ISSN 0006 **241 = Bothalia

Bothalia

’N TYDSKRIF VIR PLANTKUNDIGE NAVORSING

A JOURNAL OF BOTANICAL RESEARCH

May/Mei 1S92

Vol. 22,1

PUBLICATIONS OF THE NATIONAL BOTANICAL INSTITUTE, PRETORIA

PUBLIKASIES VAN DIE NASIONALE BOTANIESE INSTITUUT, PRETORIA

Obtainable from the National Botanical Institute, Private Bag X101,

Pretoria 0001, Republic of South Africa. A catalogue of all available

publications will be issued on request.

BOTHALIA

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

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

house journal of the National Botanical Institute, Pretoria, is devoted to

the furtherance of botanical science. The main fields covered are

taxonomy, ecology, anatomy and cytology. Two parts of the journal

and an index to contents, authors and subjects are published

annually.

Verkrygbaar van die Nasionale Botaniese Instituut, Privaatsak X101,

Pretoria 0001, Republiek van Suid-Afrika. ’n Katalogus van alle beskik-

bare publikasies kan aangevra word.

Bothalia is vemoem ter ere van Generaal Louis Botha, eerste Eerste

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

lyfblad van die Nasionale Botaniese Instituut, Pretoria, is gewy aan die

bevordering van die wetenskap van plantkunde. Die hoofgebiede wat gedek

word, is taksonomie, ekologie, anatomie en sitologie. Twee dele van die

tydskrif en ’n indeks van die inhoud, outeurs en onderwerpe verskyn

jaarliks.

MEMOIRS OF THE BOTANICAL SURVEY OF SOUTH AFRICA

MEMOIRS VAN DIE BOTANIESE OPNAME VAN SUID-AFRIKA

The memoirs are individual treatises usually of an ecological nature, ’n Reeks van losstaande omvattende verhandelings oor vemaamlik

but sometimes dealing with taxonomy or economic botany. ekologiese, maar soms ook taksonomiese of plantekonomiese onderwerpe.

THE FLOWERING PLANTS OF AFRICA / DIE BLOMPLANTE VAN AFRIKA

This serial presents colour plates of African plants with accompanying

text. The plates are prepared mainly by the artists at the National Botanical

Institute. Many well known botanical artists have contributed to the series,

such as Cythna Letty (over 700 plates), Kathleen Lansdell, Stella Gower,

Betty Connell, Peter Bally and Fay Anderson. The Editor is pleased to

receive living plants of general interest or of economic value for

illustration.

From Vol. 50, one part of twenty plates is published annually. A volume

consists of two parts. The publication is available in English and Afrikaans.

Hierdie reeks bied kleurplate van Afrikaanse plante met bygaande teks.

Die skilderye word meestal deur die kunstenaars van die Nasionale

Botaniese Instituut voorberei. Talle bekende botaniese kunstenaars het

tot die reeks bygedra, soos Cythna Letty (meer as 700 plate), Kathleen

Lansdell, Stella Gower, Betty Connell, Peter Bally en Fay Anderson.

Die Redakteur verwelkom lewende plante van algemene belang of

ekonomiese waarde vir afbeelding.

Vanaf Vol. 50 word een deel, bestaande uit twintig plate, jaarliks

gepubliseer. ’n Volume bestaan uit twee dele. Die publikasie is beskikbaar

in Afrikaans en Engels.

FLORA OF SOUTHERN AFRICA / FLORA VAN SUIDELIKE AFRIKA

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

Ciskei, Transkei, Lesotho, Swaziland, Bophuthatswana, Namibia,

Botswana and Venda. The FSA contains descriptions of families, genera,

species, inffaspecific taxa, keys to genera and species, synonymy, literature

and limited specimen citations, as well as taxonomic and ecological

notes.

’n Taksonomiese verhandeling oor die flora van die Republiek van Suid-

Afrika, Ciskei, Transkei, Lesotho, Swaziland, Bophuthatswana, Namibie,

Botswana en Venda. Die FSA bevat beskrywings van families, genusse,

spesies, inffaspesifieke taksons, sleutels tot genusse enspesies, sinonimie,

literatuur, verwysings na enkele eksemplare, asook beknopte taksonomiese

en ekologiese aantekeninge.

PALAEOFLORA OF SOUTHERN AFRICA / PALAEOFLORA VAN SUIDELIKE AFRIKA

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

southern Africa. Much of the information is presented in the form

of tables and photographic plates depicting fossil populations. Now

available:

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

like Afrika. Baie van die inligting word aangebied in die vorm van tabelle

en fotografiese plate waarop fossiele populasies afgebeeld word. Reeds

beskikbaar:

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

Molteno Formation (Triassic) Vol. 2. Gymnosperms (excluding Dicroidium), by/deur J.M. & H.M. Anderson.

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

Obtainable from/Beskikbaar van: A. A. Balkema Marketing, Box/Posbus 317, Claremont 7735, RSA.

BOTHALIA

’N TYDSKRIF VIR PLANTKUNDIGE NAVORSING

A JOURNAL OF BOTANICAL RESEARCH

Volume 22,1

Editor/Redakteur: O.A. Leistner

Assisted by B.A. Momberg

Editorial Board/Redaksieraad

D.F. Cutler

B.J. Huntley

P.H. Raven

J.R Rourke

M.J. Werger

Royal Botanic Gardens, Kew, UK

National Botanical Institute, Cape Town, RSA

Missouri Botanical Garden, St Louis, USA

Compton Herbarium, NBI, Cape Town, RSA

University of Utrecht, Utrecht, Netherlands

Editorial Committee

Redaksiekomitee

O.A. Leistner

B.A. Momberg

M.C. Rutherford

ISSN 0006 8241

Issued by the National Botanical Institute, Private Bag X101, Pretoria 0001, South Africa

Uitgegee deur die Nasionale Botaniese Instituut, Privaatsak X101, Pretoria 0001, Suid-Afrika

1992

CONTENTS — INHOUD

Volume 22,1

1. Notes on the Strumariinae (Amaryllidaceae-Amaryllideae). Six new taxa in Strumaria and Hessea

from the central and northwestern Cape, South Africa, and southern Namibia. D.A. SNIJMAN 1

2 . Three new species of Diascia (Scrophulariaceae) from the western Cape. K.E. STEINER 13

3. Studies in the genus Riccia (Marchantiales) from southern Africa. 24. R. moenkemeyeri , subgenus

Ricciella : new records. S.M. PEROLD 19

4 . Aspidonepsis (Asclepiadaceae), a new southern African genus. A. NICHOLAS and D.J. GOYDER 23

5. Notes on African plants:

Bryophyta. New and interesting records of mosses in the Flora of southern Africa area : 2. Giga-

spermaceae— Bartramiaceae. J. VAN ROOY and S.M. PEROLD 37

Euphorbiaceae. Notes on Euphorbia species from the northwestern Cape. P. BRUYNS 37

Fabaceae. The identity of Argyrolobium obsoletum and the correct names for some species of

Polhillia (Crotalarieae). B-E. VAN WYK 42

Fabaceae. A new species of Priestleya from the southwestern Cape. E.G.H. OLIVER, A.C.

FELLINGHAM and B-E. VAN WYK 47

Gomphyllaceae (Lichenes). A new species of Bullatina from the Transkei Wild Coast. F. BRUSSE 44

Moraceae. New records of Ficus species and their pollinators on Grand Comore. S.G. COMP-

TON 46

Proteaceae. The correct author citation for Paranomus reflexus. J.P. ROURKE 43

6. The occurrence in southern Africa of the hepatic, Symphyogyna brasiliensis (Pallavicinaceae). S.M.

PEROLD 52

7. Morphology, evolution and taxonomy of Wachendorfia (Haemodoraceae). N.A. HELME and H.P.

LINDER 59

8. An overview of Penicillium (Hyphomycetes) and associated teleomorphs in southern Africa. A.L.

SCHUTTE 77

9. Vegetation and checklist of Inaccessible Island, central South Atlantic Ocean, with notes on Nigh-

tingale Island. J.P. ROUX, P.G. RYAN, S.J. MILTON and C.L. MOLONEY 93

10. Salt glands in flowering culms of Eriochloa species (Poaceae). M.O. ARRIAGA Ill

11. Invasive alien woody plants of Natal and the northeastern Orange Free State. L. HENDERSON ... 119

12. The Ven. Charles Theophilus Hahn, a hitherto unknown Edwardian botanical illustrator in Natal,

1908-1916. J.P. ROURKE and J.C. MANNING 145

13. Book reviews 155

1 4 . Guide for authors to Bothalia 157

Digitized by the Internet Archive

in 2016

https://archive.org/details/bothaliavolume2222unse

Bothalia 22,1: 1-11 (1992)

Notes on the Strumariinae (Amaryllidaceae-Amaryllideae). Six new taxa

in Strumaria and Hessea from the central and northwestern Cape, South

Africa, and southern Namibia

D.A. SNIJMAN*

Keywords: Amaryllidaceae, Amaryllideae, Hessea, Strumaria, new rare species, subspecies, southern Africa

ABSTRACT

Newly described are four species and a subspecies of Strumaria and one species of Hessea. S. aestivalis Snijman from

the Langberg and S. perryae Snijman from the Bokkeveld escarpment are rare species closely allied to S. pubescens W.F.

