub in rock crevices on mountain

SOLANACEAE

W ithania somnifera (L.) Dun. Shrub in forest region pee. Ae

Physalis peruviana L. Weed in disturbed areas... ... ... 2. ree

Solanum aculeatissimum Jacq. Shrub along road : ~ es

S. geniculatum E. Mey. Shrub on forest margin . i Se teh aie es ee eee

- giganteum Jacq. Shrub in marshy area

S. indicum L. Weed on side of road .... ae Bea accor sean

S. panduraeforme E. Mey. Weed of disturbance cae te Ngee:

Datura stramonium L. Well-known weed ..... . Bey nerey she ee bs PR ee

Sano eTILCTAGDINS

Halleria lucida L. Widespread tree in mountain, forest and on river banks ....

Nemesia melissaefolia Benth. Small herb on mountain summit ... .... ....

Diclis reptans Benth. Herb on fringe of forest (2.00. ck cesses

Bowkeria cymosa MacOwan. Shrub or small tree on forest margin ee

Sutera accrescens Hiern. Suffrutescent forb _.... seal dpcteuaeies 634 Seat ap els

S. floribunda (Benth.) Kuntze. Lax suffrutescent forb .. ... ...

Zaluzianskya katherinae Hiern. Herb in rock crevices. en cece sees ses

YanlychnidcasWalps shrub) between rocks) =

Craterostigma wilmsii Engl. ex Diels. Herb in seasonal pools ‘and rocks on summit

of mountain re ee ee ee ee ee eek Ra

Ilysanthes conferta Hiern. ‘Submerged herb in seasonal pools on summit of mounta

Hebenstreitia comosa Hochst. Herb on summit of mountain 2... 0. uu. wn ;

H. dentata L. Herb on summit of mountain 20 2. : ; auth cee

Selago elata Rolfe. Herb of grassveld on forest margin o ae ey) ean ee

S. muddii Rolfe. Herb on summit of mountain Bee Wakes oes fee si

S. nelsoni Rolfe. Herb on summit of mountain ... ... .... ee eee

S. villosa Rolfe. Herb with blue flowers on top of mountain ... ... ...

S. sp. Small shrub on mountain slope... ae : ga ESE sda

Cycnium racemosum Benth. Parasite in grassveld : : : ee

Rhamphicarpa tubulosa Benth. Shrub on river-banks_ . Sih cetera see

Harveya coccinea Schtr. Widespread parasitic herb .... ....

H. huttonii Hiern. Root parasite on forest floor in mountain forest

BIGNONIACEAE

Tecomaria capensis Spach. Twining shrub in Lowveld scrub on river-banks

GESNERIACEAE

Streptocarpus confusus Hilliard. Herb on forest floor

. cyaneus S. Moore. Herb in forest on mountain slope _....._ .....

. micrantha C.B. Cl. On rocks in indigenous forest. ..... :

. parviflorus Hook.f. Small tufted herb on high forest floor ...

. pusillus Harv. In shade of rocks on mountain summit

. wilmsii Engl. Epiphyte on trees in indigenous forest

ACANTHACEAE

Thunbergia alata Boj. ex Sims. Twiner in forest margin . eR any es

T. natalensis Hook. Shrub on forest floor eee ral

Barleria gueinzii Sond. Herb in undergrowth _.... Tae ety ee ew

B. rotundifolia Oberm. Herb on floor of riparian forest 0.0 0. ee

Sclerochiton harveyanus Nees. Sprawling to subscandent shrub. Abundant

high forest undergrowth ... ..._ ..... Pa eee heen

Crossandra_ greenstockii S. Moore. Herb in grassveld : ae

Dicliptera clinopodia Nees. Undershrub in kloofs and mountain forest...

D. zeylanica Nees. Shrub on fringe of mountain forest... See ge

Hypoestes aristata R. Br. Undershrub of sour grassveld and scrub cit

H. triflora Roem. & Schult. Shrub on river-banks and forest margin :

H. verticillaris R. Br. Small shrub in undergrowth ... 0.0.0... Be yeeee nae

Mackaya bella Harv. Shrub in open parts of high forest 0... .... a eae

Duvernoia adhatodoides E. Mey. ex Nees. Shrub prominent in substrata...

Ruttya ovata Harv. Shrub on Blyde River banks ....._ .... en fee ee

Justicia petiolaris E. Mey. Undergrowth of Blyde River forest ...

J. protracta (Nees) T. Anders. Undershrub of Blyde River riparian forest

AnNKHnHN

6136

ee, 5140

6139, 6374

6780, 5496

5587, 4838

7 6794

5527, 4866

4895

ae 4756

4600, 4874

aa 4896

4798, 4575

5608, 5616

ee 5968

4361A, 4645

6202, 5536

ate 5620

4811

4 i 4517

in 6339

4833

mee 6768

4449, 6769

aes 5834

at 4522

4383, 4524

oes 6492

6460

ree 6495

et 4683

6258A, 5894

5167

6635, 4727

. 6048, 6478

_ 4507A, 4997

4971

4817, 4816

. 4951, 4329A

. 6372

5895, 5916

— 4602

oe 5498

4899, 4514

5010, 5859

6730, 5491

4463

5552

$128

Spl slo

ae 4673

6041. 5147

5494, 6400

Bie 5509

5507

498

PLANTAGINACEAE

Plantago lanceolata L. Exotic herb in grassveld at forest station .... ent aea eee 6436

RUBIACEAE

Oldenlandia rupicola (Sond.) Kuntze. Herb on summit of mountain ... ... 6764, 4821A

Conostomium natalense toch Brem. Herb in pine Plantations ee eee 2 4510

ee ce ee ee 5052, 6089

Cephalenihus natalensis Oliv. Liane in mountain forest eo aes ... 6798, 6171

Gardenia amoena Sims. Strangling shrub in scrub and scrub forest ... .... Sogn 5538, 6093

G. spatulifolia Stapf & Hutch. Small Lowveld tree ... 2.0 2. 2 ee —

Rothmannia capensis Thunb. Small to medium-sized understory | tree of high

forest Be Ee Ta ates Seria MR yl secs sr ea aii a. 5881, 4881

Oxyanthus gerrardii Sond. Large understory shrub of high forest .. .... es 4405, 4626

Tricalysia capensis (Meisn.) Sim. Small understory tree in high forest. .... ... 6277, 4400

T. lanceolata (Sond.) Burtt Davy. Small shrub in forest ... ... .. ... sae Send 6067

Pentanisia angustifolia Hochst. Herb in pine plantation 2... 0. cc cee 4904

P. prunelloides (Klotzsch ex Eckl. & Zeyh.) Walp. Herb on forest margin 4446, 4914

Vangueria cyanescens Robyns. Small understory tree) or shrub). 1. Ve. oe. —

V. infausta Burch. Variable small tree in Lowveld areas .... ... .... become Oe ae

Canthium ciliatum (Klotzsch) Kuntze. Small understory tree in ioreste: ee 5965

C. gueinzii Sond. Stout woody liane of high forest, and forest margins _..... ... 6168, 4792

C. mundianum Cham. & Schlechtd. Shrub on stony and rocky ridges in forest

AREAS)” Ree dee Gene cee nee cig Ee) EEG iN Re er ae) ce more ... 6087, 6473

C. obovatum Klotzsch. Undershrub in mountain forest .. .. ... a. 5123, 4882

C. ventosum (L.) S. Moore. Small tree on fringe of forest ... ....._.... a. 5906, 6231

Pachystigma ae ae (Sond.) Robyns. Shrub in undergrowth 2.0. un 6276

Fadogia tetraquetra K. Krause. Herb on mountain slopes .... .... .. i eees 5958, 6069

Pavetta assimilis Sond. Scrub on banks of Blyde River 0 0c, ce ee en 5489

P. lanceolata Eckl. Small tree or shrub on high forest margin ... ... .... 6785, 6042

P. schumanniana F. Hoffm. ex K. Schum. Shrub in Blyde River Lowveld ..._.... 6119

Psychotria capensis (Eckl.) Vatke var. capensis. Small wager tree in high

GOTESE. er eee yet th ee oe) aie a ee ee ge ae ee ee ee Foe eet = 38 4663, 4633

P. zombamontana . (Kuntze) Petit. Small understory tree in forest .. cue ua. 4729A, 4668

Galopina circaeoides Thunb. Herb on forest floor 2. ce cent 4994, 4324A

Anthospermum ammanioides S. Moore. Shrub on slopes of mountain... eta 6128

A. herbaceum L.f. Prostrate herb on mountain summit ...._..... Pe ee tOS41EE S596

A. hispidulum E. Mey. Prostrate herb on mountain summit 0 00. 0. un. 4345, 5995

Otiophora calycophylla (Sond.) Schltr. & K. Schum. Small shrub on summit of

MMOUNTAINY cui cosas. Geet estat ec ne Ee See eA ene lea yee eel ree aie 6318

O. cupheoides N.E. Br. Small shrub in open areas and forest margins ....._..... 4523, 4831

Rubia petiolaris DC. Straggling shrub on forest margins 00 0. 0 en tonto 4606

DIPSACACEAE

Scabiosa columbaria L. Widespread herb in mountain grassveld ....._..... .. 5009, 4983

CUCURBITACEAE

Cucumis africanus L.£. Herb in mountain grassveld _.... ee ee eee eee 6511

C. sp. Herb in tall grassveld on foothills of mountains 2... 0.0. 0 es 5877

Trochomeria hookeri Harv. Slender soft climber in erased on foothills of moun-

(1 0 nee ee tM ce en ho rane cio este Ghat coer cae cue agin 5075

Peponium mackenii (Naud.) Engl. Climbing ‘herb in’ riparian SChUD eee 6404

Coccinia adoensis (Hochst. ex A. Rich.) Cogn. Herbaceous climber in seral grass-

veld and forest scrub 0.0. ee eet ier cout aieanect curt ice tees 6385

C. palmata (Sond.) Cogn. Twiner in pine plantation Ie, eee as aw 5531, 6063

C. sessilifolia (Sond.) Cogn. Twiner in shrub forest 0.0. cc ec te —

C. variifolia A. Meeuse. Soft climber of the openings in high forest ..... .... 6133A, 6511

CAMPANULACEAE

Wahlenbergia undulata (L.f£.) A.D. Herb in mountain grassveld 0... 0 ok ce cee cos

W. virgata Engl. Herb between boulders on mountain summit ...... _ ..... vale Mone 6195

Lobelia decipiens Sond. Small herb, pioneer on bare soil in wet places. ..... .. 6561, 5868

L. filiformis Lam. var. krebsiana (Presl) E. Wimm. Small herb in marly. grassveld 6019

L. laurentioides Schltr. Herb spreading flat on ground .. wu... one sis 5087

L. pteropoda (Presl) A.D. Herb on forest floOF on. cee cee te eee 4697

Monopsis kowynensis E. Wimm. Small herb on moist places on summit of moun-

tain ese ates OR ee ener en mr arm em eames 6020, 6180

_ stellarioides (Presl) Urb. var. stellarioides E. Wimm. Herb on moist places

of forest margins _..... ie cides “ees Pa MES ere eee ee ee ee as

Vernonia ampla O. Hoffm. Shrub up to 3 m high in scrub of

SIN SNS

499

COMPOSITAE

foothills of mountain _.....

. corymbosa (Thunb.) Less. Small shrub on mountain slopes

. crataegifolia Hutch. Shrub on eastern mountain slopes _...

. oligocephala (DC.) Sch. Bip. ex Walp. Herb in grassveld on lower slopes. .....

. Shirensis Oliv. et Hiern. Tall shrub in pine plantations

umbratica Oberm. Small shrub on forest margin

Ageratum conyzoides L. Herb in bluegum plantation

Mikania cordata (Burm.f.) Robinson. Liane in scrub forest in forest margin 6811,

Mikaniopsis sp. Liane infrequently scattered in high forest

Dichrocephala integrifolia (L.f.) Kuntze. Herb in forest margin

Erigeron canadensis L. Weed in mountain grassveld

Nidorella auriculata DC. Herb in mountain and foothill — grassveld

Conyza ivaefolia (L.) Less.

Tarchonanthus galpinii Hutch. & Phill.

transvaalensis Phill. & Schweick. Large forest tree .....

Laggera pterodonta (DC.) Sch. Bip. Weed of disturbed sites .

Cassinia_ phylicaefolia (DC.) J.M. Wood. Herb in mountain grassveld.

Helichrysum acutatum DC. Herb on summit of mountain

yy ye Be

Stoebe vulgaris Levyns.

Macowania tenuifolia M.D. Hend.

appendiculatum (L.f.) Less. Small herb in mountain grassveld

decorum DC. Herb with yellow flowers on mountain slopes

. aureum (Houtt.) Merrill. Herb on side of road

. latifolium (Thunb.) Less. Herb on summit of mountain

lepidissimum S. Moore. In grassveld on slopes of mountain

mechowianum Klatt. Herb on forest floor, mountain slope

mundii Harv. Herb in mountain grassveld

forest,

Spreading small shrub on banks of Blyde River

C. hochstetteri Sch. Bip. Herb in scrub on banks of Blyde River

Brachylaena huillensis O. Hoffm.

Shrub or small tree in mountain areas

6620,

, 5099

, 5534

6057,

Small tree on rocky ridges in mountain forest

nudifolium (L.) Less. var. quinquenerve (Thunb.) Moeser. Weed in pine plantation

obductum H. Bol. Herb on summit of mountain

odoratissimum (L.) Less. Herb in marshy grassveld

platypterum DC. Herb in mountain grassveld Bee (ee

. setosum Harv. Small shrub on mountain slopes _....

. sutherlandii Harv. Herb on rocky cliffs at waterfall

. subglomeratum Less. Herb with yellow flowers, on plateau

. swynnertonii S. Moore. Herb in mountain grassveld

. undatum (Thunb.) Less. Subwoody forb-grassveld

. wilmsii Moeser. Herb in mountain grassveld :

sp. nov. aff. aff. H| wilmsii Moeser. Herb in marshy mountain grassveld

6241,

sp. nov. (= Smuts & Gillett 2375, Codd 3336 +6202). Herb on summit of moun:

tain ee ec tee cme | eres aur ree xeersev bua vatteel py lastits! Neste:

Athrixia elata Sond. Small shrub in mountain grassveld

Philyrophyllum schinzii O. Hoffm.

phylicoides DC. Mountain grassveld

Anisopappus junodii Hutch. Small shrub on summit of mountain

Acanthospermum australe (Loefl.) Kuntze.

Spreading cricoid shrub in mountain grassveld

Small shrub on summit of mountain

In grassveld on banks of Blyde River

Small shrub in mountain grassveld

genie mossambicensis (Oliv.) Wild. Shrub with yellow flowers on river bank

Tithonta tagetiflora Desf. On side of road near Blyde picnic spot

Galinsoga parviflora Cay. Weed of disturbance in moist area

Eumorphia davyi H. Bol.

Schistostephium heptalobum (DC.) Oliv. & Hiern.

summit ...._......

Artemisia afra Jacq. Herb in mountain grassveld

Lopholaena disticha (N.E. Br.) S. Moore.

Crassocephalum crepidioides (Benth.) S. Moore.

picridifolium (DC.) S. Moore. Herb in pine plantation

Cineraria sp. Liane in forest margins -

Senecio barbellatus DC. Herb on mountain summit

BRP Peg

caudatus DC. Herb in mountain grassveld

. coronatus Harv. Herb on mountain plateau

. deltoideus Less. Undergrowth on Blyde River banks

erubescens Ait. Herb in moist areas _..... ae eee

galpinii Hook.f. Herb. in mountain grassveld Lae

junodii Hutch. & Burtt Davy. Small shrub in forest margins

. natalensis (Sond.) Wild. Shrub on banks of river near disturbed area

Shrub in mountain grassveld

Herb on summit of mountain

Small shrub with white flowers on top of mountain :

Herb along road on mountain

4908,

SUA,

4580

6507

, 5164

4540

5165

5530

5541

4888

4869

4599

6268

5103

5033

5098

, 5104

6037

, 4903

5572

4403

, 5979

, 4472

5947

4616

4885

6197

4834

5096

, 4660

4351

5377

6489

; 4335

4775

6176

4905

6196

5535

6789

5490

5656

5044

6393

6096

6122

5634

4835

4394

6508

5612

4465

6504

4752

6448

6242

, 6338

5493

. 6286

5023

4347

500

S. mikanioides Otto. Slender liane in mountain forest ae es

S. orbicularis Sond. In grassveld on slopes of mountain 2... 0 1. tes ~ 5002, 6353

S. pandurifolius Harv. Tall herb in grassveld of lower slopes ..... _..... bios a Rese eae 4612

S. polyanthemoides Sch. Bip. Herb in mountain grassveld est woe ee 6488, 4563

S. pterophorus DC. Herb in marshy grassveld .... 0 0... Be eae! a. 4887, 4342

S. quinquelobus DC. Slender liane in mountain forest ... .. .... erp ete eee 4453

S. speciosus Willd. Herb im mountain grassveld 2.0 200 ee 6252, 6038

Siitarmoides) Gy 1 0) 2 ie Pee Pee be seo PieNioe iaree enc pete ee ee ee ee 4372, 5086

Euryops pedunculatus N.E. Br. Small shrub on summit of mountain... .. ... ... 5613

Haplocarpha scaposa Harv. Herb on summit of mountain .... ..._.... . 4977, 6326

Hirpicium bechuanense (S. Moore) Roessl. Weed on fallows in “Blyde River Poort 5955

Berkheya insignis (Harv.) Thell. Small herb in grassveld on earstern slopes 5879, 4976

B. latifolia Wood & Evans. Herb on banks of Klaserie River 2.0 0. es 4361

B. radula (Harv.) De Wild. Near water at Klaserie Waterfall _..... CAD Hen pee ere! 4925

Gerbera jamesonii H. Bol. ex Hook.f. On banks of BN River

ae 5512

G. kraussii Sch. Bip. On banks of Blyde River 0 0 ww. Ce setae nee wee 6386

501

Book Review

Witp Flowers or THE Natat DrakenspercG by W. R. Trausetp. Cape Town: Purnell. 1969.

Pp. xxxi + 220, 814 colour plates. Price R11.50.

This is the first book published for the layman on the flora of the Natal Drakensberg.

As such, it fills a great need. The Drakensberg is visited by thousands of holidaymakers

each year, many of whom have a special interest in the wild flowers of the mountains.

The book which runs to 220 pages contains, first of all, a list of many of the plants

collected by Mr. Trauseld in the Royal Natal National Park and the Giant’s Castle Game

Reserve and by the reviewer in the Cathedral Peak area. The list excludes grasses and

sedges. Following the list is a table of flowering times and then colour photographs of

521 species, each of which is supported by a short 2 to 5-line description of the plant,

its habitat, altitudinal range, flowering period and frequency. Mr. Trauseld points out

that the altitudes given are based on his own experience in the two areas he was concerned

with. Also, the frequency data are not applicable to the Natal Drakensberg as a whole.

It is important to remember these two points, otherwise one could be led astray.

The chief criticism I have of the work is the way in which individual flowers or

inflorescences have often been plucked and placed on rocks or inserted in holes or

crevices in boulders presumably to eliminate unnecessary movement of the subjects during

photography. The “Berg. of course, can be a windy place. The photographs of a head

of Protea multibracteata and a raceme of Greyia sutherlandii projecting from a_ rock

crevice and hole respectively are, to me, not only quite unnatural, but ecologically inaccurate

and aesthetically unpalatable — similarly the photographs of Oxalis smithiana, Vellozia

viscosa, Papaver aculeatum and Pelargonium luridum in which the plants are lying prostrate

on a rock surface with their bases weighted down by smal] stones. Surely Mr. Trauseld

could have used a tripod with clamp and eliminated the effect of wind by some more

sophisticated method?

The photographs vary in quality from extremely poor, e.g. Crocosmia aurea, Erica

woodii, E. westii, Buddleia salviifolia, Psammotropha mucronata and Peucedanum connatum

to very good e.g. some of the asclepiads. On the whole, the close-ups of flowers and

inflorescences are better than the habit photographs. In the latter the plants tend to

merge with the background, e.g. Habenaria clavata, Satyrium longicauda, Disa stachyoides

and Silene burchellii. The colour reproduction is not always true, e.g. Leonotis dysophyllus,

Protea multibracteata and P, dracomontana. The plate of Anisotoma pedunculata has been

printed upside down.

There are several unfortunate typographical errors, e.g. Crytanthus for Cyrtanthus on

p. xii, Tulbughia for Tulbaghia on p. 17, Bowkeria verticellata for B. verticillata on pp.

166 and 167, Vernonia hirsutu for V. hirsuta on pp. 194 and 195, Eulophia welwitchii for

E. welwitschii on p. 75. Diascia purpurea for D. purpurea on pp. 164 and 165, and

Solonaceae for Solanaceae on p. 126. A few mistakes have obviously escaped the scrutiny

of the editor, e.g. Aloe boyleii instead of A. boylei, Sisyranthus huttonae instead of

S. huttoniae and Beupleurum mundii instead of Bupleurum mundii (even incorrect in

the index).

A couple of misidentifications were spotted. Phygelius capensis does not, as far as

present records show, occur in Natal. Presumably Mr. Trauseld collected P. capensis

on the summit of the Drakensberg in Lesotho and assumed that the rather similar-looking

P. aequalis growing in Natal represented the same species. Trauseld 162 (p. 35) is not

Moraea natalensis Bak., but more probably M. violacea Bak. However, we need to see

the type of M. violacea before this can be confirmed.

Several other points deserve mention. The map of the Natal Drakensberg on_ the

front end paper would offend the purist cartographer: the name Drakensberg has been

printed upside down i.e. in the original version and not just an error of reproduction.

Mr. Trauseld states that Philippia evansii occurs on the summit of the Drakensberg as

well as on the Little Berg. I have not seen P. evansii on the summit and doubt whether

it has been collected there. It is surprising that Watsonia socium, Moraea pubescens and

Berkheya multijuga, common species in the Drakensberg, were not depicted in the work.

In spite of the few imperfections mentioned, Mr. Trauseld’s book is a praiseworthy

effort and will be of immense value to all lovers of the flora of the Natal Drakensberg.

D. J. B. KiLrick.

Marguerite Gertrud Anna Henrici (photograph taken about 1930)

Boihalia, 10, 4: 503—S08

Marguerite Getrud Anna Henrici (1892-1971)

M. D. Gunn

In November 1922 a Swiss scientist arrived in South Africa, a country she was to

make her home for close on fifty years. Dr. Henrici was born on 22nd February, 1892,

in the city of Basle, Switzerland, and obtained the Swiss matriculation certificate in

1912. After spending a period in France she enrolled in 1913 at the University of Basle

where she studied botany, chemistry and zoology, eventually specializing in plant-

physiology under Prof. Gustav Senn. In December 1917 she was awarded a doctor’s

degree summa cum laude.

The summer months of her later University years were spent at a small mountain

laboratory which Senn had fitted up at Muottas Muraigl above Samaden in the

Engadine Alps. Here, with her assistance, Senn worked during the vacations in-

vestigating the transpiration, respiration and assimilation of alpine plants. She

became private assistant to Prof. Senn from 1919-20 and during 1920-22 she was a

research worker in the botanical institute of the University.

At the beginning of her student career in 1913, she sat alongside a thickset

bearded man in his middle forties. He was Dr. A. Theiler, later to become Sir Arnold

Theiler, founder and director of the world-famed Veterinary Research Institute at

Onderstepoort, near Pretoria. Dr. Theiler was home in Switzerland on a refresher

course at the time. In after years, recounting her first meeting with the Veterinarian,

she referred to it as “that important day in my life’’.

Dr. Theiler noted the young woman’s progress, her capacity for work and her

achievements as a research worker. When again in Europe in 1921, he visited Basle

and invited her to come to South Africa to join his staff.

LAMSIEKTE—ARMOEDSVLAKTE

A disease among cattle known in South Africa as lamsiekte (botulism) had for

long caused severe losses to stock farmers and baffled earlier investigators who had

made a study of the disease. About 1912 the farm Armoedsvlakte, near Vryburg in the

north-western Cape Province, an area particularly noted for lamsiekte, was taken over

by the Veterinary Research Division as a field station for the investigation of the

disease. In 1917, lamsiekte was particularly rife in South Africa and Theiler was

asked to undertake the special task of lamsiekte research. Early in 1918, relieved of all

administrative duties, he took up the post of Director of Lamsiekte Research at the

Government farm Armoedsvlakte. By 1919 he was able to explain the cause of the

paralysis in animals and how the disease could be prevented. A new field in the

science of nutrition had been opened up by Theiler and his co-workers and it was to

this field that Dr. Henrici was invited to come and work as a plant physiologist on the

phosphorous deficiency of the grassveld.

From the University of Basle (its foundation dates from 1460), a centre of culture

and refinement, she was placed in charge of the Armoedsvlakte field station, in an

isolated, semi-arid region adjoining the Kalahari Desert. Not only can a greater

contrast scarcely be imagined, but it was a precedent in South African Public Service

history for a young woman, and a new-comer at that, to be put in charge of an out-

station. She entered her new position with trepidation but, with the confidence of her

chief, Sir Arnold Theiler, she immediately got down to work, showing the drive and

determination for which she was noted throughout her career.

Her laboratory had been fitted up for veterinary research work but was ill

equipped for a plant physiologist, and the only literature she had were her own

private books and journals.

504

There was also the difficulty of language. She was fluent in German and French

and had a working knowledge of English, but Afrikaans was new to her. In time she

overcame this handicap to a certain degree but her reports required an understanding

editor and, to the end of her days, she spoke both English and Afrikaans with a

heavy German-Swiss accent.

Her first South African scientific paper, written in German, dealt with the

transpiration of grasses in Bechuanaland and was published in Basle (1923). She then

began a series of publications based on her research on natural pastures. This included

work on chlorophyll, carbohydrates, phosphorous content of grasses and the cystine

and sulphur content of Karoo shrubs and grasses, and in 1927 the University of

South Africa awarded her a D.Sc. degree for a thesis on her plant physiological

studies.

From July 1926 to December 1927 she was stationed near Ermelo in the eastern

Transvaal in order to study the phosphorous content of highveld grasses, and the

results of her investigations were published in the Report of the Director of Veterinary

Services (1930).

THE VELD RESERVE, FAURESMITH

Dr. Henrici was transferred from the Division of Veterinary Services to the

Division of Plant Industry in 1929 and was appointed as Officer in Charge of the

Veld Reserve at Fauresmith. Situated near Fauresmith in the south-western Orange

Free State, the Reserve of some 70 ha (75 morgen) had been allocated by the Munici-

pality in 1926 for the purpose of studying problems connected with the Karoo veld and,

to some extent, pasture problems in other parts of the country.

Later, with an additional 25 ha, the land was purchased by the Government

and Dr. Henrici was consulted on the planning and equipping of the laboratories

and other buildings, including a residence which was to be her home for 29 years.

In this small but up-to-date laboratory, with adequate reference books and periodicals,

and assisted by competent technical staff, among whom Miss A. J. van der Walt and

Messrs. P. E. Potter, A. F. J. Visagie and L. P. Meyer deserve special mention, Dr.

Henrici entered a period of long and fruitful scientific research.

Not unnaturally, the conservative farming community did not immediately

react favourably to her and her new approach to grazing problems. However, it

is to her credit that she overcame these feelings of reserve and soon was regularly

consulted on pasture matters. She was called in to visit neighbouring farms, attended

their meetings and soon even gained the confidence of the farmers’ wives, to the

extent of being given their closely guarded cookery recipes.

She spent 1939 on vacation in Europe visiting plant physiological institutes and

meeting some of her scientific correspondents.

During the 1940’s she undertook a series of transpiration studies. Her techniques

came in for some criticism and as a result of divergent opinions on the relative effects

of indigenous and exotic trees on ground water resources, a certain amount of con-

troversy followed.

On reaching the age of retirement on 21st February, 1948, Dr. Henrici continued

to serve the Department of Agriculture in a temporary capacity. She continued with

her work at the Veld Reserve, Fauresmith, until March 1957 when her services with

the Department were finally terminated.

Dr. Henrici was not idle in retirement and had been urged to publish a book on

Karoo bushes in both Afrikaans and English. She completed the manuscript which is

as yet unpublished.

In spite of indifferent health at times, she lived actively and achieved a high

scientific output. During the last two years of her life her health deteriorated and

eventually she was accommodated in a home for the aged in Bloemfontein where she

died on 28th July, 1971.

505

SOCIETIES AND AWARDS

In 1926 she was elected a member of the South African Association for the

Advancement of Science and was a regular supporter of the Association. Their

Journal was an outlet for many of her scientific papers. In 1937 she was President of

Section C, her presidential address on the occasion being entitled ‘Transpiration of

water supply of South African plants.” In the 1950’s she was the Vice President for the

Orange Free State region.

She joined the South African Biological Society in 1926 and was awarded the

Senior Capt. Scott Memorial Medal in 1935 for outstanding scientific achievements.

She was an honorary member of Basle Botanical Society and in 1969 the University

of Basle awarded her an honorary D.Sc., acknowledging its appreciation of her

pioneer work in her field of research.

In 1971 the South African Association of Botanists honoured her by electing

her an honorary life member of the Association.

No academic award she received charmed her as much as an illuminated address

presented to her in 1968 by the farmers of Fauresmith district.

She amassed a herbarium of between 6,000 and 7,000 specimens, mainly of

Karoo plants, and is commemorated in the names Neohenricia L. Bol. and Salsola

henriciae Verdoorn.

BIBLIOGRAPHY

1918. Chlorophyllgehalt und KGhlensdure-assimilation bei Alpen und Ebenenpflanzen. Inaugural

dissertation, Basle.

1921. Zweigipflige Assimilationskurven. Mit spezieller Beriickichtigung der Photosynthese von-

alpinen phanerogamen Schatten pflanzen und Flechten. Verhandlungen der Naturfor-

schenden Gesellschaft, Basle.

— Influence de la conductibilité de l’air sur la photosynthése (note preliminaire). Archives des

Sciences Physiques et Naturelles, Geneva.

1923. Die Transpiration einiger Graser in Bechuanaland. Verhandlungen der Naturforschenden

Gesellschaft, Basle. 35: 356-373.

1925. Chromotophoren und Kohlensaure-assimilation nicht-griiner Gefasspflanzen. Berichle der

Schweizerischen Botanischen Gesellschaft.

1926. The chlorophyll content of grasses in Bechuanaland. 11th and 12th Repts. of the Dir. of Vet.

Ed. and Res., U. of S.A., 259-271.

— Preliminary report upon the occurrence of hydrocyanic acid in the grasses of Bechuanaland.

11th and 12th Repts. of the Dir. of Vet. Ed. and Res., U. of S.A., 495-498.

— Physiological plant studies in South Africa. Part 1. Wilting and osmotic phenomena of

grasses and other plants under arid conditions. 11th and 12th Repts. of the Dir. of Vet. Ed.

and Res., U. of S.A.m 619-668.

—— Physiological plant studies in South Africa. Part 2. Transpiration of grasses and other plants

under arid conditions. 11th and 12th Repts. of the Dir. of Vet. Ed. and Res., U. of S.A.,

671-702.

— Growth of veld plants under the arid conditions of Bechuanaland. S. Afr. J. Sci. 23: 325-339.

1927. Phosphormangel als Ursache von Stérungen in Leben der Pflanze. Verhandlungen der

Naturforschenden Gesellschaft, Basle. 38: 316-326.

1928. The relations between the amount of carbohydrates in the leaves of Armoedsvlakte grasses

and the meteorological factors. 13th and 14th Repts. of the Dir. of Vet. Ed. and Res., U. of

S.A., 1041-1074.

—— The phosphorus content of the grasses of Bechuanaland in the course of their development.

13th and 14th Repts. of the Dir. of Vet. Ed. and Res., U. of S.A., 1077-1208.

— Woolly finger grass in Bechuanaland. Fmg. S. Afr. 3: 711-712. i

—— Pflanzenphysiologische Probleme aus Siidafrika. Sitzungsberichte der Bernischen Gesellschaft,

Berne.

1929. The structure of the cortex of grass roots in the more arid regions of South Africa. Sci. Bull.

No. 85 U. of S. Afr. Dept. Agric. Pretoria: Govt. Printer.

1930. Mineral and feeding-stuff analyses of grasses of the Eastern Transvaal Highveld. 16th Rept.

of the Dir. of Vet. Ser. and An. Ind., U. of S.A., 421-434.

—— The phosphorus content of grasses in the Eastern Transvaal. 16th Rept. of the Dir. of Vet.

Ser. and An. Ind., U. of S.A., 435-199.

—— Phosphorus content of grassveld. Fmg. S. Afr. 5: 29-30.

1931. Grass or bush in the Karroid area. S. Afr. J. Sci. 28: 266. :

1932. Cystine and sulphur content of bushes and grasses in a Karroid area (Fauresmith). 187h Rept.

of the Dir. of Vet. Ser. and An. Ind., U. of S.A., 579-596,

506

South African Pastures: Their mineral and protein content. Fmg. S. Afr. 7: 245-248.

Namaqualand’s sheep pastures. Fmg. S. Afr. 7: 5-8.

An investigation of the content of phosphorous, calcium and protein of grasses in coastal

regions of Natal and Southern Zululand. Sci. Bull. No. 115 U. of S.A., Dept. Agriculture,

Pretoria: Govt. Printer.

Transpiration of water supply of South African plants. S. Afr. J. Sci. 34: 61-72.

Is Cystine a “Building Stone” of wool, performed in plants? Fig. S. Afr. 7: 37.

ene and feeding-stuff analyses of grass-clover experiments. Fig. S. Afr. 6: 481-483,

1-503, 526.

Propagation of Karoo bushes. Fmg. S. Afr. 8: 435-436.

(with Potter, P. E.). An investigation of pastures in the eastern province and Albany district.

Sci. Bull. No. 134 Union of S. Afr. Dept. Agric. Pretoria: Govt. Printer.

Carrying capacity and tests of palatability of Karroo bushes. Fmg. S. Afr. 9: 53-55.

Pasture deficiencies in minerals and nitrogen. Effect on plants and animals. Fg. S. Afr.

9: 181-183.

The Veld Reserve at Fauresmith. Fmg. S. Afr. 9: 400-402.

(with Potter, P. E. and Pont, J. W.). Fodder plants of the Broken Veld. Part 1. Sci. Bull. No.

142 U. of S. Afr. Dept. Agric. Pretoria: Govt. Printer.

Germination of the Karoo bush seeds. Part 1. S. Afr. J. Sci. 32: 223-234.

Transpiration and water supply of South African plants. S. Afr. J. Sci. 34: 61-72.

Karoo veld. Handbook for Farmers in S. Afr. 3rd enlarged ed., Govt. Printer, Pretoria.

384-389.

Good Karoo plants. Handbook for Farmers in S. Afr. 3rd enlarged ed., Govt. Printer,

Pretoria. 389-396.

Some physiological aspects of the genus Tribulus. Onderstepoort J. Vet. Sci. Anim. Ind.

10: 367-392

Germination of Karoo bush seeds. Part 2. S. Afr. J. Sci. 36: 212-219.

Fodder plants of the Broken Veld. Part 2. Sci. Bull. No. 213 Union S. Afr. Dept. Agric.

Pretoria: Govt. Printer.

The transpiration of different plant associations in South Africa. Part 1. Transpiration of

Karoo bushes. Sci. Bull. No. 185 Union S. Afr. Dept. Agric. & For. Pretoria. Govt. Printer.

Survey of Fodder plants in the Clanwilliam—Piquetberg Sandveld. Survey of the Tierhoek

Reserve, Calvinia. Survey of the proposed Reserve Rodewerf, between Ceres and Calvinia.

Roneo-ed.

(with Steyn, D. G. and Adelaar, T.). Poisonous plants in relation to animal industry in

Union of S. Afr. Communication No. C (K) 8. 1-3. African Regional Scientific Conference.

Johannesburg 1949.

Transpiration of large Karoo bushes. S. Afr. J. Sci. 37: 156-163.

Data on the sugar and starch content of some fodder plants under different physiological

conditions. S. Afr. J. Sci. 40: 157-161.

On the possibility of using sodium selenite for seed testing of Karoo bushes. S. Afr. J. Sci.

39: 152-154.

Transpiration of grasses in the sour mountain grassveld of the Drakensberg in comparison

with the water loss of indigenous forests. S. Afr. J. Sci. 39: 155-163.

The effect of wilting on the direct assimilates of lucerne and other fodder plants. S. Afr. J.

Sci. 41: 204-212.

Digestion experiments with fresh Karoo plants. S. Afr. J. Sci. 41: 213-217.

Effect of excessive water loss and wilting on the life of plants, with special reference to Karoo

plants and lucerne. Sci. Bull. No. 256 Union of S. Afr. Dept. Agric. Pretoria: Govt. Printer.

Transpiration of different plant associations in South Africa. Part 3. Indigenous and exotic

trees in the Drakensberg area. Sci. Bull. No. 247 Union of S. Afr. Dept. Agric. Pretoria:

Govt. Printer.

Transpiration of different plant associations in South Africa. Part 4. Parkland, forest and

sour mountain grassveld, large Karoo bushes. Sci. Bull. No. 244 Union of S. Afr. Dept.

Agric. & For. Pretoria: Govt. Printer.

The use and misuse of shrubs and trees as fodder. Jt. Publ. Commonw. Agric. Bur. 10: 57-93.

Tribulus terrestris, content of assimilates, glucoside and nitrates under different edaphic

conditions. S. Afr. J. Sci. 43: 195-202.

Transpiration of South African plant associations. Part 2. Indigenous and exotic trees under

semi-arid conditions. Sci. Bull. No. 248 Union of S. Afr. Pretoria: Govt. Printer.

Transpiration of different plant associations in South Africa. Part 3. Indigenous and exotic

trees under semi-arid conditions. Sci. Bull. No. 248 Union of S. Afr. Dept. Agric. & For.

Govt. Printer: Pretoria.

Transpiration studies. S. Afr. Sci. 2: 36-38.

Euphorbia in south west Free State. Afr. Wild Life. 2: 77-80.

The effect of grazing and cutting on Pentzia incana during drought. S. Afr. J. Sci. 44: 120-121.

The carbohydrate content of lucerne under different meteorological and physiological

conditions, Onderstepoort J. Vet. Sci. Anim. Ind. 22: 373-413.

507

Hoe ’n ooreiste Karooplaas herstel kan word. (Nauwpoort) Ldb. Weekbl. 10 Aug.

Nitrate accumulation in Tribulus terrestris. S. Afr. J. Sci. 46: 197.

The effect of cutting and grazing on Pentzia incana under different systems of veld manage-

ment. Bull. No. 292 Union of S. Afr. Dept. Agric. Pretoria: Govt. Printer.

The seasonal changes of soluble nitrogen, particularly nitrate in Tribulus terrestris on lime-

stone formation. S. Afr. J. Sci. 47: 212.

Comparative study of the content of starch and sugars of Tribulus terrestris, lucerne, some

Gramineae and Pentzia incana under different meteorological, edaphic and physiological

conditions. Paper No. 2. Carbohydrate nutrition. Onderstepoort J. Vet. Res. 25: 45—92.

Some micro-elements in Karoo plants. S. Afr. J. Sci. 49: 331.

The water soluble nitrogen of Tribulus terrestris. S. Afr. J. Sci. 48: 243.

Karroid and Karoo veld as nutrition to farm animals. Matveg. 1: 5-16.

Further nutrition studies on Tribulus terrestris. Sci. Bull. No. 348, Union of S. Afr. Dept.

Agric. Pretoria: Govt. Printer.

Influence of molybdenum and manganese on the reducase of Karoo plants. S. Afr. J. Sci.

50: 303.

Micro-elements in Karoo plants. S. Afr. J. Sci. 49: 331.

Temperatures of Karoo plants. S. Afr. J. Sci. 51: 245-248.

Temperatures of Karoo plants. S. Afr. J. Sci. 54: 119-122.

Bothalia, 10, 4: 509-516

The Genus Pithomyces in South Africa

W. F. O. Marasas* and Ingrid H. Schumann*

ABSTRACT

Descriptions are given of South African isolates of Pithomyces sacchari (Speg.) M. B. Ellis,

Pithomyces chartarum (Berk. & Curt.) M. B. Ellis and Pithomyces karoo Marasas & Schumann,

sp. nov. P. sacchari and P. chartarum were isolated from Medicago sativa L. seed. P. chartarum was

also isolated from dead leaves of Lolium perenne L. and Sporobolus capensis (Willd.) Kunth. plants

from artificial pastures in the eastern Cape Province. P. karoo was isolated from stems of Gnidia

polycephala (C.A. Mey.) Gilg and Rhigozum trichotomum Burch. from the Karoo, Cape Province

and from Avena sativa L. stubble collected in the Orange Free State.

During the course of an investigation of the seedborne mycoflora of South

African lucerne seed, Pithomyces sacchari (Speg.) M. B. Ellis and P. chartarum

(Berk. & Curt.) M. B. Ellis were isolated from Medicago sativa L. seed produced in

the Cape Province. P. chartarum and a new species, P. karoo Marasas & Schumann

sp. nov., were isolated during the course of a mycological examination of several

photosensitization syndromes in sheep in the Cape Province and Orange Free State.

Isolations were made by placing seeds or small pieces of plant tissue without

surface sterilization on 14% malt extract agar containing 100 mg/1 of sodium novo-

biocin and incubating at 25° C. Morphological descriptions were prepared of cultures

incubated on 14% malt extract agar (malt extract: 15 g; agar: 17 g; distilled water:

1 1) and potato-carrot agar (potatoes: 20 g; carrots: 20 g; agar: 17 g; distilled water:

1 1). All spore measurements are based on spores mounted in lactophenol and were

made with the aid of an oil-immersion lens at a magnification of 1250.