Barker. S. discifera Marloth ex Snijman is widespread on the Bokkeveld and Roggeveld escarpments but S. discifera subsp.

bulbifera Snijman which comprises several clonal populations, is narrowly restricted to the dolerite ridges near Nieuwoudtville.

S. villosa Snijman, a rare species, is localised on quartz hills near Kosies in the Richtersveld. H. speciosa Snijman occurs

in red sand and friable loam from southern Namibia to the central Cape.

UITTREKSEL

Vier nuwe spesies en ’n subspesie van Strumaria sowel as 'n nuwe Hessea- spesie word beskryf. S. aestivalis Snijman

van die Langberg en S. perryae Snijman van die Bokkeveld platorand is skaars soorte wat na aan S. pubescens W.F. Barker

verwant is. S. discifera Marloth ex Snijman is wydverspreid langs die Bokkeveld- en Roggeveld-platorand. S. discifera subsp.

bulbifera Snijman is beperk tot ’n aantal klonale populasies op die doleriet-heuwels naby Nieuwoudtville. S. villosa Snij-

man, ’n skaars soort, word slegs op die kwartsiet-heuwels naby Kosies in die Richtersveld aangetref. H. speciosa Snijman

kom voor in rooi sand en bros leem vanaf suidelike Namibie tot in die Kaapse Middellande.

INTRODUCTION

The Strumariinae, an exclusively southern African sub-

tribe of the Amaryllideae, is centred in the semi-arid

winter rainfall region of the Cape Province. The often

insignificant, hysteranthous leaves and short-lived autumn

flowers of the species, are phenological characteristics

which render many members of the subtribe insufficiently

collected. Thus since the last review of the Strumariinae

(Miiller-Doblies 1985) some 12 additional new taxa have

been discovered, of which five have already been published

(Snijman 1989; Snijman 1991).

The 37 known species of Strumariinae are currently

placed in eight genera ( Namaquanula D. & U. Miiller-

Doblies, Kamiesbergia Snijman, Hessea Herb. , Carpolyza

Salisb., Strumaria Jacq., Bokkeveldia D. & U. Miiller-

Doblies, Gemmaria Salisb. and Tedingea D. & U. Miiller-

Doblies). Phylogenetic studies in the Strumariinae using

cladistic analyses (Snijman in prep.) have shown that

Strumaria, Bokkeveldia and Gemmaria are weakly defined

and paraphyletic and that they are best treated as a single

genus Strumaria. Although the necessary generic redelimi-

tation will be explained and effected elsewhere, it is

important, notably for conservation purposes, to validate

the names of the undescribed species of the subtribe. Four

of the new species described here are assigned to

Strumaria, here defined according to Ker-Gawler (1814),

Bolus (1923), Barker (1943, 1944). The fifth new species

is placed in Hessea sensu Miiller-Doblies (1985), which

has proven to be a monophyletic genus with the exclusion

of the poorly known species H. spiralis Baker.

* Compton Herbarium, National Botanical Institute, Private Bag XI,

Claremont 7735, South Africa.

MS. received: 1991-06-27.

MATERIALS AND METHODS

This study was based on material from BM, BOL, K,

NBG, PRE, SAM and WIND. Additional morphological

and phenological data were gathered from the living

collection of all known members of the Strumariinae at

the National Botanical Garden, Kirstenbosch. Habitat in-

formation was derived from my own field observations.

The dates accompanying the cited specimens are field

collection dates of flowering bulbs. Specimens without

dates comprise cultivated flowering material which was

gathered over several years.

1. S. aestivalis Snijman, sp. nov., quoad tunicam

luteam bulbi, folia pubescentia et flores infundibuliformes

ad S. pubescentem W.F. Barker accedit, sed ab ea con-

cavitatibus latis inter filamenta interiora et stylum differt.

Figure 1.

TYPE. — Cape Province, 3018 (Kamiesberg): (— DB),

Farm Langberg, NW of Loeriesfontein, fl ex NBG

31-1-1984, Perry 1991 (NBG, holo.; K, PRE, MO).

Bulb solitary or occasionally forming bulblets, ovoid,

20-40 mm diam. , with the outer fibrous covering ranging

from brown to cream-coloured, fleshy and yellowish

within; neck up to 70 mm long, rarely absent. Leaves

absent at an thesis, 2 or rarely 3, recurved, lorate, 80—

280 x 15-26 mm, canaliculate, both surfaces densely

pubescent with 2 mm long, patent, silky, white hairs;

amplexicaul cataphyll shortly exserted, tipped with red,

soon withering down; non-amplexicaul prophyll hidden

in the bulb. Inflorescence widely spreading, 60—100 mm

across; scape 60—100 x 2.5— 4.0 mm, pale green to

glaucous, sometimes flushed with pink, pubescent or

glabrous, breaking off at the base in fruit; spathe valves

2

Bothalia 22,1 (1992)

FIGURE 1. — Strumaria aestivalis. 1, inflorescence; 2, vegetative habit; 3, whole flower; 4, androecium and gynoecium with free portion of foremost

stamen removed; 4A, transverse section through base of androecium and style showing nectar wells between inner filaments and style;

5, anther attachment, dorsal view; 5A, ventral view; 5B, lateral view. Drawn from Perry 1991.

lanceolate, 30—50 x 5-7 mm; bracteoles filiform, up to

20 mm long. Flowers 10—20, spreading, widely funnel-

shaped, white, with a pale pink median dorsal band on

each tepal, turning deeper pink with age, heavily scent-

ed; pedicels straight to upwardly curved, 40—55 mm long,

pale greenish-pink. Tepals free to the base, spreading,

oblong-lanceolate, 12—14 X 3—5 mm. Stamens equalling

the tepals, spreading slightly from near the base; filaments

separate, 7—10 mm long, adnate to the broadened style

base for up to 4 mm; the inner face of the inner whorl

Bothalia 22,1 (1992)

3

free with only the lateral margins adnate to the broadly

triquetrous style, forming 3 tubular nectar wells; anthers

subcentrifixed, ± 3 mm long before opening, wine-red;

pollen cream-coloured. Ovary with 1—2 ovules per locule.

Style up to 17 mm long, broadly triquetrous in the proximal

third, tapering and slender distally. Seeds fleshy, ovoid,

4—6 mm diam., green to reddish brown. Chromosome

number. 2n = 20.

Diagnostic features : Strumaria aestivalis is remarkable

in having three wide nectar wells formed by the fusion

of the lateral margins of the inner filaments with the

winged edges of the triquetrous style (Figure 1.4A). This

specialisation is also well developed in the species of

Strumaria with leaves arranged in a fan (5. truncata Jacq.,

S. hardyana D. & U. Miiller-Doblies, S. barbarae Oberm.

and S. phonolithica Dinter) and indicates parallel develop-

ment in S. aestivalis and this group. Strumaria aestivalis

is most closely related to S. pubescens with which it shares

yellow-fleshed bulbs and broadly lorate, pubescent leaves.

Yellow inner bulb tunics were previously reported as a res-

tricted character (Miiller-Doblies 1985) but the data given

here indicate that it is more widespread.

Distribution and phenology, the northwestern foothills

of the Langberg, northwest of Loeriesfontein, is the only

known locality of S. aestivalis (Figure 2). The population

is confined to the southeast-facing banks of a seasonal

stream, where the bulbs are aggregated in the shade of

rocks or low shrubs, amongst shale chips overlying heavy

loam, at elevations of 950 m. This site which lies east of

the main winter rainfall region where Strumaria is centred,

is located within a zone where the probability of rain is

greatest in March (Zucchini & Adamson 1984). S.

aestivalis responds rapidly to scattered summer thunder-

showers and flowers during January and February.

CAPE. — 3018 (Kamiesberg): Farm Langberg, NW of Loeriesfontein.

(-DB), Perry 1991 (K, MO, NBG, PRE); 20-1-1986, Snijman 1006 (MO,

NBG).

2. S. perryae Snijman, sp. nov., ex affinitate S.

pubescentis W.F. Barker et S. aestivalis Snijman, ab

utroque bulbi tunicis albidis et foliis anguste loratis differt.

Figure 3.

FIGURE 2. — The known geographical distribution of Strumaria

aestivalis, #; and S. perryae, ▲.

TYPE. — Cape Province, 3119 (Calvinia): (— AA),

between Grasberg and Theunisdrift, NW of Nieu-

woudtville, 15-5-1980, Perry 997 (NBG, holo.; K, PRE,

MO).

Bulb solitary, globose, 10—15 mm diam, with lightly

fibrous light brown outer tunics, fleshy and whitish

within; neck up to 45 mm long. Leaves absent at anthe-

sis, 2, suberect to recurved, narrowly lorate to lanceolate,

50—150 (—250) x 2.5— 5.0 mm, softly pubescent with

hairs up to 2 mm long on both surfaces, flushed with

red towards the base of the abaxial surface, subtended

by a subterranean amplexicaul cataphyll and non-

amplexicaul prophyll. Inflorescence somewhat clustered,

25-30 mm across; scape erect to flexuose, 50-165 (—240)

mm long, ± 1 mm diam. , reddish pink with a grey bloom,

rarely pubescent, breaking off at the base in fruit;

spathe valves linear-lanceolate, 15—20 x 1—2 mm; brac-

teoles filiform, up to 6 mm long. Flowers 3 — 11, more or

less ascending, widely funnel-shaped, scentless; pale

pink with a deeper pink median dorsal band on each tepal,

turning deep pink with age; pedicels straight to upwardly

curved, 20—30 (—60) mm long, pale green to reddish

pink. Tepals shortly adnate to the filaments for up to 1

mm, otherwise free, the outer spreading more widely than

the inner, oblong-lanceolate, 10—17 x 2.5— 4.0 mm.

Stamens suberect to slightly spreading, exserted beyond

the tepals; filaments separate, up to 17 mm long, with the

outer and inner whorls adnate to the style base for up to

2.5 mm and 3.5 mm respectively; anthers subcentrifixed,

± 3 mm long before opening, dark maroon; pollen cream-

coloured. Style up to 19 mm long, equalling or slightly

exceeding the stamens, slightly thickened and trigonous

proximally, tapering gradually upwards; with nectar col-

lecting in 3 droplets between the style and inner filaments;

stigma shortly trifid. Seeds fleshy, ovoid, 2.0— 2.5 mm

diam., green to reddish brown. Chromosome number.

2n = 20.

Flowering time : May, but commencing in April when

cultivated.

Diagnostic features: the long, lorate, pubescent leaves

and somewhat funnel-shaped flowers of S. perryae are

characteristics also found in 5. pubescens and S. aestivalis,

and indicate a close affinity with these species. The narrow

leaves of S. perryae are diagnostic (at most 5 mm across).

In contrast, S. pubescens and S. aestivalis have leaves more

than 10 mm wide and the synapomorphy of yellow inner

bulb tunics. The adnation of the filaments to the style is

well developed and reaches a length of 3.5 mm. This

feature is also conspicuous in specimens of S. pubescens,

S. watermeyeri L. Bolus, as well as S. aestivalis. Unlike

S. aestivalis the inner filaments of these species are closely

adnate to the style and the three efferent canals, which

conduct nectar from the septal nectary to the sinus between

the inner filaments and style, are only microscopically

visible.

Distribution and habitat: 5. perryae is known from a

single small population on the northern Bokkeveld

escarpment between Grasberg and Theunisdrift, north-

west of Nieuwoudtville (Figure 2). Plants grow in clay soil

in association with low karroid shrubs.

4

Bothalia 22,1 (1992)

EWH \\\\ '05

FIGURE 3 .—Strumaria perryae. 1, inflorescence; 2, vegetative habit; 3 & 4, whole flowers; 5, androecium and gynoecium with free

portion of foremost stamen removed and with nectar between the inner filaments and style; 6, transverse section through column

formed by fusion of the stamens to the style; 7, anther attachment, lateral view; 7A, dorsal view; 7B, ventral view. Drawn from

Perry 997.

Bothalia 22,1 (1992)

5

Etymology : the epithet honours Miss Pauline Perry of the

National Botanical Garden at Worcester, who discovered

this species. She has also located several rare and poorly

documented species of Strumariinae from Namaqualand.

CAPE.— 3119 (Calvinia): between Grasberg and Theunisdrift, NW of

Nieuwoudtville, (— AA), 15-5-1990, Perry 997 (K, MO, NBG, PRE); +7

km from Grasberg homestead towards Theunisdrift, (— AA), 8-5-1985,

Snijman 867 (NBG).

3. S. discifera Marloth ex Snijman, sp. nov., a specie-

bus ad Strumarium sensu lato pertinentibus, bulbi tunicis

interioribus albidis, foliis longis lanceolatis (ad 160 mm),

pubescentia (certe in iuvenilibus), floribus stellaribus,

tepalis canaliculatis, styli basi strumosa et strumae forma

manifeste bulbiformi vel discoidea distincta.

TYPE. — Cape Province, 3118 (Vanrhynsdorp): (-DB),

Bokkeveld Mountains, top of Koebee Pass, 18-4-1981,

Snijman 443 (NBG, holo.; K, PRE).

Bulbs solitary or forming large clumps, ovoid to sub-

globose, 10—20 mm diam., with outer tunics light brown

and softly fibrous, fleshy and white or occasionally pale

mauve within; neck (10—) 20—60 mm long. Leaves absent

or incipient at anthesis, 2 or occasionally 3, suberect to

prostrate, narrowly lanceolate, 20—160 x 3—10 mm,

pubescence variable, with long soft hairs or short hairs

covering both surfaces or the adaxial surface only, rarely

glabrous, sometimes flushed with red towards the base

of the abaxial surface, subtended by a subterranean

amplexicaul cataphyll and non-amplexicaul prophyll.

Inflorescence spreading, 25—130 mm across; scape some-

what flexuose, 50—140 mm long, + 2 mm diam., green

to reddish brown, glabrous or rarely pubescent, usually

breaking off at ground level while fruiting; spathe valves

linear-lanceolate, up to 30 x 3 mm; bracteoles filiform,

up to 5 mm long. Flowers (2—) 5-16, spreading, stellate,

glistening white, with an olive-green to pink median dor-

sal stripe on each tepal, scented or scentless; pedicels

straight to upwardly curved, 20—75 mm long, concolorous

with the scape. Tepals free to base, outspread, with the

outer whorl often deflexed, oblong-lanceolate, 4-7 x

1.5— 3.0 mm, channelled, sometimes abruptly conduplicate

in the proximal third. Stamens equalling or slightly shorter

than the tepals, spreading; filaments separate, adnate

proximally to the swollen style, with the inner whorl

usually attached slightly higher up than the outer; anthers

subcentrifixed, approximately 2 mm long and wine-red

before opening; pollen cream-coloured. Ovary with 1-3

ovules per locule. Style up to 7 mm long, equalling or

shortly exceeding the stamens, variably dilated in the

proximal half, either somewhat bulbiform or discoid with

a prominent distal irregular rim, narrowly terete in the

distal half, with nectar collecting in 3 droplets between

the base and inner filaments; stigma shortly trifid. Seeds

fleshy, ovoid 2.5— 4.0 mm diam. green to reddish brown.

.Chromosome number. 2n — 20.

Flowering time: March to May.

Diagnostic features : in comparison to the group of

closely allied pubescent-leaved species with white-fleshed

bulbs and stellate flowers, S. discifera has consistently

long, narrowly lanceolate leaves, distinctly channelled

tepals and a conspicuous bulbiform to discoid swelling at

the base of the style.

Distribution and variation: Strumaria discifera is

distributed between Vanrhynsdorp and Nieuwoudtville

eastwards to Calvinia and the Roggeveld escarpment in

the northwestern Cape (Figure 5).