A key to and morphological descriptions of the South African species of Pitho-

myces are given below.

A. Conidia produced on potato-carrot agar and 14% malt extract agar at

25° C predominantly smooth-walled or almost so, transverse septa 0-2-5,

longitudinal septa 0-0-2, 9-15-28 « 4-6-10p.............. P. sacchari

Conidia produced on potato-carrot agar and 14% malt extract agar at

gn Cadehimitelysrough=walledemncr pitas scr r eee nnn ae Eo B

B. Conidia verruculose, transverse septa 0-3-5, longitudinal septa 0-2-3,

A= 203 Om aye SID bh ee techies. een ed oe etn ee SNS P. chartarum

Conidia coarsely verrucose, transverse septa 0-2-5, longitudinal septa 0-1-3,

Op Daa ile Ce LESH] Oe rales ats cs oct iter cas eerie nnnEs tee here Oey P. karoo

1. Pithomyces sacchari (Speg.) M. B. Ellis in Mycol. Papers 76: 17 (1960);

Hughes, Mycol. Papers 50: 70 (1953); Lakshminarasimhan & Rama Rao, Curr.

Sci. 38: 74 (1969).

Figures: 1, 2.

Colonies on potato-carrot agar at 25° C are flatly appressed with little aerial

mycelium, white, beginning to sporulate after 5 days and becoming tinged black

with spores, particularly in the centre. On 14°% malt extract agar at 25° Ccolonies are

woolly, white to olive-grey with the reverse smoky-grey to black, beginning to sporu-

late after 4 weeks. Vegetative mycelium composed of hyaline, smooth or coarsely

verrucose, branching, septate hyphae, 1-3 diam., and pale brown, smooth or verru-

cose hyphae, 4-7 diam., often forming strands. Conidiophores arise laterally on

the aerial mycelium at right angles to the parent hyphae, peg-like or cylindrical,

hyaline to subhyaline, thin-walled, straight or curved, non-septate, 2-6 « 1,5-2,0u,

* Plant Protection Research Institute, Department of Agricultural Technical Services, Private Bag

X134, Pretoria.

510

frequently densely clustered and bearing conidia in sporodochium-like masses,

following spore dispersal the remains of the conidiophores appear as denticles along

the hyphae (Fig. 2). Conidia arise singly as blown-out ends of each conidiophore,

obovoid or clavate, brown, smooth-walled when immature and at maturity, rarely

slightly roughened, with 1-5 transverse and 0-2 longitudinal septa, mostly 2 trens-

verse and 0 longitudinal septa (Fig. 1; Table 1), slightly constricted at the septa

9-28 « 4-10 (Table 2). Each conidium bears a short, hyaline basal frill which is

the upper part of the conidiophore (Fig. 1).

Specimens examined: Cultures on 14° malt extract agar and potato-carrot

agar: PRE 44584, PRE 44585, PRE 44586 (Mycological Herbarium), isolated from

Medicago sativa L. seed, Oudtshoorn, Cape Province, 1970.

Considerable difficulty was experienced in assigning these three South African

isolates to a species. The spores are very similar to those of P. sacchari (Speg.) M. B.

Ellis as described by Ellis (1960), and to those of Sporidesmium bakeri Syd. var. 1

as illustrated by Hughes (1953). The latter variety was considered a synonym of P.

sacchari by Ellis (1960). The immature and mature conidia of the South African

isolates are, however, predominantly smooth-walled on the agar media used and

in this respect most closely resemble the conidia of P. graminicola Roy & Rai as

described by Roy & Rai (1968). An examination of the type collection of P. gramini-

cola (IMI 126508, on Saccharum minja, Banaras Hindu University Campus, India)

revealed that the conidia of this fungus are smooth-walled, pyriform to clavate with

0-2 transverse and no longitudinal septa, 7-12 « 4-6u (mostly 9 x 5u) (Fig. 3).

These spore characteristics are quite different from those of the three South African

isolates (Fig. 1; Table 1, 2).

Dr. M. B. Ellis of the Commonwealth Mycological Institute, Kew, England,

examined isolate PRE 44585 (— IMI 153452) and found that it produces: “short,

obovoid, clavate and pyriform conidia with rough walls and longitudinal as well

as transverse septa after 2 weeks on oat agar and I take this to be Pithomyces sac-

chari” (M. B. Ellis, personal communication).

P. sacchari has previously been recorded as a saprophyte on various plants

from a number of countries (Hughes, 1953; Ellis, 1960) and has also been reported

on leaves of Saccharum officinarum in Jamaica (Hudson, 1962) and in soil in India

(Lakshminarasimhan & Rama Rao, 1969).

This is the first record of the occurrence of P. sacchari in the Republic of South

Africa.

2. Pithomyces chartarum (Berk. & Curt.) M. B. Ellis in Mycol. Papers 76: 13

(1960); Hughes, Mycol. Papers 50: 66 (1953); Dingley, N.Z.J. Agr. Res. 5: 49 (1962).

Figure: 4

Colonies on potato-carrot agar at 25° C are flatly appressed with little aerial

mycelium, white, beginning to sporulate after one to two weeks and becoming tinged

black with spores. On 14% malt extract agar at 25° C colonies are woolly or floccose,

olive-grey in the centre with a white margin and a smoky-grey to black reverse.

beginning to sporulate after one to four weeks, becoming black with a crust of spores,

often giving rise to sectors which are more densely floccose than the rest of the colony

and sporulate more heavily. Vegetative mycelium composed of thin-walled, hyaline,

septate, smooth or verrucose, branching hyphae, 2-5 diam., and dark-brown, smooth

or verrucose, septate hyphae, 4-7 diam., which may give rise to chains of verrucose,

one-celled, dark-brown, intercalary chlamydospores, 10-20 x 8-18. Conidiophores

arise laterally on the aerial mycelium at right angles to the parent hyphae, hyaline,

thin-walled, straight or curved, non-septate or one-septate, 3-15 x 2-3y, frequently

clustered and bearing conidia in sporodochium-like masses. Conidia arise singly

as blown-out ends of each conidiophore, at first fusoid, coarsely echinulate, hyaline

Silt

to pale brown, one-celled, becoming muriform and dark-brown. Mature conidia

broadly ellipsoidal, rarely obovoid or clavate, dark-brown, verrucose, slightly con-

stricted at the septa, with 2—5 transverse and 0-3 longitudinal septa, usually 3 trans-

verse and 1-2 longitudinal septa in the central cells (Fig. 4; Table 1), 14-36 « 8-21

(Table 2). Conidia become detached through fracture of the conidiophore wall and

consequently each conidium characteristically bears a short, hyaline basal frill

which is the upper part of the conidiophore (Fig. 4).

Specimens examined: Cultures on 14% malt extract agar and potato-carrot

agar: PRE 44581 (Mycological Herbarium), isolated from dead leaves of Sporobolus

capensis (Willd.) Kunth., Robbehoek, Humansdorp Distr., Cape Province, September

1970, Marasas OP-9; PRE 44599, isolated from dead leaves of Lolium perenne L.,

Keokama, Humansdorp Distr., September 1970, Marasas OP-10; PRE 44582,

isolated from Medicago sativa L. seed, Ladismith, Cape Province, June 1970, Marasas

306; PRE 44584, isolated from M. sativa seed, Oudtshoorn, Cape Province, July

1970, Marasas 366.

P. chartarum has previously been recorded as a saprophyte on paper and dead

leaves and stems of many different plants from a number of temperate, sub-tropical

and tropical countries, including Malawi, Rhodesia and Zambia (Hughes, 1953;

Morris, 1956; Ellis, 1960; John, 1963; Sahni, 1966). This fungus has also been reported

from Australia (Hore, 1960) and New Zealand (Dingley, 1962) where it is known

to cause facial eczema in sheep; from the air on spore traps in Britain (Lacey &

Gregory, 1962; Gregory & Lacey, 1964; Pawsey, 1964); from soil in Honduras

(Goos, 1964) and Ontario (Barron, 1968); from wheat grains in Australia (Shipton

& Chambers, 1966); from groundnuts and grasses in Texas (Taber, Pettit, Taber

& Dollahite, 1968) and from human foodstuffs in Japan (Udagawa, Ichinoe &

Kurata, 1970).

This is the first record of the occurrence of P. chartarum in the Republic of

South Africa.

3. Pithomyces karoo Marasas & Schumann sp. nov.

Figures: 5, 6, 7, 8, 9, 10.

Hyphae ramosae, septatae, hyalinae vel brunneae, leves vel verrucosae, 2-6.

crassae. Chlamydosporae intercalares vel terminales, solitariae vel catenulatae,

uni- vel bi-cellulares, leves vel verrucosae, fuscae, crassitunicatae, 5-15 = 5—10p.

Conidiophora singula ex apice lateribusque hypharum oriunda, simplicia, continua

vel septata, cylindrica, recta vel flexuosa, hyalina, 3-25 x 2-4. Conidia singula

in apice conidiophori oriunda, ellipsoidea, obovoidea, obpyriformia, clavata vel

sarcinaeformia, brunnea vel fusca, verrucosa, septis constricta, septis 0-5 transver-

salibus et 0-3 longitudinalibus praedita, 10-41 = 9-20p.

Colonies on potato-carrot agar at 25° C flatly appressed with little or no aerial

mycelium, white, beginning to sporulate within 3 days and becoming tinged black

with spores, particularly in the centre; at the edge of the petri dish old colonies

sometimes develop a ring of downy white aerial mycelium or a brown ring of immersed

spores that differ morphologically from the spores produced on the aerial mycelium.

On 14% malt extract agar at 25° C colonies are slow-growing, woolly or velvety,

aerial mycelium in the centre Greyish Olive (Ridgway, Plate XLVI), beginning to

sporulate within 3 days and becoming black in the centre with masses of spores

that obscure the aerial mycelium; margin appressed with little or no aerial mycelium,

white to Light Greyish Olive (Ridgway, Plate XLVI); colony reverse Cinnamon-Buff

(Ridgway, Plate XXIX) to Vinaceous-Russet (Ridgway, Plate XXVIII) in the centre

surrounded by a Light Yellowish Olive (Ridgway, Plate XXX) zone that becomes

almost black in age and a white margin; young colonies of some single spore isolates

produce an exudate that stains the surrounding agar Vinaceous-Russet (Ridgway,

Plate XXVIII), but this exudate is not evident in old colonies. Vegetative mycelium

512

composed of several types of hyphae: hyaline, thin-walled, septate, branching,

smooth or verruculose hyphae that are occasionally coarsely warted, 2-6u diam.,

and brown, multiseptate, smooth or verruculose hyphae that are often aggregated

into strands, composed of cylindrical or ellipsoidal cells, 5-15 x 3-6; the hyaline

as well as brown hyphae occasionally give rise to intercalary or terminal, one- or

two-celled, smooth or verruculose, brown, thick-walled, ellipsoidal or ovoid to

subglobose chlamydospores that are solitary or produced in short chains, 5-15

5—10u (Fig. 10). Conidiophores arise laterally on the aerial mycelium at right angles

to the parent hyphae, hyaline, thin-walled, straight or curved, usually non-septate

but occasionally with one or two septa, short and cylindrical, 3-25 x 2-4 (Fig. 6);

occasionally long, flexuous, septate, hyaline, thin-walled hyphae indistinguishable

from vegetative hyphae, terminate in conidia (Fig. 7); following spore dispersal the

remains of the conidiophores appear as denticles along the hyphae. Conidia aleurio-

spores that arise singly as blown-out ends of each conidiophore, at first fusoid,

hyaline to straw-coloured, coarsely verrucose to warted, one-celled, becoming muri-

form and dark brown (Fig. 6). Mature conidia mostly ellipsoidal with broadly

rounded ends occasionally obovoid, obpyriform, clavate, cruciform, sarciniform or

curved, dark brown, thick-walled, coarsely verrucose to warted with obtusely rounded

warts up to 2u long and 2u wide at the base, conspicuously constricted at the septa,

with 0-5 transverse and 0-3 longitudinal septa, occasionally sarciniform with nume-

rous transverse, longitudinal and oblique septa, mostly with 2 transverse and 1 or

2 longitudinal septa in the central and/or terminal cells (Figs. 5, 8; Table 1), 10-41 x

9-20u, mostly 20-24 x 15-l6m, including warts, width measured in the broadest

part (Table 2); conidia become detached through fracture of the conidiophore wall

and consequently each conidium bears a short, hyaline basal frill which is the upper

part of the conidiophore (Fig. 5). Jmmersed conidia are produced in the agar on

slender, thin-walled, hyaline conidiophores that are poorly differentiated from the

thin-walled, hyaline, immersed hyphae, 5-15 « 1,5-2,5u; hyaline, subhyaline to

light brown, thick-walled, smooth or verruculose, ellipsoidal with broadly rounded

ends or obovoid, obpyriform or cruciform, one- to four-celled, occasionally sarcini-

form and multi-celled, mostly two- or three-celled, 10-20 x 10-15 (Fig. 9).

Specimens examined: Cultures on 14% malt extract agar and potato-carrot

agar: PRE 44605 (Holotype), isolated from dead stems of Gnidia polycephala (C.A.

Mey.) Gilg, Riekertsfontein, Colesberg Distr., Cape Province, February 1970, Mara-

sas OP-11; PRE 44606, isolated from Avena sativa L. stubble, Rusthof, Heilbron

Distr., Orange Free State, March 1971, Marasas OP-12; PRE 44607, isolated from

living stems of Rhigozum trichotomum Burch., heavily infested by scale insects and

a sooty mould, Welverdiend, Hopetown Distr., Cape Province, March 1971, Marasas

OP-13.

Type specimens in the form of dried-down cultures on agar have been deposited

in the National Herbarium, PRE (Mycological Herbarium), Department of Agricul-

tural Technical Services, P.O. Box 994, Pretoria, South Africa. A culture of the type

strain (PRE 44605) has also been deposited in the Commonwealth Mycological

Institute, Kew, England, under accession number IMI 155881.

P. karoo differs from P. chartarum in having coarsely verrucose conidia that

predominantly have two transverse septa and are much more irregular in shape.

The morphologically dissimilar immersed conidia found in cultures of P. karoo on

PCA were not found in cultures of P. chartarum. The significance of these immersed

conidia can not be interpreted at present. Marked differences between the spores of

Trichocladium opacum (Corda) Hughes produced on aerial hyphae and those on

immersed mycelium were also noted by Kendrick & Bhatt (1966).

A study is in progress to determine the effects of temperature on growth, sporu-

lation and spore morphology of the South African Pithomyces isolates.

Fic. 1-2,—Pithomyces sacchari. Fig. 1, mature conidia. Fig. 2, denticulate hypha (< 1 000).

Fic. 3.—Pithomyces graminicola (Type), mature conidia (x 1 000).

Fic. 4.—Pithomyces chartarum, mature conidia (= 1 000).

Fic. 5-10.—Pithomyces karoo (Type). Fig. 5, mature conidia. Fig. 6, immature conidium on hyaline

dium. Fig, 8, mature conidium with

conidiophore. Fig, 7, aerial hypha terminating in a coni

a longitudinal septum in the terminal cell. Fig. 9, immersed conidium. Fig. 10, intercalary

chlamydospore (* 1 000).

514

TABLE |.—Spore septation of Pithomyces species®.

Species & isolate (PRE)

Septation (% of spores?)

Potato-carrot agar

14% malt extract agar

Transverse Longitudinal Transverse Longitudinal

VBSseeae Liisa: bar sons} Q it As Wea shed oS QM i 2 gs

IPSICRAnLGnura 4599 eee eee erie OP ASS 96 ONION 145530) S65 On| Om OSLOO MOM ON LOS e560)

I, QHERHNE COST 3s cogacaaudonsodovou0n | 0 O 70 24 6 8 30 62 0 0 20 70 8 2 6 44 48 2

PA SaACChHarita 45 84 een onesie 8 60 26 4 2) 74 20 6 0 No spores No spores

IP, GHAI CSRS. 5 asacobodconoucoova0ne 20 44 30 6 0/}92 8 0O 0 No spores No spores

PK Gr OOx4 460 Gre erie cr ee er eee 40 50 10 0 0O/| 14 52 34 0} 26 56 12 4 2] 20 52 26 2

Piikaroo 4460 Ticker ee oo eee 36 48 12 4 0 6 46 46 2) 32 60 6 O 2] 18 40 38 4

@ Cultures incubated at 25° C for 21 days.

b Values based on 50 spores mounted in lactophenol.

TABLE 2.—Spore measurements of Pithomyces species®

| Spore size (uw)

if aaa = r= aL — 5 igs her =

Species & isolate (PRE) | Potato-carrot agar | 14% malt extract agar

Range Average Mean | Range Average Mean

IPNCHGNIGTUITESAS Sileee er rierte err 16-25 x 8-14 20,1x11,1 20x11 18-26 10-13 20,3x11,2 20x12

(IPRICHOYLQFUNTASS 99 ernie entrees 14-25 x 8-15 VOW Pals WP 15-24 x 9-14 19,9x11,7 19x12

IPMICNGHIALUPMAS Oo eee 17—26 x 9-15 AYA SNASS. phos 17) No spores

P. chartarum 44583................ | 19-30 x 9-16 22,6X12,8 21x12) 18-36 12-21 25,9x16,8 25x16

PSA CCHATHAAS SAME reine rens 10-22 x 5-10 If 2O55) USS No spores

P. sacchari 44585................. 9-28 x 4-8 16,8x5,8 14x6 | No spores

IPSACCHALLAAS SO One ee tieeiiets 9-25 x 5-8 15,1 x6,1 166 No spores

PN Karoor4460 seen eerie 10-27 x 9-17 19,8x12,9 20x15 16-31X12-19 22,3x14,3 23x16

I GAO OOD. cacovcnnocvooounbn 13-31*11-17 19,9x14,8 22x15 | 13-36x11-19 23,5x15,1 24x15

II MOOD LNW occocodenaos000000c | 16-41x11-17 22,0x14,8 21x15 | 13-41 x 10-20 21,1x15,0 21x15

* Cultures incubated at 25° C for 21 days.

b Values based on 50 spores mounted in lactophenol.

ACKNOWLEDGEMENTS

The authors are deeply indebted to Dr. M. B. Ellis, Commonwealth Mycological

Institute, for confirming the identity of the species of Pithomyces and for critically

reading the manuscript; to the Director, Commonwealth Mycological Institute, for

the loan of the type specimen of P. graminicola; and to Dr. G. C. A. van der Westhui-

zen, Plant Protection Research Institute, for unfailing interest and advice.

S15)

REFERENCES

BARRON, G. L., 1968. The Genera of Hyphomycetes from soil. Williams and Wilkins, Baltimore.

DINGLEY, JOAN M., 1962. Pithomyces chartarum, its occurrence, morphology and taxonomy. N.Z.J.

Agr. Res. 5: 49-61.

E.us, M. B., 1960. Dematiaceous Hyphomycetes: I. Mycol. Papers 76: 1-36.

Goos, R. D., 1964. Further observations on soil fungi in Honduras. Mycologia 55: 142-150.

Grecory, P. H. & Lacey, M. E., 1964. The discovery of Pithomyces chartarum in Britain. Brit.

Mycol. Soc. Trans. 47: 25-30.

Hore, D. E., 1960. Facial eczema. Austr. Vet. J. 36: 172-176.

Hupson, H. J., 1962. Succession of micro-fungi on ageing leaves of Saccharum officinarum. Brit.

Mycol. Soc. Trans. 45: 395-423.

HuaGues, S. J., 1953. Fungi from the Gold Coast II. Mycol. Papers 50: 1-104.

Joun, K. P., 1963. Rhizoctonia and other diseases of cover plants. R.R.J. Plant Bull. 68: 187-190

(cited in Rev. Appl. Mycol. 43: 376, 1964).

KENDRICK, W. B. & BHATT, G. C., 1966. Trichocladium opacum. Can. J. Bot. 44: 1728-1730.

Lacey, M. E. & Grecory, P. H., 1962. Occurrence in Britain of the fungus causing facial eczema

in sheep. Nature 193: 85.

LAKSHMINARASIMHAN, A. V. & RAMA RAO, P., 1969. Three new records of fungi from soils. Curr.

Sci. 38: 73-74.

Morris, E. F., 1956. Tropical Fungi Imperfecti. Mycologia 48: 728-737.

Pawsey, R. G., 1964. An investigation of the spore population of the air at Nottingham. II. The

results obtained with a Hirst spore trap, June-July 1956. Brit. Mycol. Soc. Trans. 47: 357-363.

RipGway, R., 1912. Color standards and color nomenclature. Hoen, Baltimore.

Roy, R. Y. & Rat, B., 1968. New species of Lacellina and Pithomyces. Brit. Mycol. Soc. Trans. 51:

152-155.

SAHNI, V. P., 1966. Deuteromycetes from Jabalpur. II. Mycopathol. Mycol. Appl. 29: 226-244.

SHIPTON, W. A. & CHAMBERS, S. C., 1966. The internal microflora of wheat grains in Western

Australia. Austr. J. Exp. Agr. Anim. Hush. 6: 432-436.

Taser, RutH, A., Perit, R. E., TABER, W. A. & Dovvanite, J. W., 1968. Isolation of Pithomyces

chartarum in Texas. Mycologia 60: 727-730.

Upacawa, S. I., IcHinor, M. & Kurata, H., 1970. Occurrence and distribution of mycotoxin

producers in Japanese foods, p. 174-184. In: M. Hertzberg (ed.) Proc. First U.S.-Japan Con-

ference on Toxic Microorganisms. Unnumbered Publ. U.S. Dept. of Interior and U.S.N.R.

Panels on Toxic Microorganisms, Washington, D.C.

73472—2

Sant fe }

a ces

aS ;

Fi ye ,

to naling rie iter

he STi . 9 7

Bothalia, 10, 4: 517-538.

Studies of Wood-rotting Fungi. II. Basidiomycetes

from the Wood-preservative Field Exposure Test

Plot at Kruisfontein

G. C. A. van der Westhuizen*

ABSTRACT

__ Ina survey of the fungi associated with decays of treated and untreated wood specimens partly

interred in a wood-preservative field exposure test plot, 13 species of Basidiomycetes were identified

{rom 120 test specimens. Basidiomycete mycelia isolated from a further 51 specimens could not be

identified due to the absence of fruit-bodies or lack of matching descriptions of cultural characters.

A further 179 did not yield Basidiomycete mycelia when cultured. Paxillus panuoides and Stereum

hirsutum were the most common species. Species of lower Hymenomycetes were more frequently

isolated than polypores.

The named species of fungi and unnamed mycelia are listed together with their hosts. Seven

species, viz. Neamatoloma fasciculare, Odontia bicolor, Paxillus panuoides, Peniophora aspera, Penio-

phora cinerea, Peniophora tenuis and Polyporus gilvus are described in pure culture.

INTRODUCTION

A plot for testing the efficacy of wood-preservatives against decay in the field

was established at Kruisfontein Plantation near Knysna, Cape Province, by the

Department of Forestry in 1946 as one of three field exposure test plots. Two others,

one at Pienaarsrivier, Transvaal and one in Durban Harbour, Natal, were established

for field exposure tests of wood-preservatives against termite attack and marine borers,

respectively.

The test plot at Kruisfontein consists of a rectangular fenced-in area, half an

acre (0,202 ha) in extent, situated on a slight rise with a northerly aspect and grey,

gritty loam soil, cleared of all trees and woody shrubs (Fig. 1). The plot is divided

into four quarters for experimental purposes and is laid out in exactly the same way

as the plot at Pienaarsrivier which had already been described by Coaton (1946) and

Krogh (1947).

The specimens for testing consist of treated billets of Eucalyptus saligna and

Pinus patula, 45 cm long and 7-10 cm in diameter, saplings of E. saligna of the same

length but about 3 cm in diameter, and 2,5 cm square stakes of Pinus patula wood.

Eight specimens of each type of stake and billet are treated and interred in the erect

position to a depth of about 25 cm so that a total of 32 specimens treated to a particular

loading of each type of preservative are under test simultaneously, eight specimens in

each quarter of the plot. In addition, billets of 50 x 5 x 2 cm of different species of

untreated wood, with separate billets for heartwood and sapwood, are interred along

the periphery of the plot to determine their natural durability. The layout of the

experiment and treatments of the specimens have been described by Krogh (1947).

This study was undertaken to determine the species of fungi which are associated

with decay of the specimens under test. It was furthermore attempted to determine

whether particular species of fungi are constantly associated with decayed specimens

which had been treated with any particular preservative. Any such association may

indicate the possible value of that fungus as a test organism for use in laboratory

tests of timber preservatives. Furthermore, the termite and marine borer fauna of the

respective field exposure test plots had been surveyed (Coaton, 1946; Krogh, 1958)

and it was thought desirable that a similar survey of the wood-decaying fungi of the

plot at Kruisfontein should be carried out.

* Plant Protection Research Institute, Department of Agricultural Technical Services, Private Bag

X134, Pretoria.

518

This survey was limited to the Basidiomycetes only. It has been shown by other

workers (Merrill & French, 1966; Kaarik, 1967) that Hyphomycetes play an important

role in the early colonization of untreated wood in the soil or in the detoxification

of certain wood-preservatives (Lyr, 1962; Madhosing, 1961). These organisms are,

however, replaced by Basidiomycetes which cause extensive white or brown rots

(Henningson, 1967a & b; Merrill & French, 1966) whilst some Basidiomycetes have

also been found among the early invaders of treated poles in contact with soil

(Kaarik, 1967). Because of the method used for collecting specimens for study,

however, the Hyphomycetes and Ascomycetes associated with the decayed specimens

had to be ignored for the purpose of this investigation.

In the present paper the species of Basidiomycetes that were identified, are listed

together with the test specimens on which they occurred. Unidentified mycelia

arranged in groups according to their cultural characters, are also listed together with

the specimens on which they occurred. Named isolates which have not been reported

in culture from South Africa before, are described and illustrated whilst the occurrence

and importance of other named species, previously described in culture from South

Africa, are discussed.

METHODS AND MATERIALS

The specimens under test were examined periodically as described by Krogh

(1947), usually in iate autumn. Ail specimens were examined visually for the presence

of fruit-bodies of decay fungi which would allow their identification. The specimens

were then removed from the soil and examined for signs of decay. Specimens of which

more than 50°% of the cross-sectional area was decayed, were regarded as failures and

discarded. These discarded specimens were collected for the isolation of fungi.

The specimens were split lengthwise by means of a circular saw to expose the

decayed areas. Small blocks of about 1 cm? were cut from the sound wood adjacent to

the decay. These blocks were then split by means of a chisel and surface sterilized

by dipping them briefly in 70% alcohol and flaming. After flaming they were placed

on 1,5°% Difco malt extract plates solidified with 1,5% Difco agar. These plates were

incubated at 24° C and fungal mycelium appearing on the blocks or plates was

transferred to fresh plates of the same medium for further growth.

When specimens could not be treated immediately upon receipt, large sections

of the specimens were cut out and placed on damp sterile sand in damp chambers

in order to keep the fungi alive for isolation later.

Fungi of which fruit-bodies were present on the specimens, were identified on the

characters of the fruit-bodies. Cultures were made by transferring small pieces of

tissue from freshly exposed, broken surfaces of the fruit-bodies by means of fine-

pointed, sterile forceps to plates of 1.5% Difco malt extract agar. These cultures

were used for comparison with other mycelia isolated from the specimens to determine

their identity.

Fungi isolated in pure culture only were grown on 1,5% Difco malt extract

agar in the dark at 24° C for six weeks and examined both macroscopically and

microscopically at weekly intervals according to the methods described by Nobles

(1948; 1965) and Van der Westhuizen (1958; 1971). Their characters in pure culture

were compared with existing descriptions of species of which these characters are

known in order to determine their identity.

The fungi were tested for the production of extra-cellular oxidase enzymes by

growing them on 1,5 °% Difco malt agar to which had been added 0,5 % tannic acid and

0,5°% of gallic acid respectively as described by Van der Westhuizen (1958).

Mycelia which could not be identified from descriptions of known species,

were inoculated onto a sawdust-maizemeal-malt extract mixture in order to induce

fructification on blocks of Pinus patula wood according to a method described by

Matters & Da Costa (1958).

StS

RESULTS

A total of 171 Basidiomycetes were found as fruit-bodies or isolated as mycelia

in pure culture from the 350 specimens of decayed wood examined. Of these fungi,

120 were identified and named from fruit-bodies on the specimens or from their

characters in pure culture. The named fungi as well as the specimens on which they

were found and their frequence of occurrence, are listed in Table 1.

It was soon evident, that very few fruit-bodies were to be seen on the specimens

during the annual inspections. Fruit-bodies of Stereum hirsutum were most abundant

whilst fruit-bodies of Peniophora cinerea, Peniophora setigera and Lenzites sepiaria

were occasionally seen. Most specimens, however, never showed fruit-bodies although

many of them were found to be almost completely decayed.

In the course of this investigation it also became evident that no decay fungi

were present in many of the decayed specimens despite repeated attempts to isolate

them in culture. This was especially noticeable in the 25 mm square Pinus patula

sticks. Many of these produced moulds only, when cultured, and Trichoderma viride

proved to be a most important component of the mould flora. In specimens with a

diffuse decay, this mould mostly grew out from the inoculum to the exclusion of all

others.

From the results in Table 1, it is evident that Odontia bicolor, Paxillus panuoides,

Peniophora tenuis and Stereum hirsutum are the species most frequently isolated from

these specimens. Neamatoloma fasciculare, Peniophora aspera, Peniophora cinerea,

Lenzites sepiaria and Schizophyllum commune were isolated less frequently. Among

the named isolates, species of Thelephoraceae and Agaricaceae were thus most

frequently isolated, rather than members of the Polyporaceae. Of these species,

Lenzites sepiaria and Paxillus panuoides were isolated from softwood specimens only

whilst Neamatoloma fasciculare, Polyporus gilvus, Schizophyllum commune and Stereum

hirsutum were found on hardwoods only. The other species occurred on both hard-

wood and softwood specimens.

Most of the fungi were isolated from untreated specimens. This is to be expected

since the lack of preservative treatment would allow their more rapid attack by fungi.

But Paxillus panuoides, the most frequently isolated species, occurred very frequently

on specimens which had been treated with preservatives. Since this species is known

to be very sensitive to timber preservatives (Cartwright & Findlay, 1958) it’s presence

on the treated specimens indicate that the different preservatives had been rendered

ineffective in the wood through leaching, detoxification or other similar causes during

the period of duration of the test.

Besides the fungi listed in Table 1, 51 more fungi of which the cultural characters

could not be matched with any existing descriptions, were isolated but their identity

could not be determined. These isolates, together with their culture numbers and

timber specimens on which they were found, are listed below in Table 2. The mycelia

are arranged in groups according to their major cultural characters as arranged by

Nobles (1958). a ig:

From the results presented in Table 2, it is clear that the majority of unidentified

fungi isolated from the specimens, in culture form undifferentiated nodose-septate

hyphae only. A larger number of these gave positive reactions when tested for extra-

cellular oxidase, than gave negative reactions. Only four cultures were found in which

fibre hyphae were present as well, whilst 8 cultures were found with nodose-septate

hyphae with irregularly thickened walls. Most of the remaining cultures did not display

any differentiated structures and few mycelia with swellings on the nodose-septate

hyphae or other special structures were present. Only one culture with simple-septate

advancing hyphae and clamped septa on the older hyphae, was found. ;

The majority of these unidentified mycelia were also isolated from specimens of

untreated timber, either as untreated controls or specimens tested for natural dura-

bility.

520

These cultures all displayed characters which allowed their inclusion into the

groups proposed by Nobles (1958). These characters however could not be matched

with existing descriptions of known species so that their identity could not be deter-

mined.

DESCRIPTIONS OF CULTURES

A number of decay fungi of which the cultural characters are not well known,

were found in this investigation. Some of them are not known to be widely distributed

in South Africa and the characters of the South African forms in culture have not

been described 5efore. For these reasons descriptions of their cultural characters are

given below.

Naematoloma fasciculare (Huds. ex Fr.) Karst.

Growth characters (Fig. 4)

Growth is moderately fast, the colony reaching a radius of 36 mm in 2 weeks and

covering the plate in 3 weeks. Advancing zone even, appressed for 1-2 mm, hyphae

then raised. Mat thin woolly, with small plumules of radiating hyphae, somewhat

farinaceous at the side of the dish. The plumulose areas develop into thin, felty,

rhizomorphic strands which merge and diverge to form an irregular, elongated

diamond-shaped pattern on the surface. The mat is white and remains so but specks of

“hazel” “pinkish buff” and “cinnamon buff” appear on the surface under the mat

after 2-3 weeks and gradually enlarge in size. The reverse bleaches slowly after about

2 weeks and a faint musty odour is given off. On gallic acid and tannic acid media,

strong diffusion zones are formed and colonies of 16 mm and 22 mm in diameter resp.

are formed after 7 days.

Hyphal characters.

Advancing hyphae: hyaline, more or less straight, branching, thin-walled, with deeply

staining contents, septate, with simple clamps at the septa, often branching opposite

the clamps, 2,0—4,0 » in diameter (Fig. 5).

Aerial mycelium: (1) hyphae as in the advancing zone; (ii) hyphae as in the advancing

zone but aggregated into rhizomorphic strands and often agglutinated by brown,

resinlike material (Fig. 7); (ii) dendrophyses yellowish, thin-walled, curved, with

short, lateral branches 4-6. long, arising at right angles from the curved part, slightly

widened and with deeply staining contents at first but later with dark-coloured,

slightly thickened walls and brown contents, up to 100 long (Fig. 6).

Submerged mycelium: hyphae as in the advancing zone.

Naematoloma fasciculare causes a white rot of hardwoods.

This fungus was described in culture by Zycha & Knopf (1966). The characteris-

tics displayed by the South African isolates, agree very well with their description.

The isolates in the present study were all very much alike displaying the strands or

fibrils of hyphae which give the mat the appearance of “‘a wet pelt” (Zycha & Knopf,

1966). The brown, widened, curved hyphal structures with short, lateral projections

(Fig. 6) designated as “‘dendrophyses” by Zycha & Knopf (1966), are very striking in

microscopic mounts and together with the characteristic appearance of the mat, are

useful diagnostic features for the recognition of cultures of this species.

N. fasciculare was isolated from hardwood specimens only on which it causes

white rot. Its association with white rot agrees with its positive reaction for extra-

cellular oxidase enzymes in culture as noted both here and by Zycha & Knopf (1966).

Doidge (1950) recorded this fungus on Pinus stumps and other unspecified species

of wood in the southernmost parts of the country and the present author found it also

on Eucalyptus in the vicinity of the testing plot. It has been recorded on various

conifers in Great Britain (Cartwright & Findlay, 1958) and is reported to be one of

the commonest fungi participating in the decomposition of oak roots in the U.S.S.R.

(Chastukhin & Nikolaevskaya, 1962).

521

Odontia bicolor (Alb. & Schw. ex Fr.) Bres.

Growth characters (Fig. 8)

Growth is moderately rapid to slow, the mat reaching a radius of up to 35 mm

after one week and covering the plate in three to six weeks. Advancing zone even to

slightly bayed, appressed for short distance then raised. Mat thin, white, appressed,

downy to fine woolly at first and with a combed appearance, later developing thin,

branching and anastomosing lines of more compact mycelium which radiate out

from the inoculum. After 2-3 weeks indefinite, concentric zones of raised mycelium,

20-30 mm wide, which end abruptly on zones of mycelium more closely appressed to

the agar, may develop in some isolates.

Reverse bleaching slowly, becoming milky white. No odour is emitted. On gallic

acid medium a strong diffusion zone of up to 40 mm in diameter is formed without

growth taking place. On tannic acid medium strong diffusion zones of up to 50 mm

and colonies of up to 40 mm in diameter are formed.

Hyphal characters.

Advancing mycelium: hyphae hyaline, thin-walled with deeply staining contents,

nodose-septate, branching often from the clamps, 2,0—5,0.. in diameter (Fig. 9).

Aerial mycelium: (i) hyphae as in the advancing mycelium, often aggregated into

thin strands; (11) cystidia numerous, each consisting of a short stalk up to 15 long,

arising as a lateral branch of a trailing hypha, terminally expanded into a subglobose

vesicle 5—10u in diameter, and surmounted by a cap of large jagged crystals, the cap

12-25, in diameter (Fig. 10); (ili) oidia present in some isolates, 3-10 x 1,5—3y.

Submerged mycelium: hyphae as in the advancing zone.

This fungus was isolated from 14 specimens of different species of hardwood, both

treated and untreated with preservative, and one of softwood. On all specimens it

caused a white, somewhat stringy rot.

In cultural characters, the South African isolates agree very well with the descrip-

tion of Odontia bicolor by Nobles (1953), and with two cultures of this species received

from Dr. J. H. Ginns, Ottawa. The texture of the mat, positive reaction for extra-

cellular oxidase enzymes, and the capitate cystidia with caps of large, angular crystals,

allow the easy recognition of this fungus in culture.

Nobles (1953) reported that O. bicolor causes a decay of considerable importance

in the heartwood of the butt and roots of a number of broad-leaved and coniferous

trees. In a survey by the Canada Department of Agriculture (1952) O. bicolor was

listed as one of the three important white rot fungi which caused 85% of the decay of

Abies lasiocarpa in the Prince George Forest District. This importance was emphazised

by the work of Smith (1963) who listed O. bicolor in association with root rot of Abies

spp. Basham & Morawski (1964) listed O. bicolor as one of 23 species which cause

91% of decay losses of timber species in Ontario, but considered it to be of economic

importance in balsam fir and spruce only. Duncan & Lombard (1965) listed O. bicolor

as one of the species frequently isolated from gymnospermous wood in the United

States. Harmsen (1967) described O. bicolor as one of the important fungi capable of

attacking structural timber treated with preservative.

Few previous records of the occurrence of O. bicolor in South Africa exist.

Doidge (1950) has no record of it but Talbot (1958) described two collections from

South Africa. In view of the frequent occurrence of O. bicolor on the test specimens

from Kruisfontein and the importance accorded to it in overseas reports, O. bicolor

must be regarded as a much more important agent of decay of timber in contact

with soil in South Africa than had been generally realized before.

Paxillus panuoides Fries.

Growth characters (Fig. 11)

Growth is slow to very slow, the mat reaching a radius of 30 mm after 3 weeks

while the plates are seldom covered at 6 weeks. The margin is even to bayed with

hyphae raised to the limit of growth. The mat is coarsely woolly, raised, consisting of

522

loosely intertwined rhizomorphic strands of mycelium radiating out from the inocu-

lum. Mat at first forming a yellowish ball on the inoculum but then grows out over

the agar as “‘light buff” to “cartridge buff” strands often with tinges of reddish

purple or violet developing on the inoculum.

The reverse darkens, the dark zone extending well beyond the mat into the agar

within one week after inoculation of the plates. No odour is emitted.

On gallic acid and tannic acid media dark diffusion zones are formed. Growth,

up to 5 mm in one week on gallic acid, none or trace only on tannic acid.

Hyphal! characters.

Advancing mycelium: hyphae hyaline, thin-walled, nodose-septate, branching sparingly

opposite the clamp connections, 1,5—5,0. in diameter (Fig. 12).

Aerial mycelium: hyphae as in the advancing zone, often aggregated into strands and

with walls frequently yellow (Fig. 13, 14).

Submerged mycelium: hyphae as in advancing mycelium but often wider, up to 7p

in diameter and more frequently branched.

Paxillus panuoides was associated with brown rot in stakes of Pinus patula,

some of which had been treated with wood preservatives.

Paxillus panuoides has been described in culture by Findlay (1932) and Siepmann

& Zycha (1968). The South African isolates described here agree very well with these

descriptions and with two cultures of this species obtained from Dr. Siepmann.

The South African isolates were grown in culture with some difficulty as they preferred

low incubation temperatures, 16—20° C, their growth being completeiy inhibited at

25° C while some growth still occurred at 10° C. This, together with the fact that

this species requires an acid medium (Findlay, 1932) made isolation possible only

after the species had been identified from a fruit-body which developed on a block

of wood cut from a decayed test specimen and kept in a damp chamber. Cultures of

this species may however be recognized quite readily by the woolly, dull yellow mat,

with tinges of violet near the inoculum.

The brown rot caused by P. panuoides is not in agreement with the positive

oxidase reaction shown by cultures of this fungus when grown on gallic acid and

tannic acid media. The reaction is weak on both media and the darkening in colour

of the media is possibly due to the brown pigment produced by the fungus diffusing

into the medium.

Paxillus panuoides is well-known as a cause of brown rot of coniferous wood

(Southam & Ehrlich, 1950; Cowling, 1957; Cartwright & Findlay, 1958; Duncan &

Lombard, 1965). The difficulty with which this fungus is isolated in pure culture,

also experienced by Siepmann & Zycha (1968), may have a negative influence on

observations of its importance as the cause of decay. In the present study it was

observed on 26 different test specimens but only five isolates were obtained in pure

culture, an experience which supports the previous remarks.

Paxillus panuoides was found most frequently of all species, named and unnamed,

on the test specimens. This in itself is surprising as Doidge (1950) lists only three

records of its occurrence in South Africa, none of them from the Knysna district.

Cartwright & Findlay (1958) state that this fungus is extremely sensitive to wood-

preservatives. Its presence on the specimens that had been treated with preservatives

indicates then that the concentration of these preservatives must have been reduced to

extremely low values in the specimens before they were attacked by this fungus.

Its frequent occurrence and constant association with extensive brown rot of these

test specimens, are contrary to Henningson’s (1967c) observation that fungi with

temperature optima below 25° C have low decay ability.