The species includes a polymorphic range of popula-

tions. From the dolerite ridges on the outskirts of

Nieuwoudtville the bulbs are densely clump-forming,

whereas other known populations comprise scattered soli-

tary bulbs. The shape of the swelling at the base of the

style is also variable. The clump-forming bulbs have a

pronounced discoid stylar swelling with a frilly rim. This

character state is fairly consistent within the population

and is probably maintained through recurrent vegetative

propagation. Collections east of Nieuwoudtville to the

Hantamsberg and Bloukranz Pass near Calvinia also have

disc-like swellings, but these are not as broad as those in

the Nieuwoudtville populations and lack a prominent rim.

Elsewhere in the distribution range the stylar swelling tends

to be bulbiform in shape. Since the specimens from the

clonal population on the dolerite koppies at Nieuwoudt-

ville can be adequately diagnosed, these are described here

as a new subspecies.

3a. S. discifera subsp. discifera. Figure 4.

Bulbs solitary. Leaves 20-120 x 4—10 mm, with 1—3

mm long, soft patent white hairs, occasionally both

surfaces glabrous but then juveniles pubescent. Scape

glabrous. Tepals 5-6 x 1.5— 3.0 mm, channelled evenly

throughout. Style smoothly bulbiform or irregularly thick-

ened and longitudinally ridged in the proximal quarter.

Distribution and habitat: the known distribution extends

from near Vanrhynsdorp, eastwards onto the Bokkeveld

escarpment, across the high-lying plateau to Calvinia, then

southwards along the edge of the Roggeveld escarpment

to near Middelpos (Figure 5). Occupying gentle slopes

and depressions, the taxon inhabits heavy loamy soils, most

commonly derived from Nama and Ecca shales. The bulbs

often grow in association with renosterbos ( Elytropappus

rhinocerotis (L.f.) Less.).

CAPE. — 3019 (Loeriesfontein): Kafferdam, about 6 km NW of Loeries-

fontein on road to Kubiskouw Mountain, (—CD), Lavranos 27602 (NBG).

3118 (Vanrhynsdorp): N banks of Wiedourivier, near bridge between

Klawer and Vanrhynsdorp, (—DA), Snijman 261 (K, MO, NBG, PRE);

top of Koebee Pass, Bokkeveld Mountains, (— DB), 26-4-1988, Snijman

1172 (NBG, PRE), Snijman 443 (K, NBG, PRE). 3119 (Calvinia):

Glenridge, (—AC), Barker 4672 (NBG); Glen Lyon, (—AC), 3-4-1982,

Perry 1824 (K, MO, NBG, PRE); Mauve & Oliver sub G.N. 19699 (PRE);

5 miles E of Nieuwoudtville towards Calvinia, (—AC), 18-4-1969, Barker

10613 (NBG); 11 km E of Nieuwoudtville towards Calvinia, (—AC), Perry

1014 (MO, NBG, PRE); Akkerdam, lower slopes of Hantam Mountains,

(— BD), Barker 9344 (NBG); Bloukranz Pass, (— DA/DB), Bayer 1853

(NBG); Farm Blomfontein, Roggeveld escarpment, (-DD), 10-5-1985,

Snijman 876 (NBG).

3b. S. discifera subsp. bulbifera Snijman, subsp.

nov., a subspecie typica bulbo prolifero, styli basi stru-

mosa, strumae forma discoidea et margine irregulari

prominenti supra strumam distincta.

TYPE. — Cape Province, 3119 (Calvinia); (—AC), Nieu-

woudtville Wildflower Reserve, 19-4-1983, Perry &

Snijman 2042 (NBG, holo.; K, MO, PRE, S). Figure 6.

Bulb producing bulblets and forming dense clumps.

Leaves 6.5—150.0 x 3—10 mm, both surfaces covered with

6

Bothalia 22,1 (1992)

FIGURE 4. — Strumaria discifera subsp. discifera. 1, inflorescence; 2, vegetative habit; 3 & 4, whole flowers; 5, androecium and style with nectar

droplets between the inner filaments and style; 6, anther attachment, ventral view; 6A, dorsal view; 6B, lateral view. Drawn from Snijman 261.

2 mm long, white, patent hairs; adaxial surface flushed

with red proximally. Scape minutely pubescent or

glabrous. Tepals 5—7 x 2—3 mm, abruptly conduplicate

at a point almost a third from the base, otherwise channel-

led; outer whorl slightly deflexed. Style discoid proximally,

with a prominent irregular rim on the disc distally, abruptly

narrowed into a slender column above.

Distribution and habitat : subsp. bulbifera inhabits slopes

and hollows of low exposed dolerite ridges on the Bokke-

Bothalia 22,1 (1992)

7

FIGURE 5. — The known geographical distribution of Strumaria disci-

fera subsp. discifera , O; S. discifera subsp. bulbifera , ▲; and

S. villosa, •

veld escarpment near Nieuwoudtville (Figure 5). The

densely aggregated bulbs grow in deep, red loamy soils,

in association with open low, succulent shrubland.

CAPE. — 3119 (Calvinia): Meulsteen Vley, (-AC), 2-5-1927, Water-

meyer in Herb. Afr. Bol. 18648 (BOL); top of Vanrhyn’s Pass, (—AC),

30-4-1946, Smith 6490 (NBG); Farm Glen Lyon, (—AC), 8-5-1985,

Snijman 863 (NBG, PRE); Nieuwoudtville Wildflower Reserve, (—AC),

19-4-1983, Perry & Snijman 2042 (K, MO, NBG, PRE, S).

4. S. villosa Snijman , sp. nov. , a speciebus ad

Strumariam sensu lato pertinentibus, bulbi tunicis interiori-

bus luteis, foliis glaucis, pubescentia in pagina adaxiali

folii, floribus stellaribus, styli basi amplificata distincta.

Figure 7.

TYPE. — Cape Province, 2917 (Springbok): (—BA),

Richtersveld, 29° 10.05'S, 17° 41.49'E, E of Kosies, 3200

ft, 29-3-1981, Van Berkel 311 (NBG, holo.; K, PRE).

Bulb solitary, subglobose, 15—25 mm diam., with light

brown lightly fibrous outer tunics, fleshy and yellowish

within; neck up to 35 mm. Leaves absent at anthesis, 2

prostrate, narrowly elliptical to lorate, 30—85 x 10—15

mm; adaxial surface glaucous, covered with 2.5 mm long,

soft white, patent hairs; abaxial surface glabrous, shiny

green, subtended by a subterranean amplexicaul cataphyll

and non-amplexicaul prophyll. Inflorescence spreading,

30—100 mm across; scape slightly flexuose, 60—140 x

2—3 mm, pale green to pink with a grey bloom, breaking

loose at the base in fruit; spathe valves linear-lanceolate,

15—20 x 4 mm; bracteoles filiform, up to 5 mm long.

Flowers 8—14, spreading, stellate, pure white or white to

pale pink with a pale pink median dorsal stripe on each

tepal, scentless; pedicels straight to upwardly curved,

35-80 mm long, concolorous with the scape. Tepals free

to the base, outspread to slightly deflexed, oblong-

lanceolate, 6.5— 8.5 x 2—3 mm, distinctly channelled with

slightly undulate margins proximally. Stamens equalling

or slightly shorter than the tepals, spreading; filaments

separate, adnate to the broadened style base, with the inner

whorl attached higher up than the outer, broad but not

bulbous basally, tapering slightly upwards; anthers sub-

centrifixed, 1.5 mm long and wine-red before dehiscing;

pollen whitish. Ovary with 1—4 ovules per locule. Style

up to 5 mm long, more or less equalling the stamens.

tapering smoothly upwards from a broad obscurely

conoidal base, with nectar collecting in 3 droplets between

the style base and inner filaments; stigma shortly trifid.

Seeds fleshy, ovoid, approximately 2 mm diam. green to

reddish brown. Chromosome number-. 2n = 20 + 2— 3B.

Flowering period extends from March to April.

Diagnostic features: the leaves of S. villosa are softly

villous on the adaxial surface and are characteristically

glaucous. Unlike other pubescent-leaved species of the

Strumariinae with white, stellate flowers and filaments

adnate to the style, S. villosa is specialized in having yellow

inner bulb tunics.

Distribution and habitat: this rare species is known from

only one locality in the Richtersveld, near Kosies, (Figure

5). Locally abundant on low hills, the species is confined

to exposed, east-facing slopes amongst quartz pebbles

which overlie weathered granite soil.

CAPE.— 2917 (Springbok): 29° 10.05'S, 17° 41.49'E of Kosies, (-BA),

Van Berkel 156 (NBG); 29-3-1981, Van Berkel 311 (K. NBG, PRE); Perry

1544 (K, MO, NBG, PRE, S).

5. H. speciosa Snijman, sp. nov. , quoad tubum bre-

vissimum perigonii et tepala plana ad Hesseam pilosulam

D. & U. Miiller-Doblies et H. incanam Snijman accedit,

sed ab ambobus foliis glabris et staminibus longioribus

(aequantibus vel superantibus tepala) satis differt. Figure 8.

TYPE.— Namibia, 2818 (Warmbad): (— CA), Warmbad

District, Farm Witpiitz, 15-5-1963, Giess, Volk & B. Bleiss-

ner 6960 (WIND, holo.; PRE).

Bulb solitary, deep-seated, subglobose, 25—60 mm

diam., covered with several layers of cream-coloured

cottony fibrous tunics, extended into a stout neck 100—170

mm long. Leaves absent at anthesis, 2, recurved, lorate,

up to 120 x 4—6 mm, plane, glabrous, dark green and

flushed with red towards the base; amplexicaul cataphyll

remaining subterranean; prophyll unknown. Inflorescence

dense, hemispherical to spherical, 70—120 mm across;

scape erect to somewhat flexuose, 60-160 x 3—5 mm,

initially green, breaking off at the base in fruit; spathe

valves linear-lanceolate, 20—40 x 3— 7 mm; bracteoles

filiform, up to 25 mm long. Flowers (20—) 30—65,

spreading, stellate, white to delicate pink with deep pink

or greenish median stripes on the undersurface, ageing

to light brown, with a heavy coconut-like scent; pedicels

straight, 20—50 mm long, becoming straw-coloured.

Tepals almost free to the base or very shortly adnate to

the staminal tube for up to 0.25 mm, otherwise outspread,

oblong-lanceolate, 8—15 x 2— 4 mm, with plane edges.

Stamens equalling or up to 2 mm longer than the tepals,

becoming outspread; filaments connate proximally into a

tube protruding from the perigone throat by (1.0—) 1.5— 4.0

mm, subulate above, occasionally shortly toothed in the

axils between adjacent filaments; anthers centrifixed, 3

mm long and dark wine-red before opening; pollen cream-

coloured. Style up to 15 mm long, narrow throughout, with

nectar collecting in a well around the base; stigma shortly

trifid. Seeds not known. Chromosome number: 2n = 22.

Flowering time: from late March into May.

Diagnostic features are the deep-seated bulb with a long

neck (up to 170 mm); the somewhat spherical inflores-

Bothalia 22,1 (1992)

tw« iaas

FIGURE O.—Strumaria discifera subsp. bulbifera. 1, inflorescence; 2, vegetative habit; 3 & 4, flowers with tepals removed to show variable

style sculpturing and nectar droplets between the inner filaments and style; 5, whole flower; 6, anther attachment, dorsal view; 6A, ventral

view; 6B, lateral view. Drawn from Perry & Snijman 2042.

Bothalia 22,1 (1992)

9

_ 20 m m

EWH 1991

FIGURE 7. — Strumaria villosa. 1, plant with inflorescence; 2 & 3, flowers indicating the attachment of the filaments to the style base and nectar

droplets between the inner filaments and style; 4, anther attachment, dorsal view; 4A, ventral view; 4B, lateral view; 5 & 6, leaves. Drawn

from Van Berkel 156.

10

Bothalia 22,1 (1992)

FIGURE 8. — Hessea speciosa. 1, inflorescence; 2 & 3, whole flowers; 4, partial section of flower; 5, anther attachment, lateral view; 5A, ventral

view; 5B, dorsal view; 6, vegetative habit. Drawn from Snijman 1163.

Bothalia 22,1 (1992)

cence; and the very short perigone tube (0.25 mm or less).

In these respects H. speciosa is similar to the shortly

pubescent-leaved species, H. pilosula D. & U. Miiller-

Doblies and H. incana Snijman, with which it also shares

plane tepals. However, the glabrous leaves and the rela-

tive length of the stamens to the tepals distinguish it from

these species. The stamens equal to or up to 2 mm longer

than the tepals in H. speciosa , whereas they are distinctly

shorter than the tepals (by 3 mm or more) in H. pilosula

and H. incana. The inflorescence of H. speciosa may

sometimes be confused with flowering material of H.

breviflora Herb, from Namaqualand, but unlike this

species the bulbs are without a conspicuous, exserted, red

cataphyll which sheathes the foliage leaves.

Distribution and habitat : Hessea speciosa is recorded

from red sand dunes and flats of friable loam, associated

with the extensive drainage system of seasonal rivers from

Warmbad in southern Namibia to Fraserburg in the central

Cape. The associated vegetation is predominantly grass-

veld (Figure 9).

Variation : often the northerly populations have a distinct

staminal tube (1.5— 3.5 mm long), whereas specimens from

the south of the distribution range have only a shortly

developed staminal tube (less than 1.5 mm). Both white

and pale pink flower forms occur, as well as the occa-

sional novelty of small teeth in the axils between adjoining

filaments.

NAMIBIA.— 2818 (Warmbad) Warmbad District, Farm Witpiitz,

(-CA), 15-5-1963, Giess, Volk & B. Bleissner 6960 (PRE, WIND);

15-5-1963, S. Bleissner 268 (PRE).

CAPE. — 2918 (Gamoep): Farm Eendop, SW of Klipvlei, (—AC),

1-5-1981, Van Berkel 331 (NBG); Aggenys Mine, (-BB), 23-5-1989, S.

Dean 655 (NBG); Banke, Pofadder, (— DB), 3-5-1988, S. Dean s.n.

(NBG). 2919 (Pofadder): Farm Kykgate, along road between Springbok

and Pofadder, (—AC), 13-5-1969, Van Breda 4147 ( PRE). 2921 (Kenhardt):

Kenhardt, (—AC), 9-5-1927, Long sub NBG 947/27 ( BOL); 14-5-1936,

Martin sub NBG 1188/36 (BOL). 3120 (Williston): 40 miles N of

Calvinia, (-AA/AB), 30-3-1953, Hall 684 (NBG); 36 km N of Downes

towards Brandvlei, (—AC), 3-4-1988, Snijman 1163 (NBG. PRE). 3121

(Fraserburg): 49 miles from Fraserburg towards Williston, (—AC), Smith

6491 (NBG).

ACKNOWLEDGEMENTS

I wish to thank Mrs J.E. Ward-Hilhorst for the botanical

illustrations.

REFERENCES

BARKER, W.F. 1943. Strumaria picta, S. pubescens in Plantae novae

africanae. Journal of South African Botany 9: 145—149.

BARKER, W.F. 1944. Strumaria salteri in Plantae novae africanae.

Journal of South African Botany 10: 131—133.

BOLUS, L. 1923. Novitates africanae. Annals of the Bolus Herbarium

3 : 70 - 85.

KER-GAWLER, J. 1814. Strumaria gemmata. Jewelled-flowered Struma-

ria. In J. Sims, The Botanical Magazine 39: t. 1620. Sherwood,

Neely & Jones, London.

MULLER-DOBLIES, D. & U. 1985. De Liliifloris notulae 2. De taxon-

omia subtribus Strumariinae (Amaryllidaceae). Botanische

Jahrbucher fur Systematik 107: 17—47.

SNIJMAN, D. 1989. New species of Hessea (Amaryllidaceae) from the

western Cape. South African Journal of Botany 55: 349—357.

SNUMAN, D. 1991. Kamiesbergia. a new monotypic genus of the

Amaryllideae-Strumariinae (Amaryllidaceae) from the north-

western Cape. Bothalia 21: 125—128.

ZUCCHINI. W. & ADAMSON, P.T. 1984. The occurrence and severity

of droughts in South Africa. Water Research Commission Report

No. 91/1/1984.

Bothalia 22,1: 13-18 (1992)

Three new species of Diascia (Scrophulariaceae) from the western Cape

K. E. STEINER*

Keywords: breeding system, Diascia , new species, Scrophulariaceae

ABSTRACT

Three annual Diascia species are described from the western Cape. Two species, D. maculata K.E. Steiner and D humilis

K.E. Steiner have small flowers and are closely related. D. maculata is characterized by a slightly gibbous corolla with

no spurs or sacs, erect stamens, and a distinct patch of oil-secreting trichomes in the tube at the base of each upper and

lateral corolla lobe. D. humilis is characterized by a bisaccate corolla, forward projecting stamens and oil-secreting trichomes

localized in corolla sacs. The third new species, D. hexensis K.E. Steiner, is most similar to D. sacculata Benth., but it

differs from that species by its larger flowers and much longer spurs which curve downward instead of upward.