Peniophora aspera (Pers.) Sacc.

Growth characters (Fig. 15)

523

Growth is slow, the colony reaching a radius of 10 mm after one week, but does

not cover the plate after 6 weeks. Advancing zone even, mat thin, hyphae raised to

limit of growth. Mat at first downy, hyaline-white, gradually becoming slightly more

compact to thin woolly. After 3 weeks more compact, small patches of mycelium

appear, scattered over the older parts of the mat and gradually developing into more

compact lumps of mycelium.

The reverse remains unchanged and a faint mushroomy odour is given off.

No growth occurs on gallic acid and tannic acid media but small, weak diffusion

zones are formed after 7 days.

Hyphal characters.

Advancing mycelium: hyphae hyaline, simple or branching, nodose-septate with

large clamps at the septa, thin-walled, with deeply staining contents, 2,5—4,5y in

diameter (Fig. 16).

Aerial mycelium: (i) hyphae as in the advancing zone; (ii) chlamydospores hyaline,

thick-walled, globose or sub-globose, terminal 6,0-9,0u in diameter.

Submerged mycelium: hyphae as in the advancing mycelium.

Peniophora aspera had apparently not been described in culture before and,

despite its world-wide distribution, had not received much attention as a decay fungus.

In culture it displays no character which might distinguish it from the many other

species which form slow-growing, white mycelia lacking in distinguishing features.

Not one of the cultures examined produced the characteristic septate cystidia which

characterize fruit-bodies of this species (Slysh, 1960).

Isolation of this species from six specimens of treated and untreated wood,

indicate that it may be of more importance as a decay organism or detoxicating

agent of certain types of wood-preservative than had been suspected hitherto. This

view is supported to some extent by the report by Bergman, Nilson & Jerkeman (1970)

who found P. aspera as one of the white-rot fungi at test points in chip piles where

temperatures had not exceeded 40° C. In these piles the effect of P. aspera was not

very marked at the points of isolation. This was thought to be due to inhibition by

Trichoderma viride because laboratory tests had shown P. aspera to be capable of

causing high losses in dry mass of test blocks.

Peniophora cinerea (Fries) Cooke.

Growth characters (Fig. 17)

Growth is moderately rapid to rapid, the mat reaching a radius of up to 45 mm

in one week and covering the plate in 2-3 weeks. Margin even with the hyphae raised

to limit of growth. Mat woolly at first, white, with sectors of dense, more felty mycelium

becoming gradually more dense with age and remaining so or developing irregular,

scattered patches of dense, finely farinaceous mycelium over the surface and sides of

dish, white at first and remaining so, or, becoming ““warm buff” but soon changing to

“sayal brown’’, later darkening to “warm sepia’’ or “bister’’. Mycelium on the sides

of the dish form white, felty lumps which soon change colour to “warm buff” or

“pale ochraceous salmon” and enlarge, coalesce and gradually darken to “‘russet”’,

“warm sepia’, ‘“mars brown” or “‘bister’’, oozing droplets of dark, reddish-brown

liquid.

The reverse is bleached at first but darkens later due to the diffusion of a brown

pigment. No odour is given off. On gallic acid and tannic acid media, strong diffusion

zones up to 85 mm in diameter and colonies of up to 80 mm in diameter are formed in

one week.

Hyphal characters. ee

Advancing mycelium: hyphae hyaline, branching, thin-walled with deeply staining

contents, nodose-septate, 2,0—4,5. in diameter (Fig. 18).

524

Aerial mycelium: (i) hyphae as in the advancing mycelium; (ii) nodose-septate hyphae

with brown, thickened walls, often with swellings and short, lateral projections and

encased in drops of brown resin-like material which apparently bind them together in

the dark brown aerial parts of the mat (Fig. 19).

Submerged mycelium: hyphae as in the advancing mycelium.

Cultures of Peniophora cinerea lack special structures which may be of value in

establishing their identity. They lack the conical, thick-walled, heavily incrusted

cystidia which are present in the carpophores of this species (Slysh, 1960). The

brownish, nodose-septate hyphae embedded in droplets or sheaths of brown resin-like

material in the brown-coloured, felty patches of the otherwise white, woolly-felty

mat together with the rapid growth rate and strong positive reaction when tested for

extra-cellular oxidase enzymes, may however serve to distinguish cultures of P. cinerea

from those of otherwise similar species.

Peniophora cinerea is not well known from previous records of its occurrence in

South Africa. Doidge (1950) lists only three collections. In the present investigation

however it was recorded on ten specimens affected by white rot, which includes both

treated and untreated hardwoods and softwoods. Very little is known about its

importance as the cause of decay of wood however. It was not listed in the United

States by Cowling (1957) and Duncan & Lombard (1965) but Nilsson (1965) found

this species to be one of the important Basidiomycetes causing decay of birch chip

piles in Sweden. Its relatively frequent occurrence on test specimens from Kruisfontein,

may indicate that this fungus has more importance as a wood destroyer than had

been generally realized.

Peniophora tenuis (Pat.) Massee.

Growth characters (Fig. 20)

Growth moderately slow to slow, the mat reaching a radius of 30-45 mm after 2

weeks and covering the plates in 4 to 6 weeks. Advancing zone even, thin, appressed

with thin, sparse, radiating strands of hyphae. The mat is thin, downy at first and

somewhat farinaceous, white, with thin, sigmoid strands of mycelium radiating from

the inoculum towards the margin. The mat gradually thickens towards the inoculum

where it becomes thick, felty. Minute droplets of clear liquid appear on the mat

especially on the rhizomorphic strands. The mat gradually thickens with time.

The reverse is bleached but no odour is given off. On gallic and tannic acid

media no growth or a trace of growth takes place but fairly strong diffusion zones of

about 20 mm in diameter are formed on both media.

Hyphal characters.

Advancing mycelium: hyaline, branching thin-walled, nodose-septate, with simple

clamps at the septa, 2,0-5,0,. in diameter (Fig. 25).

Aerial mycelium: (1) hyphae as in the advancing mycelium; (i1) cystidia hyaline,

elongate-ovoid, to cylindrical (Fig. 22, 23); (ii) capitate cystidia globose 5-8u in

diameter, with deeply staining contents, pedicellate on clamped hyphae, lateral or

terminal (Fig. 21); (iv) stephanocysts ovoid, hyaline, two-celled with deeply staining

contents and with a row of minute spines along the median septum, 12-15 x 6-7p,

sessile on short lateral protuberances of clamped hyphae (Fig. 24).

Submerged mycelium: nodose-septate hyphae as in the advancing zone, often inflated

to up to 8 diameter.

Peniophora tenuis was isolated from 18 specimens of different species of hardwood

and softwood, which includes specimens both treated and untreated with preservative,

on which it caused white stringy rot.

Peniophora tenuis was described in culture by Boidin (1950) who figured and

named the stephanocysts which are also present in fruit-bodies of this species. Similar

structures have been reported from fruit-bodies of a few other species of Peniophora,

S25)

closely related to P. tenuis, by Boidin (1950; 1958), Cunningham (1963) and Burdsall

(1969). The stephanocysts described here agree closely with those of Peniophora

tenuis as described by Boidin (1950; 1958) and Burdsall (1969). In other characters,

the cultures described here also agree well with Boidin’s description and, as fruit-

bodies of Peniophora tenuis were present on some of the specimens from which

these isolations were made, there can be no doubt about the identity of this species.

The presence of the characteristic two-celled stephanocysts in the thin, white, felty

mycelium of cultures which give a positive reaction when tested for extra-cellular

oxidase enzymes, serves to distinguish this species in culture.

Stephanocysts identical to those described above, were reported by Burdsall

(1969) from cultures and carpophores of Hyphoderma tenue and from carpophores

only of H. guttuliferum and H. puberum. He also stated that the cystidia may be (i)

subulate, embedded and slightly thick-walled, to long subulate, or, (ii) cylindrical

and slightly thick-walled, and embedded, or, (ii1) cylindrical, thin-walled and pro-

truding beyond the hymenium. These three types may occur in the same carpophore

but none react with sulfobenzaldehyde. Species which possess stephanocysts are

included in the genus Hyphoderma Wallr. emend. Donk by Parmasto (1968).

Little is known about the ability of Peniophora tenuis to decay the wood in which it

grows. Harmsen (1967) listed this species as one of the Corticiaceae capable of

breaking down timber treated with wood preservatives. The frequent isolation of

this species in the present study together with Harmsen’s (1967) report indicates the

importance of Peniophora tenuis in the earlier stage of the decay of timber.

Polyporus gilvus Schw. ex Fr.

Growth characters (Fig. 26).

Growth is moderately rapid to slow, the mat reaching a radius of up to 20 mm

in one week and covering the plate in 3-5 weeks. Advancing hyphae even, raised to

limit of growth. Mat at first thin, white, cottony, minutely striate with striae radiating

from the inoculum, with irregular white, cottony patches around the inoculum.

Mat darkens gradually to areas of “warm buff” “honey yellow” to “yellow ochre”,

the patches of mycelium around the inoculum increasing in size and number and

coalescing to form irregular rounded lumps, at first “warm buff” but darkening to

“buckthorn brown” and often developing minute pores. Against the sides of the

dish, thin, lacquer-like areas of “buckthorn brown” to “‘russet” develop after five

weeks.

Reverse darkens gradually, finally assuming a mottled appearance due to forma-

tion of dark-coloured areas on the mat. A faint, fragrant, mushroomy odour is given

off.

On gallic and tannic acid media strong diffusion zones are formed. Little or

no growth occurs on gallic acid medium but colonies up to 20 mm in diameter may

form on tannic acid.

Hyphal characters.

Advancing mycelium: hyphae hyaline, branching, thin-walled, with deeply staining

contents and simple septa, 2,0-4,5 in diameter (Fig. 27).

Aerial mycelium: (i) hyphae as in the advancing zone; (ii) fibre hyphae reddish brown,

branching or unbranched, thick-walled, aseptate, 2,5—4,5 in diameter and variable

in length, arising from thin-walled, septate hyphae (Fig. 28).

Fructification: (i) thin-walled, septate and reddish-brown aseptate, fibre hyphae as

in the aerial mycelium; (ii) setae dark reddish-brown, subulate, conical or somewhat

ventricose, 15-35 x 2,5-6,0.. (Fig. 29).

Submerged mycelium: hyphae as in the advancing zone but often somewhat distended,

up to 7,0 in diameter.

526

Polyporus gilyus had been described in culture before by Davidson, Campbell

& Blaisdell (1938), Davidson, Campbell & Vaughn (1942), Hirt (1928), Refshauge &

Proctor (1936) and Nobles (1948; 1958; 1965). The isolates of this species from

South Africa agree very well with the descriptions by these authors. This species may

be recognized fairly easily in culture if the small-pored fructification, bearing the setae,

are formed. Cultures which lack these are however rather featureless and may be

confused with a number of other species with very similar cultural characters.

_ Overholts (1953) reported that the basidiospores and setae formed in fructifi-

cations in culture are identical to those present in fruit-bodies found in nature.

The micromorphological characters of the structures formed in cultures of

Polyporus gilvus agree very well with those of structures present in its carpophores as

described by Fidalgo & Fidalgo (1968).

Polyporus gilvus is well known as an important white rot fungus of timber of

various species of broad-leaved trees (Cartwright & Findlay, 1958). It is also a common

and widely distributed fungus in South Africa and had been reported on wood of

various species from the Knysna area on many occasions (Doidge, 1950). Despite its

widespread distribution, it has been isolated from two specimens only during this

present investigation.

Fungi isolated from the test specimens but previously described from S. Africa.

Coniophora arida (Fries) Karst.

This fungus was isolated in culture only once in the course of this study. Its

cultural characters closely agreed with the author’s earlier description (Van der

Westhuizen, 1958). It is recognizable by the pale greyish-brownish mycelial mat

which tended to soften the surface of the agar and the presence of whorls of clamps

at the hyphal septa with branches often arising from the clamps. No diffusion zones

were formed on gallic acid and tannic acid-malt agar although colonies of up to

45 mm and 15 mm in diameter respectively, formed on the two media in 7 days.

This fungus was associated with a brown rot of wood of Quercus palustris.

Kemper (1937) described the morphology of the fruit-bodies and cultural charac-

ters of Coniophora arida. He reported that C. arida caused more extensive disintegra-

tion of spruce and pine test blocks than C. puteana. Southam & Ehrlich (1950) found

C. arida to be one of the fungi most frequently associated with brown rot of Thuja

plicata poles. Under experimental conditions it was capable of causing up to 57,5%

loss in dry mass after 6 months when inoculated in test blocks of western red cedar

sapwood. Duncan & Lombard (1965) listed C. arida as one of the 10 most prevalent

fungi on softwoods as well as one of the common species on hardwoods. It was also

isolated frequently from the underground decayed portions of experimental pine

sapwood stakes at Madison, Wis., and Corvallis, Oreg., that had been treated with

various preservatives. No information on the tolerance limits of C. arida to various

preservatives are available however.

Lenzites sepiaria (Wulf. ex Fr.) Fr.

This fungus was isolated from three specimens of wood of Pinus spp. on which

it caused a brown rot. In cultural characters the isolates agree closely with the descrip-

tions by Cartwright & Findlay (1958), Nobles (1948; 1965) and Van der Westhuizen

(1971). The cultures also formed fructifications on blocks of Pinus patula, according to

the method described by Matters & Da Costa (1958) which allowed their identification.

Although this fungus is one of the important species causing brown rot of

timber in the United States (Duncan & Lombard, 1965) and Europe (Cartwright &

Findlay, 1958) it is known in South Africa only since 1961 (Van der Westhuizen,

1971). Its comparatively frequent occurrence on these test specimens is thus in con-

trast to its brief history in South Africa.

527

Polyporus adustus Willd. ex Fr.

The characters of this isolate in culture agreed very well with those described by

Nobles (1948; 1965), Zycha & Knopf (1966) and Van der Westhuizen (1971) for this

species. Despite its association with a white rot, no diffusion zones were formed on

gallic acid and tannic acid media. Colonies of up to 35 mm in diameter formed on

gallic acid but no growth occurred on tannic acid. These characters together with the

general texture of the mycelial mat which lack strong distinguishing micromorphologi-

cal characters, serve to identify this isolate with this species.

Polyporus adustus was isolated only once and from an untreated hardwood

specimen on which it was associated with a white rot. This species is not very common

in South Africa, only 10 collections having been recorded (Doidge, 1950). Most of

these are from the cool moist, belt of the southern Cape Province. It is one of the

few isolations of a species of polypore in the present study.

Polyporus sanguineus L. ex Fr.

The characters of the cultures and carpophores of this species as well as the other

two orange-coloured species of polypores included in the genus Pycnoporus Karst.,

were described and compared in great detail by Nobles & Frew (1962). They demon-

strated by means of interfertility tests that two of these species viz., P. sanguineus

and P. coccineus (Fr.) Bond & Sing. occur in South Africa and that they are very

similar in cultural characters. The isolate studied here displayed the texture and

colours of the mat associated with cultures of P. sanguineus. For this reason and

because of the fact that carpophores of this species were very abundant on slashings

and prunings in the immediate vicinity of the Test Plot, this isolate is assigned to

this species.

This fungus was isolated from one specimen only, an untreated test stake of

Quercus mexicana in which the orange-red mycelium was clearly evident in the white,

decayed parts. This is one of the commonest and most widely distributed species of

polypore in South Africa. Its carpophores were frequently seen in great numbers on

prunings in the plantations around the Test Plot. Its low frequency of occurrence on

these stakes is therefore rather surprising but it is not listed as a cause of decay of

living oaks by Davidson, Campbell & Vaughn (1942) or of wood products by Duncan

& Lombard, (1965). This indicates that the conditions prevailing in the underground

portions of wooden stakes under test may not be suitable for the development of

this species.

Schizophyllum commune Fries.

Cultures of this fungus are readily recognized by the raised, woolly to felty,

white mat, the formation of a weak diffusion zone on tannic acid-malt agar but not

on gallic acid-malt, and the presence in the mat of hyphae with numerous, minute

lateral projections (Nobles, 1948; Van der Westhuizen, 1958). Fruit-bodies which

allow the identification of the fungus, often develop on most new isolates.

S. commune is one of the commonest and most widely distributed decay fungi in

South Africa (Doidge, 1950). It is listed by Duncan & Lombard (1965) as a frequent

invader of wood products in the United States and of birch and aspen pulpwood in

Sweden by Henningson (1967b). Cartwright & Findlay (1958) however maintain that

it does not cause extensive decay despite its frequent occurrence on timber in England.

Stereum hirsutum ( Willd.) Pers.

This species had been described in culture by Van der Westhuizen (1958) and

the cultures isolated from the test specimens in the present study agreed very well

with these descriptions. In culture this species is readily recognized by the presence

of wide hyphae, 6-10 in diameter, in the advancing mycelium with whorls of large

528

clamps at the septa, and the formation of a thick, felty mat which develops tough,

smooth felty pads of “pinkish buff”, “light buff” to “ochraceous tawny” colour.

Strong diffusion zones and colonies up to 60 mm and 50 mm in diameter are formed

after one week on gallic acid-malt and tannic acid-malt agar respectively.

This species, which is very common and wide spread in South Africa, was isolated

from 24 test specimens, which ranks it as second in the frequency of the species

encountered. Its fruit-bodies were also very numerous on decaying prunings and

other woody debris in the vicinity of the Test Plot and it was one of the very few

species of which fruit-bodies were present on the test stakes. It was always associated

with extensive creamy white rot of both treated and untreated test specimens. Cart-

wright & Findlay (1958) regard this fungus as the most important cause of decay in

sapwood of oak logs after felling in England. Duncan & Lombard (1965) do not list it

as a cause of decay of wood products in the United States but Cowling (1957) reported

it on stored hardwood lumber and pulpwood logs. Henningson (1967a, b) recorded

Stereum hirsutum as one of the first Basidiomycetes to appear on birch and aspen

pulpwood and it remained active for the entire period under observation (30 months),

fruiting abundantly in autumn. He also reported it to be one of the most aggressive

species in the decay of stored chips (Henningson, 1967c).

DISCUSSION

A relatively small number of species of Basidiomycetes were recorded in this

survey but some interesting facts and observations emerged nevertheless.

A surprising observation was the almost complete absence of fruit-bodies of

decay fungi on the test specimens. The reasons for this are not at all clear. As the

specimens were examined and collected only once per year in late autumn it is not

unlikely that the prevailing conditions could have been unfavourable for fruit-body

formation for many species. Removal of the specimens before they were completely

decayed may also have been an important factor. But the absence of fruit-bodies

made the identification of the mycelia obtained in culture almost impossible. Conse-

quently only those species for which adequate descriptions of cultural characters

exist or for which cultures could be made from fruit-bodies, could be identified

reliably.

Another interesting fact is the almost complete absence here, of those species

which have been used traditionally as test organisms in laboratory tests devised to

evaluate the toxicity and efficacy of various wood-preservatives. Closely allied to

this observation is the very low incidence of polypores, only Lenzites sepiaria, Poly-

porus adustus, Polyporus gilvus and Polyporus sanguineus being present. This latter

species was found on one specimen only despite the fact that its conspicuous fruit-

bodies were present in large numbers on prunings in the vicinity of the Field Exposure

Test Plot.

The absence of polypore species is the more striking if viewed in the light of the

micromorphological characters of the unidentified mycelia isolated from these

specimens. The great majority of these cultures show no differentiation of hyphae

into fibre hyphae or special structures. This indicates that these mycelia must belong

to species of the lower Polyporaceae and Thelephoraceae, in which such structures

are not present, and the Agaricaceae. The higher polypores with complex, tough

fruit-bodies were thus almost entirely absent from the specimens under test.

Of the fungi identified on the test specimens, Stereum hirsutum proved to be one

of those encountered most frequently. This is one of the commonest and most widely

distributed species in South Africa (Doidge, 1950). Two other species of lower

Hymenomycetes, Odontia bicolor and Peniophora cinerea were also of frequent

occurrence though both were virtually unknown in South Africa before (Doidge, 1950;

Talbot, 1958). Of the species which occurred most frequently however, Paxillus

529

panuoides, had been recorded in South Africa on three previous occasions only

(Doidge, 1950). Their frequent presence on the specimens studied here, in comparison

with their few previous records of occurrence, together with the observed general

absence of fruit-bodies on the test stakes, indicate that these species must have been

overlooked previously because of inconspicuous or suppressed fruit-bodies.

The species of fungi isolated and identified in this investigation agree in general

with those of Henningson (1967a, b, c) who found that wood in chip piles were

invaded first by species of lower Basidiomycetes with low wood-destroying activity,

resulting in slow initial decay. Stereum hirsutum was found to be one of the early

and very aggressive invaders which were followed much later by polypores. The

results thus indicate that the wood specimens from Kruisfontein, examined here,

had been removed while they were, in general, still in the early stages of decay,

despite the fact that they have been discarded as failures. This could also account to

some extent for the almost total absence of fruit-bodies on the specimens. The presence

of species such as Peniophora cinerea which are not generally regarded as severe wood-

destroyers, indicate, however, that they must play an important part in the early

stages of decay of timber in contact with soil.

ACKNOWLEDGEMENTS

It is a pleasure to express my sincere thanks to Mr. J. H. van Wyk former Chief

of the Forest Research Institute, Pretoria, for permission to undertake this work at

the Wood Preservative Field Exposure Plot at Kruisfontein, to Mr. P. M. D. Krogh,

Assistant Director of Research of the Forest Research Institute for bringing this

problem to my attention, for supplying specimens and valuable information and

discussions, to Dr. J. H. Ginns of the Mycology Section, Plant Research Intsitute,

Ottawa and Dr. R. Siepmann, Institut fiir Forstpflanzenkrankheiten, Biologische

Bundesanstalt fiir Land und Forstwirtschaft, Miinden, Hann., for kindly supplying

cultures of Odontia bicolor and Paxillus panuoides respectively.

TABLE |.—Species of wood-rotting Basidiomycetes identified, the affected timber

species, preservative treatment and number of specimens affected in the

Wood-preservative Field Testing Plot at Kruisfontein.

No. of

Fun

gus

Coniophora arida...........

Lenzites saepiar

eR Bio erecta

Neamatoloma fasciculare.....

Odontia bicolor

———

ee Cry

Timber

Quercus palustris.......

Rinusspatulasyeeeeeee.

Pinus palustris..........

| Pinus taeda...:..-.....

WeBerlintatsprerremmerrnce

| Eucalyptus capitulata....

Eucalyptus pillularis.....

Gymnosporia acuminata.

Heywoodia lucens.......

Eucalyptus saligna......

Rinusspatulasseeeeeree

Adina macrocephala....

Afrormosia angolensis...

Cordiaicafirdseenee

Eucalyptus botryoides...

Eucalyptus corymbosa...

Eucalyptus globulus.....

Eucalyptus propingua...

Mellitia caffra..........

Syncarpia laurifolia.....

Treatment

Natural durability.......

Di. GelcureyAy.

Natural durability... ....

Natural durability.......

Dy/aiGelcuresener

stitute

5% Magnesium

fluoride

| Natural durability.......

Wisnolite:. oss cc |

SATMAR creosote sub- |

silico- |

| Natural durability.......

Natural durability... ....

Natural durability.......

| Natural durability.......

Natural durability.......

Natural durability... ....

| specimens

530

TABLE |.—Species of wood-rotting Basidiomycetes identified, the affected timber

species, preservative treatment and number of specimens affected in the

Wood-preservative Field Testing Plot at Kruisfontein (Continued).

Fungus

Peniophora aspera..........

Peniophora cinerea..........

Peniophora tenuis...........

Polyporus adustus...........

Polyporus gilvus............

Polyporus sanguineus........

Schizophyllum commune.... .

(eRinusypatulaae erie

Pinus patulann sess

| Pinus patulacar see

| Pterocelastrus

| Pinus michoacana.......

| WES THIS. coon oop ocn eee

| Albizzia gummifera.....

| Fraxinus pennsylvatica. .

| Populus serotina........ |

| Vepris lanceolata.......

Timber

Treatment

No. of

| specimens

Pinus patula............

| Pinus patula............

Pinus patula............

ins patulasseeee eens

Curtisia dentata........

Pinus pinaster..........

tricuspi-

data

Eucalyptus saligna......

Pinus patula............

Ekebergia capensis......

Fraxinus americana.....

Ptaeroxylon obliquum. ..

Quercus mexicana.......

Eucalyptus saligna......

Pinusipatulasensee eee

Pinus patula............

Pinus luchuensis........

Eucalyptus sideroxylon. .

Quercus mexicana.......

Zyzygium cordatum..... |

SY, Calewind oonaceodas

| Copper - 3 - phenylsali- |

cylate

0,257 °% Dieldrin........

5 % Monochloronahptha-

lene

Rosin amine “‘D”’ acetate

Rosin amine “‘D”’ penta-

chlorophenate

Sodium orthophenylphe-

BENE

1% Triolith + 0,68%

Copper sulphate

Untreated control.......

Natural durability.......

SATMAR creosote sub-

stitute B

Sodium orthophenyl phe-

nate

Untreated control.......

Natural durability.......

Albolineum............

0,257% Dieldrin........

Sodium orthophenyl] phe-

nate

Untreated control.......

Natural durability.......

Natural durability... .

Sodium orthophenylphe-

nate

Copper naphthenate, 5%

IDJOCIIM, soonccouecdaac

IDGEAMONEsooco0occccc008

10% Metanate zinc naph-

thanate

Rosin Amine ““D” acetate

AWW Kkamoleeneeeee

Untreated control...

; Magnesium silicofluoride

Orthopheny! phenol..

SATMAR creosote... wart

Warne WW ooccacaccco

Natural durability.......

| Natural durability.......

Natural durability.......

| Natural durability.......

Natural durability....... |

Natural durability.......

— —_

NNEwWeEeE PN

—— ee

ON Oa 5

531

TABLE 1.—Species of wood-rotting Basidiomycetes identified, the affected timber

species, preservative treatment and number of specimens affected in the

Wood-preservative Field Testing Plot at Kruisfontein (Continued).

Fungus

Stereum hirsutum...........

: | No. of

Timber Treatment | specimens

Eucalyptus saligna...... Sa GelcuremAuern are 2

Eucalyptus saligna...... 5% Copper naphthanate 2;

+ 1% Sodium dichro- |

mate |

Apodytes dimidiata..... | Natural durability....... 1

Betulaysp=a eee eine Natural durability....... 2

Eucalyptus paniculata... | Natural durability....... 1

Eucalyptus pedunculata.. | Natural durability....... I

Eucalyptus saligna...... | Natural durability....... 1

Fraxinus americana..... | Natural durability....... | 3

Gymnosporia acuminata. | Natural durability....... | 2

KNOG cosocusocbuocs | Natural durability....... 1

Nuxia floribunda....... | Natural durability....... 4

Quercus mexicana....... | Natural durability....... | 2

Rapanea melanophloeos. | Natural durability....... 2

TABLE 2.—Unidentified mycelia of Basidiomycetes together with the species of

timber and preservative treatment from which they were isolated, arranged

according to their main cultural characters.

Cultural characters

| Isolate no.

il, Extra-cellular

reaction negative;

1.1 Thin-walled hyphae

nodose-septate;

1.1.1 Hypha undifferentia-

ted, hyaline:

1.1.2 Clamped hyphae with

irregularly thickened

walls also present:

1.1.3 Swellings on clamped

hyphae:

oxidase |

Timber species

Eucalyptus saligna......

Albizzia gummifera.....

| Zyzygium cordatum.....

Pinusipatulasseracmree ater

Eucalyptus saligna......

Pinus)patulaseeeeeore

Vepris lanceolata.......

Pinusspatulayceee sacs

Eucalyptus saligna......

| Curtisia dentata.........

Pinus patula............

Populus serotina........

Binusipatulasnrecercrce

Pinusepatulasssnen ace

Pinus patulahpere coc

Pinusspatulaseeyeeers ae

Preservative treatment

Creosote.

Natural durability.

5% Zinc silicofluoride.

Permatox W.R. +- Varno-

lene.

Coppernaphthanate +

Varnolene.

Ortho-phenylphenol.

Natural durability.

25% Permatox W.R. +

75% Varnolene.

Natural durability.

Sodium ortho-phenylphe-

nate.

Natural durability.

Relysol.

Untreated control.

2% Sodium orthophenyl-

phenate.

73472—3

532

TABLE 2.—Unidentified mycelia of Basidiomycetes together with the species of

timber and preservative treatment from which they were isolated, arranged

according to their main cultural characters

(Continued).

Cultural characters Isolate no. Timber species Preservative treatment

283 | Eucalyptus saligna...... | Albawax in kerosene.

287 Eucalyptus saligna...... Copper-3-phenylsalicylate.

1.1.4 Differentiated — thick-

walled fibre hyphae

also present: | |

lh 2B) | Pinus patula............ | Untreated control.

1.2 Thin-walled hyphae |

simple-septate, undif- |

ferentiated:

188 Eucalyptus saligna...... | 10% Coppernaphthanate

| __in diesel oil.

203 Pinus patula............. Untreated control.

1.2.1 Advancing hyphae | |

simple septate, older |

hyphae nodose-sep- | | |

tate: | |

229 Quercus mexicana....... | Natural durability.

Des Extra-cellular oxidase

reaction positive;

2.1 Thin-walled hyphae

nodose-septate;

2.1.1 Hyphae undifferentia-

ted, hyaline: |

100 Eucalyptus saligna...... | 2% Copper sulphate +-

1% Sodium dichromate.

103 Apodytes dimidiata..... Natural durability.

104 Eucalyptus saligna...... Cresoleum.

105 | Eucalyptus obliqua...... Natural durability.

106 Eucalyptus saligna...... | Creosote.

190 Eucalyptus saligna...... | Albolinium.

202 Pinus patula............ | Untreated control.

205 Pinus patula............ Untreated control.

206 Pinus patula............ | Untreated control.

272 Eucalyptus saligna...... | Coppernaphthanate in

| diesel oil.

302 | Populus serotina........ Natural durability.

396 Eucalyptus gomphocep- | Natural durability.

hala

399 Gymnosporia acuminata. | Natural durability.

402 | Pinus patula............ | 3,75°% Tim-Bor.

405 Pinus patula............ | Watco Timber guard.

408 Pinus oocarpa.......... Natural durability.

410 Gymnosporia acuminata. | Natural durability.

2.1.2 Differentiated fibre

hyphae also present: | |

145 Fraxinus americana..... | Natural durability.

150 Harpephyllum caffrum... | Natural durability.

235 | Rapanea melanophloeos. | Natural durability.

2.2 Thin-walled hyphae |

simple septate;

2.2.1 Hyphae undifferentia- |

ted:

109 | Nuxia floribunda....... | Natural durability.

234 | Fraxinus americana..... | Natural durability.

288 Eucalyptus saligna...... | Sodium orthophenyl phe-

| | nate.

412 | Pinus patula............ Untreated control.

413 Heywoodia lucens.......

| Natural durability.

533

Fics. 1—3.—General views. Fig. 1, the wood-preservative field exposure test plot. Fig. 2. Test specimen

with fruit-body of Peniophora sp. at groundlevel. Fig. 3. Fruit-bodies of Stereum hirsutum on

billet under test for natural durability.

534

Fics. 4-7.—Naematoloma fasciculare. Fig. 4, culture at two weeks. Fig. 5, nodose-septate hyphae

from advancing zone, x 1000 phase contrast. Fig. 6, dendrophysis from aerial mycelium, x 500

phase contrast. Fig. 7, strands of aerial hyphae, x 1000 phase contrast.

Fics. 8-10.—Odontia bicolor. Fig. 8, culture at four weeks. Fig. 9, nodose-septate hyphae and capitate

cystidia, x 1000 phase contrast. Fig. 10, capitate cystidia with encrusting crystals, x 1000 in

lactophenol with Cotton Blue.

Fics. 11-14.—Paxillus panuoides. Fig. 11, culture at 4 weeks. Fig. 12, nodose-septate advancing

hyphae, x 1000. Fig. 13, hyphal strand from aerial mycelium, x 1000 phase contrast. Fig. 14,

crystalline incrustations on hyphae, x 1000 phase contrast.

Fics. 15-16.—Peniophora aspera. Fig. 15, culture at 4 weeks. Fig. 16, nodose-septate hypha from

advancing zone, x 1000 phase contrast.

Fics. 17-19.—Peniophora cinerea. Fig. 17, culture at 4 weeks, Fig. 18, nodose-septate hyphae from

advancing zone, x 1000 phase contrast. Fig. 19. brown hyphae with drops of brown, resin-like

material, x 500.

Fics. 20-25.—Peniophora tenuis. Fig. 20, culture at 4 weeks. Fig. 21, capitate cystidia, x 1000 phase

contrast. Fig. 22 and Fig. 23, cystidia with crystalline incrustations, x 1000 phase contrast. Fig.

24, stephanocyst, x 1000 phase contrast. Fig. 25, nodose-septate hyphae from advancing zone,

x 500 phase contrast.

Fics. 26-29.—Polyporus gilvus. Fig. 26, culture at 4 weeks Fig. 27, simple-septate hyphae from ad-

vancing zone, x 1000 phase contrast, Fig. 28, brown, aseptate fibre hyphae, x 500. Fig. 29, setae,

x 500,

SB)

REFERENCES

BasHaM, J. T. & MoraAwskl, Z. J. R., 1964. Cull studies, the defects and associated Basidiomycete

fungi in the heartwood of living trees in the forests of Ontario. Canada Dept. of Forestry Publi-

cation No. 1072, 69 pp.

BERGMAN, O., Nitsson, T. & JERKEMAN, P., 1970. Reduction of microbial deterioration in outside

chip storage by alkali treatment. Svensk Papperstiding 73: 653-666.

Bown, J., 1950. Sur l’existence de races interstériles chez Gloeocystidium tenue (Pat.) H6hn. & Litsch.

Bull. Soc. Mycol. Fr. 66: 204-221.

Borin, J., 1958. Essai biotaxonomique sur les Hydnés résupinés et les Corticiés. Théses no. 202,

Faculté des sciences, Univ. de Lyon 387 pp.

BurpsA_LL, H. H., 1969. Stephanocysts: Unique structures in the Basidiomycetes. Mycologia 61:

915-923.

CANADA DEPARTMENT OF AGRICULTURE, 1952. Annual Report of the Forest Insect and Disease

Survey, 1952.

CARTWRIGHT, K.StT. G., & FINDLAY, W. P. K., 1958. Decay of Timber and its Prevention. Her

Majesty’s Stationary Office, London.

CHASTUKHIN V. Y. & NIKOLAEVSKAYA, M. A., 1962. The microflora of decaying root residues of the

rhizosphere of Oak. Vestv. Leningr. Univ. Ser. Biol. 17: 43-53. (In Rev. Appl. Mycol. 42, p. 55,

1963).

CoaTon, W. G. H., 1964. The Pienaars River complex of wood-eating termites. J. Entom. Soc. S. Afr.

9: 130-177.

Cow Lina, E. B., 1957. A partial list of fungi associated with decay of wood products in the United

States. P/. Dis. Reptr. 41: 894-896.

Davipson, R. W., CAMPBELL, W. A. & BLAISDELL, D. J., 1938. Differentiation of wood-decaying

fungi by their reaction on gallic or tannic acid medium. J. Agr. Res. 57: 683-695.

Davipson, R. W., CAMPBELL, W. A. & VAUGHN, D. B., 1942. Fungi causing decay of living oaks in

the eastern United States and their cultural identification. U.S. Dept. Agr. Tech. Bull. 785.

Dowce, E. M., 1950. The South African Fungi and lichens. Bothalia 5: 1-1094.

Duncan, C. G. & LomBarD, F. F., 1965. Fungi associated with principal decays in wood products in

the United States. U.S. Forest Service Research Paper W O-4, 31 pp.

FIDALGo, O. & Fipatco, M. E. P. K., 1968. Polyporaceae from Venezuela. I Mem. N.Y. Bot. Gdn.

17; 1-34.

Harmsen, L., 1967. Uber einige bauholzzerstérende Corticiaceen. Mater. & Org. 2: 207-213.

HENNINGSSON, B.. 1967a. Interactions between micro-organisms found in birch and aspen pulpwood.

Studia Forestalia Suecica, Stockholm. Nr. 53, 31 pp.

HENNINGSSON, B., 1967b. Microbial decomposition of unpealed birch and aspen pulpwood during

storage. Studia Forestalia Suecica, Stockholm. Nr. 54, 32 pp.

HENNINGSSON, B., 1967c. The physiology, interrelationships and effects on the wood of fungi which

attack birch and aspen pulpwood. Institutionen for Virkeslara, Skogshégskolan Uppsatser Nr. U

19, 10 pp.

Hirt, R. R., 1928. The biology of Polyporus gilvus (Schw.) Fr. N. Y. State C ollege of Forestry, Syracuse

University, Tech. Pub. No. 22, 47 pp.

KAarik, A., 1967. Colonization of pine and spruce poles by soil fungi after six months. Mater. & Org.

2: 97-108.

Kemper, W., 1937. Zur Morphologie und Zytologie der gattung Coniophora, insbesondere des

sogenannten Kellerschwams. Zb/. Bakt., Abt. 2, 97: 100-124.

Krocu, P. M. D., 1947. The comparative efficacy of preservatives in wood exposed to termites and

decay. Fifth British Empire Forestry Conference, Great Britain,

538

Kroau, P. M. D., 1958. A marine exposure test of chemically impregnated poles and untreated

controls in South African Harbours. Record of the Annual Convention of the British Wood

Preservers Association.

Lyr, H., 1962. Detoxification of heartwood toxins and chlorophenols by higher fungi. Nature 195:

289-290.

MADHOSINGH, C., 1961. The metabolic detoxification of 2,4-dinitrophenol by Fusarium oxysporum.

Can. J. Microbiol. 7: 553-567.

MERRILL, W. & FRENCH, D. W., 1966. Colonization of wood by soil fungi. Phytopath. 56: 301-303.

NILSSON, T., 1965. Mikro-organismer i flisstackar. Svensk. Papp. Tidn. 68: 495-499. (in Rey. Appl.

Mycol. 45: 1553, 1966).

Nos tes, M. K., 1948. Identification of cultures of wood-rotting fungi. Can. J. Res. C, 26: 281-431.

Nostes, M. K., 1953. Studies in wood-inhabiting Hymenomycetes. I. Odontia bicolor. Can. J. Bot.

31: 745-749.

Nos-es, M. K., 1965. Identification of cultures of wood-inhabiting Hymenomycetes. Can. J. Bot.

43: 1097-1139.

Nos es, M. K. & Frew, B. P., 1962. Studies in wood-inhabiting Hymenomycetes. V. The genus

Pycnoporus Karst. Can. J. Bot. 40: 987-1016.

OvERHOLTS, L. O., 1953. Polyporaceae of the United States, Alaska and Canada. Ann Arbor, University

of Michigan Press.

PARMASTO, E., 1968. Conspectus Systematis Corticiarum. Tartu, 261 pp.

REFSHAUGE, L. D. & Proctor, E. M., 1936. The diagnosis of some wood-destroying Australian

Basidiomycetes by their cultural characters. Proc. Roy. Soc. Vict. 48: 105-123.

SIEPMANN, R. & ZycuHA, H., 1968. Artdiagnose einiger holzzerstérender Hymenomyzeten an Hand

von Reinkulturen. Nova Hedwigia 15: 559-570.

Stysu, A., 1960. The genus Peniophora in New York State and adjacent regions. State University

College of Forestry, Syracuse, N.Y. Technical Publication No. 83, pp. 47-49.

SmitH, R. B., 1963. Annual Report of the Forest Entomology & Pathology Branch, Canada Depart-

ment of Forestry, for the year ended March 31, 1963.

SouTHAM, C. M. & ExRLICH, J., 1956. Etiology of some sap rots of Western Red Cedar poles. Phyto-

path, 40: 439-444.

TAsor, P. H. B., 1958. Studies of some South African resupinate Hymenomycetes. Part II. Bothalia

7: 131-187.

TAMBLYN, N. & DA Costa, E. W. B., 1958. A simple technique for producing fruit-bodies of wood-

destroying Basidiomycetes. Nature 181: 578-579.

VAN DER WESTHUIZEN, G. C. A., 1958. Studies of wood-rotting fungi. 1. Cultural characteristics of

some common species. Bothalia 7: 83-108.

VAN DER WESTHUIZEN, G. C. A., 1971. Cultural characters and carpophore construction of some

poroid Hymenomycetes. Bothalia 10: 137-328.

Zycua, H. & Knorr, H., 1966. Cultural characteristics of some fungi which cause red-stain of

Picea abies, Lloydia 29: 136-145,

Bothalia, 10, 4: 539-546

A New Species of Encephalartos from Swaziland

IN. JA, Dyer

ABSTRACT

Encephalartos heenanii R. A. Dyer is described for the first time. It is indigenous in Swaziland

and is nearest related to E. paucidentatus Stapf & Burtt Davy.

The Cycads of Southern Africa seem to have endless surprises in store for those

whose investigations penetrate deeper into relatively unexplored territory. On this

occasion, in addition to the record of a previously undescribed species, it is possible to

give for the first time descriptions and illustrations of fresh male and female cones of

Encephalartos paucidentatus Stapf & Burtt Davy from Swaziland. This is appropriate

because E. paucidentatus is the nearest related species to E. heenanii. The records are

made possible by the field work of Mr Denis Heenan and his son David.

Encephalartos heenanii R. A. Dyer, sp. nov., E. paucidentato Stapf & Burtt Davy

affinis, sed planta e basi plus ramosa, truncis minoribus, foliis brevioribus, foliolis

non recurvis, strobilis dense brunneo-hirtolanatis, microsporophyllarum bulla

crassiore et breviore differt.