UITTREKSEL

Drie eenjarige Duzsaa-spesies van die Wes-Kaap word beskryf. Twee spesies, D. maculata K.E. Steiner en D. humilis

K.E. Steiner het klein blomme en is na verwant. D. maculata word gekenmerk deur die effens bulterige blomkroon sonder

spore of sakkies, die regop meeldrade, en ’n duidelike gebied met olie-afskeidende trigome in die kroonbuis aan die basis

van elke boonste en laterale blomkroonlob. D. humilis word gekenmerk deur die tweesakkige blomkroon, die meeldrade

wat na vore gerig is, en die olie-afskeidende trigome wat tot die sakkies van die blomkroon beperk is. Die derde nuwe spe-

sies, D. hexensis K.E. Steiner, kom in baie opsigte met D. sacculata Benth. ooreen, maar verskil daarvan in die groter

blomme en veel langer spore wat afwaarts in stede van opwaarts krul.

INTRODUCTION

Diascia section Diascia comprises approximately 41

species of annuals centred in the western Cape of South

Africa. This group is currently under revision and has been

found to contain many undescribed species. Three new

species are described below.

D. maculata K.E. Steiner, sp. nov., D. humili K.E.

Steiner proxima, sed differt corolla gibbosa (nonbisacca-

ta), staminibus porrectis (non erectis), trichomatibus

oleum secernentibus in tubo corollae basi loborum

superorum lateraliumque corollae (non in sacculis

corollae).

TYPE. — Cape, 3119 (Calvinia); 300 m north of

Nieuwoudtville Caravan Park, (-AC), 740 m, 21-viii-1990,

Steiner 2165 (NBG, holo.; K, MO, PRE, US).

Annual herb, glabrous, simple or branching from the

base. Stems up to 220 mm long, erect or decumbent,

tetragonal in cross section, up to 1.5 mm on a side. Basal

leaves few to many, rosulate or clustered, simple, petiolate,

spreading or ascending, lamina 4—23 mm long, ovate to

elliptic, apex rounded to acute, base attenuate, margins

nearly entire to irregularly lobed or divided, divisions up

to ± 3 mm long, oblong-ovate to triangular, entire, acute

to acuminate; petioles up to ± 21 mm; stem leaves smaller,

becoming reduced upwards, alternate, opposite or verticil-

late. Flowers axillary, one or two flowers open per stem,

nodding in bud, pedicels 22-41 mm long, ascending,

broadened and dorsiventrally flattened specially where

attached to flower, recurving in fruit. Calyx lobes five,

± equal, ± 2.0-2. 5 x 0.8-1.0 mm, spreading, or the

* Compton Herbarium, National Botanical Institute, Private Bag X7,

Claremont 7735.

MS. received: 1991-07-09.

lateral two slightly reflexed, lanceolate, acuminate, mar-

gins white ciliate. Corolla bilabiate, limb ± 7.4—10.5 x

8.6—11.0 mm; upper lobes ± 2.2— 3.0 x 2.0— 2.8 mm,

oblong-ovate, rounded; lateral lobes + 2.7— 3.6 x 2.3— 3.4

mm, ovate, rounded, bases oblique; lower lobe ± 3.0— 3.9

x 2.9— 5.0 mm, obovate, rounded to emarginate, upper

lobes yellowish pink distally, bases purplish red with red

veins or uniformly reddish purple, other lobes similar in

colour but lacking veins, all lobes with dark purple-tipped

glandular trichomes, especially on inner surface near the

base; tube ± 1.0— 2.0 mm, reddish purple with yellow

spots below upper and lateral lobes, yellow patch below

each upper lobe consisting of a single spot or 3 or 4

separate or partially coalesced spots, patch below each

lateral lobe a single ± 1 mm long elliptical patch cor-

responding to a gibbous portion of the tube, yellow patches

below lateral and upper lobes -usually separated by a nar-

row strip of reddish purple tissue, but sometimes consist-

ing of a single spot spanning the sinus between upper and

lateral lobes; gibbous portion of tube with two patches of

oil-secreting trichomes, a + 1 mm long elliptical patch

below each lateral lobe and a ± 0.3 mm patch at base of

each upper corolla lobe near the sinus with the lateral lobe;

spurs or sacs absent; central portion of tube turned out-

ward to form a boss bearing the stamens, boss ± 1.2— 1.5

mm high, connected to the upper lip by a septum. Stamens

four, projecting forwards, anticous filaments (appearing

posticous due to twisting of the bases) ± 1.5— 1.7 mm long,

curved and bearing a few scattered trichomes, posticous

filaments (appearing anticous) ± 1.2— 1.5 mm long, ±

straight, widened and bilobed or simply strongly bent

backward just below the anthers, anthers ± 0.2— 0.5 mm,

strongly cohering, yellow, attached to posticous lobes of

filaments if bilobed, pollen usually orange. Ovary ± 1.2

x 0.7 mm, ovate in outline with purple markings on upper

two thirds to one half, style ± 0.7— 0.8 mm long, straight,

stigma subcapitate, surrounded by anthers, ovules ±

50—55. Capsule ± 4.7— 6.0 x 2.7— 3.0 mm, falciform-

14

Bothalia 22,1 (1992)

FIGURE 1. Diascia maculata, Steiner 21651.4 : A, habit, X 1; B, flower, front view, x 4.2; C, flower, rear view, X 4.2; D, calyx,

x 7.8; E, gynoecium, x 16.2; F, capsule, x 6; G, seed, ventral view, x 25; H, seed, side view, x 25; I, androecium, x 16.2.

Bothalia 22,1 (1992)

15

ovate in outline, exceeding calyx at maturity; seeds +

0.65—0.85 mm long, dorsal surface ridged, ventral surface

with seed coat extended to form a cupule with an oblong

elliptical opening. Figure 1.

Flowering time : August— October.

Diagnostic features: Diascia maculata is characterized

by a slightly gibbous corolla with no spurs or sacs, stamens

that project forwards, and a distinct patch of oil-secreting

trichomes (cf. Vogel 1974) in the tube at the base of each

upper and lateral corolla lobe. In most populations, there

is a yellow spot corresponding to the trichome patches.

The pair of spots below one upper and one lateral lobe

can be partially confluent or separated by a small strip

of reddish purple tissue. The spotting pattern at one

locality, the Matroosberg, is slightly different. There, the

single elliptical yellow spot below each lobe characteristic

of other populations is broken up into several smaller

vertically oriented narrowly elliptic spots that are either

distinct or only partially confluent. The specific epithet

refers to the yellow spotting pattern of the corolla tube.

Diascia maculata is most closely related to D. humilis

which has flowers of a similar size and colour. These two

species also have very similar capsules, seeds, and pollen

colour and can occur sympatrically on recently (less than

one year old) burned fynbos (Farm Welkom) and renoster-

veld (Farm Matroosberg). D. maculata is distinguished

most easily from D. humilis by the position of the stamens

in relation to the upper corolla lobes, the presence of four

distinct oil-secreting trichome patches and the absence of

corolla sacs or spurs. The corolla tube of D. maculata is

very similar to that of D. gracilis Schltr., since both species

are somewhat gibbous; however, in D. gracilis the stamens

are erect rather than projecting forward and the long linear

capsules are usually more than four times as long as wide,

not falciform-ovate and only ± twice as long as wide.

Distribution and habitat: D. maculata ranges from the

Hex River Mountains in the southwestern Cape north to

the Kamiesberg in Namaqualand (Figure 2). It ranges in

elevation from 640 to 1 130 m. Despite its broad geo-

graphical range, D. maculata is known from relatively few

collections. This may be due to overall rarity, a dependence

on first year burns or small, easily overlooked, flowers.

Breeding system: D. maculata is autogamous; however,

because it secretes floral oil, it may be visited and cross-

pollinated, at least occasionally, by small short-legged

oil-collecting Rediviva bees (Melittidae) (cf. Steiner &

Whitehead 1988, 1990, 1991; Whitehead & Steiner 1985).

CAPE.— 3018 (Kamiesberg): Farm Welkom, 6.4 km south of junction

with Garies— Platbakkies Road, (—AC), 1 130 m, 30-ix-1988, Steiner

1852 (NBG). 3119 (Calvinia): old Nieuwoudtville road, 3.5 km west of

main road in Nieuwoudtville, (—AC), ± 820 m, 10-ix-1986, Steiner 1359

(NBG); Farm Lokenburg, ± 7.1 km southwest of Oorlogskloof road,

(— CA), ± 640 m, 25-ix-1986, Steiner 1393 (NBG). 3319 (Worcester):

Farm Matroosberg, Hex River Pass, + 5 km east of turnoff to De Dooms

on the National Road (Nl), (-BD), ± 710 m, 3-X-1989, Steiner 2037

(NBG).

D. humilis K.E. Steiner, sp. nov., D. maculatae K.E.

Steiner proxima, sed differt corolla bisaccata (non

gibbosa), staminibus erectis (non porrectis), trichomatibus

oleum secernentibus in sacculis corollae (non in tubo

corollae basi loborum superorum lateraliumque corollae).

TYPE. — Cape, 3119 (Calvinia); Farm Koerdemoe-

fontein, 5.9 km east of Oorlogskloof road on road to

Clanwilliam (R364), (—CD), + 720 m, 15-ix-1989, Steiner

2005 (NBG, holo.; E, K, MO, PRE).

Annual herb, glabrous, simple or branching from the

base. Stems up to 250 mm long, erect or decumbent,

tetragonal in cross-section, up to 1.5 mm on a side. Basal

leaves few to many, rosulate or crowded, simple, petio-

late, spreading or ascending; lamina 4-41 mm long, ovate

to elliptic, apex rounded to acute, base attenuate, margins

nearly entire to irregularly lobed or divided; divisions up

to ± 3 mm long, oblong-ovate to narrowly triangular,

entire, acute; petioles up to ± 12 mm; stem leaves smaller,

becoming reduced and more dissected upwards, 3-verticil-

late. Flowers axillary, one or two open per stem, nodding

in bud, pedicels 16.0—20.5 mm long, ascending, broadened

and dorsiventrally flattened especially where attached to

the flower, recurving, but with the capsule turned up slight-

ly, in fruit. Calyx lobes five, ± equal, ± 1.8— 2.8 x

0.7-1.0 mm, spreading, lanceolate, acuminate, margins

white ciliate. Corolla bilabiate, limb 6.2— 7.6 x 7.0-8.6

mm; upper lobes ± 1.4— 2.5 x 1.7— 2.0 mm, falciform-

oblong to oblong-ovate, rounded; lateral lobes ± 2.3— 3.0

x 1.9— 2.4 mm, broadly ovate, rounded, bases oblique;

lower lobe 2. 4-3.0 x 2.5— 3.1 mm, obovate, rounded to

emarginate, upper lobes yellowish pink or reddish purple,

with red veins at the base, other lobes similar in colour

but lacking veins, all lobes with sessile or semi-sessile dark

purple glandular trichomes, especially on inner surface

near the base; tube + 1.0-1. 5 mm, distended at base of

each lateral lobe into a shallow yellow sac, ± 1.8 mm long,

sacs containing yellow oil-secreting trichomes, especially

near the tips; central portion of tube turned out to form

a boss bearing the stamens, boss ± 1.2— 1.5 on anticous

side and 0.2— 0.3 mm on posticous side, reddish purple

in front and back and yellow on sides, connected to the

upper lip by a septum. Stamens four, erect, anticous fila-

ments (appearing posticous due to twisting of the base),

touching the upper corolla lip, + 2.0 mm long, reddish

purple, falciform, glabrous or with a few scattered purple

clavate trichomes, posticous filaments (appearing anticous)

± 1.5— 1.6 mm, reddish purple, + straight, except just

FIGURE 2.— Known geographic distribution of Diascia maculata , A;

and D. humilis, •, in South Africa.

16

Bothalia 22,1 (1992)

below the anther where there is a sharp nearly 90° bend

and broadening of the filament sometimes with a sterile

outgrowth in the opposite direction, bend or outgrowth

with purple clavate trichomes; anthers ± 0.20—0.30 mm,

strongly cohering, pale pink or whitish, pollen usually

orange. Ovary + 1.3— 1.4 X 0.6— 1.0 mm, ovate in outline,

sometimes with purple markings on upper half, style

± 0.70—0.75 mm long, straight, stigma surrounded by

anthers, ovules ± 40—60. Capsule ± 5.0— 6.5 x 2.6—

3.0 mm, falciform-ovate in outline, exceeding calyx at

maturity; seeds ± 0.65—0.90 mm long, falciform in

outline, dorsal surface ridged, ventral surface with seed

coat extended to form a cupule with an oblong-elliptical

opening. Figure 3.

Flowering time : August— October.

Diagnostic features : D. humilis is characterized by

small, usually yellowish pink flowers, its erect stamens,

orange to red-orange pollen and small yellow sacs. It is

most similar to D. maculata with which it can often be

FIGURE 3. — Diascia humilis, Steiner 2119.4: A, habit, X 0.7; B, flower,

front view, x 3.5; C, flower, rear view x 3.5; D, calyx x 5.5;

E, seed, ventral view, X 17.5; F, seed, side view, X 17.5; G,

capsule, x 4.2; H, gynoecium, X 11.4; I, androecium, X 11.4.

found on burned sites. It can be distinguished from that

species most easily by the configuration of its stamens.

In D. maculata the stamens project forward away from the

upper corolla lip, whereas in D. humilis , they are erect

with the anthers in a position very close to the upper

corolla lip. The yellow markings on the two species also

differ. The sacs and sides of the boss in D. humilis are

yellow, but there is no discrete spotting pattern as in D.

maculata. In D. maculata there are one or several yellow

spots at the base of each upper corolla lobe in addition

to the yellow spot(s) at the base of the lateral lobes.

Associated with the yellow spots at the base of each upper

lobe in D. maculata is a patch of oil-secreting trichomes.

D. humilis does not have such trichomes in this position.

In D. humilis they are contained within each yellow sac.

Distribution and habitat: D. humilis ranges from the

Kamiesberg south to Worcester and east through the Little

Karoo to Uniondale (Figure 2). It ranges in elevation from

320 to + 1 200 m and is most commonly found on first

year renosterveld or fynbos burns, however it also occurs

in short karroid shrublands.

Breeding system: D. humilis is autogamous; but like D.

maculata it secretes floral oil and may therefore be visited

and cross-pollinated, at least occasionally, by small, short-

legged, oil-collecting Rediviva bees (cf. Steiner & White-

head 1988, 1990, 1991; Whitehead & Steiner 1985).

CAPE. — 3018 (Kamiesberg): Farm Welkom, 6.4 km south of junction

with Garies— Platbakkies Road, (—AC), 1 B0 m, 29-ix-1988, Steiner 1837

(NBG). 3119 (Calvinia): Lokenburg, (— CA), 29-viii-1941, Compton 11501

(NBG); Botterkloof Pass, (—CD), 16-viii-1983, Batten 660 (E); Farm

Koerdemoefontein, 5.9 km E of Oorlogskloof road on road to Clanwil-

liam (R364), (-CD), ± 720 m, 15-ix-1989, Steiner 2005 (NBG). 3219

(Wuppertal): road to Wuppertal, 3.9 km south of turnoff to Doringrivier

in Bidouw Valley, (— AA), 500 m, 29-viii-1990, Steiner 2204 (NBG);

pass into Bidouw Valley, 6.2 miles from turnoff from Clanwilliam—

Calvinia road, (— AA), ± 470 m, 24-viii-1967, Thompson 348 (STE);

Krom River, S Cedarberg, (— CB), 2-X-1952, Esterhuysen 20469 (BOL).

3220 (Sutherland): Farm Driefontein, 7.2 km east of turnoff to Ouberg

Pass on road to Ladismith, (— CB), 850 m, 6-viii-1990, Steiner 2119

(NBG); ± 15 m north of road 356, 86.7 km NE of road 355, Farm

Thyskraal, (— CC), 840 m, 26-ix-1984, Steiner 793 (NBG). 3319

(Worcester): Farm Matroosberg, Hex River Pass, ± 5 km east of turnoff

to De Dooms on Nl, (— BD), ± 710 m, 3-X-1989, Steiner 2034 (NBG);

ibid., 18-X-1989, Steiner 2049 (NBG); Breede River flood plain behind

shooting range on Worcester commonage, (— CA), 220 m, 9-ix-1985,

Steiner 1010 (NBG). 3320 (Montagu): Touwsrivier, on hill + 2 km west

of Tweedside railway station, (— AB), 1 200 m, 12-viii-1988, Vlok 1990

(NBG); Farm Driefontein, 7.2 km east of turnoff to Montagu on

Touwsrivier-Ladismith road, (— CB), 850 m, 6-viii-1990, Steiner 2119

(NBG); Farm Rietvlei, Montagu -Barrydale Road (R62), + 10.5 km SE

of Montagu, (— CC), ± 370 m, 18-viii-1987, Steiner 1501 (NBG);

Kogmanskloof, 100 m north of tunnel on Ashton-Montagu road, (— CC),

180 m, 8-ix-1984, Steiner 740 (pressed ex hort), 2-xi-1984 (NBG);

Montagu -Barrydale road, (R62), 16.5 km west of turnoff to Tradouw’s

Pass in Barrydale, (-DC), ± 540 m, 18-viii-1987, Steiner 15(17 (NBG).