Planta e basi ramosa; trunci erecti vel ascendentes vel aetate provecta decum-

bentes, ad 2,25 m alti, 25-35 cm diam.; cataphylla acuminato-lanceolata, brunneo-

lanosa. Folia initio dense hirto-lanata glabrescentia, oblongo-lanceolata, 100-130 cm

longa, 15-20 cm lata; petiolus circa 20 cm longus, pulvino dense hirto-lanato. Foliola

media oblongo-lanceolata, pungentia, 12-15 cm longa, plus minusve 1,5 cm lata,

marginibus edentatis, infra manifeste 20-30-nervata. Srrobili pilis brunneis hirto-

lanatis. Strobilus masculinus solitarius, subcylindricus, compactus, 25-30 cm longus,

15-17 cm diam., pedunculo 8-9 cm longo, 2,5—3,5 diam. ; microsphorophylla mediana

late cuneata, patentia, 6,5—7 cm longa, 5,5—6 cm lata; bulla subcylindrica, 1,3-1,5 cm

longa, circa 2,5 cm crassa, verrucosa, copiose brunneo-lanata; vulticulus terminalis

circa 1,75 cm latus et altus. Strobilus femineus solitarius late ovatus, 23-30 cm longus,

17-18 cm latus, pedunculo 5-6 cm longo, 4 cm diam. obconico, tomentoso; mega-

sphorophylla mediana 6-6,5 cm longa, 5—5,5 cm lata; bulla subcylindrica, circa 2,5 cm

longa et 2,5 cm crassa, verrucosa, dense brunneo-lanata; vulticulus terminalis 2—2,5

cm latus, 1,7—2 cm altus.

Type: Swaziland.—Piggs Peak, on krantzes, Sept. 1969, D. Heenan in PRE

30904 (PRE, holo.).

Plants branched form the base; trunks erect or ascending, sometimes becoming

procumbent with age, up to about 2,25 m tall, 25-35 cm diam.; bracteate leaves

linear-lanceolate, very densely woolly with long brown hairs, tips recurving and

becoming dry and brittle. Leaves with long brownish woolly hairs when young,

gradually glabrescent except pulvinus, oblong-lanceolate in outline, 100-130 cm long,

15-20 cm broad, slightly contracted to apex and base; rhachis somewhat twisted;

petiole about 20 cm long; leaflets (pinnae) slightly spaced towards base, denser above,

sometimes slightly overlapping; upper ones slightly curved towards apex; basal

ones not reduced to prickles but 1 or 2 sometimes lobed and somewhat prickle-like;

* Botanical Research Institute, Department of Agricultural Technical Services, P.O. Box 994,

Pretoria.

PLATE 1.—Encephalartos heenanii; top, habit with David Heenan; bottom left, male cone; right,

female cone.

541

margin entire, except very rarely with | or 2 prickles in juvenile forms; median

leaflets fairly widely spreading or reflexed from the rhachis, oblong-lanceolate,

12-15 cm long, about 1,5 cm broad, becoming rigid, undersurface drying with 20-30

prominent nerves. Cones single, sometimes terminal, usually lateral to the terminal

leaf-whorl, densely shaggy brown woolly, rarely glabrescent, yellowish, male and

female alike in general shape, but male much lighter in mass per unit volume than

female. Male cones pedunculate; peduncles 8-9 cm long, 2,5-3,5 cm diam. below

cone, tapered to base, velvety woolly, without stipules except subtending ones;

cones subcylindric, 27-30 cm long, 15-17 cm diam., broadest slightly below the middle

and narrowed slightly to apex and broadly rounded base, shaggy woolly with brown

hair about | cm long, with only the uppermost weathered scale-faces slightly exposed;

scale-faces smallest towards apex; median scales spreading more or less horizontally,

6,5—7 cm long, 5,5—-6 cm broad at the shoulder, cuneately narrowed to base, shortly

stalked; stalk about 1 cm long; limb glabrous, 4-6 mm thick, with sterile margin 3-4

mm wide; bulla 1,3-1,5 cm long, subcylindric, about 2,5 cm thick vertically, abruptly

expanded into lateral wings, densely woolly, verrucose; upper facet humped and

apparently more or less rounded, 1 cm higher than sporangial spine; lower facet

rounded; terminal facet not clearly defined, flattish. verrucose, about 1,75 cm broad

and 1,5 wide vertically. Female cones pedunculate; peduncles obconic, 5—6 cm long,

4 cm diam. at top, velvety woolly; cones broadly ovate, 23-30 cm long, 17-18 cm

broad, shaggy woolly as with male cones; median scales 6—6,5 cm long, 5—5,5 cm

broad at the shoulders, 2,5—-3 cm thick vertically, with incurved toothed lateral lobes

5-10 mm long; bulla 2,5 cm long, about 2,5 cm thick vertically, abruptly expanded

into lateral wings, shaggy woolly, verrucose; upper facet humped with | or 2 ridges;

lower facet rounded; terminal facet 2—2,5 cm broad, 1,7—2,0 cm wide vertically;

seeds angled by compression 3,75—4 cm long with fleshy beak (immature).

SWAZILAND.—2531 CC (Komatipoort): Piggs Peak district, on krantzes, Sept-

1969, D. Heenan in PRE 30904; Nov. 1969, D. Heenan in PRE 30953; June, 1971,

D. Heenan in PRE 31825a (male cone); 31825b (female cone).

The first notification of this species came from Mr. Denis Heenan in August, 1969.

At this time no fresh cones were available and the remains of old ones were severely

damaged by insects and were largely disintegrated. Mr. Heenan had located E.

paucidentatus Stapf & Burtt Davy in quantity a kilometer and a half away and E.

laevifoluis Stapf & Burtt Davy slightly further. It seemed prudent not to ignore the

possibility that E. heenanii had originated by hybridisation between these two species

known to occur in the same neighbourhood. No supporting evidence for this idea was

found, however. But the more knowledge gathered of F. heenanii the more definite it

became that it should have separate specific status. The development and collection of

fresh cones in 1971 was the final proof required and was a fitting reward for Mr.

Heenan’s patient field work with the able assistance of his son David.

The length and density of the shaggy brown hair on emerging leaves is pheno-

menal (Fig. 1) but gradually most of it is shed before the leaves reach maturity.

A most unusual feature of the loosely brown woolly cones is that the male and female

are similar in size and shape. They are not readily distinguishable on sight because of

the similarity of the scale-faces which are almost invariably covered with copious

brown shaggy hair. On handling the cones it is at once evident that the male is far

lighter than the female per unit volume: the male has considerable air space between

the pollen-bearing surfaces and shrinks appreciably on drying, whereas the female

cones are nearly solid and shrink far less. The observation by Mr. Heenan of one

practically glabrous cone on a plant with one normal shaggy cone was a distinct

surprise and at this stage no explanation can be presented. This type of variation has

not been observed in any other species with normally woolly cones.

542

PLATE 2.—Encephalartos heenanii; top, ad- and abaxial views of male cone-scales; bottom, ad- and

abaxial views of female cone-scales with young seeds.

543

PLATE 3.—Encephalartos heenanii, emerging young leaves.

544

D om

RST rococo CC

AT MARCH

HERR

,b,c, side-, ab- and adaxial views of male cone-scales; d,e, ad- and

abaxial views of female cone-scale.

Fic. 1.—Encephalartos heenan

545

Encephalartos paucidentatus Stapf & Burtt Davy. As in other species of the genus,

there is appreciable variation between individual specimens over the full range of

distribution. Unless one is able to make regular field observations over a number of

seasons there is little chance of noting all the possible variations. Recordings are

thus done piecemeal.

The number of cones produced by a trunk of E. paucidentatus was thought to be

1-3. In the mountain kloofs near Piggs Peak, where the species is sometimes common,

David Heenan found female trunks with up to 3 large cones and male trunks with up

to 5 cones. Unlike EF. heenanii, where the male and female cones are almost indis-

tinguishable, the difference in male and female cones of FE. paucidentatus is very

marked. For one thing the male cones are far more slender than the female cones

and for another the median male cone-scales have a relatively long narrow beak to the

bulla with a small terminal facet.

Since no description has been published of fresh male and female cones of

E. paucidentatus, this opportunity is taken of recording them with illustrations

(see Plate 4).

Cones yellowish, sometimes slightly woolly when young, irregularly papillate-

pubescent with reddish-brown and colourless hairs; hair falling with age but bulla-

face not becoming altogether glabrous.

Male cones 1-5 together at the apex of stems, subcylindric, slightly narrowed to

base and apex, 40-50 (60) cm long, 12-15 cm diam., with cylindric peduncle about 10

cm long and subtended by slender acuminate woolly bracts; median scales narrowly

oblong, up to 5,5 cm long, 2,5 cm broad, nearly flat on upper surface and with micro-

sporangia extending nearly to the lateral margins and base; bulla decurved with

sharp lateral angles, 1,5-2 cm long, 1,2-1,5 cm thick vertically; upper facet with

median ridge; lower facet more or less continuous with sporangial surface; terminal

facet 1-1,2 cm broad and slightly less vertically, with minutely irregular margin;

the width of the terminal facets from base to apex of cone fairly constant while the

vertical measurement decreases to about half.

Female cones 1-3 together at apex of stems, oblong-oval, 35-50 cm long, 20-22

cm in their greatest diam., with short stout obconic peduncle and subtended by slender

acuminate woolly bracts; median scales 6,5 cm long; bulla verrucose, about 3 cm long,

projecting 1,5-2 cm, 44,5 cm broad, 2,7-2,8 cm thick vertically, with lateral ridges

extending into incurved lateral lobes about | cm long; upper facet rounded, indis-

tinctly ridged; lower facet slightly convex; terminal facet concave, verrucose, 2-3 cm

broad, 1-2 cm wide vertically; terminal facets becoming gradually smaller towards

the apex of the cone.

546

PLATE 4.—Encephalartos paucidentatus; left, male cone; right, female cone.

Bothalia, 10, 4: 547-553

Notes on Acacia Species in Southern Africa: II

Eee Nosss

ABSTRACT

Some information concerning miscellaneous Acacia species is presented. The typification of

A. galpinii Burtt Davy is discussed, attention is drawn to an unusual specimen of A. giraffae Willd.

and to the seedling development of A. haematoxylon Willd., and the identity of A. inermis Marl. is

discussed. The continued confusion over the identity of A. heteracantha Burch. is considered and the

misapplication of this name in the literature is traced. The location of type specimens of A. spirocar-

poides Engl. and of A. maras Engl. is recorded.

ACACIA GALPINII Burtt Davy

Burtt Davy in Kew Bull. 1922: 326 (1922) based his description of A. galpinii

on Galpin 483M which was collected in the Waterberg district of the Transvaal on

19th Sept. 1920. The date of collection of the type specimen is important as Galpin

returned to the type locality, or probably to the original tree, in later years and

collected further specimens which he also numbered 483 M. In addition to the speci-

mens of 483 M collected on 19th Sept. 1920 specimens of 483 M collected on the

following dates have been examined: 21st Sept. 1923 (PRE); 22nd March 1924

(PRE); 25th Sept. 1927 (BOL, SRGH); Dec. 1927 (SRGH); 10th April 1928 (BM,

BOL, PRE, SRGH).

Galpin’s continued use of the number 483M over a period of years has led to

confusion in several herbaria. Only those specimens of 483 M collected on 19th

Sept. 1920 can be regarded as forming part of the type collection. All of the specimens

of 483 M collected subsequently were collected after A. galpinii was described and

cannot therefore be regarded as type specimens even although they were probably

collected from the type-tree.

In addition, it appears that Galpin, in at least one instance, gave two sets of numbers

to some of his specimens for specimens in BM, BOL, PRE and SRGH collected on

10th April 1928 are numbered 483 M while specimens in K and NH collected on the

same day are numbered 14009.

Acacia galpinii Burtt Davy in Kew Bull. 1922: 326 (1922). Type: Transvaal,

Waterberg district, banks of Bad-Zyn-loop River, Mosdene Estate, Naboomspruit,

19th Sept. 1920, Galpin 483 M (K, holo., BM, GRA, PRE, iso.).

Syn. A. dulcis sensu Henkel in Woody Plants of Natal and Zululand: 233 (1934).

ACACIA GIRAFFAE Willd.

The greyish velvety pods of A. giraffae, although varying somewhat in size

and in shape, are very characteristic. During the examination of material of A.

giraffae, attention was drawn to a specimen with atypical pods, namely, Strey 2292

from the Rehoboth district of South West Africa. The pods, although not quite

mature, are very distinctly coiled and are only up to 2,2 cm wide which is much

smaller than those of typical A. giraffae (see Fig. 1). However, the pods are not in

any way similar to those found on plants regarded as hybrids between A. giraffae and

A. haematoxylon Willd. (Ross in Bothalia 10: 359-362, 1971). A photograph of the

plant from which Strey 2292 was collected reveals that all of the pods on the tree

were of a similar shape and size. Apart from the pods Strey 2292 is otherwise indis-

tinguishable from specimens of typical A. giraffae. The pods of A. erioloba E. Mey.,

* Botanical Research Institute, Department of Agricultural Technical Services, P.O. Box 994,

Pretoria.

73472—4

548

which is a synonym of A. giraffae, were described as “‘semilunate”’. These are probably

similar to those of typical A. giraffae but unfortunately I have not succeeded in

tracing the type specimen of A. erioloba.

essen

A Icom B

Fic. 1.—A, the outline of a “typical” pod of Acacia giraffae (Meeuse 10143); B, the outline of a pod

from Strey 2292.

ACACIA HAEMATOXYLON Willd.

A. haematoxylon is easily distinguished from all other Acacia species in southern

Africa by its fine greyish foliage, the leaflets being very tightly compressed laterally so

that the leaves appear superficially simply pinnate. Some years ago seeds collected in

the Kalahari were germinated in Durban and the seedlings were watched. The leaves

produced during the first three years were distinctly bipinnate and quite unlike those

found on more mature plants. The leaflets on these juvenile leaves were quite discrete

and were up to 3 mm long and 1,3 mm wide in contrast to the small laterally com-

pressed leaflets up to 0,8 mm long and 0,4 mm wide found on adult leaves. The single

surviving plant of A. haematoxylon in Durban has grown very slowly and is only

0,6 m high after five years.

ACACIA INERMIS Marl.

Marloth based the name A. inermis on his specimen number 1317 which he

collected near Otjimbingwe in Hereroland in May 1886. A. inermis is a nomen nudum

for although the name appears on the specimen, Marloth 1317, Marloth never validly

published this name. This is explained in a paper read by Marloth on 26th Oct. 1887

549

and subsequently published in Trans. S. Afr. Phil. Soc. 5: 267-274 (1889). Marloth

(1.c.: 269) wrote: “... I have to mention another new species from Damaraland,

which I had named inermis, on account of its having no spines or prickles whatever,

but the name of which has been changed by Professor Engler to that of A. marlothii.”’

Engler’s description of A. marlothii appeared in Bot. Jahrb. 10: 19 (1888). A further

reference to A. inermis may be found in a paper by Wordsworth, Hutchinson, F. Bolus

and L. Bolus in Ann. Bol. Herb. 3: 21 (1920). Examination of Marloth 1317, the type

of A. marlothii, revealed that the species is an A/bizia and that it must be regarded as a

synonym of Albizia anthelmintica (A. Rich.) Brongn.

ACACIA TORTILIS (Forsk.) Hayne subsp. HETERACANTHA (Burch.) Brenan

Brenan in Kew Bull. 1957: 88 (1957) regarded A. heteracantha Burch. and

A. litakunensis Burch. as synonymous with each other and referred both to A. tortilis

subsp. heteracantha. However, the status and identity of A. heteracantha and A.

litakunensis remained a matter of controversy. In response to a request Brenan re-

investigated the matter and his findings were published in Kew Bull. 13: 409-411

(1959). Despite Brenan’s full and convincing explanation some workers remain

unconvinced and maintain that the use of the epithet “‘heteracantha’”’ is unfortunate in

view of the past confusion and uncertainty over its identity. Evidence is now led in

support of Brenan’s conclusions and the confusion over the name A. heteracantha,

and the subsequent misapplication of this name, will be traced in the literature.

The confusion over the identity of A. heteracantha rests on Burchell’s description

of the pods as ““Legumen lineare”’ in his type description in his Travels in the interior of

Southern Africa I: 389 (1822). The type specimen of A. heteracantha (Burchell 1710

in K) is a sterile twig (see Plate I) and was collected at Springslangfontein between

Griquatown and the Orange River. This type specimen has very small leaves with

rhachides up to 8 mm long, rhachillae up to 8 mm long, leaflets 1-2 mm long and the

straight stipular spines are very slender being only 1,5 mm in diameter at the base.

The type specimen agrees well with Burchell’s description of A. heteracantha except

for the words “‘legumen lineare”’.

As A. luederitzii Engl. and A. hebeclada DC. (syn. A. stolonifera Burch.) also

grow in the northern Cape Province, it is fortunate that this sterile type specimen of

A. heteracantha can be positively identified. Like A. heteracantha, A. luederitzii

and, less frequently, A. hebeclada often have a mixture of short recurved and long

straight stipular spines and sterile specimens of A. /uederitzii and A. hebeclada some-

times superficially resemble those of A. heteracantha. However, A. /uederitzii has

linear legumes. Burchell’s description of A. heteracantha, particularly when the

comment “‘Iegumen lineare”’ is considered, actually fits plants in the A. /uederitzii

complex more accurately. Consequently it is quite understandable why some workers,

especially those who never saw the type specimen of A. heteracantha, applied the

name A. heteracantha to plants in the A. /uederitzii complex.

Sterile material of A. heteracantha can be distinguished from material of A.

luederitzii and A. hebeclada without much difficulty. Indeed, it is usually much simpler

to distinguish sterile specimens of A. heteracantha from A. luederitzii or from A.

hebeclada than it is to distinguish sterile specimens of A. /uederitzii and A. hebeclada

from one another. Material of 4. heteracantha from the northern Cape differs from

A. luederitzii and A. hebeclada (corresponding dimensions of these two species

respectively are given in brackets after those of A. heteracantha) from the same

locality in having smaller leaves with rhachides 0,2-1,8 (1-3,4; 1,5-4,2) cm long,

rhachillae 0,3-1,4 (0,8-2,8; 0,8-2,8) cm long and leaflets 1-2,5 x 0,6-1 (2-4,5 x

0,5-1,5; 2,2-5,3 x 0,9-1,5) mm. The short recurved and long straight spines of

A. heteracantha are more slender than those of A. /uederitzii and A. hebeclada, the

straight spines of the two latter species being usually 2-3 mm in diameter basally.

550

KEW NEGATIVE

No. 12509

DATI i

17 AUG 1971

tl HH

: 9

FD A

OL A

HH iyi

lyre leacew

Aeterna cantha, Berebedl

Yiawels f,p- 289 (820

vier Speer er vale

PLATE 1.—Burchell 1710, the type specimen of Acacia heteracantha (x 1). (By permission of the

Director, Royal Botanic Gardens, Kew).

515)

These characters when considered collectively enable A. heteracantha to be distinguis-

hed. There is a difference in the growth form of the three species and Burchell particu-

larly commented on his A. heteracantha having a “thick clear simple stem (frequently

crooked) and may be distinguished by its growth form half a mile off”’.

Brenan (1959) mentions that there is evidence in Burchell’s MS ‘‘Catalogus

Geographicus Plantarum Africae Australis Extratropicae” at Kew that the comment

about the pod of A. heteracantha was not written at the same time as Burchell described

the type-tree. Brenan states: ‘There is in fact nothing in the catalogus except the

added phrase (Legumen Acaciae capensis) to indicate that the type-tree was in fruit,

and it seems probable that the phrases ““Legumen Acaciae capensis” in the catalogus

and “‘Legumen lineare’”’ in the Travels were not derived from the type-tree of A.

heteracantha’’.

Burchell collected his type specimen of A. heteracantha on 25th October 1811 and

this date of collection is very important. In Southern Africa A. tortilis usually starts

flowering in November or in December and pods are found from January onwards

until August at the latest. Examination of all available specimens from the northern

Cape and from Botswana has not revealed a single specimen with pods that was

collected as late as October and only one specimen with pods collected in August.

It seems, therefore, extremely unlikely that Burchell would have found pods on the

tree or on the ground (the pods are relished by game and are usually eaten soon after

falling) when he collected his type specimen as there is no record of a fruiting specimen

collected as late as October. This supports Brenan’s contention that the description

of the pods was not written at the same time as Burchell described the type-tree.

Dr. L. E. Codd has suggested that after Burchell collected the sterile type specimen

of A. heteracantha on 25th October 1811 he may have later encountered a plant

exhibiting a mixture of short recurved spines and long straight spines with straight

pods (a member of the A. /uederitzii complex) and that Burchell may have mistaken

this plant for his A. heteracantha. This could explain why Burchell added the comment

“‘legumen lineare” to his type description. However, if this did happen Burchell

apparently never collected a specimen for there is no fruiting specimen of a member of

the A. /uederitzii complex in his collection now.

There is evidence that Burchell was uncertain of the identity of the taxon he

named A. heteracantha because he later collected sterile specimens, 2397 and 2402, of

another species under the name A. heteracantha. However neither of these specimens

was cited with the type description of A. heteracantha. Unfortunately Burchell 2397

and 2402 cannot be identified with absolute certainty; they are either A. hebeclada or

A. luederitzii. Burtt Davy in his Fl. Transv. 2: 340 (1932) cites these specimens under

A. hebeclada but I feel that they resemble A. /uederitzii more closely. It may be argued

that, if the specimens are A. hebeclada, Burchell should have recognized them as

such for A. hebeclada was based on his specimen number 2267 and his own A. Stolo-

nifera (a synonym of A. hebeclada) on Burchell 2138. Burchell 2397 was collected from

a shrub 1,2 m high and 2402 from a shrub 1,5 m high.

In support of the view that Burchell was uncertain of the identity of the taxon he

named A. heteracantha it has also been pointed out that Burchell later described

A. litakunensis which is now regarded as a synonym of A. heteracantha. Now Burchell

2205, the type specimen of A. litakunensis, is a sheet consisting of three sets of speci-

mens collected in 1818, 1819 and 1820, representing stages in growth of the young

plants grown by Burchell from seeds of A. litakunensis. Burchell collected the seeds

from Takun (Litakun) between 24—29th July 1812. No specimen of the original tree of

A. litakunensis seems to have been preserved by Burchell. In his type description in his

Travels 2: 452 (1824), the pods are described correctly. It is difficult to believe that

Burchell would have described A. litakunensis as a new species had he ever seen the

similar spirally twisted pods of A. heteracantha. Obviously Burchell believed that

A. heteracantha and A. litakunensis were distinct species. It must be recalled that the

552

type of A. heteracantha is a sterile shoot which displays both short recurved and long

straight spines and, apparently, Burchell never saw the spirally twisted pods. On the

other hand the armature in the type description of A. /itakunensis is recorded as

“‘Spinae stipulares geminae breves recurvae’”’. Burchell’s type-tree of A. litakunensis

apparently had only short recurved spines. This is quite possible as not all specimens

of A. tortilis display a mixture of short recurved and long straight spines. Despite the

similarity in growth form of A. heteracantha and A. litakunensis Burchell obviously

saw no reason to believe that his two species, one exhibiting a mixture of short

recurved and ‘ong straight spines (in itself an unusual feature) and the other exhibit-

ing short recurved spines only but with curiously twisted pods, were one and the

same species.

Burchell’s comment ““Legumen lineare” at the end of his description of A.

heteracantha was indeed unfortunate for it was this phrase which gave rise to the

confusion and uncertainty over the identity of this species. This uncertainty in turn

led to yet other species being described which has further complicated the synonymy.

The uncertainty over the identity of A. heteracantha and subsequent misapplication of

the name will now be traced.

Harvey in his key to the Acacia species in Fl. Cap. 2: 279 (1862) recorded the

pod of A. heteracantha as linear and this is repeated under his description of the

species on p. 280.

Engler in Bot. Jahrb. 10: 19 (1888) keyed out A. heteracantha under the species

with linear pods. As Engler believed that A. heteracantha had linear pods he considered

it necessary to describe A. spirocarpoides (1.c.: 23) and A. maras (1.c.: 24). He added

under his description of A. spirocarpoides that this species differed from A. heteracantha

in having spirally contorted pods. This is confirmed by Marloth in Trans. S. Afr.

Phil. Soc. 5: 270 (1889): “There is, however, no specimen known which with safety

can be referred to this name (A. heteracantha). | thought first that a species pretty

common in Griqualand West should be considered to be Burchell’s A. heteracantha,

but the shape of its legumes differs so widely from B’s description, that it has been

necessary to give it another name (A. spirocarpoides Engler)”. Engler maintained that

A. maras differed from A. spirocarpoides in that the pods were constricted between the

seeds.

Dinter in Deutsch-Stidwest-Afrika Flora Forst-und land-wirtschaftliche Frag-

mente: 76 (1909) applied the name A. heteracantha to plants in the A. /uederitzii

Engl.—A. reficiens Wawra complex. This is clear from his description of the pods as

3-4 cm long and ¢ cm wide and by his use of the Herero name “Omungondo” for

this taxon. All subsequent misapplications of the name A. heteracantha to plants in the

A. luederitzii—A. reficiens complex appear to have originated here.

Glover in Ann. Bol. Herb. 1: 151 (1915) included A. heteracantha under “‘Imper-

fectly known and doubtful species”. Glover noted: “Flowering branchlets of A.

spirocarpoides Engl. seem to me to be identical with those of A. heteracantha Burch.,

but as Burchell’s type has no fruit and as he in his notes describes the legume as

“linear”, I hesitate to unite these two species”’.

Engler in Die Pflanzenwelt Afrikas 3 (1): 355-357 (1915) discussed A. /itakunensis

and A. heteracantha and mentioned that he had not seen the type specimens of either

of these species. He expressed doubt as to whether the plant referred to as A.

heteracantha by Dinter was in fact A. heteracantha. Engler concluded that Dinter’s

plant could just as well be A. uncinata Engl. (which it was!). Dinter in his Index

Fedde Rep. 15: 80 (1917) once more misapplied the name A. heteracantha to plants

in the A. luederitzii—A. reficiens complex.

E. G. Baker in Leg. Trop. Afr.: 822 (1930) keyed out A. heteracantha under those

species with annular or spirally contorted pods. Bak. f. was unable to separate A.

heteracantha from A. tortilis except by that last report of taxonomists, the geographical

discontinuity.

553

Ponnighaus in J. S. W. Afr. Sci. Soc. 6: 13 (1933) further perpetuated the mis-

application of the name A. heteracantha for plants in the A. /uederitzii—A. reficiens

complex.

Walter and Volk in Grundlagen der Weiderwirtschaft in Siidwestafrika 211, t.

68B (1954) did likewise. The illustration t. 68B shows A. heteracantha quite distinctly

as having a linear-oblong pod.

The identity of A. heteracantha was finally resolved by Brenan in Kew Bull.

(1957, 1959). Brenan (1957) regarded A. spirocarpoides Engl. and A. maras Engl. as

synonyms of A. fortilis subsp. heteracantha. In doing so Brenan mentioned that the

type specimens of these two species were destroyed in the Berlin Herbarium and that

he was therefore interpreting the two species from their descriptions.

While examining specimens on loan from the Albany Museum, Grahamstown an

isosyntype of A. spirocarpoides, Marloth 839, was found. This confirmed that the

species is correctly regarded as a synonym of A. tortilis subsp. heteracantha. Also in

the Albany Museum collection is a specimen of Marloth 1260 which is the type number

of A. maras. However, although the specimen carried the type number, there are

certain discrepancies between the information published in the type description in

Bot. Jahrb. X: 24 (1888) and the information written on the label. The information

published is “‘Otjimbingue, alt. 900 m—Fructifera m. Junio 1886”, while the infor-

mation on the label is ‘“‘ad ripas fluminis Kan, pr. Usakos, 860 m., Majo 1886”.

The date is slightly different as is the locality although admittedly the river Kan runs

between Usakos and Otjimbingue. The specimen agrees well with the description of

A. maras and although I felt somewhat hesitant initially about accepting this specimen

as an isotype it seems safe to regard it as a probable isotype.

: ea the gill nen ae 4 ] are A

J | q erie Pee Bi i fe Va) lbh. Ste

x ious

it tis TA, its: apeshie, Tai

4 i cn ean Aged: ie

i : 44 bf ay Sias eta a Uae

' a oeeit tke

b hen

t y i

tat 4 vars ai

. oe

: : . ole oa fr te

oe } ‘i T'

eee wher} ve age > =

Bothalia, 10, 4: 555-564

Contribution to the caryological study of the African

grass Aristida rhiniochloa Hochst., based on specimens

from the Southern Hemisphere

Pierre Bourreil*, Alain Geslot*, Monique Gorlier*

and Bernard de Wintert

ABSTRACT

The chromosome number established for Aristida rhiniochloa Hochst. by the study of material

from three localities in the Southern Hemisphere confirms the results obtained on material from

North Africa. The haploid complement (n = 11) and the diploid number (2n = 22) of this species

conform to the basic number x = 11 typical for the Aristideae. It is shown that the caryotype of

this species is sub-symmetrical. Preliminary studies of material from two localities show that the

meiotic behaviour conforms to that found in the diploid species with n bivalents.

INTRODUCTION AND ACKNOWLEDGEMENTS

The study of A. rhiniochloa Hochst. in the Southern Hemisphere follows similar

investigations started on North African material of this species, of which the distribu-

tion coincides for the greater part with the Sudan-Angolan phytogeographic region.

We are grateful to O. J. Azancot de Menezes, L. E. Codd and B. K. Simon who

kindly made material available for this investigation.

METHODS

The study was partly done on material cultivated in a greenhouse in the Botanical

Garden of the “‘Centre de St.-Jér6me”’ of the University of Provence. The root tips of

plants (originally from Angola), were fixed in bichromate of potassium and stained in

crystal violet (3). The young panicles of plants (originally from Rhodesia) were fixed

in formula 3 of Carnoy (9) and stained in carmine-haematoxylin (1, 2, 5 and 9).

The method of application of heat as suggested by Cauderon (10) was used. It is

similar to the method usually applied by us, but the material is pre-stained at room

temperatures and is only heated over a spirit lamp after squashing in 45° acetic acid

or in acetic haematoxylin (2) diluted to 50°%. Boiling the liquid must be avoided by

passing the slide over the flame fairly rapidly, otherwise chromosome damage may

result. Staining proved to be good with both methods and it was possible to study and

photograph divisions after more than two months storage in a refrigerator. Slides

stored in this way should be repeatedly passed above the flame of a spirit lamp

immediately before study of the divisional plates under the microscope. This technique

shows the colourless spindle clearly in prometaphase I and metaphase I of the divisions

of the microspores (Fig. Rle). The photographs of the prometaphase and metaphase

plates were taken by focussing on the chromosomes not the whole spindle. A beautiful

spindle-vestige in telophase I of a pollen mother cell is visible in Fig. Rlg.

RESULTS

Caryology of the Angolan material

Origin of material: Mucope (Lat. S., 16° 25’, Long W., 14° 50”), Angola. Collected

by O. J. Azancot de Menezes. Date of collection: 1968.

* Université de Provence (Centre de St. Jérdme), Marseille, France.

+ Botanical Research Institute, P. O. Box 994, Pretoria.

556

The chromosome number of 2n = 22 was established and confirmed on five

different metaphases found in root tip material. (See Fig. A.) The length of the chromo-

somes vary from 0,6-0,8 » for short and up to 1,4 for long chromosomes.

Caryology of two collections from Rhodesia

Origin of material: Districts Wankie (near Victoria Falls) and Chiredzi. Collector:

J.C. R. Hill. Dates of collection: August, 1969 and March, 1970 respectively.

1. Chiredzi material

In a study of pollen mother cells the number n = 11 was established 19 times

with certainty on prometaphase I plates. (See Fig. Rlc, Rld, Rif), four times on

metaphase I plates (See Fig. Rle) and three times on anaphase I plates. The comple-

ment 2n = 22 was established seven times on metaphases in ovaries. (See Fig. Rla

and R1b). The length of the chromosomes varies from between 0,6 to 1 » for the

shortest chromosomes and 1,3 to 2,8 » for the longest chromosomes.

2. Wankie material

Studies of pollen mother cells made it possible to establish the haploid number

n = 11 with certainty on 22 diplotene plates, 91 times in diakinesis (See Fig. R2c),

seven times in prometaphase I (See Fig. R2d, R2f, R2e) and six times in metaphase I.

In the homeotypic mitosis of the meiotic divisions this haploid number was established

25 times with certainty on prophases II and five times on metaphase IJ. The chromo-

some complement of 2n = 22 was established with certainty, at least three times on

metaphases in the ovary (See Fig. R2b) and three times on metaphases in the stamens

(See Fig. R2a). The length of the chromosomes vary from 1-1,3 » for the shortest

to 2,1-2,8 » for the longest.

DISCUSSION OF THE CARYOLOGY

Both the haploid and diploid chromosome complements agree with the numbers

established in somatic meristems of cultivated plants of this species from North Africa.

The caryotype

The study of the relative size and configuration of the chromosomes will be

continued by at least one of us (P.B.) using more appropriate techniques, such as

immersion, if necessary, in alphamonochloronapthalene (10), treatment with pectinase

l.e. rapidase C (17), staining in Feulgen followed by staining in acetic haematoxylin

(9) or in acetic orcein (10). Methods of measuring used by Essad (18), Essad and

Najcevska (19) will in addition be used. The measurements set out in Tables 1 and 2

can in the interim be supplied as an indication of the maximum and minimum lengths

of the chromosomes treated with different fixatives.

TABLE 1.—Length of the longest and shortest somatic chromosomes of the root tips

(Cr207K2 fixative).

(a) Length in » | (b) Length in p “Ratio” Number of dif-

Locality of autosomes of | of autosomes of b | ferent caryotypes

minimum size maximum size | a | measured

Attakou. Ennedi septentrional.. | 0,6 il 7/ Pe eg ara |

Airport El Obeid (Sudan)...... 0,7 | 2 Ilene ey7eaa| I

Airport El Fasher (Sudan)...... | 0,7-1 l= 2) lon eo 2

Angolaccedicec sere ee ee 0,6-0,8 | 1,4 2,0 | 2

557

TABLE 2.—Lengths of the longest and shortest somatic chromosomes of the root tips

(Carnoy type fixative).

| |

Organ (a) Length in » | (b) Length in » | Ratio | Number of dif-

Locality Saicicde| of the autosomes | of the autosomes | b ferent caryotypes

| of minimum size | of maximum size| 4 measured

Airport of El Fasher | basal |

(Sudan) eee nee | meristem |

| of leaf... | 12, 2), 1—2),6 Wena beat? Fee 2

Chiredzi, District | |

(Rhodesia)....... | ovary.... 0,6-1 | 1,3-2,8 eS fi

Wankie, District | stamen |

| and

(Rhodesia)....... | ovaries. . 1-1,3 2,1-2,8 51 2

An analysis of Tables 1 and 2 show that the relationship of the lengths of chromo-

somes of different pairs of autosomes varies from 1 to 2,8. The study of somatic

metaphases, furthermore, showed that chromosomes may be distinguished as having

meta- or submetacentric centromeres. The caryotype is therefore subsymmetrical.

For a definition of the symmetrical caryotype the reader is referred to Stebbins (26).

If the width of the somatic chromosomes are compared it is seen that those of

the roots are + of the width of those of the leaves the stamens and the ovary. This

may almost certainly attributed to the fact that two different fixatives were used.

The meiotic behaviour

The meiotic behaviour of the Rhodesian representatives is of the regular type

found in diploid species (i.e. 11 bivalents in prophase I). The differences that exist

are only those of relative frequency of the types of bivalents (II straight, II rings,

II intermediates: angled, open rings). The frequency of the different types of bivalents

and the chiasmata, will be supplied later in tables modelled on those used by Cauderon

(10).

In prometaphase I and metaphase I certain bivalents are sometimes separated

on the same spindle fibre (Fig. Rld and Ric). Such pseudo-univalents have in fact

been observed in certain diplotene stages of two collections from Rhodesia. Our

interpretation agrees with that given by Cauderon for a metaphase reduction division

of Agropyrum intermedium (Host.) P. Beauv. (10) and by Geslot for certain metaphases

I in the microspogenesis of Campanula recta Dul. (20).

In material from the Wankie district in nearly full prophase I a small spheroidal

body, staining as deeply as the nucleolus and the chromosomes, was observed. It is

not visible in any prometaphase, metaphase or anaphase plates, but it reappears in

telophase. This cycle is shared by the nucleolus and it must therefore be considered

as a nucleolus-satellite. This interpretation is supported by the presence of two

nucleoli in some somatic cells. Although we have not examined meiotic prophases in

the material from the Chiredzi district, it seems probable that this peculiarity is

characteristic of the species.

CONCLUSIONS

This study has confirmed that Aristida rhiniochloa Hochst. is a diploid with

2n = 22 chromosomes and has contributed information on its caryotype and the

meiotic behaviour of the diploids. The chromosome count 2n = 38 for Aristida

rhiniochloa in Darlington and Wylie (13) does not agree with our results and needs

further investigation based on material of the species from the Sudan.

558

The present study is a preliminary one and should be extended to include a

wider range of material to determine whether there are areas in which the species

has developed polyploidy. Polyploid series in Aristida with 2-4 ploids (A. fendleriana,

A. glauca, A. longiseta), 2-4—6-ploids (A. wrightii), 2-4-6-8 ploids (A. purpurea) have,

as far as we know, for the first time been mentioned by de Lisle (14) in studies of

material from the south-west United States.

REFERENCES

. BourreiL, P., 1964. A propos de deux techniques anatomique et caryologique de traitement de

certains végétaux. Feuille Inf. Prof. Biol. & Géol., C. R. D. P. Marseille: 4—7.

. BOURREIL, P., 1968. Etude pratique de la méiose male du Vicia faba L. Ann. C. R. D. P., Marseille:

4-8, 13-19.

. BourrelL, P., 1970. Adaptation des techniques de coloration au Violet cristal et a l’hématoxyline

aux coupes de méristémes radiculaires difficilement colorables de graminées africaines des

genres Aristida et Stipagrostis. Techniques, C. R. D. P. Marseille, 1: 7-18.

. BourREIL, P., 1970. Recherches taxinomique sur les Aristides (graminées) de l’Ancien-Monde.

Article de coordination et de synthése déposé en 1970 au C.N.R.S. sous le no. A.O. 1386.

. BourreIL, P. & Geslot, A., 1970. Contribution a l’étude caryologique de diverses graminées

africaines des genres Aristida L. et Stipagrostis Nees, Adansonia, ser. 2, 11, 1: 125-135.

BourrelL, P., Gestot, A. & GILLET, H., 1971. Contribution a l’étude caryologique d’ Aristida

rhiniochloa (graminée) d’aprés des spécimens d’Afrique boréale. Adansonia, ser. 2, 11, 4

. BourreIL, P. & Give, H., 1965. Sur la présence d’un Aristida d’ Ethiopie et d’Afrique australe

dans le massif de l’Ennedi (Nord Tchad). J. Agric. Trop. Bot. Appl. 7, 1-2: 108-110.

BourreIL, P. & GILLeET, H., 1971. Synthése des connaissances et des recherches nouvelles sur

Aristida rhiniochloa, graminées africaine amphitropicale. Mitt. Bot. Staatsamml. Munchen 10:

309-340.

BourreIL, P. & TrRouIN, M., 1970. Techniques rapides de fixation et de coloration pour le

dénombrement de chromosomes de graminées. Techniques, C. R. D. P. Marseille, 2: 21-26.

. CAUDERON, Y., 1958. Etude cytogénétique des Agropyrum francais et le leurs hybrides avec les

blés. Thése de Doct. Etat. Ann. Amélior Pl. 8, 4: 414-422, 431-440, 530.

. CAUDERON, Y. & SAIGNE, B., 1962. Sur la présence et l’étude cytogénétique de Brizes diploides et

tétraploides dans le Massif Central. Rev. Cytol. Biol. Vég. 25: 461-464.

. DARLINGTON, C. D., 1965. Cytology. London, J. & A. Churchill Ltd.: 84-119.

. DARLINGTON, C. D. & Wyte, A. P., 1945. Chromosome atlas of flowering plants. London:

Allen G. & Unwin Ltd.: 417, 434-435.

. De Liste, D. G. 1969. Chromosome number and pollen size in the genus Aristida. Proc. Iowa

Acad. Sci., 76: 74-81.

. De Rosertis, E., NOwWINSKI, W. & SAEZ, F., 1956. General cytology. London: W. B. Saunders

& Co.: 243-279, 295-304.

. De Winter, B., 1965. The South African Stipeae and Aristideae (Gramineae). An anatomical,

cytological and taxonomic study. Bothalia 8: 232-234.

. DUMAS DE VAULX, R., 1970. Mise au point d’une méthode de dénombrement des chromosomes

chez le Melon (Cucumis melo L.) Ann. Amélior P1. 20, 3: 375-378.

. Essa, S., 1962. Etude génétique et cytogénétique des espéces Lolium perenne L., Festuca pratensis

Huds. et de leurs hybrides. Thése de Doct. Etat, sér. A, n° 8, I.N.R.A.: 43-50.

. Essap, S. & NAJCEVSKA, D., 1969. Analyse statistique d’un caryotype de Festuca pratensis Huds.,

Ann. Amélior P1. 19, 1: 15-22.

. Gestot, A., 1971. Contribution a I’étude cytotaxinomique de Campanula recta Dul., C. scheuch-

zeri Vill. et C. ficarioides Timb.- Lagr. en Pyrénées centrales et Orientales. Colloque interdis-

ciplinaire sur les milieux naturels supra-forestiers des montagnes du bassin occidental de la

Méditerranée, organisé au nom de la Soc. Bot. Fr. au C. S. U. de Perpignan: 271-298.

21.

24.

7},

26.

559

GUINOCHET, M., 1965. Notions fondamentales de Botanique générale. Paris: Masson & Cie:

335-337.

Lamoter, M. & L’HERITIER, PH., 1965. Structure de la cellule. Reproduction sexuée. Biologie

Générale, I. Paris: Doin, 122-147.

Petit, CL. & Prevost, G., 1967. Génétique et évolution. Paris: Hermann 66-66.

QUEZEL, P.—Flore et végétation des plateaux du Darfour Nord-occidental et du Djebel Gourgeil,

Dossier, 5, R.C.P. 45, C.N.R.S.: 5-14.

S.R.B., AM., Owen, R. D., EDGAR, R.S., 1965. General genetics. London: W. H. Freeman & Co.:

76-88.

STEBBINS, G. L., 1957. Variation and evolution in plants. New York: Columbia Univ. Press:

442-475.

Wuite, M. J. D., 1964. Les chromosomes (traduit par A. Berkaloff) Paris: Dunod: 68-87.

560

] Ey & I m4 Ric

10 p

PLATE 1.—Drawings of the chromosome complement of Aristida rhiniochloa obtained from cultivated

plants.

Origin of the material: A, Angola. R, Rhodesia: R1, Chiredzi district; R2, Wankie district.

Explanation of the phases: Somatic metaphases (2n = 22) of: ovary (Rla, R1b, R2b); the

stamens (R2a); the root meristem, (A). Meiosis: prometaphase I (RIc).

Type of association of the chromosomes.—Fig. Ric, 11. Ifa (bivalents in rings, association very

homogenous).

Note on the drawings. Certain chromosomes are drawn finely speckled to distinguish them from

others with which they are in contact; vertical arms are outlined but left white. In heterotypic mitosis,

drawing in black makes it difficult to give an accurate presentation of the chromosomes and each

bivalent is drawn partially in black and partially in a hatching of black stipples. The vertical line to

the left of each drawing represents | p.

561

PLATE la.—Photographs of the chromosome complements of Aristida rhiniochloa, which correspond

with the drawings on Plate 1.

ErrRATA—For Rib read R2a and for R2a read RIb.

N.B.—Note the absence in Ric of the membrane of the nucleus and nucleolus, as well as the

absence of the achromatic spindle, also the bivalents which are aligned close together lengthwise in

the cell corresponding to the position of the future spindle, not on the equatorial plane (characteristic

of prometaphase I). Note in R2c (diakenesis, n = 11) the trace of the nuclear membrane and the

deeply stained nucleolus which is much larger than the chromosomes.

562

Rid eae **

10p

—= =

PLATE 2.—Drawings of the haploid chromosome complement (n = 11) of Aristida rhiniochloa

obtained from cultivated specimens.

(For origin of material and notes on the drawing see Plate 1).

Explanation of the phases of the meiotic divisions of the pollen mother cells: Prometaphase

(Rid, Rif, R2d, R2e, R2f); metaphase (Rle). N.B. In Rid the lowest bivalent forms a very open

“elbow”’.

563

Type and association of the chromosomes:

Fig. Rid, 4IIa + 1 Ia. dis. + 6 dr.c.... = 11 bivalents (association less homogeneous).

Fig. Rle, 6Ila. + 4Ila.o. + 1 I dr.dis.... = 11 bivalents (association fairly homogeneous).

Jens, IRUINE, GUE SS BELO eocceocca0eo0 = 11 bivalents (association homogeneous).

Tey RAGE IO Wer, 45 IUGR oooeoco0deds = 11 bivalents (association fairly homogeneous).

Pig Re eral Mae serait cae Gace ceeaoie = 11 bivalents (association very regular).

Jeli, VAR LIE ooo oa oop encooooeso oo = 11 bivalents (association very regular).

Explanation of synbols: If = bivalent; a = a ring; a.o. = an open ring (the ring may be open

from the start or more often the chiasmata may have been released in the twisting during prophase);

a. dis. = ring separated into two chromosomes (pseudounivalents); dr. = straight; dr.c. = arms

straight forming an “elbow’’; dr. dis. = straight, separated.

Rif

* Rie

Rid

Rig es Rof

PLATE 2a.—Photographs of the haploid chromosome complements (N = 11) of Aristida rhiniochloa

corresponding to the drawings in Plate 2. Note in addition a telophase I (R1g).

73472—5

Bothalia, 10, 4: 565-574

New and Interesting Records of African Plants

Various Authors

CELASTRACEAE

Pseudosalacia Codd, gen. nov., Salaciae L. affinis, sed staminibus 5 prope margine

disci insertis differt; a Cassini L. fructibus 2—5-seminalibus demum dehiscentibus

differt.

Arbor. Folia alternantia, petiolata, exstipulata. Inflorescentia axillaris pauciflora

cymosa, breviter pedunculata vel fasciculata, bracteis minutis deciduis. Flores her-

maphroditi pedicellati. Sepala 5, imbricata. Petala (4—) 5 (—6), patentia. Discus

applanatus. Stamina 5, prope margine disco inserta. Ovarium superum, in disco

immersum, 3-loculare, loculis 2-ovulatis, ovulis collateralibus erectis, stylo 1, stigmate

non distinguibili. Fructus capsularis coriaceus demum loculicidaliter dehiscens.

Semina 2-5. Endospermum praesens.

Type species: Pseudosalacia streyi Codd.

Pseudosalacia streyi Codd, sp. nov., species unica.

Arbor 3-5 m alta, ramulis cinereis glabris. Folia petiolata coriacea cinereo-:

viridia glabra; lamina oblongo-elliptica vel ovato-elliptica, 7-14 cm longa, 4-8 cm

lata, penninervia apice rotundata vel emarginata, basi obtusa, margine integra,

petiolo 6-12 mm longo. /nflorescentia breviter pedunculata vel fasciculata 1—-7-flora,

pedicellis 1-2 cm longis. Sepal/a 5, ad basim conjuncta persistentia. Petala (4—) 5 (—6),

viridi-flava vel flava patentia subrotundata, ca. 3 mm diam., subintegra. Discus

carnosus obscure 5-angulatus. Stamina 5, petalis alternantia, subsessilia, prope

margine disco inserta. Ovarium 3-loculare, in disco immersum, stylo brevissimo.

Fructus globosus, 2-3 cm diam., pericarpio crasso coriaceo brunneo minute ver-

ruculoso. Semina 2-5, brunnea glabra trigona, 8-12 mm longa.

Type: Natal, Port Shepstone District, Uvongo River, about 5 km from the sea,

22 Oct. 1969, Strey 9150 (PRE, holo.).

Tree 3-5 m tall; bark greyish, smooth; branchlets glabrous. Leaves petiolate,

coriaceous, grey-green, glabrous; blade oblong-elliptic to ovate-elliptic, 7-14 cm long,

4-8 cm broad, penninerved; apex rounded to emarginate, base obtuse; margin entire,

slightly thickened; petiole 6-12 mm long, channelled above. Inflorescence shortly

pedunculate or fasciculate, 1—7-flowered; pedicels 1-2 cm long; bracts minute, scale-

like. Sepals 5, united at the base, imbricate, triangular, 1-1,5 mm long, persistent and

reflexed at the fruiting stage. Petals (4—) 5 (—6), greenish yellow to yellow, spreading,

subrotund, ca. 3 mm in diameter, subentire. Disc fleshy, obscurely 5-angled. Stamens

5, alternating with the petals, subsessile, arising from small protuberances near the

margin of the disc. Ovary 3-locular, almost completely immersed in the disc; locule

2-ovulate; ovules collateral, erect; style 1, very short; stigma not distinguishable.

Fruit capsular, globose, 2-3 cm in diameter, eventually dehiscent; pericarp thick,

leathery, dark brown, minutely verruculose. Seeds 2-5, brown, glabrous, trigonous,

8-12 mm long; endosperm present.

For some years sterile specimens of this interesting small tree have been known

from the south coastal area of Natal and it is largely due to the persistence of Mr.

R. G. Strey, Curator of the Natal Herbarium, that the fruits (in 1965) and flowers

(in 1966) became known. More recently, excellent flowering and fruiting specimens

have been collected by Mr. Strey and by Mr. H. B. Nicholson of ‘Skyline’, St.

Michaels-on-Sea, Port Shepstone District.

566

PLATE 1.—Pseudosalacia streyi: Above, flowering twig, x4; below, fruits, x.

567

PLATE 2.—Pseudosalacia streyi, fruiting branch, x1 (Strey 9150).

The fruits are hard, leathery, globose, 2—5-seeded capsules, eventually dehiscing

in three valves and are reminiscent of some Salacia species. The flowers differ from

Salacia in having 5 stamens which arise from small protuberances near the margin of

the disc. In Sa/acia there are usually 3 stamens (rarely 2 or 4) which are borne inside

the disc, around the base of the ovary.

Flowering specimens resemble some species of Cassine sens. lat., and the alternate

leaves suggest an affinity with C. aethiopicum Thunb. The fruits of Cassine are,

however, drupaceous and almost invariably 1-seeded, while the five stamens are

borne outside the disc.

Specimens were sent to our Liaison Officer at Kew who, after consultation with

Dr. N. K. B. Robson, reported that it did not satisfactorily fit into a known genus.

The tree is relatively rare, occurring on river banks in the Port Shepstone District

not far from the sea, usually among rocks.

NATAL.—3030 (Port Shepstone): Uvongo River, about 5 km from the sea (—CD), Strey 9150;

9240; Nicholson s.n.; Izotsha River (CD), Strey 6860; 7559; Ross 1887. 3130 (Port Edward): Umtam-

vuna River (—AA), Strey 5835; 7250.

L. E. CoDD

568

CUCURBITACEAE

A NEw COoMBINATION IN ZEHNERIA

Jeffrey in Kew Bull. 15: 343 (1962) realigned some of the generic limits within

the tribe Melothrieae Endl. The most important consequences of this realignment

were the re-establishment of the genus Zehneria Endl., which is characterized by its

three 2-thecous stamens, So/ena Lour. by its peculiar obliquely triplicate anther-

thecae, and /fukia Arn. by its tumid seeds and clustered flowers, as distinct from

Melothria L. nto which they had been sunk by Cogniaux (Cogniaux in De Candolle,

Monographie 2 Phanerogamarum 3, 1881). Melothria L. is then left as an entirely New

World genus of plants with long-stalked fruits and male racemes, compressed seeds,

and three stamens, two of which are 2-thecous and the other 1-thecous.

Of the five southern African species of Melothria dealt with by Meeuse in Bothalia

8: 13-21 (1962), four have already been transferred either to Mukia or to Zehneria

by various authors. However, Melothria parvifolia Cogn. appears to have been over-

looked. Examination of specimens indicated that the species should be placed in the

genus Zehneria and this opportunity is taken of effecting the necessary combination.

Zehneria parvifolia (Cogn.) J. H. Ross, comb. nov.

Melothria parvifolia Cogn. in Bull. Herb. Boiss. 3: 420 (1895); Burtt Davy in

FI. Transv. 1: 225 (1926); Meeuse in Bothalia 8: 18 (1962). Syntypes: Natal, Durban,

Rehmann 8839, 8842 (Z).

Zehneria parvifolia has been recorded from Mozambique, the Transvaal and

Natal. As Meeuse (/.c.) cited an adequate selection of specimens it is not considered

necessary to cite the specimens again here,

J. H. Ross

FLACOURTIACEAE

TRIMERIA ROTUNDIFOLIA OR T. GRANDIFOLIA ?

For a long time Trimeria rotundifolia (Hochst.) Warb. and T. grandifolia (Hochst.)

Warb. have been considered conspecific and there seems to be no doubt that this is

the case. Monospora rotundifolia Hochst. and M. grandifolia Hochst. were published

simultaneously in the same publication, namely Flora 24: 661 (1841). In choosing one

or other of the epithets, it is obligatory, according to Article 57 of the Code (1966), to

accept the choice of the first author who sunk one of the species under the other.

Milne-Redhead in Kew Bull. 1939: 34 (1939) presented a good case for the adoption

of the epithet rotundifolia, since this was the epithet chosen by Gilg in Engl., Pflanzenw.

Afr. 3,2: 582 (1921) who, according to Milne-Redhead, was the first to unite Trimeria

rotundifolia and T. grandifolia. Subsequently most botanists seem to have used the

name T. rotundifolia.

However, during discussions with Dr. H. Sleumer of the Ruijksherbarium,

Leiden, who is revising the Flacourtiaceae for the Flora of Tropical East Africa, it

was learned that Gilg was not the first to unite 7. rotundifolia and T. grandifolia: it

was actually Durand and Schinz in their Consp. FI. Afr. 1: 226 (1898), 23 years

earlier, and they chose the epithet grandifolia. The name T. grandifolia must therefore

be used for the combined species. The author is indebted to Dr. Sleumer for pointing

this out to him,

D. J. B. KILLick

569

MYRTACEAE

A NEw SPECIES OF EUGENIA FROM PONDOLAND AND SOUTHERN NATAL

Eugenia erythrophylla Strey, sp. nov., ab speciebus africanis australibus omnibus

distincta.

Arbor vel arbuscula, 3-10 m alta; rami brunnei, adultis griseis, ramis et foliis

juvenilibus roseis albo-pilosis glandulosis. Folia opposita, usque 12 cm longa, viridia,

coriacea, obovata vel obovato-lanceolata, glandulosa, basi cuneata, margine revoluta,

apice acuminata vel obtusa vel rotundata, nervis superne prominentibus, lateralibus

6-8 prope marginem conjunctis, nervis subtus leviter prominentibus; petioli 2-6 mm

longi, melano-rugulosi. Flores masculi sessiles vel subsessiles; inflorescentia albo-

pilosa, pauciflora vel pseudo-racemosa abbreviata vel congesta; bracteae 2-4 mm

longae, ciliatae, caducae, receptaculum turbinatum, 2 mm longum, tomentosum;

sepala 4, late rotundata, 3-5,5 mm longa, superne pauci-pilosa, subtus subglabra;

petala 4, obovata, 6-7 mm longa, glabra, margine ciliata; discus subcupulatus, dense

pilosus; stamina numerosa, filamentis 4-5 mm longis libris omnibus antheriferis;

antherae 2-thecae, 1-1,3 mm longae, versatiles, loculis longitudinaliter dehiscentibus ;

ovarium abortivum; stylus | mm longus; stigma abortivum. Flores hermaphroditae

subsessiles vel breviter pedicellatae, solitariae, axillares, valde ramis abbreviatis;

bracteae 2,5 mm longae, caducae; bracteolae 2, glabrae; receptaculum adnatum;

sepala et petala ut in floribus masculis vel aliquid majora; receptaculum obconicum,

4 mm longum; gynoecium in receptaculum profunde immersum; discus planus,

carnosus, papillosus vel pilosus; stamina numerosa; filamenta 3-5 mm longa, fili-

formia, glabra, libra, omnibus antheriferis; antherae 2-thecae, 1 mm longae, versatiles,

loculis longitudinaliter dehiscentibus; ovarium 2-loculare, ovulis in quoque loculo 2

evolventibus 1 vel 2; stylus filiformis, 8 mm longus, glaber; stigma discoideum

minutum; baccae subglobosae, apice calycum lobis 4 persistentibus, 2—2,5 cm longae,

1,8-2,5 cm diam., juvenilibus pilosis vel puberulis, adultis glabrescentibus; semina

globosa, 1-1,5 cm diam., brunnea.

Type: Cape, 3129 (Port St. Johns): Goss Point (-BD), Strey & Nicholson 10100¢

(NH, holo.). Fig. 1.

Tree 3-10 m tall; branches brown, becoming grey when mature; young shoots

and leaves pinkish, densely whitish pilose, becoming glabrous with age. Leaves

opposite, petiolate, coriaceous, green at maturity, gland-dotted, obovate to obovate-

oblanceolate or elliptic, 6-12 cm long; 3-7 cm broad, cuneate at the base, apex

abruptly acute or obtuse to rounded, margin revolute, midrib conspicuous, lateral

nerves in 6-8 pairs, spreading, slightly prominent below, joined near the margin;

petiole 2-6 mm long, blackish, rugulose. Male flowers sessile or subsessile, congested on

abbreviated shoots which occasionally develop as leafy shoots; bracts 2-4 mm long,

ciliate, caducous. Calyx united at base; tube broad, more or less saucer-shaped, 2 mm

long, densely pubescent; lobes 4, subrotund, 3-3,5 mm long, sparingly pubescent

below, subglabrous above. Petals 4, obovate, 6-7 mm long, glabrous, margin ciliate.

Disc saucer-shaped, densely pubescent. Stamens numerous, arising from the disc;

filaments of various lengths, 4-5 mm long, free to the base; anthers 2-thecous, versa-

tile, 1-1,3 mm long, all fertile. Ovary abortive; style rudimentary, 1 mm long; stigma

absent. Hermaphrodite flowers subsessile to shortly pedicellate (pedicels up to8 mm

long), usually solitary, axillary, occasionally on short abbreviated shoots; bracts

2,5 mm long, caducous; bracteoles 2, attached at the base of receptacle, 2,5 mm long,

glabrous. Calyx and petals as in male flowers, but somewhat larger; calyx tube

obconical, 4 mm long. Disc flat, fleshy, densely pubescent. Stamens numerous,

arising from the disc; filaments of various lengths, 3-5 mm long, free to the base;

anthers 2-thecous, 1 mm long, with white membranous fringe, broadly rectangular.

Ovary immersed in the disc, 2-celled ; ovules 2 per cell, 1 or 2 developing; style filiform,

570

2, longitudinal

longitudinal

yx and style of hermaphrodite flower, x 3; 4,

gitudinal section of hermaphrodite flower, x 3;

section of male flower, x 3; 3, bracts, cal

Fic. |.—Eugenia erythrophylla. 1, lon

section of fruit, x 2.

571

glabrous, 8 mm long; stigma small, discoid. Fruit obovoid to subglobose, 2-2,5 cm

long, 1,8-2,5 cm diam., glabrescent, with persistent calyx lobes at the apex. Seeds

globose, 1-1,5 cm diam., brown.

This new species of Eugenia was discovered only recently in coastal montane

forest at several localities in Pondoland and southern Natal. The plant is usually

found in rocky situations near streams or along the upper edge of Table Mountain

Sandstone cliffs.

NATAL.—3130 (Port Edward): Farm Beacon Hill (-AA), Strey 7225; Ross 1850; Umtamvuna

Forest Reserve (-AA), Nicholson s.n. 3030 (Port Shepstone): Izotsha (-CD), Strey & Nicholson 7171;

Mgongono (—CD), Strey & Nicholson 7611; 80-6; 9295; Cooper 27; Uvongo Nature Reserve (—CB),

Strey & Nicholson 103489.

Cape.—3129 (Port St. Johns): Goss Point (-CB), Strey & Nicholson 101003; Lupatana (-BD),

Strey & Nicholson 102403; Mkambati Waterfall (-BD), Strey 8575; Magwa Falls (-BD), Jenkins s.n

E. erythrophylla is an erect, medium-sized tree, 3-10 m high, with a straight bole

which, in older trees, is sometimes fluted and twisted. The branches are rather widely

spaced. The bark is thick, light grey, scaling irregularly and very finely striate. On

older branchlets the bark is ash-grey, exfoliating in thin strips. The young shoots are

very distinctly reddish-velvetty at the growing tips. The slash is rich dark brown.

The species appears to have no close affinity in South Africa.

R. G. STREY

STERCULIACEAE

A NEw SPECIES OF HERMANNIA

Hermannia muirii Pi//ans, sp. nov. distinctissima, speciebus stipulis foliaceis

pertinens.

Frutex effusus 3-5 dm altus; rami sparse hispidi et stellato-pubescentes. Folia

0.6-1 cm longa oblongo-oblanceolata vel oblonga, integra, sessilia, pilis simplicibus

vel stellatis ciliata, supra et infra pauce pilis simplicibus munita. Stipula foliis simu-

lantibus sed 5-8 mm longa. Flores fasciculati, terminales. Pedunculi subsessiles, 3-flori.

Pedicelli ad 3 mm longi, sparse setosi. Bracteae 0.4-1 cm longae, lineari-lanceolatae

pilis setaceae simplicibus ciliatae. Calyx 6 mm longus, extus pilis rectis setaceis stellatis,

1-1.5 mm longis paratus; tubus cyathiformis; lobi tubo aequilongi anguste deltoidei,

acuminati, subacuti, sparse intus pubescentes. Petala 6-7 mm longa, albida, lamina

3.5 mm longa, obovata glabrata; unguis lamina subequilongus, cuneatus basi angus-

tatus dorsalis partim minute stellato-pubescens. Filamenta 3-3.5 mm longa, oblongo-

obovata glabrata. Anthera 2.5 mm longa lanceolata, obtusa, minute ciliata. Ovarium

ellipticum angulis pilis setaceis, inter angulos pilis brevioribus pubescentibus. Sty/i

basi setosi. Capsula 4 mm longa, late elliptica vel subrotunda angulis pilis setaceis,

inter angulos pilis brevioribus. Semina minute scabrida.

CapE.—Riversdale District, Droogeveldvlakte, Muir 1882 (BOL, holo.; PRE, iso.).

A much branched spreading shrub, 3-5 dm high; stems woody about the base,

rigidly wiry with spaced long or short stellate or simple hairs. Leaves mostly 0.6—1 cm

long, oblong-oblanceolate or oblong, shortly acute or obtuse, entire, sessile, ciliate

with coarse stellate or simple hairs, with a few similar hairs on the dorsal or ventral

surface. Stipules mostly 5-8 mm long, broadly oblong- or ovate-lanceolate, obtuse or

subacute, coarsely ciliate. Flowers in dense terminal clusters. Peduncles subsessile,

usually 3—flowered. Pedicels up to 3 mm long, with dispersed coarse hairs. Bracts

0.4-1 cm long, linear-lanceolate, with simple coarse cilia. Ca/yx 6 mm long, with

abundance of straight, coarse, pale, stellate hairs 1-1 .5 mm long on the outer surface;

tube cyathiform; lobes as long, narrowly deltoid tapering to a subacute apex, sparsely

pubescent on the ventral surface. Petals 6-7 mm long, white; limb 3.5 mm long,

obovate, glabrous; claw almost as long, cuneate, tapered to the base, minutely

stellate-pubescent behind the upper margins. Filaments 3-3.5 mm long, oblong

SIZ

about the base gradually widening upwards to shortly above the middle thence

gradually tapering to the apex, glabrous. Anthers 2.5 mm long, lanceolate obtuse,

with short cilia on the cells. Ovary elliptical, with coarse erect hairs on the angles

and apex with fewer small hairs between the angles. Styles setose about the base.

Capsule 4 mm long, broadly elliptic or subrotund with coarse stellate hairs on the

angles, with small stellate hairs between. Seeds minutely scabrid.

This distinct species belongs to the comparatively small group having leaves and

stipules very similar in shape. It is readily distinguished from H. orophila Eckl. &

Zeyh., a species which is in the same group and which occurs in the same area, by the

absence of distinct serrations or teeth on the leaves.

The above description and notes were taken from the manuscript of the late Mr

N. S. Pillans of the Bolus Herbarium, whose work on a revision of the genus Herman-

nia was interrupted by his untimely death.

Investigation undertaken subsequently has shown that the species is locally

common in successive valleys south of Albertinia in the Riversdale district.

Two New VARIETIES OF HERMANNIA FILIFOLIA

Hermannia filifolia L.f., Suppl. 302 (1781). Type: Cape, Thunberg s.n. (S, holo. ;

PRE, photo.).

(1) var. grandicalyx Verdoorn, var. nov.

H. linifolia sensu Eckl. & Zeyh., Enum. 371 (1934); sensu Harv. in FI. Cap.

1:194 (1860), non Burm.f.

Haec varietas a typica floribus majoribus, calycibus fere petalis equilongis et

quam petalis palidioribus saepe cremis, foliis subcarnosis vel non-nunquam sub-

succulentis et glaucis, internodiis longioribus plurumque glabris et nitidis differt.

Type: Cape, Prince Albert, Acocks 17098 (PRE, holo.).

The most diagnostic feature of this variety is the large calyx which is about as

long as, and much paler than, the usually dark red petals. It is pale pink or more

often cream coloured, finely stellate pubescent and lobed to just beyond the middle

with the lobes long-acuminate to the apex. The leaves are inclined to be succulent and

glaucous but this feature is not constant for in some of the intermediate specimens the

leaves are firmly fleshy as in the typical variety. The plant is usually heavily browsed

and the shoots that spring from the woody stump are characterised by long, usually

glabrous and shiny internodes. Another feature that may assist in distinguishing this

variety is that the capsule is usually somewhat longer, up to 12 mm long, as against

the 6-7 mm long capsule of var. filifolia, and 7-8 mm long in var. robusta.

The description in the Flora Capensis of H. Jinifolia fits this variety and the two

specimens cited, Drege 7285 (LE, W) and Ecklon & Zeyher., Enum. No.371 (L, S, W)

belong here. Harvey doubtfully ascribes the species to Linnaeus but Burmann filius

is the author. An examination of Burmann’s type, kindly sent on loan to this Institu-

tion by the Director of the Conservatoire Botanique at Geneva, revealed that it was

conspicific with the type of Mahernia scoparia Eckl. & Zeyh.

Harvey mentions that his concept of H. linifolia is “nearly allied to H. filifolia’’.

A study of the complex has led to the conclusion that it is merelya variety of H. filifolia.

While it is readily distinguished in the main area of its distribution there are bordering

specimens that are not so distinct.

The area of distribution of this variety stretches from Montagu eastwards to

Oudtshoorn and northwards to Sutherland and Beaufort West. The largest concentra-

tion has been recorded from the vicinity of Prince Albert.

S15

Cape.—3220 (Sutherland): Klein Roggeveld (-DC), Marloth 9584. 3221 (Merweville): Vin-

dragersfontein (-DB), Acocks 14328; about 5 km N. of Prince Albert Road Station (-DC), Acocks

17098. 3222 (Beaufort West): Doornboomfontein (-AA), Van Breda 527. 3320 (Montagu): Josephs-

kraal (-BA), Van Breda 2123; about 18 km W. of Laingsburg (-BA), Acocks 24351; Matjesfontein

(-BA), Cannon 166; about 6 km N. of Matjesfontein (-BA), Comins 1090; Jakkalsfontein (-DA),

Van Breda 1238. 3321 (Ladysmith): southern foot of Sevenweekspoort (-AD), Edwards 2260; 32 km

W. of Prince Albert (-BC), Leistner 243; about 10 km S. of Ladysmith (-CB), Acocks 14611; Klein

Karroo (-CD), Muir 3754. 3322 (Oudtshoorn): S.W. of Prince Albert (-AA), Taylor 7330; Zwartberg

(-AA), Bolus 11443; Tygerberg (-AB), Marloth 4453; N. of Oudtshoorn (-AC), Bolus 11722; Zebra

(-CB), Van Breda 510.

(2) var. robusta Verdoorn, var. nov., a var. typica omnibus partibus robustioribus

differt; a var. grandicalyce foliis densioribus firmioribus ramulis et calycibus scabriori-

bus differt.

Type: Cape, Port Elizabeth, 2 km N.W. of Donkin Memorial, Olivier 458 (PRE,

holo.).

This variety is so much more robust than the majority of specimens in the typical

variety that at first glance it appears to be specifically distinct. But on examination it is

found to agree in the characteristic leaves, pubescence and colouring of var. filifolia,

differing only in the size of all its parts. The plant is up to 1 m tall and the branchlets

are rough with minute scales or the tubercle-bases of fallen hairs. The fascicled ericoid

leaves (appearing narrowly linear in pressed specimens), together with their leaf-like

stipules are 10-20 mm long and are more crowded on the branches. From var. grandi-

calyx it differs mainly in that the leaves are always firm (not subsucculent) and

crowded, without long glabrescent and shiny internodes, and the calyx is not so

markedly pale nor finely stellate pubescent but is rough with fairly sparse minute

scales or tubercles.

Except for a few intermediate specimens in the surroundings, var. robusta has

to date been recorded only from the Port Elizabeth district. The specimen of Zeyher

2004 in the National Herbarium, Pretoria, labelled as from the ‘‘Wintersbergen’’,

is this variety. It is cited in Flora Capensis under H. filifolia. However, the same num-

ber, Zeyher 2004, in the Stockholm Herbarium is rather H. flammea and is labelled

to be from ““Adow & Zwartkopsrivier’’. In some cases these old specimens are rather

confusing and must be treated with caution.

Capre.—3325 (Port Elizabeth): Victoria Park (-DC), Long 792; Baakens River (-DC), Long 475;

Galpin 9930; 2 km N.W. of Donkin Memorial (-DC), Olivier 458; 458a.

I. C. VERDOORN

WP eRE

7 i a

rye ib i

ie nie

dae ——

i 4 i De aor foal

sh U

' seach

Bothaiia, 10, 4: 575—578

Vegetative Multiplication of Strelitzia reginae

and its Allies

IR, JAN Dyer*

ABSTRACT

In the past it has been assumed that individual plants of Strelitzia reginae Ait. and its allies

increase in size by suckering from the base. This proves to be incorrect and it is seen to be by the

dichotomous branching of the rudimentary stem axis.

The ideal study of Strelitzia reginae Ait., would be to follow the development of a

seedling to its maturity. It is intended that this should be done as part of a compre-

hensive project on the genus Stre/itzia which is being undertaken at the University of

Port Elizabeth under Prof. J. G. Small.

The present note is the outcome of observations stimulated by the ““Comments

on Strelitzia” published in Baileya Vol. 17, p. 65., 1970, by H. E. Moore and Peter A.

Hyypio of the Bailey Hortorium, Cornell University.

The distichous production of leaves is to be observed in quite a number of genera

of Monocotyledons. Strelitzia is one of them. The leaves arise distichously from the

rootstock of the acaulescent species and from the stems of caulescent species, that is,

the youngest leaf is produced opposite the previous one to form a pair. In bulbs and

the arborescent species of Strelitzia the process is more or less indefinite, whereas in

the acaulescent species of Stre/itzia it is finite.

In Strelitzia reginae Ait. and S. juncea Link (sensu stricto) and their allied forms,

2-4 pairs of leaves are produced distichously in the usual manner to form a ‘fan’. At

the crucial stage of subdivision the innermost two leaves emerge back to back from the

axils of the next older leaves. The two innermost back-to-back leaves then become the

first leaves (eventually the outermost) of 2 new shoots and so the process of dicho-

tomous subdivision is initiated and continues by repetition. There is no continuation

of growth between the back-to-back leaves. Once subdivision begins, displacement

of the older leaves takes place and obscures their distichous origin. Inflorescences may

emerge from the axis of one or more of the median leaves of the ‘fan’.

In one exceptional case the petioles of the two innermost leaves of a “fan” of

S. reginae were fused along their backs, but normal leaves emerged from the axis of

each to initiate the growth of 2 new fans.

It is hoped that the two illustrations will make the principle of the vegetative

multiplication clear.

* Botanical Research Institute, Department of Agricultural Technical Services, P.O. Box 994,

Pretoria.

$76

PLate 1.—Strelitzia reginae: aa, outside pair of leaves of the “fan”; bb, next pair of leaves facing

each other; cc, next pair of leaves (innermost pair) back to back.

PLATE 2.—Strelitzia juncea: aa, outside pair of leaves of the ‘fan’; bb, next pair of leaves facing each

other; cc, next pair of leaves facing each other; dd, next pair of leaves (innermost pair) back to

back; ee, first young leaves of 2 new ‘fans’; x, scape of the inflorescence.

Bothalia, 10, 4: 579-582

An Apparatus for Facilitating the Manual

Tabulation of Phytosociological Data

P. J. Muller* M. J. A. Werger,* B. J. Coetzee,* D. Edwards* and

N. G. Jarman*

ABSTRACT

Attempts by various continental workers to mechanize and facilitate compilation of Braun-

Blanquet phytosociological tables from raw data are shortly reviewed and a new apparatus for this

purpose is described. The apparatus consists of a table with aluminium channels in which different

headed rivets can be placed, symbolizing cover-abundance values. With this apparatus phytosociolo-

gical tables can be compiled in two stages.

In phytosociological techniques, especially those developed by the Ziirich-

Montpellier School, the manual ordering of stands and species for preparing the

synthesis tables by rewriting has always been a time consuming procedure and a

potential source of errors.

The usual method of table preparation in the Ziirich-Montpellier School was

described in detail by Ellenberg (1956). Shortened English and French translations

of the procedure were given by Kiichler (1967) and Gounot (1969), respectively.

Attempts to simplify and mechanize the tabulation procedure have been made by

several workers. Wilmanns (1959) described a method using small wooden blocks with

holes bored through them. Cover-abundance figures were written on each block

and the sequence of stands or species was changed by putting a rod through the holes

and moving the entire row or column. The method appears to be useful only with

small numbers of stands and species, up to about 60 of each. Another disadvantage is

that during tabulation the set must be handled very carefully, because one block out of

place could easily result in disarrangement of the table structure. This is particularly

so when several rows or columns are simultaneously removed from their positions

and movability of all those remaining has to be retained. Similar considerations also

apply to Margl’s (1967) table.

Ellenberg & Christofolini (1964) suggested the use of visual ‘punch cards’. The

species present in each relevé were punched on one card. Cover-abundance values

were marked by symbols around each hole. On a light-table the cards were then

compared one by one with a ‘standard’ card on presence alone. The ‘standard’ card

can be chosen at random. Ellenberg (1968) described a modification of this technique,

where cover-abundance values for each of the species were punched as logical sums.

These techniques proved to be useful where the vegetation has already been largely

classified and the vegetation units described, and where only additional relevés

needed to be arranged and included in the existing data material. They may, however,

not be readily applicable to data of virtually unknown areas.

Sophisticated techniques for tabulation by computer were recently developed by

Stockinger & Holzner (1970) and by Spatz & Siegmund (1971). These methods seem

to be particularly useful for the objective and rapid processing of a vast amount of

stand and species data. When the amount of data is not too large, and the time spent

in preparing the data and the cost of operating the computer are important considera-

tions, the apparatus presented here should prove useful.

* Botanical Research Institute, Department of Agricultural Technical Services, P.O. Box 994, Pretoria

73472—6

580

We developed an apparatus which can be used to rearrange and order a matrix

without the need to rewrite the various stages described by Ellenberg (1956) in the

normal tabulation method. In the present case the apparatus was designed to handle

a matrix of up to 124 « 130.

As columns or rows in the usual table matrix, we used aluminium channel, 8 mm

wide, 5 mm high and 100 cm long, in which every 8 mm a 3 mm diameter hole, was

drilled, thus resulting in 124 holes per channel. Rivets, 3 mm in diameter, --12 mm

long, with cheese-heads of 7 mm diameter, fitted into the holes. The heads of the

rivets were sprayed in different colours, symbolizing different cover-abundance

values. The advantage of using aluminium channels with rivets in lieu of blocks is

that, if one rivet is accidently removed, then the rest of the arrangement cannot be

affected as a result.

The channels fitted into the raised edges of a table measuring 100 cm wide and

120 cm long. At one end the channels fit under a flat aluminium bar fixed to that

edge of the table, and at the other end they can be fitted under a moveable flat alumi-

nium bar, thus providing extra stability when an arrangement of the channels has

been completed or is interrupted. To enable the channels to slide smoothly, they are

supported by five metal edges let into the table surface.

For tabulation of the data two such tables should be available. From the field

data the raw phytosociological table is set up on the first table, using a channel for

either a species or a stand, and the different coloured rivets for cover-abundance

values. The channels are then arranged in the desired sequence. The second table is

placed so that the direction of the channels is perpendicular to that of the one in the

first table, and the pattern of the first table is copied on to the second table. If in the

first table the species (rows) were arranged, stands (columns) can now be shifted on the

second table or vice versa. In this way the complete phytosociological table can be

quickly arranged in two stages, without rewriting being necessary, thus avoiding an

important potential source of errors. The end result can be copied in writing from the

second table. The apparatus can also be used for processing other data as the sprayed

rivets can be used to represent symbols for life forms, habitat features, and other

meaningful ecological variables.

We wish to acknowledge the technical assistance of Mr. N. L. Galpin.

SAMEVATTING

Pogings deur verskillende kontinentale werkers om die samestelling van Braun-

Blanquet-plantsosiologiese tabelle uit veldgegewens, te meganiseer en te vergemaklik,

word kortliks bespreek. ’n Nuwe apparaat vir hierdie doel word beskryf. Dit bestaan

uit ’n tafel met aluminiumgeute, waarin klinknaels geplaas kan word, met verskillend

gekleurde koppe, wat verskillende bedekking-abundansiewaardes voorstel. Met

hierdie apparaat kan plantsosiologiese tabelle in twee stappe saamgestel word.

REFERENCES

ELLENBERG, H., 1956. Aufgaben und Methoden der Vegetationskunde. Phytologie IV, I, ed. H. Walter

Stuttgart: Ulmer.

ELLENBERG, H., 1968. Sichtlochkarten zur Ordnung, Klassifikation und Analyse pflanzensozio-

logischer Waldaufnahmen. in: Pflanzensoziologische Systematik, Int. Symp. Stolzenau, 1964,

ed. R. Tiixen. Den Haag: Junk.

ELLENBERG, H. & CHRISTOFOLINI, G., 1964. Sichtlochkarten als Hilfsmittel zur Ordnung und Auswer-

tung von Vegetationsaufnahmen. Ber. Geobot. Inst. ETH, Stift. Riibel, 35; 124-134.

Gounot, M., 1969. Méthodes d’étude quantitative de la végétation. Paris: Masson.

Kucuter, A. W., 1967. Vegetation mapping. New York: Ronald Press.

Marc1, H., 1967. Ein Geraét zum raschen Ordnen einer Tabelle. Forstl., BVA Wien. Informations-

dienst 109.

Spatz, G. & SIEGMUND, J., 1971. Eine Methode zur tabellarischen Ordination, Klassifikation und

dkologischen Auswertung pflanzensoziologischen Bestandsaufnahmen durch den Computer.

Vegetatio: in the press.

STOCKINGER, F. J. & HoLzner, W. F., 1970. Rationelle Methode zur Auswertung pflanzensozio-

logischer Aufnahmen mittels Elektronenrechner. 14th Int. Sym. Rinteln: in the press.

WILMANNS, O., 1959. Ein Gerat zur Mechanisierung von Tabellenarbeit. Ber. dtsch. bot. Ges. 72:

419-420.

581

PLATE 1.—Two views of the apparatus designed for facilitating the manual tabulation of phytoso-

ciological data.

Bothalia, 10,4: 583-594.

Species-Area Relationship and Plot Size: with Some

Examples from South African Vegetation

M. J. A. Werger*

ABSTRACT

A short review is given of mainly recent literature dealing with the problem of minimal area

and plot size for sampling vegetation.

__ A procedure to determine optimal plot size on the basis of the information content given by

different plot sizes is demonstrated on 15 examples from eight different South African veld types.

INTRODUCTION

At the end of the last century in Europe and North America, the first attempts

were made to consider vegetation ecologically. This new approach made it neccessary

to define fundamental concepts on which the new science was to be based.

_ Soon a number of “schools” developed, differing in the concept of what vegeta-

tion basically is, and what possibilities there are to study and classify vegetation.

The development of the concepts and ideas of these various ecological schools

are well documented by Whittaker (1962), McIntosh (1967) and Langford & Buell

(1969). Very broadly one may divide the schools on bases of their approach: the

individualistic and the classificatory. Schools of the individualistic approach regard

variations in vegetation to be continual, although some pattern exists. This pattern

can be investigated by sampling the vegetation. Schools of the classificatory approach

regard vegetation as being composed of basic units of groups of plant species with

sociological relations. These units are usually called communities, and can be classi-

fied in a system. Again, by sampling the vegetation, one can investigate these com-

munities.

Most vegetation studies use a plot technique for sampling. A series of plots is

layed out over the vegetation, according to a certain principle, and information within

these plots is recorded. The plot technique, however, gave rise to a problem, namely,

the size of the plot. It was reasoned that a pattern, or community, being composed of

plant species, needs a certain area to manifest itself. This area was generally called

the “minimal area’. For adequate sampling, a plot must be large enough to cover the

pattern or community to be investigated, and must, therefore, be at least the size

of the ‘“‘minimal area”. On the other hand, in studying vegetation it is important,

from the economic point of view, that no effort is wasted in collecting maximum

information. Therefore, the ideal plot size will be the one nearest to the “minimal

area”, giving the most favourable balance between information obtained and effort

expended.

* Botanical Research Institute, Department of Agricultural Technical Services, P.O, Box 994,

Pretoria.

584

The problem, however, has proved to be in the definition of “‘minimal area’’.

The object of this paper is to show in a short review of relevant literature, largely

after 1952 when Goodall published his extensive review, that the use of the concept of

“minimal area” is impractical, mainly because it is impossible to define. It will be

suggested that optimal plot size is a more useful concept, and that this can be expressed

in terms of information required. It is clear that large plot sizes give more information

but require much effort.

Some examples from South African vegetation types will demonstrate the

applicability of this concept.

SPECIES-AREA RELATIONSHIP

In the Braun-Blanquet school of phytosociology the determination of “minimal

area” is usually based on the species-area curve. A series of nested quadrats is layed

out in a homogeneous piece of vegetation and the increase in number of species in the

successively enlarged area is recorded. The number of species is then graphically

plotted against the area. In 1913, Braun-Blanquet (quoted in Goodall, 1952, and

Hopkins, 1957) defined minimal area (Mindestausmass) as “‘the area above which no

new species occur in the association’’. Later (1928, 1964) he modified this definition to

“the area at which the species-area curve became more or less horizontal.” Ellenberg

(1956) gives the same definition: ““Als Minimal-Areal der Gesellschaft gilt diejenige

Flachengrosse, bei der die anfangs steil ansteigende Kurve in den fast waagerecht

verlaufenden Ast umbiegt.” Recently Tiixen (1970) reinterpreted this curve, regarding

it as consisting of three phases: (a) a strongly curved phase; (b) a slanting straight

line; and (c) a horizontal line. Minimal area is taken as the area at the point where the

horizontal line starts.

It has often been pointed out (for example, Goodall, 1952; Cain & Castro, 1959;

Van der Maarel, 1966; Daubenmire, 1968) that on the species area regression curve

the point of inflexion depends on the relative scales of abscissa and ordinate axes.

Cain & Castro (1959) showed that, depending on the ratio of these axes, they could

find three different minimal areas for an American grassland association. They then

tried to develop a more accurate and independent method to determine the point of

inflexion. A tangent to the curve was constructed, parallel to a line through zero and a

point (x, y), where x is 10°% of the ultimate area that is surveyed, and y is 10% of the

number of species for that area. The tangent “point” gives then the minimal area.

This type of method has the great disadvantage, however, as pointed out by Goodall

(1952), that the resulting minimal area depends closely on the size of the largest area

that is surveyed—the larger this area, the larger the minimal area.

Du Rietz et a/. (1920) (quoted in Goodall, 1952; Hopkins, 1957) and later again

Du Rietz (1954), defined minimal area as the area above which there was no increase

in constant species, constant species being those species of a community which have a

percentage frequency greater than 90% on an area of sufficient size, that is, above the

minimal area. They pointed out that there was a step in the constancy-area curve

above which only extremely large areas would add new constant species. Other

Scandinavian ecologists, like Nordhagen (1923) and Kylin (1926), doubted this.

Nordhagen (1923) argued that the definition of minimal area must be of a practical

kind and should be the area that includes all the important constant species, par-

ticularly the dominant ones.

585

In his analysis of twelve British plant communities, Hopkins (1955, 1957) came to

the conclusion that neither a break in the species-area curves, in the sense of the

Braun-Blanquet school, nor a step in the constancy-area curves in the sense of the

Uppsala school, can be shown with enough evidence, and thus that a minimal area

cannot objectively be defined on these bases. He introduced (Hopkins, 1955) the term

“characteristic area”, which can objectively be defined as ae [where a is the index of

diversity, and N the number of individuals (“plant units’’?) on the sample area (“unit

area’’)], but whose ecological meaning is not clear.

Poore (1964) constructed species-area curves for a tropical rainforest in Malaysia

and found “‘little sign of flattening out at large areas.” Later (Poore, 1968) he con-

structed species-area curves from the same data for a number of constant species

and for tree species represented by more than ten and twenty trees. Apart from the

curve for more then ten trees, which continues to rise, all these curves flattened at

about 4 ha. Poore concluded that “‘if it is considered adequate to define forest types on

constant species of large trees, sample areas amounting to between 2 and 5 ha should

be large enough.”

Van der Maarel (1966) concluded from a detailed study that minimal area cannot

satisfactorily be defined as an absolute intrinsic character of the vegetation and,

therefore, must be interpreted “‘pragmatically” as the minimal size of area that must

be analysed to get a representative view of a vegetation. The term “representative” is

not further defined. He based his practical definition then on frequent species (Van der

Maarel, 1966; 1970).

Other definitions by investigators of the Braun-Blanquet school are given by

Meyer Drees (1954), who distinguished qualitative and quantitative minimal area for

applied survey in tropical rainforests. For qualitative minimal area practically all

plant species are present, whereas in the quantitative sense all timber species reach

such dimensions that it can serve as a basis for timber estimations. Beeftink (quoted

in Van der Maarel, 1966) modified the concept of quantitative minimal area, as the

area where all the species present get a rating on a combined abundance-dominance

scale that is characteristic of the particular vegetation. Calléja (1962) studied a

Brachypodietum, relating the increase in the number of species per increase of area

[XSI ;

(~) to the surface area. He thus obtained hyperbolic curves whose parameters he

Avs é Nos ft he :

considered characteristic of the particular vegetation. With this method, it is possible

to determine objectively and practically a minimal sample size, but even then only

after certain conventions have been adopted, for example, the choice of the system of

co-ordinates to be used (Segal, 1969). Calléja found that a community does not have a

strict floristic minimal area. Thus methods of studying minimal area based on species

number are inadequate, or subjective, and that apart from floristics the structure

of the vegetation should also be taken into account (Calléja, 1962). Gounot & Calleéja

x 100, in which a and b

(1962) suggested the use of co-efficient of similarity (P = ey

a —

are the number of species in stand A and B respectively, and c is the number of

species common to A and B), to define the minimal area, this being the area where the

average co-efficient between four samples of the same size is significantly higher than a

given value. Again one has to agree on the “given value”. Segal (1969) pointed out the

importance of taking into account the structure of the vegetation in defining the

586

minimal area. He distinguished qualitative minimal area, as the area, which even

after a progressive increase, at most yields a relatively small increase in the number of

species, and representative minimal area, as the smallest area that provides sufficient

space for a combination of habitat factors to develop its characteristic vegetation

composition and structure both in a qualitative and in a quantitative sense. He added

that the practical difficulty of defining representative minimal area is to quantify it and

that its estimation is much more subjective than is the case with the qualitative

minimal area.

All these definitions, however, do not result in an objective method to determine

minimal area. The pragmatic, rather than statistically determined, definitions are

based on the fact that an association or community must be well known in its species

and structure, before minimal area can be determined. Especially in areas where the

vegetation has not yet been described in associations, these concepts of minimal

area cannot serve as the basis for determining the optimal plot size.

Arrhenius (1920) (quoted in Gleason, 1922, 1925; Goodall, 1952; Van der

Maarel, 1966) was the first worker, who presented a mathematical expression for

the relations between number of species and area:

size of area | oe of species in 1\n

+7 ?

size of area 2 number of species in 2,

where n is a constant.

Gleason (1922, 1925) showed that this formula does not give a true picture of the

relationships, especially in large areas where the number of species to be expected is

much too high. Gleason (1925) presented then the formula:

log B—log A b—a

log C—log A c—a

where A and B are representative parts of area C, and a, b and c are the number of

species on these areas respectively. This formula can also be written as:

y =a-+ blog, x,

where y is the number of species to be expected on area x, and a and b are constants

(Goodall, 1952). Pidgeon & Ashby (1940) empirically derived a similar equation.

Fisher (Fisher et a/., 1943) derived from biological data, which he compared with

the logarithmic series, the equation:

Se—sar loge f +=},

where S is the number of species observed, N the number of individuals and a a

constant. Except for small areas, this curve fitted well the one produced by Gleason’s

(1925) equation, as was shown by Williams (1943). Williams (1943, 1944, 1947 a, b,

1950) called a the index of diversity, and he and others (forexample, Von Broembsen,

1966) showed that this and similar formulae, and others, derived from this logarithmic

series, fit well a wide variety of natural biological situations. Kilburn (1966) presented

another formula for species-area relationships:

Va ekexe

where y is the number of species in area x, and k and z are constants. The value of k is

taken as the number of species in one square metre, thus it should reflect species size,

whereas Zz reflects the species richness of the community. Only on small areas, up to

circa 900 m2, does this formula fit the observed data.

587

Without defining the term minimal area, Goodall (1952) pointed out that the

size of the minimal area will be smaller when using rectangular plots, than when one

uses square or circular plots to survey a vegetation. This is because the sampling

variance is usually less between narrow rectangles than between squares or circles

of the same area. In a later paper Goodall (1954) argued that if the minimal area is not

purely arbitrary, and since the idea of minimal area implies that one is sampling

vegetation homogeneous in some sense, it must be reasonable and possible to base it on

the concept of homogeneity. Thus he defined minimal area as the smallest sample area

for which the expected differences in composition between replicates are independent

of their distances apart. Tests done on a very uniform salt marsh vegetation in South-

east Australia and on a semi-desert mallee scrub suggested that a minimal area does

not exist, neither for single species, nor for a complete community.

In 1961 Goodall published a paper on pattern and minimal area, defining the

latter as the smallest sample for which, for all species, the variation between replicate

samples is independent of the distance between them. In this way no minimal area

could be determined for mallee scrub in Australia and an Uganda rainforest.

In 1963 after some further studies in Western Australia, he mentioned, however,

that it must be admitted that it is possible to find areas of vegetation which may

satisfactorily be regarded as homogeneous even by a fairly vigorous test, and that in

such areas a minimal area can be identified by appropriate techniques, such as analysis

of variance at different spacings. Grid analysis failed to reveal any significant differen-

ces in variance at spacings greater than the average diameter of the dominant indi-

viduals.

English workers on statistical ecology studied the interrelated problems of pattern,

homogeneity and minimal area in vegetation and concluded that there is “no objec-

tive significance for the idea of minimal area” (see Greig-Smith ef al., 1963; Greig-

Smith, 1964; Kershaw, 1964).

We may conclude that an objective definition of minimal area seems impossible.

OPTIMAL PLOT SIZE

The optimal plot size to be used in sampling vegetation for phytosociological

studies will be one giving the most favourable balance between information obtained

and effort expended, as has already been pointed out. This suggests a pragmatic

approach.

The regression equations of Gleason (1925) and Fisher (Fisher et al/., 1943) based

on the logarithmic series, are generally regarded as best fitting the observed data

(see Goodall, 1952; Hopkins, 1955; Dahl, 1957; Von Broembsen, 1962). The ratio of

increase of information (here the increase of species per area) to increase of time

needed to survey that area, was used by Scheepers (1968) to determine the most

efficient plot size in a survey of the Highveld. The amount of time necessary to sample a

plot is not an intrinsic character of the vegetation, however, and will depend on a

number of factors. It will vary from observer to observer and from day to day, due to

factors such as wind, rain, temperature and topography. Time measurement is, there-

fore, rejected here as a means for determining plot size.

In 1943 Williams plotted the number of species against the size of the area In

which they occurred, both on a logarithmic scale, for areas from as small as 1 cm* up

to the total landsurface of the earth using check lists and floras. He found that up to an

area of circa one hectare the curve “follows the expected increase in species due to

588

increase in size of sample within a uniform population, or within a single ecological

association. In other words, neither the climate nor the rest of the environment

changes very rapidly on an average,” within an area of this size. ““As soon as we pass

beyond this limit we begin to include new ecological conditions in our sample, with the

result that the number of species will increase more rapidly than would be expected if

the population sampled remained uniform.”

__ Thus on the average an uniform population or community can manifest itself

within an area of about one hectare and this area can thus be regarded as giving

the typical species and structure of the community.

It would be very uneconomic to sample vegetation by means of plots of one

hectare*, apart from the fact that in most cases it is difficult to find a reasonably

homogeneous area of that size. With the regression equation:

y =a-+ blog, x

(Gleason, 1925; Goodall, 1952), where a and b will be calculated from observed data,

the expected number of species in one hectare of the sampled vegetation can be

calculated. Regarding number of species as amount of information, the | ha value can

be taken as the 100% level of information. Arbitrarily and individually one can then

decide, what percentage of information one requires per plot, depending, for example,

on the scale of the survey, and so calculate the required plot size.

SOME EXAMPLES FROM SOUTH AFRICAN VEGETATION

Fifteen samples were taken in eight South African Veld Types (Acocks, 1953) in

order to test whether this procedure helps to determine optimal plot size. Twelve

samples were taken in concentric circles with successive increase in radius (0,5; 1; 2; 3;

4; 6; 8; 12; 16; 20 m) i.e. the plot sizes varied between 0,8 m? and | 256 m?. Three

samples in Fynbos vegetation were taken with rectangular nested quadrats of respec-

ively 1; 2; 4; 8; 16; 32; 64; 128 and 256 m?. Care was taken that the plots covered

vegetation where the physiographic features and the vegetation structure were as

homogeneous as possible. The importance of ecological homogeneity in this type of

studies was emphasized by Dahl (1957). Every time the increase in number of species

of permanently recognizable plants was noted. The samples were spread as follows:

Sample 1.—Riverine woodland; along Orange River at Goedemoed. Total cover

estimated at 95°; tree layer up to 8 m, shrub layer up to 4 m, under-

growth up to 0,30 m.

Sample 2.—Themeda—Festuca Alpine Veld (Veld Type 58); in dense grassland

about 16 km from Jouberts Pass near Lady Grey. Total cover estimated

at 95%; soil loamy; very gentle slope (--3°); aspect SSW; one stratum

up to 0,40 m.

Sample 3.—Dry Cymbopogon—Themeda Veld (50); in grassland near Morgenzon,

between Lady Grey and Aliwal North. Total cover estimated at 40%;

soil sandy; gentle slope (+5°); aspect SSW; Elyonurus argenteus,

Cymbopogon plurinodus, Themeda triandra and other grasses dominant.

Sample 4.—False Arid Karoo (35); in open dwarfscrub near Kraankuil. Soil loamy

sand; on plain. One stratum up to 0,50 m; Pentzia incana dominant.

Sample 5.—Central Upper Karoo (27); open dwarf scrub near Houtkraal north of

De Aar. On calcrete rich plain. One stratum up to 0,45 m; Pentzia

incana dominant.

* In tropical rainforest one could use plots larger than one hectare, which are then usually sampled

by means of subplots. The problem still remains to determine the optimal size of the subplot.

589

Sample 6.—False Upper Karoo (36); in open dwarf scrub about 24 km from Norvals

pont on way to Bethulie. Total cover estimated at 25°%; dwarf shrub

layer up to 0,40 m; layer of annuals, rosette plants, etc., up to 0,06 m;

Chrysocoma tenuifolia dominant.

Sample 7.—False Upper Karoo (36); in open dwarf scrub about 13 km West of

Bethulie. Total cover estimated at 60 °%; dwarf shrub layer up to 0,40 m;

layer of annuals, rosettes, etc. up to 0,10 m; Chrysocoma tenuifolia

dominant.

Sample 8.—False Upper Karoo (36); in open shrub and dwarf shrub vegetation

at Tussen die Riviere near Bethulie. Total cover estimated at 30%; on

dolerite; slope 15°; aspect SSE; dwarf shrub and grass stratum up to

0,30 m; shrub stratum up to 2,50 m; tree +6 m. Chrysocoma tenui-

folia, Rhus ciliata and Rhus erosa dominant.

Sample 9.—False Upper Karoo (36); in open scrub between Petrusville and Coles-

berg. Total cover estimated at 65°; on fine-grained sandstone and

mudstone; slope 25°; aspect WSW; dwarf shrub and grass stratum up to

0,90 m; shrub stratum up to 4 m. Rhus undulata and Euclea crispa

dominant.

Sample 10.—False Orange River Broken Veld (40); in open dwarf shrub vegetation

near old road bridge across Orange River at Hopetown. Total cover

estimated at 35%; on andesitic lava; slope 14°; aspect ESE; dwarf

shrub and grass layer up to 0,40 m; very sparse shrub layer up to 2,5 m;

Chrysocoma tenuifolia dominant.

Sample 11.—False Orange River Broken Veld (40); in open scrub-dwarf scrub

between Hopetown and Douglas. Total cover estimated at 30%; on

andesitic lava; on plain; dwarf shrub and grass layer up to 0,50 m;

shrub and low tree layer up to 3 m. Acacia mellifera subsp. detinens and

Rhigozum trichotomum dominant.

Sample 12.—Orange River Broken Veld (32); in open scrub-dwarf scrub, about

32 km from Douglas on way to Prieska. On andesitic lava with slight

sand cover; on plain; dwarf shrub and grass layer up to 0,50 m; shrub

layer up to 4 m. Acacia mellifera subsp. detinens and Phaeoptilum

spinosum dominant.

Sample 13.—Fynbos (69); in dense Protea neriifolia—Protea repens scrub at Jonkers-

hoek near Stellenbosch. Total cover estimated at 100%; on Table

Mountain sandstone; slope 24°; aspect E; undergrowth up to 0,50 m;

shrub layer up to 3 m.

Sample 14.—Fynbos (69); in Protea arborea pseudo-savannah at Jonkershoek near

Stellenbosch. Total cover estimated at 95°%. on Table Mountain sand-

stone; slope 36°; aspect N; undergrowth up to 0,60 m; tree layer up to

4m.

Sample 15.—Fynbos (69); in dense Restionaceous vegetation at Jonkershoek near

Stellenbosch. Total cover estimated at 95°; on granite; slope 32°;

aspect SE; one vegetation layer up to 0,60 m with isolated emergents

up to 2 m.

Table 1 shows the observed numbers of species of each plot at different plot

sizes, the calculated constants a and b of the expression y = a + b log, x, and the

expected number of species in | ha of the population. Calculated values for number

of species fitted the observed values closely for the different plot sizes, indicating that

reasonably homogeneous populations were sampled.

590

“,U ¢‘OZS] Pore “UI ZZ = 14 PaAJOsgO SEM oNjRA SIU], »

S ‘811 €TET IT 0000‘ ST = a 69 19 | SS or 6£ CE IZ SI 69 ST

1°8L 8hE6*L 6666 ‘1 = 9¢ €V 9€ | 82 €T Og || OM El 8 69 | val

L‘09 8597S 8LLI ‘TI a (47 Or TE | O€ SZ Ec | 0 91 er | 69 | el

| Le |

ey | ur saroads —= | 986 || SGI | 1 ce | OT Bf é I

jo Jaquinu q e | : Z adApjaA ajdures

poioce xa zUl 3e saiseds Jo IoquUINNy

¢°99 96199 666b S$ LS WS | Gp | Se | Se | ee | a | on | L | ¢ cE ral

b°89 O£06°S | 9F10°HI ID |) AS | eS) Ne Or SE Ie ST 07 Ol Or Ol

ecb 866£ 7 L818 ‘1 oe | CS | ile | ve GC I | ial 8 G I 9€ 6

€°8L LSL6°L SEL8‘Y L9 ID | @ | e | Gr | te | Ge LI 9 S 9€ 8

€°8s 6998'r | Mare | SS | US |} Sy | gE | es 8% | 97 v7 81 ra 9€ L

6 Ih | SEO || ESS a | AE || GG | xe IZ 07 LI SI €l 6 9€ 9

€°6¢ SET8 “7 OOLE“€ OG | SE IZ LI 91 El OI 8 ¢ € Ka iS

4 LL7‘b ZO8E “9 8E gs || ee O€ VT 8 | OL iA! 8 9 SE v

€°LS IveT 7 8667 81 — €S oF (64 LE SE-a1 SSIs 97 G6 7! os €

S‘€P 6r8I°€ LLIT ¢I a =e le ve (a3 O€ 82 | 92 LI SI ZI 8S C

€°87 ZO6E “TZ 60L7°9 et | ms in 61 81 LT | St O1 L v MU I

| | eiaced eee

ey | ul saioods | | 9° 9S7I| € 708 | p°cSh | I 107 Vert, €°OS | €°82 | 9c | TE | 80

josoqunu q e adi plea ajdwies

payodxg UI 1e soioads Jo Joquinyy

[ alavL

591

Taking the calculated number of species in 1 ha of the population as the 100°/

level of information, the plot sizes belonging to respectively the 40°%, 50 °/, 55°%, 60%

and 70% levels of information were calculated. The results are shown in Table 2.

TABLE 2

Plot size in m? for % of ha-information

Sample Veldtype |————_— —_— sues = sees

Tepes ratl n 5024 tres Rp a55°7 | CD |” TO%

( ) | i

r Ry oh Re al aR 48,6 | 9950), 1 |aa 287

2 Sy BG TOE 21.6 42.6 167

3 50 3.0 11,5 2277 44.6 173

4 Bi ng 45.3 11.7 133.0 392

5 27 19.7 55.6 93,5 157.0 444

6 36 974 30.1 53,8 96.2 307

7 36 oe ee SKO 45.5 82.9 274

8 36 MS | SABA 120,0 197.0 525

9 36 RO ap Bikes 132.0 2120 557

10 AD | OG I 20,5 54.4 9771 309

ih AO) ani) aeiies 35.6 62.5 110.0 339

12 2unaiae 124 66.0 109.0 181.0 492

13 69 9.9 31,4 56,0 99.6 315

14 69 Me Th BO | MS.O 195.0 522

15 69 17.8 | 51.3 | 86.9 | 147.0 423

Comparing the two grassland samples from the Themeda—Festuca Alvin2 Veld

(sample 2) and the dry Cymbopogon—Themeda Veld (sample 3) with the two dwarf

scrub samples from the False Upper Karoo (samples 6 and 7) at the 40% level, it

is necessary in karoid dwarf scrub to have a plot of about three times that in grassland

in order to get the same amount of information. At the 60% level this value has

become about two times. The same features are shown by comparing the two dwarf

scrub False Upper Karoo samples (6 and 7) with the two open tall scrub and tree

amples from the same Veld Type (8 and 9).

These differences are probably mainly due to the structure of the vegetation types.

In a tall scrub and tree vegetation a number of plants have larger dimensions than in a

dwarf scrub vegetation, thus the average plant interspacing is necessarily larger. The

same applies to differences between dwarf scrub and grass vegetation types, although

here the Karoo dwarf scrub vegetation is also usually more open than the Highveld

grasslands.

Similar results were shown when the procedure was tested on a set of nested

quadrats from 1 to 1 024 m2, taken by Mr. J. C. Scheepers in Transitional Cymbo-

pogon—Themeda Veld (49) near Kroonstad in an overgrazed, trampled, harvester

termite infested, patchily denuded grassland. The number of species expected on one

hectare of this grassland (52,7) compares well with the values of the other grasslands

(samples 2 and 3; Table 1). Much larger plot sizes are necessary for this grassland

than for the two others, however, to get a similar percentage of information (for 40%

of ha-information 24,7 m2, for 50°% 67,3 m2, for 55% 110,1 m?). These results show,

that although the pattern in the grassland is similar to other grasslands, the structure

is much coarser here, and one should sample it with much larger plots due to the

patchily denuded veld.

The importance of the structural factor for plot size and minimal area have

already been pointed out by Calléja (1962), Gounot & Calléja (1962), Segal (1969) and

others.

592

Values for the False Arid Karoo and the Central Upper Karoo samples (4 and 5)

are intermediate between the dwarf scrub (samples 6 and 7) and tall scrub and tree

samples (samples 8 and 9) of the False Upper Karoo. Values for the False Orange

River Broken Veld (10 and 11) compare well with the dwarf scrub of the False Upper

Karoo (6 and 7) and values for the Orange River Broken Veld (12) with the tall

scrub of the False Upper Karoo (8 and 9). Also, values for the open Protea arborea

vegetation (Fynbos; sample 14) compare well with the tall scrub of the False Upper

Karoo (8 and 9). Those for the Protea neriifolia—Protea repens scrub (sample 13) agree

better with the dwarf scrub values from the False Upper Karoo (6 and 7), whereas the

values for the Restionaceous vegetation (sample 15) compare with those of the False

Arid Karoo (4) and Central Upper Karoo (5).

Of course, structure of the vegetation is not the only important factor. Floristic

richness is also important. The sample from the riverine woodland (sample 1) shows

values comparable with those of dwarf scrub False Upper Karoo (6 and 7) vegetation,

although its structure is much coarser than these. This is probably due to the floristic

poorness of the riverine woodland. On a rather small area most of the species are

already present, and very few new ones appear on larger areas.

The structure of the vegetation of sample 15 is comparable with those of the

grasslands (2 and 3). Still, in sample 15 a much larger plot size, similar to the False

Arid and Central Upper Karoo, is needed to get an equal percentage of information.

The floristic richness of the Restionaceous vegetation can be regarded as the main

factor for this phenomenon.

Although this approach does not give a specific value for an optimal plot size for a

certain type of vegetation, it allows one to form an idea of the percentage information

obtained with different plot sizes. One can then decide arbitrarily what increase in

information is worth the extra effort needed to sample a larger plot.

Arbitrarily, the author has regarded the optimum plot size as between 50% and

55% of the hectare-information for a phytosociological survey of the Orange River

Valley.

If one reckons that an area of one hectare is insufficient for a certain community

to manifest itself, one can calculate in the same way the expected number of species

for any size area that is regarded as sufficient. The same procedure can then be

followed for determining the optimal plot size.

ACKNOWLEDGEMENTS

The author wishes to thank Dr. D. Edwards for his critical comments on the

manuscript, and Mr. J. C. Scheepers for kindly allowing the use of unpublished data

from his Kroonstad survey.

REFERENCES

Acocks, J. P. H., 1953. Veld types of South Africa. Mem. Bot. Surv. S. Afr. No. 28.

ARRHENIUS, O., 1920. Distribution of the species over the area. Medd. K. Vetensk. Akad. Nobelinst.

4 (7): 1-16.

BRAUN-BLANQUET, J., 1913. Die Vegetationsverhaltnisse der Schneestufe in den Ratisch-Lepontischen

Alpen. Denkschr. Schweiz. Naturf. Ges. 48: 1-347.

BRAUN-BLANQUET, J., 1928. Pflanzensoziologie. Grundziige der Vegetationskunde. Berlin: Springer.

BRAUN-BLANQUET, J., 1964. 3rd ed. Wien-New York: Springer.

CALLéyaA, M., 1962. Etude de la courbe aire-espéce et l’aire minimale. Bull. Serv. Carte Phytogéorg.,

Série B, 7: 161-179.

593

Cain, S. A. & Castro, G. M. DE O., 1959. Manual of vegetation analysis. New York: Harper &

Brothers.

Dau, E., 1957. Rondane: Mountain vegetation in South Norway and its relation to the environment.

Skr. Norske Vid.—Akad. i Oslo. I. Mat.-Natury. Klasse 1956 (3): 1-374.

Nee R., 1968. Plant communities. A textbook of plant synecology. New York: Harper &

rothers.

Du Retz, G. E., 1954. Vegetation analysis in relation to homogeneousness and size of sample area.

8th Congr. Int. Bot. Paris. Sect. 7: 24-35.

Du RIETZ, G. E., Fries, T. C. E., OsvALD, H. & TENGWALL, T. A., 1920. Gesetze der Konstitution

natiirlicher Pflanzen-gesellschaften. Vetensk. Prakt. Unders. Lappl. (Flora och Fauna) 7: 1-47.

ELLENBERG, H., 1956. Grundlagen der Vegetationsgliederung. 1. Aufgaben und Methoden der Vege-

tationskunde. in H. Walter: Einfiihrung in die Phytologie, 4. Stuttgart: Ulmer.

FISHER, R. A., Corset, A. S. & WILLIAMS, C. B., 1943. The relation between the number of species

eng ie number of individuals ina random sample of an animal population. J. Anim. Ecol. 12:

GLEASON, H. A., 1922. On the relation between species and area. Ecology 3: 158-162.

GLEASON, H. A., 1925. Species and area. Ecology 6: 66-74.

Gounot, M. & CALLEJA, M., 1962. Co-efficient de communauté homogénéité et aire minimale.

Bull. Serv. Carte Phytogéogr., Série B, 7: 181-200.

GoopDALL, D. W., 1952. Quantitative aspects of plant distribution. Bio/. Rev. 27: 194-245.

GooDALL, D. W., 1954. Minimal area: a new approach. 8th Congr. Int. Bot. Paris. Sect. 7: 19-21.

GooDALL, D. W., 1961. Objective methods for the classification of vegetation. IV. Pattern analysis

and minimal area. Aust. J. Bot. 9: 162-196.

GooDALL, D. W., 1963. Pattern analysis and minimal area—some further comments. J. Ecol. 51:

705-710.

GreiG-SmitH, P., 1964 Quantitative plant ecology. London: Butterworths. 2nd ed.

GreiG-SmitH, P., KERSHAW, K. A. & ANDERSON, D. T., 1963. The analysis of pattern in vegetation:

a comment on a paper by D. W. Goodall. J. Ecol. 51: 223-229.

Hopkins, B., 1955. The species-area relations of plant communities. J. Eco/. 43: 409-426.

Hopkins, B., 1957. The concept of minimal area. J. Ecol. 45: 441-449.

Kersuaw, K. A., 1964. Quantitative and dynamic ecology. London: Edw. Arnold.

KILBURN, P. D., 1966. Analysis of the species-area relation. Ecology 47: 831-843.

Kyun, H., 1926. Uber Begriffsbildung und Statistik in der Pflanzensoziologie. Botaniska Notiser.

1926: 81-180.

LANGrorD, A. N. & BUELL, F. M., 1969. Integration, identity and stability in the plant association.

In Advances in ecological research, ed. J. B. Cragg, Vol. 6: 83-135. London-New York: Acade-

mic Press.

McIntosh, R. P., 1967. The continuum concept of vegetation. Bot. Rev. 33: 130-187.

MEYER Drees, E., 1954. The minimum area in tropical rain forest with special reference to some types

in Bangka (Indonesia). Vegertatio S—6: 517-523.

NorbDHAGEN, R., 1923. Om homogenitet, konstans og minimiareal. Bidrag til den plantesociologiske

diskussion. Nyt. Magaz. for Naturvidenskaberne 61: 1-51.

PmpGEoN, I. N. & Assy, E., 1940. Studies in applied ecology. I. A statistical analysis of regeneration

following protection from grazing. Proc. Linn. Soc. N.S.W. 65: 123-143.

Poore, M. E. D., 1964. Integration in the plant community. J. Ecol. 52 (suppl.): 213-226.

Poore, M. E. D., 1968. Studies in Malaysian rainforest. I. The forest on triassic sediments in Jengka

Forest Reserve. J. Ecol. 56: 143-196.

594

SCHEEPERS, J. C., 1968. Plant ecological survey of the Highveld Region: Preliminary work and testing

of methods and procedures. Unpubl. Rep. (A)P-Pr. 23/1. Bot. Res. Inst. Pretoria.

SEGAL, S., 1969. Ecological notes on wall vegetation. Den Haag: Dr. W. Junk N.Y.

TUXEN, R., 1970. Einige Bestandes-und Typenmerkmale in der Struktur der Pflanzengesellschaften.

in R. Tuxen (ed.): Gesellschaftsmorphologie. Int. Symp. Rinteln 1966: 76-107. Den Haag:

Dr. W. Junk N.V.

VAN DER MAAREL, E., 1966. Over vegetatiestructuren-relaties en systemen, in het bijzonder in de

duingraslanden van Voorne. Proefschrift. Utrecht.

VAN DER MAAarEL, E., 1970. Vegetationsstruktur und Minimum-Areal in einem Diinentrockenrasen.

in R. Tiixen (ed.): Gesellschaftsmorphologie. Int. Symp. Rinteln 1966: 218-239. Den Haag:

Dr. W. Junk N.V.

VON BROEMBSEN, H. H., 1962. Some recommendations on sampling procedures for the objective

classification of vegetation. Unpubl. Rep. Bot. Res. Inst. Pretoria.

VON BROEMBSEN, H. H., 1967. The logarithmic series applied to point samples drawn from natural

pastures. Proc. Xth. Int. Grassl. Congr. Finland. 1966. Sect. 4: 913—916.

WHITTAKER, R. H., 1962. Classification of natural communities. Bot. Rey. 28: 1-239.

WILLIAMS, C. B., 1943. Area and number of species. Nature 152: 264-267.

WILLIAMS, C. B., 1944. Some applications of the logarithmic series and the index of diversity to

ecological problems. J. Ecol. 32: 1-44.

WILLIAMS, C. B., 1947a. The logarithmic series and the comparison of Island floras. Proc. Linn. Soc.

London. 158: 104-108.

WILLIAMS, C. B., 1947b. The logarithmic series and its application to biological problems. J. Ecol.

34: 254-272.

WILLIAMS, C. B., 1950. The application of the logarithmic series to the frequency of occurrence of

plant species in quadrats. J. Eco/. 38: 107-138.

Bothalia, 10, 4: 595—598.

The Current Status of Mistbelt Mixed Podocarpus

Forest in Natal

Ep Moll&

ABSTRACT

The known distribution and history of Mistbelt Mixed Podocarpus Forest in Natal, and its

utilization and destruction are discussed. It is suggested that there may be a general drying of the

forest climate, which is supported by evidence from canopy tree growth and regeneration. That this

generally drier period has contributed to the rapid rate of forest degradation is postulated, and the

need for the immediate implementation of conservation measures to ensure the safety of a represen-

tative area of forest is stressed.

DEFINITION AND DISTRIBUTION

Mistbelt Mixed Podocarpus Forest was the name given by Edwards (1967) in his

survey of the vegetation of the Tugela Basin, for the climatic climax forest vegetation

of the Natal Mistbelt. Previously this forest type had been variously called High

Timber Forest (Fourcade, 1889), Yellow Wood Bush (Bews, 1912) and Temperate

Forest (Pentz, 1945; Acocks, 1953). It generally occurs between about 3 500 ft (1 000

m) and 4500 to 5000 ft (1 300 to 1500 m), on steep south-facing slopes. These

slopes are subject to relatively frequent mist, particularly in summer, (hence the name

“mistbelt’), and the rainfall is good (at least 1 000 mm a year), so the region is rela-

tively moist. Temperatures are equable, with low maxima (about 37°C), high minima

(about —4°C), and an annual mean of about 16°C (Weather Bureau, 1954). Mode-

rately severe frosts occur on level ground, but probably not on the steep slopes on

which the forest is situated. Snow does occur occasionally, and the rare heavy falls

can cause great mechanical damage (Moll, 1965).

As the name suggests, the most important tree species are, or were, Podocarpus

spp. Many associated tree species occur, such as Ptaeroxylon obliquum, Celtis africana,

Calodendrum capense, Olea capensis, Cussonia chartacea, Cryptocarya myrtifolia,

Prunus africanus, Xymalos monospora, Kiggelaria africana and Combretum kraussii.

Mistbelt Mixed Podocarpus Forest represents one of a series of three montane

forest types found in Natal. Montane Podocarpus Forest occurs at higher altitudes,

it is physiognomically and structurally reduced, and floristically depauperate (Moll,

1965; Edwards, 1967), The Inland Sub-tropical Forest types (Acocks, 1953) occur at

lower altitudes and further north, and have greater floristic affinity with the Tropical

Forests.

In Natal, Mistbelt Mixed Podocarpus Forest occurs from Qudeni in the north,

to the Ingeli and Impetyne Forests in the south (Fig. 1).

UTILIZATION

We know from various historical accounts, summarized relatively recently by

Rycroft (1942), Moll (1965) and Edwards (1967), that the Mistbelt Mixed Podocarpus

Forests of Natal were intensively worked for timber until about 1940. The chief

species cut were Podocarpus spp., Ocotea bullata, Ptaeroxylon obliquum and Olea

capensis. In addition poles, laths and saplings were taken out by the thousand for

hut-building timber by the Bantu. Exploitation was not limited to severe tree cutting,

the results of which are still exhibited by the irregular forest canopy, but also to the

utilisation of forest areas as winter grazing for cattle—a practice which has a definite

detrimental effect on regeneration (Taylor, 1961, 1962). The practice of burning the

grassland surrounding the forest, without due precaution for the protection of the

forest margins, has also contributed to forest destruction.

* Botanical Research Institute, Department of Agricultural Technical Services, Durban Unit, Botanic

Gardens Road, Durban.

73472-7

596

\CRONNYBROOK =)

30° Be

oll, SBS

WANGWANE “S,

G A

Sn 4 500 feet

LAND ABOVE

1 400 metres

nee ys -

IMPETYNE A @m = APPROXIMATE SIZE AND DISTRIBUTION

OF FOREST PATCHES

0 50 miles

d 80kilometres

Fic. 1.—Map showing size and distribution of patches of Mistbelt Mixed Podocarpus Forest in Natal.

Today it is generally accepted that the major factor causing the reduction in

size of our Mistbelt Mixed Podocarpus Forests is man. Taylor (1961) pleaded for the

protection of the Karkloof Forest, which covered an estimated 80 000 acres in 1880

(Fourcade, 1889) and had greatly diminished to an estimated 15 000 to 20 000 acres

by the early 1940’s (Rycroft, 1944). Rycroft suggests that even if the 1880 estimate was

high, it does mean that there might have been as much as 80% reduction of forest area

in only 60 years, the chief factors causing this reduction being fire and over-exploita-

tion. Taylor (1963) in a report on the Nxamalala Forest states that this forest, which

was about 8 000 acres in area in 1880, had been reduced to a mere 1 500 acres 70 years

later.

OBSERVATIONS ON CANOPY TREE GROWTH AND REGENERATION

In 1929 the resident forester at Xumeni Forest, near Donnybrook, laid out a line

through the forest and on it recorded the circumference at breast height of all trees in

40 one chain square, systematically placed plots. In 1966, 37 years later, the same

trees were re-measured (Moll & Woods, 1971). The results showed that the mean

increment rate was very slow, 0,201 --0,05 (n-+160) inches (5,10 =0,127 mm) in

circumference per year. Common trees at Xumeni are Podocarpus henkelii, Kiggelaria

africana, Xymalos monospora, Podocarpus falcatus and Fagara davyi. Of these species,

F. dayyi, P. falcatus and K. africana grew the fastest.

Moll & Haigh (1966) wrote of Xumeni that, “regeneration was poor, and it

would appear that, under natural conditions, regeneration is not sufficient to main-

tain the forest”. Xumeni Forest has been protected by the Department of Forestry

since 1910.

In 1967/1968 Moll (in preparation) collected density data of all woody plants

from twelve 40 x 40 m stands in the Karkloof Forest. The data indicated that the

species regenerating were not those species that are presently important in the canopy.

Common canopy trees in the Karkloof at present are, Xymalos monospora, Podo-

carpus latifolius, Celtis africana, Calodendrum capense, Fagara davyi, Kiggelaria

africana, Podocarpus falcatus and Ptaeroxylon obliquum. Canopy tree species most

common as young trees are Ptaeroxylon obliquum, Podocarpus latifolius, Olea capensis,

Podocarpus falcatus, Vepris undulata and Celtis africana, while canopy tree species

most common as seedlings and young saplings are V. undulata, P. obliquum, P.

latifolius and C. africana. This means that species such as X. monospora, C. capense,

F. dayyi and K. africana are not regenerating and in future years will be unimportant

canopy trees.

Taylor (1961) observed that on Miss Morton’s farm in the Karkloof there were

many seedlings of Celtis africana, Cussonia chartacea and Ptaeroxylon obliquum. He

also noted that where cattle grazed the forest the tree seedlings were unable to ad-

vance, and that the two species common at Morton’s, Podocarpus henkellii and

Ocotea bullata, were not regenerating.

Moll (1965) observed that regeneration in the Dargle forests was restricted to

Podocarpus falcatus, P. latifolius and Ptaeroxylon obliquum.

The general conclusion which can be drawn from these observations is that the

species regenerating are those capable of tolerating a drier climate. Added to this, in

areas of forest which have been protected, such as Xumeni which has been protected

for the last 60 years, regeneration is poor. Seedling density in the Karkloof of 48

Vepris undulata, 32 Ptaeroxylon obliquum and 16 Podocarpus latifolius per hectare, is not

indicative of active regeneration, not when one compares this to seedling densities in

actively regenerating forests on the coast, such as at Hlogwene (Moll, in preparation)

where, for example, there are 131 Olea woodiana and 94 Strychnos decussata seedlings

per hectare. In addition, seedlings of trees which prefer a cool moist environment,

such as Ocotea bullata and Podocarpus henkelii, are extremely rare.

FACTORS CONTRIBUTING TO A DRIER FOREST CLIMATE

Moreau (1966) states that in the last 18 000 years the temperatures in Africa have

risen by 5°C. Stuckenberg (1969) quotes Van Zinderen Bakker (1963), who states,

“Tt has often been said that changes in temperature of the magnitude of only 5°C are

of minor importance in a tropical continent such as Africa. These changes have,

however, been of very great significance........ Little but consistent changes of this

nature can have an enormous influence on the distribution of plants and animals.”

Stuckenberg also quotes Bailey (1960), who states that these temperature changes

affect maritime climates most. The Mistbelt Forests in Natal are influenced to a

considerable degree by weather from the Indian Ocean.

598

Accepting a rise in temperature of 5°C during the last 18 000 years means that

evaporation alone would be greatly increased. The mountain biomes which were more

extensive are now much reduced; the montane limit, according to Moreau (1966), was

about 2 300 ft (700 m), and is now 5 000 ft (1 500 m). Acocks (1953) also suggests

that forest and scrub forest has largely disappeared in Natal, and that the drier

vegetation types of bushveld and grassveld have greatly increased (see Acocks’s

Maps 1 & 2).

In addition to climatic changes, natural fires, and more especially man-made

fires, have become more numerous and these too have contributed to forest destruc-

tion; both directly and also indirectly, by increasing runoff. Furthermore, cattle

grazing in the forests not only eat and trample the vegetation, but also open up the

margins, allowing wind to penetrate beneath the canopy and fires to enter protective

marginal vegetation.

PRESERVATION REQUIREMENTS

Referring again to Acocks (1953), we are warned that unless our vegetation is

scientifically managed the drier vegetation types will expand further. If we are to

preserve an example of Mistbelt Mixed Podocarpus Forest we will, therefore, have to

manage it.* It has been shown by a few conservation minded farmers who live in the

Karkloof and Dargle areas that indigenous trees, such as Ocotea bullata and Podocar-

pus henkelii, if planted and cared for, grow relatively rapidly. However, the first

priority is to have a sufficiently large area of forest proclaimed as a Nature Reserve.

Once this has been achieved active management must include tree planting, run-off

retention and protection of the margin from fire. Also large grazing and browsing

animals must be excluded from the forest. Such management would have to be linked

to a scientific monitoring programme, designed to measure which management

practices are most beneficial in insuring maximum forest development.

REFERENCES

Acocks, J. P. H., 1953. Veld types of South Africa. Mem. Bot. Sury. S. Afr. No. 28.

Bews, J. W., 1912. The vegetation of Natal. Ann. Natal Mus. 2: 253-331.

Epwarps, D., 1967. A plant ecology survey of the Tugela Basin. Mem. Bot. Surv. S. Afr. No. 36.

FourcApDE, H. G., 1889. Report on the Natal forests. Natal Blue Book. Pietermaritzburg: W. Watson.

Mott, E. J., 1965. An account of the plant ecology of the Upper Mgeni Cathcment. Unpublished

M.Sc. thesis, University of Natal, Pietermaritzburg.

MOLL, E. J., (in preparation). A quantitative ecological study of the Karkloof Forest, Natal.

MoLtL, E. J., (in preparation). A quantitative floristic comparison of four Natal forests.

Mo tt, E. J. & Woops, D. B., 1971. The rate of forest tree growth and a forest ordination at Xumeni,

Natal. Bothalia 10: 451-460.

Moreau, R. E., 1966. The bird faunas of Africa and its Islands. Academic Press.

Pentz, J. A., 1945. An agro-ecological survey of Natal. Dept. Agric. and For. Bull. No. 25, Govt

Printer, Pretoria.

Rycrort, H. B., 1942. The plant ecology of the Karkloof Forest, Natal. Unpublished M.Sc. thesis,

University of Natal, Pietermaritzburg.

Rycroft, H. B., 1944. The Karkloof Forest, Natal. J. S. Afr. Forestry Association. 11.

STUCKENBERG, B. R., 1969. Effective temperature as an ecological factor in Southern Africa. Zoo/lo-

gical Africana. 4 (2): 145-197.

TAyior, H. C., 1961. The Karkloof Forest, a plea for its protection. Forestry in S. Afr. 1: 123-134.

TAYLOR, H. C., 1962. A report on the Nxamalala Forest. Forestry in S. Afr. 2: 29-S1.

WEATHER BurREAU, 1954. Climate of South Africa. Part 1. Climate Statistics. Goyt. Printer, Pretoria.

* Only 0,8% of the total area of Acocks’s (1953) ““Temperate & Transitional Forest and Scrub

Types” is conserved (Edwards, 1971).

Bothalia, 10, 4: 599-614

A Phytosociological Study of the Cape Fynbos and

other Vegetation at Jonkershoek, Stellenbosch

ML |: A. Were 18, |, Ieqeery ail Ie C, llandon

ABSTRACT

The Braun-Blanquet phytosociological method was tested in the complex Fynbos vegetation of

the South-western Cape Region of South Africa. In the Swartboschkloof Nature Reserve, Jonkers-

hoek, the Fynbos, riverine scrub and forest vegetation was classified preliminarily into eight com-

munities, which are described floristically and related to habitat. The results hold promise, and the

possibilities of classifying the Cape Fynbos in a formal phytosociological system are discussed.

INTRODUCTION

The Braun-Blanquet phytosociological method commonly used in Europe has

remained relatively unknown in Southern Africa. Possible reasons for this are

language difficulties, the need for more general, rather than more detailed information

on the vegetation, and the general non-acceptance of the method by English and

American plant ecologists. For many years the only source of information in the

English-speaking world was Fuller & Conard’s (1932) authorised translation of

Braun-Blanquet’s first edition of Pflanzensoziologie (1928), a work that omitted certain

important details of the phytosociological technique. More detailed German works

of the phytosociological school were largely inaccessible (for example Ellenberg, 1956;

Braun-Blanquet, 1951. 1964).

In recent years an English evaluation of the method was given by Poore (1955, 1956),

although his main criticisms were shown by Moore (1962) to be largely unfounded.

Becking (1957) reviewed the phytosociological school and its concepts, and Kiichler

(1967) translated the tabulation techniques from Ellenberg (1956).*

Originally, nearly all vegetation surveys undertaken in Southern Africa were on

a physiognomic or on a non-formal descriptive basis, with the exception of Acocks

(1953) who used his own floristic technique to construct a system of veld types.

Recently, multivariate analysis and ordination techniques have been applied by a

number of ecologists in the Republic of Southern Africa (see Killick, 1966-67).

In the Portuguese territories physiognomic classifications were created, and in each

unit one sample was taken. The species list from each sample was rated on the Braun-

Blanquet scale of cover-abundance and sociability (e.g. Gomes Pedro & Grandvaux

Barbosa, 1955). In Central Africa, Belgian ecologists, traditionally familiar with the

Braun-Blanquet method, have set up a hierarchy of syntaxa for the Congolese rain

forests (Lebrun & Gilbert, 1954).

In Southern Africa it has often been suggested that the flora is too rich to apply

successfully a floristic technique. In the Cape Fynbos, particularly, such techniques

were thought to be impracticable. Outside Southern Africa also, Donselaar (1965), ina

classification of the Northern Surinam savannas, stated that to use successfully the

Braun-Blanquet method the number of species must be moderate.

* Botanical Research Institute, Department of Agricultural Technical Services, P.O. Box 994,

Pretoria.

+ Jonkershoek Forest Research Station, Department of Forestry, Stellenbosch.

+ Botanical Research Institute, Department of Agricultural Technical Services, Stellenbosch Unit,

P.O. Box 471, Stellenbosch.

1The percentage classes of Braun-Blanquet as listed in Table 10, p. 231, Kichler (1967) are

not correctly reproduced, as they give the impression of being based purely on cover. In reality they

are based on cover and abundance.

600

From systematically distributed quadrats used for an association analysis of the

vegetation of the Cape of Good Hope Nature Reserve, Taylor (1969) also prepared a

Braun-Blanquet synthesis table. Due to the rigid sampling technique, a number of

quadrats had to be excluded from the table as they were located on community

transitions and were not representative of more or less homogeneous stands of

vegetation. Nevertheless, he obtained associations that were recognizable in the field.

This stimulated the authors to undertake a survey according to the phytosociological

techniques described by Braun-Blanquet (1964) and Ellenberg (1956), in order to

test its usefulness in the floristically rich Fynbos vegetation. The area chosen was the

Swartboschklcof—Sosyskloof Nature Reserve in the Jonkershoek valley near Stellen-

bosch, as the results could also be used for the International Biological Programme

survey of conservation sites.

THE STUDY AREA

The Swartboschkloof—Sosyskloof Nature Reserve, 373 hectares in extent, lies in

the Jonkershoek Forest Reserve in the Hottentot Holland Mountains some 15 km

from Stellenbosch, at 34°00’S latitude and 18°57’E longitude. It was proclaimed a

nature reserve in terms of the Forest Act in 1936, to conserve vegetation representative

of the Jonkershoek valley.

The vegetation is chiefly Fynbos (Acocks 1953, veld type 69). It consists mainly

of sclerophyllous scrub and dwarf scrub, in which amongst other, Restionaceae and

Cyperaceae also play an important role. A dense scrub grows along the river banks,

and upstream, where the valleys narrow, small patches of forest occur. Van der Merwe

(1966) analyzed vegetation-site units in the Reserve.

Topography, Geology and Soils

The Reserve is a fan-shaped valley at between 285 m and | 200 m altitude. The

steep slopes average about 30°, ranging from about 5° to 50°. Aspect is largely northerly

with only about 5 per cent of southerly facing slopes.

Porphyritic granite of the Pre-Cape System forms the undulating floor of the

valley, but is often buried by sandstone talus fans from the medium-grained homo-

geneous sandstones of the Table Mountain Series. The sandstones are very resistant to

erosion so that the valley is bounded by precipitous cliffs and knife-edge ridges.

Continuous creep characterizes the talus on the slopes below the cliffs.

The valley was formed by a series of secondary faults, roughly at right angles

to those which gave rise to the main Jonkershoek valley. Streams follow the courses of

these faults. The jointed layers with a northerly tilt store a fair proportion of the

precipitation which is released in conspicuous seepage steps.

Soils are generally poorly developed, rocky and acid, the granite soils being

less acid than the sandstone. Those soils derived from granite are often over 1 m deep

and more fertile than those from sandstone, which are coarse-textured, skeletal and

rarely over 1 m deep. Mixtures of granite and sandstone debris give rise to inter-

mediate soils. Humic and organic soils of varying depth (0,05-2 m) occur locally

on wet sites.

Climate

Wicht ef al. (1969), in describing the climate of the Jonkershoek valley in some

detail, notes that ““The climate is Etesian of the Mediterranean type,........ with a

dry summer and the average temperature of the warmest month below 22°C........

[t conforms to Walter & Leith’s (1960) Climate Type IV, and to Képpen’s (1931)

humid-mesothermal Erica-climate. Precipitation is usually associated with strong

cyclonic winds from the north-west. In summer, strong anti-cyclonic winds from the

south-east prevail, These are generally dry, but frequently deposit moisture at highe

601

altitudes. From a rain-gauge at the mouth of the Swartboschkloof valley the mean

annual rainfall over 20 years is about | 600 mm. Fifty per cent of this fell during the

months May to August, and only 12 per cent from December to March, when the

greatest moisture deficits occur. Snow falls rarely at higher altitudes.

History of anthropogenic influence

For millenia the Jonkershoek Valley was visited intermittently by Khoisan

tribes, but they probably seldom settled there (Schapera, 1930; Seddon, 1966, 1967).

Colonists first settled in the Valley in the late 17th century, but the effects of their

agricultural practices on Swartboschkloof was probably first evident after 1832, when

the Reserve formed part of land transferred to a farmer living close by. The early

farmers used these lands as grazing for sheep, goats and cattle, exploiting the forests

for timber, the Protea arborea stands for firewood, and such shrubs as Agathosma

crenulata for medicinal uses.

Early man probably had little effect upon the vegetation, but European coloniza-

tion would have initiated rapid change. Veld-burning to improve grazing was fairly

standard practice, and Swartboschkloof was probably burnt at 4-10 year intervals—

more frequently than probably occurred naturally. Grazing would have been restricted

largely to the lower slopes owing to the steep topography. The area does not appear

to have been cultivated. Mammal populations diminished or disappeared as a result of

hunting and the destruction of habitats in the surrounding lowlands.

The acquisition of Jonkershoek by the Department of Forestry in 1933 and

proclamation of the Nature Reserve introduced radical changes in land-use. A

policy of complete fire protection was adopted, and only recreation and non-destruc-

tive research were permitted in Swartboschkloof. The Reserve is surrounded by

firebreaks, burnt in spring or autumn on a 4-year rotation. In spite of policy, the

whole reserve burnt accidentally in December 1942 and February 1958. Two small

fires also occurred during this period.

METHODS

Gleason’s (1925) regression equation, as reformulated by Goodall (1952), was

applied to data from three sets of nested quadrats, | m? to 256 m? in size, to estimate

the quadrat size-information ratio (Werger, 1970). On this basis a quadrat size of

100 m? was selected. The Braun-Blanquet method does not require uniform sized

quadrats, but as many as possible were 100 m? for the sake of consistency. Only

two quadrats (nos. 14 and 37) were taken at 50 m? to avoid obvious heterogeneity

due to a sharp change in slope aspect and to a conspicuous vegetational difference.

Quadrat nos. 42, 43 and 44 were 128 m?, being a stage in a set of nested quadrats.

Altogether 44 quadrats were layed out at sites selected on the basis of visually asses-

sed homogeneity of vegetation structure and habitat. Species in each quadrat were

listed and rated on the Braun-Blanquet cover-abundance scale. Only permanently

recognizable species were recorded, thus omitting annuals and most geophytes. Site

features such as slope angle, aspect and altitude were measured, soil depth was

estimated, and geological and geomorphological characteristics were noted.

The data were entered in a raw table and from this an association table was

compiled (Tables 1 and 2), according to the prescribed methods. Tables 1 and 2 list

character species for each community, together with those species with over 50 percent

presence in a group of two, three or four related communities. The remaining species

are grouped in a “‘tail”’ of the tables. Of this “tail”, species occuring only once or twice

and with low cover-abundances in the quadrats, are listed for convenience in the

Appendix. Complete species-quadrat lists are considered important for Braun-

Blanquet phytosociological work, because some of the species will assume greater

importance when further data are acquired, and are necessary if current concepts of

the communities are to be revised.

602

The communities are named after one or two faithful and conspicuous species, so

that they are easily recognizable in the field.

To obtain complete pictures of the communities and their relationships, more data

are needed, preferably from other areas. Only then will it be possible to decide with

certainty whether some of the below mentioned species are true character species.

THE COMMUNITIES

Based on floristic composition we have distinguished five Fynbos, one riverine

scrub and two forest communities.

Fynbos

The Fynbos communities usually consist of two or three layers: a graminoid

and dwarf shrub layer, a shrub layer and, in some communities, a tall shrub or small

tree layer (Table 1).

1. Protea arborea—Rhus angustifolia Community

This community consists of many of the more common Fynbos species. The

vegetation usually consists of three layers: a tall shrub and small tree layer from

2-3 m high; a shrub layer I-1,5 m high; and an undergrowth of sedges, grasses,

restionaceous plants, other herbs and dwarf shrubs from about 0,10—0,60 m high.

The middle layer usually has the highest cover, although sometimes the upper layer

may dominate.

There are a number of possible character species, which are, however, not fully

constant, like Podalyria myrtillifolia, Euphorbia genistoides and Helichrysum zeyheri.

Constant but not strictly faithful character species appear to be Diospyros glabra and

Rhus angustifolia. A number of species that clearly typify this community and the

Restio perplexus Community are Protea arborea (which has its optimum in this

community), Psoralea obliqua, Watsonia pyramidata, Themeda triandra and Ursinia

filiformis.

The Protea arborea—Rhus angustifolia Community is found on relatively deep,

detrital, sandy loam soils; slope varies, but aspect is generally between north and east.

It is limited to lower altitudes (up to ca. 600 m) in Swartboschkloof.

2. Brunia nodiflora—Psoralea rotundifolia Community

This community generally has the usual three strata: a tall shrub and small tree

layer (1,5-3 m), a shrub layer (0,60—-1,20 m) and an herbaceous dwarf shrub under-

growth (0,10-0,60 m). High cover values are obtained, especially in the undergrowth,

although sometimes they are high in the upper layer as well. The community has a

number of good character species: Brunia nodiflora, Psoralea rotundifolia, Helichrysum

teretifolium, Corymbium scabrum, Danthonia lanata, Osteospermum tomentosum and

Tetraria burmannii. Abundant here, but common also to the Thamnochortus gracilis-

Hypodiscus aristatus Community, is Restio filiformis (Table 1).

Species common to this community and the Protea arborea—Rhus angustifolia

Community are Anthospermum aethiopicum, Montinia caryophyllacea, Asparagus

thunbergianus, Diosma hirsuta, Ficinia filiformis, Lichtensteinia lacera and a number

of other species.

The community occurs in a very distinct habitat, with generally steep slopes

(16°-32°) and relatively deep detrital soils, with a definite south-eastern aspect. It

occurs mainly at altitudes of 300-500 m.

3. Thamnochortus gracilis—Hypodiscus aristatus Community

The shrub layer (0,75—1,50 m high) is the most important stratum in this commu-

nity. There is an undergrowth of about 0,20-0,50 m high, but the upper layer of ca.

2 m high tall shrubs and small trees if often wanting.

ae AS She CRO Oae Steno can ss a6 dws aks e

Be ee | oO Ae ene ap Ves eos came Ba eK weer

Character species Prot. arb.-Rhus ang. Comm.

Diospyros glabr. ie Waete.; 6. ns cheat ee

oy en Ri at — Et eee eee

Character species Thamn. grac.-Hypod. arist. Comm.

lia

TD OER MNES. bo ks Fickle

Sy ee VS. SO? Cee athe” 5 is

Character species Restio perpl. Comm.

Cina svete. ..0: a... 2. te +

Pestie Sepler Meath... . .... dee sacs den ke

Asparagus

Venidium semipapposum DC........ 1.2.2.2 2222772710

Character species Berz. lan.-Osm. ast. Comm.

Gomes qhmeenta Mite: :. 665526 scess ce. Ne. *

seoraeeaphpla Vos. |... «Ben. oe 3

Osmitopsis asteriscoides (L.) Cass... . . e

pt pom ate yy ae SE ER S2 TAS. t

leg. n. E EPR Pe SS. as

ucade: icifolium (Salisb.) Williams... . ‘.

Restio gr i MAME, co Stare oe ~

Cliffortia graminea Lf.................... ~

cesenenn painciiries GOB OV Ig oa ca Oa ce, 2

Oe oe se lg SR RR sane 2 Sere «

ptocarpus paniculatus Pillans..................2000...- -

Species common to Prot. arb.-Rh

rot., Thamno., grac.-Hypod. edit.

PIROOPOROR INT QUMMME TORE. .ic cc ccc cc ccwcbacveccsess

Aristea th

sete eeeeeeseeeeeeeeesseessessesesen

: ++ pt: oo eG

SGhaceete > <statt cote

Cliffortia ruscifolia L.

Lie es

+e be tt

eee eee eresceseeeseesaesesessesesseesses

Anthospermum aethiopicum L..... 2.0.0.2 =

Montinia caryophyllacea Thunb,................. : .

Restio gaudi -

common to Brun. nodifl.-Psor. rot. and Thamn. grac.-

Hypod. arist.

Restio filiformis Poir.

Tetraria

O.0 gt O06. oli

* w?

‘thor th

Ca 2

RS at ae t Faye

+mmt iy te te te tenet oe ce

Cig tt 88 8 fee e 6 eh

I A EC a a Oe Se

2 SE? Pete Reese 6. 8 5 ae Bar,»

‘H+ pwtttpot

+++: =

ee OES EMR OF rs a 8 OEM wale ig Cevs 6 Boek D abghe-s

bit te

Ce oi i st Oe Oe a A OM a a |

+) t4++u: ttt 1 pe +) ttt: ++.) Pete eee eens

Ja od < ren

+ +

eee RR I is ae I RC OO SO oe eo CNET

Se eee ae

tr eee

tee + eee ne ae Sete eee es

Hh E+ be tomes

4 +mty

ee a

ee et LUE Re TO on ce arene

+ t+: +++ pre wett 4) @aee yl) ae ° 4.w ge oo 97,0" 9 Paes a eee ee bee fs Se ls SC

CR oe eee

ge tats hE > Cage ae Nk: CO OR lg ee rma

HaZortu

git

+) tet

be wt: tia?

net eeton aiedien =f

ree he

Peg e-y

FI fe ee se ee

ee hats sa Gat

Bye

Susvz

i ot oi

the bie

+++ t= met te ttt teens

totttuttete

* A at ee

Oe es ee

Ore 670le 6, 60 ieee Ue éley.e & 8's

"hee +++»

mt EB aH ees Tras! Grtd mae

Set Orr

ee a

CT ie PO eC ea ee ree? Pee

ean! Syeda al doll Ie SO Se See 8

. ++: a

+: wtt: + os?

++: Fy

i So Cee aes

Or iy OTL NOT nao 6. atiabittn es (0 ae. of «

ot oa Pt stat “Ho “He "2 8 ie ie od .

' amt t mete a,

+++ ttt ++

“

se es he a ee Oe ee

+++++

$8 et matt tt++ ++++ te mt

Seva enck is

dike be ha th en Lee

a i i ie er A

mttttte tet

Hebe bt

tf tttt++uttet+

a oo er aA he’

le dai erat SUBCL OMC” Oe aoae Y e e e

ee ee

¢ yt 28456 tones

+e ttt: tet tees

be pat

+? wm! F mms tas -" ++ pt+++: +++++

+b mts +

. ache him

Pe a OE Be aero ees

sae) a RT A RU a a a

® +++ a! +: i

eve she + mpt tHtyot

es ee

“°F

mt?

Re ee” cease, eR) eM eee wabh. hele ol abe 8 vs k tcebobs

nt bitte

tA Male a pe Me ER?

sh ae as $e te tt ttt

+ ant om * ++

om’ + ty

Cte Pet abe eee dt

Diet eee COGS ER Oe CMe MRR vA tS bah re o's, obey oe

* te ms yy? “b tae

. -

- .

. .

. -

- .

- -

- .

- -

tebe tte

tHe tte tye + t+t+ ute

+utt: *

: a ee

eo ae oe

SN ah os hale, Ce Co aoa ae oe 8

i i

Fin! + a t+

SR edie tt ote Sm

per Se tee, ene

9.6 0 Ben wee 0 Wd sue e's

Wise @reng « 2

* mt+: ++ tp’ was me “fof

‘ator +

a ee

re | OC ae eet ee

a op! $+: mt + ae ome tee te te fe

bY tb ttt

ie oie a ++ ws

te be be

+ oofe

, C7 9."s see

Ble ta gee TENTS SRR TOV 8 Grey ew Als das As ele laldyve

> P16, 9 Cae arty we

° nN" ++: at f Pe -* Ott es et +

oa te

tte te

ual

it Mie ee Spee aa Ua Oe a Oe at

wehaee

atapZsey

ad ee Ma

Pet eH Pt P+

eee ee Te

o 0 oo of

er tk a + vs * oes =m? ++

pA 8 ee ieee

+) ttt

ee eile a are One re Oe a ae a

”

Po ae wt! be bp ae

te +

a a

a ee

+. ° +++ +

NEE a Oe Ear r ager? eb Sede 6 oe ore he dies ocebedy

atutar

w’' + om es ome EE ny

ye? ot 7

2 ee oe ee

ewe 8 Lae abe

BE pas ae OR Ore. a a A olee Figae gi te

aneSep

Ih ZB ry

Btn

me ma t+ tw +:

‘meth

a a

pi a ee RITE ae I Oe a Se So

oe hi

SRE SSS ee Cn Re hT gg RL ys

mmm! t+ +++ he Lae

8 eae

man’ * A tas

eee 00 ates oe

fut

a0... wees s

og te DAE * SS Seay, 8, 9 alah.

. + ms nme

Wa Fetes. S66 ea a

teats $s $44

++++:

+++.,° os

Ane a SORE SEPA GEN IR: OA aa O'R A Shey nie aly s AY age ob Conia 0 6, oeohhip

++ rrp 6 298, hee ye

ie: tran’ =wt

ge SPP ERS esti Sa Ge Ue ee Gedub ee Gs

rv)

Oye eee eh

+++

‘ mtiat: ++:

‘bin

Game © *

See ey en Cee WANES. 66 Fir 608 a CAA ely sb pie oe ble 0 ete

omens

ae AS LS i Coe Oe le ee pk, £18 ia @

‘Ww Nanw' vn +

& 7am 6 NOW, 8 88 & ee Ee ae

BO im a 6h een ees DS

vi) 8 2 ele «6

Oe © Te me 6

rece oes

I ae ae ee ee ee

Pp ee Ph ee 6 Re 6 Ome. ee 2 es ee Oe ates tr O Or 2's 6 hee oe Ce oo

* Rw’ ww Note

i Oe 0)

Cer MARR Ge Se OOP Cletatele ewe 46 Bk ok, hee 0 baie ans

>. et He

fo.) ¢ Bie ¢

i ee Uh Le

I ere a a a |

605

A number of good character species typify this community: Thamnochortus

gracilis, Hypodiscus aristatus, Anthospermum ciliare, Staberoha cernua, Prismato-

carpus diffusa, Tetraria capillacea, Tetraria fasciculata, Willdenowia sulcata, Blaeria

dumosa, Clutia polygonoides and others (see Table 1). Restio sieberi is highly constant

but common also to the Restio perplexus Community.

Many of the species common to the Protea arborea—Rhus angustifolia and the

Brunia nodiflora—Psoralea rotundifolia Communities, are rare or absent here, but as

Table 1 shows, there are many others, especially Leucadendron adscendens and

Tetraria bromoides, that are common to all these communities.

The community occurs at fairly high altitudes (--500-800 m), on Table Mountain

Sandstone soils of various depths. Slopes are moderate to steep (13°-34°). There is no

prevalent aspect.

4. Restio perplexus Community

Although this community is only represented by three samples, it is, nevertheless,

quite distinct from the other communities. The community is usually two-layered

(undergrowth layer of herbs, sedges, grasses and dwarf shrubs 0,20-0,40 m high;

shrub layer 0,60-1,50 m high), although there may be an open small tree layer 3-4 m

high.

Restio perplexus appears to be a good character species. Cliffortia polygonoides

and Euryops abrotanifolius are abundant and/or constant, but not strictly faithful

(see Table 1).

Again, some species, for example Clutia alaternoides, are common to all four

communities so far described, while many of the species generally occurring in either

the first two communities or the first three communities are absent in this Restio

perplexus Community. As already pointed out, this latter community has a number

of species in common with the first Protea arborea—Rhus angustifolia Community.

The relationship between these two communities is, however, not yet clear.

Relevé 34 cannot be unequivocally placed in this Community, because it contains

species typical of the Thamnochortus gracilis—Hypodiscus aristatus Community. This

relevé may be more heterogeneous than was thought, due to its position in a firebreak.

The community occurs at high altitudes (-+-800-1 100 m) on soils of ca. 50 cm

depth on Table Mountain Sandstone. Steep slopes (25°—40°), with a general north-

eastern aspect, seem typical.

5. Berzelia lanuginosa—Osmitopsis asteriscoides Community

This Fynbos community on seepage areas along drainage lines is very different

from all others. The vegetation is dense, with an undergrowth of ca. 0,50 m high, a

middle layer of sedges and shrubs 1-2 m high, and an open tree layer 3-5 m high.

The permanently moist soil often contains a high percentage of organic matter.

Character species are thought to be Berzelia lanuginosa, Osmitopsis asteriscoides,

Carpha glomerata, Leucadendron salicifolium, Elegia capensis, Restio graminifolius,

Cliffortia graminea, Tetraria punctoria, Elegia thyrsifera and Leptocarpus paniculatus.

Very few of the species common in the other Fynbos communities are present in this

Community.

Riverine Scrub and Forest

The stream channels are generally fringed by woody vegetation. This is largely

a dense scrub, but locally real forest occurs, especially where the valley narrows.

Small patches of forest also occur on coarse, loose scree. These woody communities

have a few faithful species in common (e.g. Secamone alpini and Restio subyerticil-

Jatus) and there is a mutual overlap in species between the different communities

(Table 2).

606

TABEL 2.—Forest and Riverine Scrub Communities

IRAE TUN OSs oococcaeoeoncbooocuoce 31 37 39

|, Oo 2 | i 6 2%

Total number of species................ DY 30 34 24 18 18 15 10

Altitude (mm) ieee ae een cise: 350 330 320 | 560 700 | 640 640 510

IAS DEC tras eee ER ee Lee | WN N Ww | ww NB | 6B E N

Sloper) terse ere hear a erceieriee are | 0 0 @ | 15 25 35 35 15

Geology (g = Granite, T = Table Mt. |

Sandstone) eee eee | gT gT eileen od wr T ‘I T

SOM GENIN GBB), occ 009000000000000000C 1 I ty il

motalkcoven (CA) sae saa eee / 100 100 100 ; 95 85 90 90 85

Character species Brab. stell. Comm. | |

Podalyria calyptrata Willd.............. eas 3

Brabeium stellatifolium L............... 3 4

Halleria elliptica Thunb................ ar 1

Blechnum capense (L.) Schlecht.......... I |

Pentameris thuarii Beauv................ r iF

Freylinia oppositifolia Spin,............. 2) AF

Restio quadratus Mast.................. +

Asparagus asparagoides (L.) Wight....... r

Mettrosideros angustifolia Smith......... :

IMOTHOD SAAR LEWIN 55566050000008000" : 1

Character species Rap. mel. Comm.

Rapanea melanophloeos (L.) Mez......... : ; ¥ 3

Ka gcelariqvasricananle eee reer : : : 1

Pellaea viridis (Forsk.) Prantl........... |

Asplenium aethiopicum Backer........... |

Character species Heer. arg. Comm.

Heeria argentea (E. Mey.) Kuntze....... 6 0 :

Rumohra adiantiformis (Forsk.) Ching... . |

Aloe mitriformis Mill...................

Oftia africana (L.) Bocq................ é é : : 6

Species common to Brab. stell., Rap. mel. | |

and Heer. arg. Comms.

Maytenus oleoides (Lam.) Loes.......... 1 I Be |

Secamone alpini Schultes............... : : aE |

Hartogia schinoides (Thunb.) C. A. Smith ‘ ; r |

Restio subverticillatus Mast............. 2D + : : + |

Blechnum punctulatum Sw............... | . | 2 .

IMiynsinenafricanapls eee ee Ns hee . . SF . ar

Knowltonia capensis (L.) Muth.......... + : 0 ar =F =F . .

GChironiaibaccif crane ern eE ert ee + : ; : r r

Species common to Rap. mel. and Heer. |

arg. Comms.

Podocarpus elongatus (Ait.) Pers.........

Oleaiafricana Nill eempreneeereeeenoe

Maytenus acuminatus (L.f.) Loes........

CNA RA UWTCTE soo cc 00c0a900000000000 : ; :

Zantedeschia aethiopica Spreng.......... | - ; : . |

Olmaieymosagihunb see eeee eee > : : 1

Species common to Brab. stell. and Rap.

mel. Comms.

Asparagus scandens Thunb.............. + : ty a +

Grinoniaicapensisnlee ere eee 1 r : : +e

Ilex mitis (Jacq.) Radlk................. | : 2 2 :

Brachylaena neriifolia (L.) R. Br......... 4 2 2 | ar . |

Oplismenus hirtellus (L.) Beauv.......... | E r

Other intruding species

Ehrharta ramosa Thunb................ ; 1 + . > |

Pteridium aquilinum (L.) Kuhn.......... sr 1 se . ; |

Asparagus thunbergianus Schult. f........ ++ ae ++ |

REDS CAGISITOWA ssn oaoo0000000000008 ar +

GassythareiliolarayNecsteeene eee ++ AF qe ||

Diospyros glabra (L.) de Winter......... | ae af ° ar

Ui oiHi@) QUAGOO /Nilioccg00o00000000000 | I se Samad | eb

Aristea thyrsiflora (Delar.) N.E. Br........ { +

Restio gaudichaudianus Kunth........... :

Elegia capensis (Burm.f.) Schelpe........ | : 1 +

- ptt

mot =

Sater

Pee

We Ww

NT eNVwr

sie

INSP SE SF to

oo ee SS ee

607

6. Brabeium stellatifolium Community

This dense scrub up to 5 m high fringes the lower, less steep parts of the streams.

Character species include Brabeium stellatifolium, Podalyria calyptrata, Halleria

elliptica, Blechnum capense, Pentameris thuarii, Freylinia oppositifolia and others.

More information might necessitate subdivision into two communities, one with the

character species Meterosideros angustiolia, and the other lacking it.

7. Rapanea melanophloeos Community

This community constitutes the largest part of the forest vegetation. It occurs at

higher altitudes along stream channels, with steep gradients. Further downstream the

rivers are fringed by the Brabeium stellatifolium Community.

There are two tree layers. The highest reaches 10-12 m, the other 5-8 m, beneath

which is a “layer” largely formed by saplings. There is no shrub layer and very little

undergrowth. Character species of the community are Rapanea melanophloeos and,

possibly, Kiggelaria africana, Pellaea viridis and Asplenium aethiopicum.

8. Heeria argentea Community

This forest community occurs in small patches on coarse, loose screes. There is

usually a singly tree layer, with an average height of 7-8 m, sometimes reaching 10 m.

The undergrowth consists of herbs and small shrubs up to 0,60 m high with a low

cover, The trees are covered with lichens and mosses, indicating that moisture con-

densation often occurs.

Character species are Heeria argentea, which has a high cover, and possibly

Rumohria adiantiformis. The species Aloe mitriformis and Oftia africana may also be

considered character species of the undergrowth where the tree canopy is open.

The two Rapanea and Heeria forest communities in Swartboschkloof have some

species in common of which Podocarpus elongatus is the most abundant. This species

and other common species, such as Olea africana, Maytenus acuminatus, Hartogia

schinoides and Olinia cymosa, all form canopy trees, while Halleria lucida and Myrsine

africana occur regularly in the understory.

DISCUSSION

The main purpose was to test the Braun-Blanquet phytosociological method on

the floristically rich Fynbos vegetation. In that a practical classification into com-

munities based on floristic criteria has been possible, and because these communities

can be correlated with definite environmental factors, we may conclude that the

Braun-Blanquet method can be successfully applied to the analysis of Fynbos vegeta-

tion. This contradicts Donselaar’s statement (1965) that the species number must be

moderate if the method is to be successful.

Walter (1968) states that, on the South-west Cape mountains near the coast,

aspect plays a more important réle than altitude. The rather strict aspect preferences

of the Protea arborea—Rhus angustifolia and the Brunia nodiflora—Psoralea rotun-

difolia Communities seem to confirm this, although the Thamnochortus gracilis—

Hypodiscus aristatus Community does not show any aspect preference, and is corre-

lated in Swartboschkloof with altitude and geological substrate. However, it is

difficult to distinguish whether altitude or geological substrate is the more important,

since these two factors are correlated with each other.

Adamson (1931) sampled the vegetation of north and west slopes of Table

Mountain between 300 and 360 m, on granite with overlying Table Mountain Sand-

stone talus, in order to find “some explanation of the apparent variability of the

floras’ of the communities which he floristically distinguished. Although he noted that

“the floristic composition of a community must ultimately be its most important

characteristic”, he could not prove this when he tried to express it in terms of species

dominance, concluding that there was life-form dominance rather than species

dominance. Nevertheless, our results show that when floristic composition 1s used asa

criterion, it is indeed possible to characterize communities.

608

It is remarkable that conspicuous vegetational structures do not always coincide

with the boundaries of communities defined upon floristic composition. We find the

dense sclerophyllous Protea neriifolia and Protea repens dominated scrub distributed

in the three most common Fynbos communities, although mainly concentrated in the

Protea arborea—Rhus angustifolia and the Brunia nodiflora—Psoralea rotundifolia

communities. The relationships between physiognomic structure and phytosociological

classification have been discussed by Westhoff (1968). He uses the term “Twin Forma-

tion” for “‘strata which vary independently on [of] each other’’, or “stands which are

floristically hardly, but structurally considerably different.”’ These situations occur

where extreme habitat factors prevail. Burning 1s considered to be such an extreme

factor. Donselaar (1965) found that in the Surinam savannas the floristic composition

of some treeless areas was so similar to the undergrowth of a certain type of tree

savanna, that they should both be placed in the same association, arguing that the

floristic principle should predominate over the structural one. Dansereau & Arros

(1959), who favour the principle of structure, argue that decisions of the kind similar

to the above are too arbitrary and they doubt the reality of such associations.

Since Fynbos is recurrently burned and the structure of the Fynbos is very often

closely related to the elapse of time since the last burning, we feel, like Donselaar (1965),

that a floristically based system of communities is more consistent and practical.

It is interesting to note that Tables ! and 2 show the intrusion into the Brabeium

stellatifolium and Berzelia lanuginosa—Osmitopsis asteriscoides Communities of some

species that usually occur in other Fynbos. Such species are Asparagus thunbergianus,

Rhus angustifolia, Rhus rosmarinifolia, Diospyros glabra, Aristea thysiflora, Cliffortia

cuneata, Erica hispidula, Protea neriifolia, Anthospermum aethiopicum and Ehrharta

ramosa. Both the Brabeium stellatifolium and the Berzelia lanuginosa—Osmitopsis

asteriscoides Communities occur only as long narrow strips of vegetation along

stream channels and seepage lines. The Communities are thus characterized by a

high ratio of margin to surface area, or a large “‘edge effect.’ Because of this large

“edge effect,” intruding species from surrounding communities are more likely to be

found throughout such long narrow communities compared with communities of

another shape. Species that are most likely to intrude are those with a wide ecological

amplitude, as appears to be the case for some of the species listed above. Other

intruding species, like Pteridium aquilinum and Cassytha ciliolata, seem to be ones with

their optimal occurrence in ecotones, especially ecotones of a shrubby nature. In

Leeuwen’s (1966) terminology they can be called species of the “limes convergens”’,

that is species living under unstable conditions, usually as large numbers of individuals.

Adamson (1927) and Walter (1968) point out that on the Table Mountain Plateau

Schizaea pectinata can be very abundant and proves to be a good indicator of fire.

In Swartboschkloof this species was not abundant, however, and was not present in

the quadrats sampled on the firebreaks. Adamson (1935) found that after a fire in the

vegetation on the slopes of Table Mountain, Euryops abrotanifolius increased quickly

during the first three years and was dominant for the following two years. From then

on the species decreased rapidly. Our results from Swartboschkloof also link Euryops

abrotanifolius with the fire factor. In quadrats 34 and 35, situated on the firebreak, it

scored cover-abundance values of 2 (5-25°%), whereas it occurred in only two other

quadrats, in old stands, with low values.

Floristic data on Fynbos vegetation in the literature are limited, and complete

species lists of quadrat samples are unavailable. It is, therefore, difficult to link the

Communities presently distinguished, with communities described in other Fynbos

literature. As early as 1908, Marloth presented an outstanding description of the

flora and vegetation of the Cape, describing the vegetation on a structural and geogra-

phical basis. In his general descriptive account of the vegetation of Table Mountain,

Adamson (1927) mentions a number of species from the “....plateau. communities

on sandy soils....”, which are typical of the Thamnochortus gracilis—Hypodiscus

aristatus Community described here. The habitat and altitude of these two communi-

609

ties also show similarities. Further study may show closer correlations. In addition, the

Communities he briefly described from seepage areas and along stream channels,

show floristic similarities, through species such as Berzelia lanuginosa and Osmitopsis

asteriscoides, with those of Swartboschkloof. The same is true for gorge forest and

riverine scrub communities in both areas.

Taylor (1969) mentions, among others, Pentaschistis colorata and Staberoha

cernua as preferential species for one of two communities forming a mosaic in the

“plateau fynbos association” in the Cape of Good Hope Nature Reserve. It is probable

that this community is related to the Thamnochortus gracilis—Hypodiscus aristatus

Community. Certainly, his ““Osmitopsis seepage scrub association” is closely related

to our Berzelia lanuginosa—Osmitopsis asteriscoides Community. His description of

the Protea arborea pseudo savannah association” (waboomveld; Taylor, 1963), with

typical species Protea arborea, Montinia caryophyllaceae, Rhus rosmarinifolia, Leuco-

spermum conocarpodendron, Themeda triandra, Hermannia cuneifolia, Cymbopogon

marginatus and Bobartia incica, strongly suggests it belongs to the Protea arborea—

Rhus angustifolia Community.

Relevés from the Fynbos are expected to accumulate and will enable ecologists to

classify the vegetation in a hierarchical system of associations, alliances, orders and

classes. A special difficulty in classifying the complete Fynbos vegetation lies in the

fact that many species have a very limited distribution. This will make it necessary to

distinguish geographical races of an association, or regional associations with a

limited geographical extension. Communities earn the status of an independent

regional association only if they have their own character species and alliance charac-

ter species, as pointed out by Oberdorfer (1968). If they differ only in the accompany-

ing species, they are geographical races of one association. Problems of this nature

will need to be handled very carefully in classifying Fynbos vegetation.

It is inevitable in such a survey that some quadrats will be badly sited. They

may either fall on an ecotone (probably Quadrat 34), or give an unrepresentative

picture of the vegetation type to which they belong (Quadrat 27), or represent a

different vegetation which is clearly undersampled (Quadrat 22). All such samples are

usually left out of the association table, because they do not contribute to an under-

standing of the vegetation. Quadrats which are wrongly sited cannot be used, but if

they represent an undersampled community, as in Quadrat 22, they should be kept

until more information becomes available. We have here included these stands in

Table | to give an example of sampling errors.

Many more relevés, especially from elsewhere in the Fynbos are required for a

successful phytosociological classification (see Tiixen, 1970). We have not presented a

formal hierarchical system here because of a lack of data; this report should be

regarded as a challenge to South African ecologists to prove and improve the classifi-

cation and so to build up an understanding of the vegetation.

ACKNOWLEDGEMENTS

We thank the Secretary of Forestry for allowing publication of this paper and

for permitting the investigations on a State Forest Reserve. We are indebted to Dr.

D. Edwards and Prof. Dr. R. Tiixen for their critical comments and to Mr. C. van

der Merwe for assistance in the field.

SAMEVATTING

Die fitososiologiese metode volgens Braun-Blanquet is getoets in die komplekse

Fynbos vegetasie van die Suidwes-Kaap in Suid-Afrika. In die Swartboschkloof

Natuurreserwe, Jonkershoek, is die Fynbos, die oewerstruikgewas en die woud-

vegetasie voorlopig in agt gemeenskappe ingedeel. Hierdie gemeenskappe is floristies

beskryf en die verband tussen die gemeenskappe en sekere habitatfaktore 1s aange-

dui. Die resultate is belowend en die moontlikhede om die Kaapse Fynbos te kan

klassifiseer in ’n formele fitososiologiese sisteem is bespreek.

610

REFERENCES

Acocks, J. P. H., 1953. Veld types of South Africa. Mem. Bot. Sury. S. Afr. No. 28.

ADAMSON, R. S., 1927. The plant communities of Table Mountain. I. Preliminary account. J. Ecol.

15: 278-309.

ADAMSON, R. S., 1931. The plant communities of Table Mountain. II. Life-form dominance and

succession. J. Ecol. 19: 304-320.

ADAMSON, R. S., 1935. The plant communities of Table Mountain. III. A six years’ study of regene-

ration after burning. J. Ecol. 23: 44-55.

BECKING, R. W., 1957. The Ziirich-Montpellier School of phytosociology. Bot. Rey. 23: 411-488.

BRAUN-BLANQUET, J. 1928. Pflanzensoziologie. Berlin: Springer.

BRAUN-BLANQUET, J., 1932. (Transl. by G. D. Fuller & H. S. Conard). Plant sociology. New York—

London: McGraw Hill.

BRAUN-BLANQUET, J., 1951. Pflanzensoziologie. 2nd ed. Wien: Springer.

BRAUN-BLANQUET, J., 1964. Pflanzensoziologie. 3rd ed. Wien—New York: Springer.

DONSELAAR, J. VAN, 1965. An ecological and phytosociological study of northern Surinam savannas.

Wentia 14: 1-163.

DANSEREAU, P. & ARROs, J., 1959. Essais d’applications de la dimension structurale en phytoso-

ciologie. I. Quelques exemples europeéns. Vegetatio 9: 48-99.

ELLENBERG, H., 1956. Aufgaben und Methoden der Vegetationskunde, in H. Walter. Einfiihrung in die

Phytologie, Bd. IV. Teil I. Stuttgart: Ulmer.

GLEASON, H. A., 1925. Species and area. Ecology 6: 66-74.

GooDALL, D. W., 1952. Quantitative aspects of plant distribution. Bio/. Rev. 27: 194-245.

Gomes PEDRO, J. & GRANDVEAUX BARBOSA, L. A., 1955. Esbogo da vegetagao de Mocambique. in:

Esbogo do reconhecimento ecologia-agricola de Mogambique. Centro de inyestigagao cientifica

algodeira, Memorias e Trabalhos No. 23: 67-226.

Kitiick, D. J. B., 1966-67. Fifty years of plant ecology in South Africa. J. S. Afr. Biol. Soc. 7, 8:

11-28

KOpPEN, W., 1931. Grundriss der Klimakunde. 2° Aufl. Berlin: W. de Gruyter.

KUcHLer, A. W., 1967. Vegetation mapping. New York: Ronald Press.

LEBRUN, J. & GILBERT, G., 1954. Une classification écologique des foréts du Congo. Publ. INéAC,

sér. scient. No. 63.

LEEUWEN, C. G. VAN, 1966. A relation theoretical approach to pattern and process in vegetation.

Wentia 15: 25-46.

Mar.otTH, R., 1908. Das Kapland, insondrheit das Reich der Kapflora, das Waldgebiet und die

Karroo, pflanzengeographisch dargestellt. Wiss. Ergebn. Deutsch. Tiefsee-Exped. “‘Waldivia’’.

1898-1899, Bd. II. T. 3. Jena: Fischer.

MERWE, P. VAN DER, 1966. Die flora van Swartboschkloof, Stellenbosch en die herstel van die soorte

na ’n brand. Ann. Univ. Stellenbosch. 41 Serie A, No. 14: 691-736.

Moore, J. J., 1962. The Braun-Blanquet system: a reassessment. J. Eco/. 50: 761-769.

OBERDOREER, E., 1968. Assoziation, Gebietassoziation, geographische Rasse. in: Pflanzensoziolo-

gische Systematik. Symp. Stolzenau 1964, ed. R. Tuxen, Den Haag: W. Junk. p. 124-141.

Poore, M. E. D., 1955. The use of phytosociological methods in ecological investigations. I. The

Braun-Blanquet system. J. Ecol. 43: 226-244.

Poorg, M. E. D., 1955. id. If. Practical issues involved in an attempt to apply the Braun-Blanquet

system. J. Ecol. 43: 245-269.

Poore, M. E. D., 1955. id. III. Practical applications. J. Ecol. 43: 606-651.

Poorg, M. E. D., 1956. id. [V. General discussion of phytosociological problems. J. Ecol. 44: 28-50.

SCHAPERA, J., 1930. The Khoisan peoples of South Africa: Bushmen and Hottentots. London: Routledge

SEDDON, J. D., 1966. The Early Stone Age at Bosman’s Crossing, Stellenbosch. S. Afr. Archaeol.

Bull. 21: 133-137.

SEDDON, J. D., 1967. Some Early Stone Age surface sites around Stellenbosch, S.W. Cape. S. Afr.

Archaeol. Bull. 22: 57-59.

Tayior, H. C., 1963. A bird’s eye view of Cape mountain vegetation. J. Bot. Soc. S. Afr. 49: 17-19.

TAyior, H. C., 1969. A vegetation survey of the Cape of Good Hope Nature Reserve. Unpubl. M.Sc.

thesis, Univ. Cape Town.

TUXEN, R., 1970. Pflanzensoziologie als synthetische Wissenschaft. Meded. Bot. Tuinen Belmonte

Arboretum Landbouwhogeschool Wageningen, 12: 141-159.

Walter, H., 1968. Die Vegetation der Erde. Bd. II. Die gemassigten und arktischen Zonen. Jena:

VEB Fischer.

Wacter, H. & Lietu, H., 1960. Klimadiagramm-Weltatlas. Jena: VEB Fischer.

Wercer, M. J. A., 1970. Minimal area and plot size. Unpubl. ms., Pretoria.

WestHorFr, V., 1968. Einige Bemerkungen zur syntaxonomischen Terminologie und Methodik,

insbesondere zu der Struktur als diagnostischen Merkmale. Meded. No. 260. R.I.V.O.N., Zeist.

Wicut, C. L., MeysBurGu, J. C. & BousTEAD, P. G., 1969. Rainfall at the Jonkershoek Forest Hydro-

logical Research Station. Ann. Univ. Stellenbosch. 44. Serie A, No. 1.

611

APPENDIX

Species excluded from Table I (relevé numbers and cover-abundance values in

brackets)

Ehrharta capensis Thunb. (1: +)

Chironia baccifera L. (2: +)

Hibiscus aethiopicus L. (2: +)

Hartogia schinoides (Thunb.) C. A.

Smith (3: r)

Myrsine africana L. (3: +)

Stoebe aethiopica L. (3: +)

Crassula obyallata L. (4: +)

Senecio umbellatus L. (4: +)

Apodytes dimidiata E. Mey. ex Benth.

(Ss 19)

Halleria elliptica Thunb. (5: 1)

Stoebe aethiopica L. (8: +-)

Tetraria compar (L.) Lestib. (8: +)

Cliffortia subsetacea Diels ex Bol. (10: r)

Halleria elliptica Thunb. (12: +)

Struthiola ciliata (L.) Lam. (13: ++)

Adenandra uniflora Willd. (14: +)

Centella virgata (L.f.) Druce (14: +)

Danthonia lupulina R. & Schult. (14: r)

Eragrostis curyula (Schrad.) Nees (14: +)

Heterolepis aliena (L.f.) Druce (14: +)

Osteospermum spinosum L. (14: +)

Carpacoce spermacocea (Reichb.) Sond.

(18: +)

Epischoenus villosus Levyns (18: +)

Gleichenia polypodioides Sm. (18: +)

Pelargonium angulosum Ait. (19: r)

Struthiola ciliata (L.) Lam. (19: r)

Protea cynaroides L. (20: r)

Apodytes dimidiata E. Mey. ex Benth.

Glen)

Berzelia intermedia Schld. (21: +)

Elegia asperiflora Kunth (22: 1)

Elegia vaginulata Mast. (22: 5)

Chrysanthemoides monolifera (L.) T. Norl.

@3=5)

Ehrharta calycina Sm. (23: +)

Pelargonium pinnatum (L.) L’Her. (23: +-)

Rhynchosia capensis (Burm.) Schinz (24:

+p)

Erica racemosa Thunb. (25: -+-)

Leptocarpus sp. (25: +)

Pseudognidia anomala Phillips (25: r)

(7/2 30)

Ficinia capillaris Levyns (28: -+-)

Leptocarpus sp. (28: +)

Pelargonium tabulare (L.) L’Her. (28: +)

Peucedanum sieberianum Sond. (29: -+-)

Asplenium aethiopicum Bacher (33: r)

Chironia baccifera L. (33: 1)

Myrsine africana L. (33: 3)

Pentameris thuarii Beauv. (33: +)

Psoralea pinnata L. (33: 1)

Zantedeschia aethiopica Spreng. (33:

Athrixia heterophylla (Thunb.)

G425)

Castalis nudicaulis (L.) T. Norl. (34:

Erica longifolia Ait. (34: +)

Ficinia zeyheri Boeck. (34: +)

Gnidia juniperifolia Lam. (34: +)

Leontonyx glomeratus (L.) DC. (34: +)

Peucedanum sieberianum Sond. (34: ++)

Selago serrata Berg. (34: +-)

Tetraria involucrata (Rottb.)

(34: +)

Ursinia dentata (L.) Poir. (34: +)

Ficinia trichodes (Schrad.) Benth. ex

Hk.f. (35: 1)

Pelargonium tabulare (L.) L’Her. (35: +)

Pentaschistis aristidoides (Thunb.) Stapf

(35: 1)

Senecio bipinnatus (L.f.) Less. (35: +)

Ehrharta calycina Sm. (36: r)

Olea africana Mill. (36: r)

Asparagus scandens Thunb. (38: r)

Blechnum capense (L.) Schlecht. (38: +-)

Centella eriantha (Rich.) Druce (38: +)

Halleria lucida L. (38: 1)

Kiggellaria africana L. (38: +-)

Mpyrica serrata Lam. (38: +)

Psoralea pinnata L. (38: +)

Rapanea melanophloeos (L.) Mez. (38: +-)

Rubus cf. rigidus Smith (38: +)

Tetraria cf. cuspidata (Rottb.) C.B. Cl.

(38: 2)

Zantedeschia aethiopica Spreng. (38: +-)

Danthonia cincta Schrad. (40: 1)

Elegia asperiflora Kunth (40: 2)

Erica intervallaris Salisb. (40: +-)

Juncus capensis Thunb. (40: r)

Restio quadratus Mast. (40: +-)

Tetraria flexuosa (Thunb.) C.B. CL.

(40: r)

Erica cerinthoides L. (41: +-)

Crassula obvallata L. (43: r)

Less.

E313}, le

Crassula scabra L. (43: +)

Eragrostis capensis (Thunb.) Trin. (43: +-)

Eroeda capensis (L.) Levyns (43: r)

Heterolepis aliena (L.f.) Druce (43: +)

Heteropogon contortus (L.) Beauv. (43: +4

Pellaea pteroides (L.) Prantl (43: 1)

Pentaschistis juncifolia Stapf (43: r)

612

Plagiochloa uniolae (L.f.) Adams. &

Sprague (43: +.)

Anemone capensis L. (44: r)

Aspalathus crenata (L.) R. Dahl. (44: +)

Eroeda capensis (L.) Levyns (44: r)

Metalasia cephalotes (Thunb.)

(44: 1)

Less

Species excluded from Table II (relevé numbers and cover-abundance values in

brackets).

Ficinia capillaris Levyns (15: r)

Heterolepis aliena (L.f.) Druce (15: +)

Sutera hispida (Thunb.) Druce (15: +)

Viscum rotundifolium L.f. (16: +)

Anthospermum aethiopicum L. (30: r)

Leonotis leonurus R. Br. (30: +)

Schoenoxipheum lanceum __ (Thunb.)

Kukenthal (30: +)

Leptocarpus paniculatus Pillans (31: +)

Struthiola myrsinites Lam. (31: +)

Cliffortia atrata H. Weim. (37: 1)

Erica plukeneti L. (37: +)

Phylica pubescens Ait. (37: +)

Protea arborea Houtt. (37: r)

Protea neriifolia R. Br. (37: r)

Schizaea tenella Kaulf. (37: r)

Stoebe plumosa (L.) Thunb. (37: +)

Thesium strictum Berg. (37: r)

Centella eriantha (Rich.) Drude (39: +)

Cyperus congestus Vahl (39: +)

Helichrysum odoratissimum (L.) Sweet

(39: +)

Osteospermum ciliatum Berg. (39: +-)

Othonna quinguedentata Thunb. (39:

Pelargonium vitifolium Ait. (39: --)

Psoralea cordata (L.) Salter (39: +)

Psoralea fruticans (L.) Druce (39: +-)

Scirpus sp. (39: r)

=P)

PLATE 1.—General view of Swartboschkloof Nature Reserve. Foreground vegetation mainly Protea

arborea—Rhus angustifolia Community; slopes of centre mountain ridge mainly Brunia nodi-

flora—Psoralea rotundifolia Community; right centre Heeria argentea Community on scree

(Photo: F. J. Kruger).

613

PLATE 2.—Protea arborea—Rhus angustifolia Community. Tree on right Protea arborea; foreground

mainly Tetraria bromoides. Note Rapanea melanophloeos Community in ravines in background

(Photo: F. J. Kruger).

PLATE 3.—Brunia nodiflora—Psoralea rotundifolia Community. Left flowering Brunia nodiflora;

right restionaceous plants (Photo: F. J. Kruger).

73472—8

PLATE 4.—Thamnochortus gracilis—Hypodiscus aristatus Community showing restionaceous charac-

ter (Photo: F. J. Kruger).

PLATE 5.—Rapanea melanophloeos Community in steep ravines (Photo: F. J. Kruger).

Bothalia, 10, 4: 615-626

A Preliminary Account of the Dune Communities at

Pennington Park, Mtunzini, Natal

E. J. Moll*

ABSTRACT

A general description of Pennington Park is given, and some of the more important environ-

mental factors affecting the plant communities are discussed. The structure, distribution and ecology

of the various dune communities, from pioneers to Dune Forest, is given. The importance of Penning-

ton Park as a conservation area is discussed in the light of the paucity of comparable sites and the

human pressure on this type of environment.

INTRODUCTION

During 1960 and 1961 two parties of students from Natal University, Pieter-

maritzburg, visited Pennington Park, on the farm ““Twinstreams” in Zululand, to

study the dune vegetation. Various ecological surveys and physiological experiments

were conducted on the dune vegetation, and two preliminary reports were compiled

(Anon, 1960, 1961). These reports, which contain some interesting data, were not

published so the results of this work have hitherto not been available. The area studied

is the only place in Natal where there is significant active and extensive sand deposi-

tion, dune formation, dune stabilization and colonization. It was decided, therefore, to

extract data from the two University reports and to add additional and recent obser-

vations in an attempt to describe, in reasonable detail, the ecology of the dune pioneers

and their successors. There is a need for these observations to be generally available

because, with the development of Richard’s Bay harbour, 35 km to the north, Mtun-

zini has been ear-marked for considerable tourist development. Such development

would certainly have a severe detrimental affect on the dune communities, which are

highly susceptible to disturbance. Thus a unique kind of vegetation is threatened.

That the dune communities of the area are still well protected is due almost

entirely to the conservation efforts of a local farmer, Mr. I. F. Garland, on whose

farm ““Twinstreams”’ most of the Dune Forests occurs. Also, part of the coastral strip,

between the Dune Forest and the sea, is administered by the Natal Parks, Game and

Fish Preservation Board and is included in the Umlalazi Nature Reserve (Fig. 1).

This latter area is open to the public so the dune communities are threatened unless

adequate measures can be taken to prevent inappropriate utilization of the dunes.

LOCATION AND PHYSIOGRAPHY

The area under consideration is situated on the east coast of Natal at the southern

extremity of the Mocambique coast plain, approximately 130 km _ north-east of

Durban, at lat. 28°58’S and long. 31°46’E. The beach profile slopes gently up from the

sea to the extreme high tide mark, then rises sharply in a series of steeply undulating

dunes (Plate 1), which are almost parallel to the coast and up to 10 m high. Sand is

actively deposited off-shore and new dunes are continually being found. Under

present conditions it takes approximately 10 years to stabilize a dune (Plate 2).

* Botanical Research Institute, Department of Agricultural Technical Services, Durban Unit,

Botanic Gardens Road, Durban.

616

—

TUNZINK

a i Wiss x

PLATE 1.—A general view of the beach showing the gently sloping

beach and the line of steep, Scaevola covered, dunes. Note the

Scaevola seeds at the foot of the dune in the foreground, and

also the scattered Scaevola plants on the beach—these plants

mark the line of the next new dune.

Umialazi Naturo

Reserve

Boundary Umlolazi

Natura Roserve

Roads

Railway

Rivors and Streams

OCEAN

Tracks and Paths

9 ANE SE OSS

PLATE 2.—The stable dune in the foreground is mainly covered by

Scaevyola, but other dune pioneers also occur. The fore-dune

H shown, is about 10 years old and stabilized almost exclusively

by Scaevola.

_INDIAN

Fic. 1.—Map showing the location of the study area, the extent of the Umlalazi Nature Reserve, the

important wind roses (Weather Bureau, 1960), and the water balance (Thornthwaite & Mather,

1962) for Mtunzini area.

617

The soils on the more recent dunes are a light yellowish brown fine sand, with no

humus and a pH of about 7,8. On the older dunes, which support a closed woody

community of Dune Forest, the soils are very dark greyish-brown sands, with a

relatively high humus content and a pH of between 7,2 and 7,4. Along the banks of the

Siayi Lagoon the alluvial soils are a grey to black clay-loam, with a high gritty con-

tent and a pH of about 6,4.*

CLIMATE

Detailed climatic data for the area are available only from Durban, which is on

the coast, but is 130 km SW of Mtunzini. The only data available locally are rainfull

records.

The prevailing winds are mainly from the NE and SW quarters. Winds are an

important factor influencing plant growth, particularly those strong winds which

blow off the sea, carrying salt-spray. The effect of this wind-borne salt-spray has been

discussed frequently in the literature (e.g. Bews, 1920; Henkel, Ballenden & Bayer,

1936; Bayer, 1938, 1952; Hillary, 1947; Edwards, 1967; and Moll, 1968), but has

not been studied in detail. It is generally considered that the wind-borne salt-spray is

deposited on the windward side of plants killing growing points, resulting in the

lopsided growth of woody plants and the pruned-hedge effect of the Dune Forest

canopy (Plate 3).

PLATE 3.—An example of the pruning effect of wind carried salt-spray on the Dune Forest canopy at

Mapelana (about 90 km NE of Mtunzini), where old, steep dunes come right down to the beach.

* Soil descriptions follow the terminology of Loxton, 1962. Initial soil pH determinations were

made using a Lovibond Colorimetric Comparator and checking these with a pH meter,

618

Winds which bring most salt-spray blow chiefly from September to January

(see wind rose insets in Fig. 1) and often reach a speed of 16-25 MPH (26-40 km/h),

gusts of up to 88 MPH (140 km/h) having been recorded (Weather Bureau, 1960).

These strong winds usually blow between 8 a.m. and 10 p.m.

Rainfall data are available from Mtunzini and have been summarized in Table I.

TABLE I.—Mean monthly and annual rainfall recorded at Mtunzini over a 53 year

period eae Bureau, 1954).

| | |

Month J | 18 M fA a | M J | J | A | S : O | N | D | Year

final B) | | | |

ei | Sees fs

Rainfall inmm 134,8 146, a 0 98, 6 me 8 59,7) 51 sale 285.2 103 2.9 9 141,3) 1 204,5

cece omar oer eee 69

| | es 7

Summer is the wettest season, although a reasonable amount of rain occurs

throughout the year. Temperatures are relatively mild, and data available from

Durban, covering a period of 76 years, record an absolute minimum of 4,1°C in

July 1947 and an absolute maximum of 41,9°C in September 1946. The mean annual

temperature is 20,5°C with a mean daily range of 8,3°C. Under this climatic régime

of moderate temperatures and good rainfall there is seldom, if ever, a soil water

deficit (see Thornthwaite & Mather, 1962; and diagram inset Fig. 1), so conditions

for plant growth are very good. Mtunzini, being 130 km NE of Durban, is rather

more subtropical, so minimum and maximum temperatures are probably higher.

VEGETATION

The most important first pioneer of the shifting sand on the beach is Scaevola

thunbergii. Some other pioneers of minor importance are Launnaea sarmentosa and

Arctotheca populofolia. The Scaevola colonies form an open, scattered community up

to about one metre tall. Scaevola seeds are round and light, and roll down the dunes

(Plate 1) from whence they are readily blown about. Scaevola is capable of continuous

stem elongation and adventitious root production from sand-covered stems, and

thrives where moving sand is continually covering existing communities. Scaevola is

tolerant of salt-spray and at Mtunzini seldom occurs beyond the fourth dune. Healthy

Scaevola colonies are dominant on the first and second dunes, and in areas on the

third dunes where disturbance has caused shifting sand. On the third and fourth

dunes, where Scaevola is also common, the plants are weak as a result of the dunes

being relatively stable. This weakened community, where the lateral stems are usually

exposed, is readily invaded by other pioneers such as Ipomoea biloba (Plate 4), Cana-

valia pes-capre, Gazania rigens, Chrysanthemoides monilfera and Tephrosia canescens,

all species which require a relatively stable substrate.

Once the dunes have been stabilized and there has been sufficient modification

of the soil by the pioneer plants, shrub species, particularly Passerina rigida, invade

the pioneer strand communities. The shrub species nearest the sea grow in the dune

troughs, usually 60 to 70 m from the high tide mark (Plate 5). Other shrub species

usually occur further inland, such as Eugenia capensis, Colpoon compressum, Carissa

bispinosa and Brachylaena discolor. Certain Dune Forest canopy trees occur as

shrubs in the Dune Scrub Community, especially Mimusops caffra, Apodytes dimidiata,

Allophylus natalensis and Canthium obovatum (Plate 6). The Dune Scrub near the sea is

up to 2 m high forming a widely scattered community, which becomes progressively

denser and taller (up to 3,5 m) further inland towards the Dune Forest (Plate 7). This

trend is well illustrated in Figs. 2 and 3.

73472,

ee tee Oe ee ee ee eee ee ee ee

Oe me em ewe me me ee me oe

FEF rmetodes scolopendria EES canyenthemsides monilifera

BB arochyucs natalensisa BB cris cam

or Nemorosa Se discolor Stipagrostis zeyheri

Horizontal scale: one division = 1 ft, Verticle scale twice the horizontl scale

Fic. 2.—Land profile and extent of vegetation cover on a 3 m belt transect (Anon, 1960).

Barringtonia racemosa Mc Mimusops caffra c Canthium ventosum M Maytenus spp. tee Gazmia_rigens

Hibiscus _tiliaceus ae Various woody climbers fb Ficus burtt-davy! Eo Eugenia capensis ae seers thunbergii

Haemanthus natalensis Ad —.Apodytes imidiata E —Eucles natalensis Le Stipagrostis zeyheri AL, Ipomces biloba

Isogossa woodii P Peddiea africana T —“Teicatysiasonderiana A imperata cylindrica

Carisse —_bispinces Pe Psychotria. capensis ¥ Vepris lanceolata EL Passerins rigida Verticle scale twice the horizontal scale

eS ae ois Phymatodes scolopendria An Allophylus natslensis Be Citatiteass tonitaes Horizontal scale: one division = 10ft,

Fic. 3.—Profile diagram of the vegetation from the sea to the Siayi Lagoon. The depth of the profile was 1 m in the pioneer strand and Dune Scrub Communities, and 4,5 m in the Dune Forest (Anon, 1961)

621

PLATE 4.—Showing a healthy Scaevola community growing in shifting sand on the left. Scaevola is

also common on the stable dune on the right, but many lateral stems are exposed; note the long

runners of Ipomoea in the dune trough.

PLATE 5.—The first stage of Dune Scrub Community development is the invasion of the pioneer

strand communities by Stipagrostis (the tuft grass pictured) and Passerina (right). Note the

scattered Scaevola plants, particularly on the dune (left).

622

PLATE 6.—In the later stages of the Dune Scrub Community Imperata is the chief understorey species.

The shrubs are more dense, with Dune Forest species, such as Mimusops (left), being more

common. Note the dead Passerina (right-centre).

PLaTe 7.—A general view of the Dune Scrub Community from the Scaevola dominated foredune’

illustrating the invasion of the strand community by Stipagrostis and Passerina, followed by

other shrub species and finally Dune Forest being established.

PLATE 8.—Dune Forest with Phymatodes scolopendria forming the dominant understorey herbaceous

layer.

Another distinct change in the Dune Scrub Community, as one moves inland, is that

first Stipagrostis zeyheri (Plate 5), and then Jmperata cylindrica, both grasses, form a

fairly dense understorey up to 0,75 m high, being most dense away from the sea.

Various other herbaceous species occurring scattered through the Dune Scrub

Community include Helichrysum kraussii, Kalanchoe sp., Senecio sp., Carpobrotus

dimidiata, Chironia baccifera and Gloriosa sp. In addition, various lianes also occur,

such as Scutia myrtina, Dalbergia armata, Asparagus falcatus and Rhoicissus spp.,

particularly R. digitata. In the shade of the shrubs on the dune just seaward of the

Dune Forest, Phymatodes scolopendria is a common understorey herb.

The margin of the Dune Forest is sharply demarcated, beginning just behind the

crest of a dune (see Figs 2 and 3). When walking at right angles to the sea one is inside

the forest one instant and then outside within a couple of paces. The tree species

forming the closed canopy (some 7 to 8 m high) nearest the sea are Mimusops caffra,

Allophylus natalensis, Eugenia capensis, Maytenus nemorosa, Euclea natalensis and

Canthium obovatum. Shrub species beneath the closed canopy are not common, nor

are lianes. What is apparent, forming a very dense field layer, is Phymatodes scolopen-

dria (Plate 8), up to 0,5 m high, which is the dominant field layer species over the

next four dunes, when it suddenly gives way to a dense community of Jsoglossa

woodii (Plate 9)*. Further from the sea, and to a very limited extent in the dune

troughs, the canopy tree species are taller, reaching a maximum height of about 15 to

17 m. An unexpected feature of the forest is that the external appearance of the canopy

does not exhibit the usual dense, pruned-hedge effect of the Dune Forest as in other

parts of Natal, because the forest is a fair distance from the sea and the salt-spray

effect is much reduced.

* The soil pH where Phymatodes dominates is the same as under Dune Scrub Communities, namely,

7,4, while the soil pH where Jsoglossa dominates is 7,2.

PLATE 9.—Dune Forest with Isoglossa forming the dominant understorey herbaceous layer.

The older the Dune Forest, the more complex is the structure and the richer is the

floristic composition. Some canopy tree species which occur further away from the sea

are Dovyalis longispina, Olea woodiana, Vepris lanceolata, Scolopia zeyheri, Trichilia

emetica and Ekebergia capensis. Beneath the continuous canopy is a rather ill-defined

intermediate small tree and shrub layer of scattered individuals such as Carissa

bispinosa, Peddiea africana, Turraea floribunda, Psychotria capensis, Bersama lucens,

Acokanthera oblongifolia, Tricalysia capensis and Teclea gerrardii. Additional species

which may occur scattered through the herbaceous field layer are Haemanthus,

Crocosmia aurea, Eulophidium sp. and various species of Acanthaceae. Where /sog/ossa

occurs there are usually few herbaceous or intermediate species present due to the

performance of /soglossa. This species grows in extremely dense stands up to 3 m

high, the whole population then flowers in the same year, dies off, and then regrows

over a period of approximately seven years to complete the cycle. Thus Dune Forest

with /soglossa understorey can look either extremely dense, when the /soglossa us

2 to 3 m high, or extremely open, when the /sog/ossa seedlings carpet the ground.

The presence of the Dune Forest means that there is a great enrichment of the

soil by humus, which is usually 10 cm and more thick. In this subtropical climate the

rate of breakdown of vegetable matter on the forest floor is rapid, and numerous

species of saprophytic fungi occur.

Lianes are not common in old, established Dune Forest, except where a natural

tree fall has created a gap. In such gaps there develops a dense tangled mass of vegeta-

tion, but eventually a canopy tree grows through this to close the canopy. Lianes that

do occur fairly frequently in the forest are Dalbergia armata, Rhoicissus sp., Cypho-

stemma sp., Acacia kraussiana and Scutia myrtina. The absence of Flagellaria guineen-

sis is worthy of note.

625

Running through the Dune Forest at Pennington Park, almost parallel to the

coast, is the Siayi Lagoon (Fig. 1). On either bank there is a dense fringing tree

community of mainly Barringtonia racemosa and Hibiscus tiliaceus (Plate 10).

Climbers such as Derris uliginosa and Dalbergia armata are particularly common in

this community, the former growing in the wet muddy soils of the lagoon bank.

PLATE 10.—A view of the Siayi Lagoon where the fringing tree community of Barringtonia forms an

arch over the water.

DISCUSSION

In Natal there are no examples of viable Dune Forest communities represented in

a Nature Reserve, except for this example in the Umlalazi Nature Reserve and the

adjoining farm ‘“‘Twinstreams”. With the development of Richard’s Bay and the

proposed development of a giant marina on the Umlalazi River, with a possible link

to the Siayi Lagoon, this area is threatened. It has been demonstrated in other parts of

Natal that the Dune Forest and pioneer strand communities are highly susceptible

to human interference. Therefore, a special attempt should be made by conserva-

tionists to preserve a viable representative of this community. It should also be the

aim of conservationists to preserve other examples of dune vegetation to the north

and south of Mtunzini to include the full spectrum of types occurring. Much of this

type of community has already disappeared in Natal, and the time to act is now.

ACKNOWLEDGEMENTS

I am particularly grateful to Mr. I. F. Garland for his generosity and active

assistance while I have worked on this report.

626

REFERENCES

ANONYMOUS, 1960. Account of the symposium on the expedition to Pennington Park. Unpublished

report, University of Natal, Pietermaritzburg.

ANONYMOUS, 1961. Account of the symposium on the second expedition to Pennington Park. Un-

published report, University of Natal, Pietermaritzburg.

Bayer, A. W., 1938. An account of the plant ecology of the Coast-belt and Midlands of Zululand.

Ann. Natal Mus. 8: 371-454.

Bayer, A. W., 1952. Notes on the vegetation of Natal. Natal Society for the Preservation of Wildlife

and Natural Resources. 1 (6).

Bews, J. W., 1920. The plant ecology of the coast-belt of Natal. Ann. Natal Mus. 4: 367-469.

Epwarps, D., 1967. A plant ecology survey of the Tugela Basin, Mem. Bot. Surv. S. Afr. No. 36.

HENKEL, J. S., BALLENDEN, S. StC, & BAyer, A. W., 1936. An account of the plant ecology of the

Dukuduku Forest Reserve and adjoining areas of the Zululand Coast Belt. Ann. Natal Mus.

8: 95-125.

Hivary, O. M., 1947. An account of the plant succession on Tongaat Beach, Natal. Unpublished

M.Sc. thesis, University of Natal, Pietermaritzburg.

Loxton, R. F., 1962. A simplified soil-survey-procedure for farm planning. Dept. Agric. Tech. Services,

Pretoria.

MOLL, E. J., 1968. The vegetation of the Three Rivers Region Natal. Unpublished report to the Nata

Town and Regional Planning Commission.

THORNTHWAITE, C. W. & MATHER, J. R., 1962. Average climate water balance data of the continent.

Part 1, Africa. Centerton, New Jersey.

WEATHER BurREAU, 1954. Climate of South Africa. Part 1. Climate Statistics. Govt. Printer, Pretoria.

WEATHER BurEAU, 1960. Climate of South Africa. Part 6. Surface Winds. Govt. Printer, Pretoria.

Bothalia, 10, 4: 627—636

The Distribution, Abundance and Utilization of the

Lala Palm, Hyphaene natalensis, in Tongaland,

Natal

E. J. Moll*

ABSTRACT

The distribution of the Lala Palm, Hyphaene natalensis, in Tongaland and Northern Zululand, is

mapped; the Palm occupies an area of about 156 000 ha. The total number of individuals is estimated

at approximately 10 500 000 and the total yield in leaves per year is estimated at about 33 000 000.

The exploitation of the leaves for fibre could be an economic proposition, but communications in

the region are poor and the area is extremely large. Present utilization of the Lala Palm, by the Bantu

is considered.

INTRODUCTION

Recently it has been suggested that the leaves of the Lala Palm, Hyphaene

natalensist, in Tongaland, which yields an excellent fibre, should be reaped and

processed for the fibre. However, before embarking on such a scheme it was con-

sidered desirable to study the distribution, abundance and present utilization of the

Lala Palm with a view to assessing the feasibility of the scheme.

Tongaland has been defined as that part of Natal which lies east of the Lebombo

Mountains, north of the Mkuzi River and St. Lucia, and south of Mozambique

(Campbell, 1969). For the most part it is flat to gently undulating country about

14,7 to 44 m above sea level, until it rises sharply in the west to the Lebombo Moun-

tains. A feature of Tongaland that is uncommon in Natal is the presence of numerous

pans and lakes, namely the Pongola flood plain and environs (Coke & Pott, 1970),

the Mosi Swamp running more or less north to south, and the series of coast lakes

including the Kosi System and Lake Sibayi (see Fig. 1). In addition, there are numerous

ephemeral, as well as a few permanent pans and swamps scattered through the

country east of the Mosi Swamp to the sea.

The Tongaland or Mozambique Plain, as the flat coastal plain is called, is a

recently uplifted area of marine sands. From the coast to the Lebombo the soils

change from white to grey sands, to red sands across the Pongola, and to Cretaceous

soils in the Lebombo foothills. The soils occupied by the Lala Palms are the grey ands

white sands.

Few climatic data are available from the area. It is known, however, that the average

summer rainfall along the coast is about | 200 mm and that this falls off steadily as

one moves inland to about 600 to 700 mm along the Pongola River. There is little, if

any, rainfall in winter. It is fairly safe to assume, therefore, that the average annual

rainfall between the Mosi Swamp and the sea, which is the main palm belt (see Fig. 1),

is between 900 and | 200 mm. No temperature data are available from the area,

but it is known that in winter minimum temperatures are not sufficiently low to

allow even light frost, and that maximum temperatures in summer are in the region

of 45° C. The climate, according to Képpen’s classification is “tropical with summer

rainfall’, and according to Thornthwaite’s classification ‘““sub-humid warm, with

sufficient moisture in all seasons” (Schulze, 1947).

* Botanical Research Institute, Department of Agricultural Technical Services, Durban Unit

Botanic Gardens Road, Durban.

+ Formerly known as Hyphaene crinita, which is apparently a tropical West African species (Furtado,

1970).

628

THE LALA PALM BELT

The area covered by Lala Palms is shown in Fig. 1. This distribution has been

plotted from aerial photographs, and ground checks in 1968 and 1969. South of St.

Lucia the palm belt stops, though small isolated patches occur down the coast as far

south as the Umtamvuna River.

The structure of the Lala Palm community varies quite considerably from east to

west. Along the east coast the palms occur mainly as widely scattered individuals in

grassland (Plate 1). The most important grass species, which form a moderately dense

tussocked community up to 0,75 m tall, are Aristida junciformis, Elyonurus argenteus,

Tristachya hispida and Trachypogon spicatus. The individuals of Hyphaene natalensis

are often mixed with another palm, Phoenix reclinata, the Wild Date Palm, and both

palms exhibit the same ability to produce several stems from a single rootstock or

plant. Further inland the grassland is invaded by other woody plants and the density

of Hyphaene natalensis increases (Plate 2), while that of the Phoenix reclinata decreases

markedly. Some of the more common associated woody plants are Dichrostachys

cinerea, Acacia burkei, Maytenus heterophylla, M. senegalensis, Vangueria infausta,

Sclerocarya caffra, Strychnos spinosa, S. madagascariensis, Combretum molle, and

Syzygium cordatum. In these areas bush clumps tend to form where there is some

protection from fire, though woody plants are also found scattered through the grass-

land. These woody species vary greatly in density, height and occurrence, depending

on local edaphic and biotic factors.

In those areas where the water table is near the soil surface for most of the year,

herbaceous species only occur. So, although Hyphaene natalensis is capable of tolerating

waterlogged soils, it is unable to withstand continuous waterlogging and only occurs

in seasonally inundated areas.

In general then, the Palm Veld near the coast is fairly open, becoming more

dense towards the Mosi Swamp where it is a mosaic of different types from open

herbaceous communities to dense bush clumps.

For mapping purposes a density of at least 10 plants per acre (0,4 ha) was taken

as constituting Lala Palm Veld, though nearer the coast Phoenix reclinata tended to

confuse the air photo interpretation. Thus, Lala Palms do occur east and west of the

area mapped as Lala Palm Veld, but in these areas individuals are widely scattered.

QUANTITATIVE METHODS AND RESULTS

In the field Lala Palms were counted at 36 sites, each site being 70 < 70 paces, or

approximately one acre (0,4 ha) in size. The location of these sample sites was

governed primarily by ease of access.

Sample sites 1 to 24 were in the vicinity of the Maputa-Ingwavuma road. The

first site was on the north side of the road 1,6 km from Maputa Post Office, there-

after sample sites were taken at intervals of 1,6 km on alternate sides of the road.

Sample sites 25 to 29 were located on a track joining the Maputa-Ingwavuma road

to the Maputa—Nseleni road, and were spaced in the same way as above. Sample

sites 30 to 33 were placed 16 km from Maputa on the Maputa—Nseleni Road in the

same way as above. Sample sites 34 and 35 were located on the Mbazwana—Lower

Mkuzi road, and site 36 on the main road south of Hluhluwe (see Fig. 1).

At each sample site the number of individual plants of Hyphaene natalensis was

counted and the number of stems per plant recorded. The number of leaves per stem

was also counted. These data are summarized in Table 1.

— 5 —— — _SIHANGWAG

> ~n23 1)

24 > 9

22,

Fic. {.—Map of eastern Tongaland showing the distribution of Hyphaene natalensis, swampy ground,

and coastal lakes in the study area.

study area

4+ Sampling sites

—-—— 6 International boundary (Natal & Mocam-

bique)

SINR fivers

Pans and Lakes

Swamps; as per 1:250,000 TOPO -

CADASTRAL map(1959)

Hyphaene distribution: es per AIR

PHOTO. interpretotion

MILES — = = — i =4

KILOMETRES & § 2. Wt WH

630

TABLE |.—Summarized field data of the number of plants, stems and leaves of

Hyphaene natalensis counted in 36 one-acre (70 x 70 paces) (0,4 ha)

sample sites.

| Total No. | Total No. | Total No.

Sample of plants of stems | BEEN | of leaves | Mv elage No. ey ge No.

sites per acre per acre | oo r ‘i | per acre | g a ed OS AEENIES

(0,4 ha) (0,4 ha) per plant — (0,4 ha) | per plant | per stem

IL 1 1 1,00 1 1,00 1,00

2 7 12 il 7 45 5,00 3,75

3 10 50 5,00 185 18,50 3,70

4 17 17 1,00 63 3,71 3,71

5 11 16 1,45 35 3,18 DMS

6 11 18 1,64 39 3,55 2,06

7 34 115 3,38 426 12535 3,70

8 15 37 2,47 76 5,07 2,06

9 13 21 1,62 41 3,15) 1,95

10 11 18 | 1,62 29 2,62 1,61

11 69 168 2,44 711 10,30 4,24

12 33 91 1,32 398 12,07 4,35

13 64 454 7,10 891 14,02 1,97

14 71 246 3,47 687 9,68 2,39

IS 7 25 3,57 157 22,43 6,28

16* 9 19 2,11 96 10,68 5,05

17 109 1 289 11,74 1 904 17,47 1,48

18 179 1 789 9,99 2 674 14,94 4,50

19 7/ 43 DS 131 WoW! 3,05

20* 2 3 1,50 12 6,00 4,00

Ail > I 1 1,00 4 4,00 4,00

Dn 3 7 | 2,33 26 8,67 3,71

23* ie ea maa. =| 5 5,00 5,00

24* I 2 2,00 | 6 6,00 3,00

25 33 103 3,43 317 9,60 3,08

26 5 9 1,80 24 4,80 2,64

27 11 21 1,18 58 5,27/ 2,76

28* I I 1,00 3 3,00 3,00

29* 1 i 1,00 3 3,00 3,00

30 15 23 1,53 65 4,33 2,83

31 17 31 1,82 74 4,35 2,39

32 14 25 eG | 65 4,64 2,60

33 12 21 L758) 37 3,08 1,76

34 83 182 2,19 772 9,30 4,24

35 71 147 2,04 689 9,70 4,69

36 30 144 4,80 668 22,27 4,64

* Sample sites outside the area mapped as Hyphaene Palm Veld.

The area of Lala Palm Veld in Tongaland and Northern Zululand is approxi-

mately 156 000 ha (600 square miles, about 375 000 acres). From the field data

obtained from sites within the area mapped as Hyphaene Palm Veld, excluding data

from the 10 sample sites outside the mapped area, we find the following:

Average number of plants per acre (0,4 ha) = 37,00

Average number of stems per acre (0,4 ha) = 195,77

Average number of stems per plant — 5,28

Average number of leaves per acre (0,4 ha) = 427,08

Average number of leaves per plant = 11,25

Average number of leaves per stem = 2,18

631

The total number of individual Hyphaene Palms may be estimated as 14 000 000

individuals, comprising approximately 73 500 000 stems and bearing a total number

of about 160 000 000 leaves. However, from the data in Table | it is apparent that the

figures have been affected by the two sample sites, 17 and 18. From field experience

and from a close study of the air photographs, it is clear that areas where Hyphaene

natalensis is dense are extremely limited, and have been estimated at less than 5°%.

A more realistic estimate of the total number of individual Lala Palms is, therefore,

about 10 500 000 individuals comprising approximately 31 000 000 stems and bearing

a total number of about 100 000 000 leaves. It should be noted that owing to utiliza-

tion by the local Bantu (see later) the average number of leaves per plant is probably

considerably lower than normal (compare Plate 4).

PROPOSED UTILIZATION OF LALA PALM

As already stated, it has been suggested that the Lala Palm be commercially

exploited for its fibre. With the statistics obtained, the economics of the scheme can be