3322 (Oudtshoorn): Farm Swartberg, lower northern slopes of Swart-

berg Mts, (—AD), ± 1 030 m, 12-ix-1986, Vlok 1605 (NBG). 3323

(Willowmore): Farm Misgund, at southern base of Antoniesberg, (-AD),

± 905 m, 24-viii-1990, Vlok 2383 (NBG).

D. hexensis K.E. Steiner, sp. nov., D. sacculatae

Benth. affinis , sed differt floribus maioribus, calcaribus

corollae deorsum curvis (non sursum curvis) et longiori-

bus (9.5-11.5 mm, non <4.0 mm).

TYPE. — Cape, 3319 (Worcester); near top of Hex River

Pass, 17.3 km west of junction with road R46 to Ceres,

(— BD), + 780 m, 20-ix-1985, Steiner 1042 (NBG, holo.;

MO). Figure 4.

Bothalia 22,1 (1992)

17

Annual herb, glabrous, simple or branching from the

base. Stems up to 220 mm long, erect or decumbent,

tetragonal in cross section, up to 1.5 mm on a side.

Basal leaves few to many, rosulate or crowded, simple,

petiolate; lamina 8—34 mm long, ovate or obovate to

elliptic, apex rounded to acute or apiculate, base attenuate,

margins irregularly toothed or occasionally lobed or

cleft, lobes or teeth narrowly to broadly triangular, acute

to apiculate; petioles up to ± 20 mm long; stem leaves

opposite, alternate or verticillate, becoming reduced

upwards. Flowers axillary, one flower open per stem,

unscented, nodding in bud, pedicels 25-80 mm long,

ascending, broadened and dorsiventrally flattened

especially where attached to the flower, Elongating and

ascending or recurved with only the apical portion

ascending in fruit. Calyx lobes five, upper three ± equal,

± 3.1— 4.0 x 0.8— 1.2 mm, reflexed, lower two somewhat

broader ± 3.1— 4.1 x 1.2— 1.4 mm, spreading, all lobes

lanceolate, attenuate with white ciliate margins. Corolla

bilabiate, limb ± 11.0—15.3 x 12.9— 17.0 mm; upper lobes

± 3.6— 4.7 x 4. 2-5.0 mm, widely ovate to oblong-ovate,

rounded to emarginate, bases oblique, lateral lobes ±

4. 9-5.1 x 3.8— 4.8 mm, oblong-ovate, rounded to emar-

ginate, lower lobe ± 4.8— 6.4 x 5.8— 8.3 mm, obcordate,

all lobes purple to reddish purple on front and pale

purple on back, upper lobes with several darker reddish

purple lines at the base, glandular pubescent especially

on inside surface; tube, ± 1.0 mm deep, purple to

reddish purple inside with 2 or 3 small fusiform to elliptic

yellow spots at the base of each upper corolla lobe, spots

on outside larger, one or two small ones and one large

one or sometimes all confluent and forming a single large

G

FIGURE 4. — Diascia hexensis, Steiner

2117.2: A, habit x 0.7; B,

calyx, x 2.9; C, flower, front

view, x 2.6; D, flower, rear

view, X 2.6; E, capsule, x

2.9; F, seed, ventral view, x

18.4; G, seed, side view, x

18.4; H, gynoecium, x 8.8; I,

androecium, x 8.8.

18

Bothalia 22,1 (1992)

spot, tube drawn out below upper lip into 2 spurs ±

9.5—11.5 mm long (measured along the inseam), diverging,

projecting backwards and downwards, attenuate, purple

to reddish purple, containing clear to light purple, spher-

ical, multicellular, oil-secreting trichomes in the distal half,

sparsely glandular pubescent outside, tube at base of lower

lip turned out to form a boss bearing the stamens, boss

± 1.0— 1.3 mm high on the anticous side and 1.3— 1.6 mm

high posticous to the stamens, with scattered glandular

hairs, connected on the posticous side to the upper lip by

a septum. Stamens four, projecting forwards, filaments

light purple, anticous filaments (appearing posticous due

to twisting of the bases) ± 3.0— 3.5 mm long, curved at

the base, with scattered purple clavate trichomes on the

upper portion, posticous filaments (appearing anticous)

± 2.5— 2.9 mm long, ± straight, glabrous or with a few

purple clavate trichomes; anthers ± 0.3— 0.5 mm long,

strongly cohering, yellow. Ovary ± 1.75—2.1 x 1.0— 1.1

mm, oblong-ovate in outline, style ± 1.5— 1.8 mm long,

± straight, stigma subcapitate, surrounded by anthers,

ovules ± 90—100. Capsule ± 8.7—10.5 x 2.2— 2.6 mm,

narrowly falcate, + twice as long as the calyx at maturi-

ty, seeds ± 0.7— 0.8 mm long, falciform in outline, dorsal

surfaced ridged, ventral surface with seed coat extended

to form a cupule with an oblong opening.

Flowering time : August— October.

Diagnostic features: D. hexensis is most easily recog-

nized by its moderately long attenuate corolla spurs, its

long narrowly falcate capsule (± 3.5 to 4.5 times as long

FIGURE 5. — Known distribution of Diascia hexensis in South Africa.

as wide) and its ovate broadly toothed leaves. Although

capsule and leaf shape are similar in D. sacculata, the

much longer spurs of D. hexensis clearly distinguish it

from that species.

Distribution and habitat : D. hexensis is known only from

the southern part of the Roggeveld Mountains west of

Sutherland, south to the northern slopes of the Langeberg

(Figure 5). It ranges in elevation from 510 m to 1 300 m

and occurs in karoo shrublands on relatively moist south-

facing slopes. The specific epithet refers to its occurrence

next to the Hex River Pass.

Breeding system: D. hexensis is autogamous; but because

of its long spurs containing floral oil, it is probably visited

and cross-pollinated, at least occasionally, by medium-

sized oil-collecting Rediviva bees with long forelegs.

CAPE. — 3220 (Sutherland): Ouberg Pass road, 0.7 km west of Vis

River rd, I 300 m, (-AD), l-x-1986, Steiner 1409 (NBG); ± 15 m north

of road 356, 86.7 km NE of R355, Farm Thyskraal, (-CC), 840 m,

26-ix-1984, Steiner 790 (NBG); below road to Sutherland (R354), 45.1

km north of junction with National Road (Nl) at Matjiesfontein, (—DC),

± 900 m, 20-ix-1985, Steiner 1053 (NBG). 3319 (Worcester): Farm

Matroosberg, Hex River Pass, ± 5 km east of turnoff to De Dooms

on Nl, (-BD), ± 710 m, 3-X-1989, Steiner 2032 (NBG). 3320 (Mon-

tagu): Whitehill, (-BA), 18-viii-1941, Compton 11249 (NBG); Farm

Driefontein, 7.2 km east of turnoff to Ouberg Pass on road to Ladismith,

(-CB), 6-viii-1990, Steiner 2117 (NBG); Farm Sewefontein, 32 km west

of Ladismith— Laingsburg road, (—DA), 510 m, 6-viii-1990, Steiner 2125

(NBG). 3322 (Oudtshoorn): Farm Frisgewaagd, northern slopes of the

Swartberg (—AD), ± 1 000 m, 12-ix-1986, Vlok 1604B (NBG).

ACKNOWLEDGEMENTS

I thank C. Handforth for the illustrations, J.P. Rourke

for checking the Latin diagnoses, J.P. Roux for the

Afrikaans translation of the abstract, and two reviewers

for their helpful comments.

REFERENCES

STEINER, K. E. & WHITEHEAD, V. B. 1988. The association between

oil-producing flowers and oil-collecting bees in the Drakensberg

of southern Africa. In P. Goldblatt & P. P. Lowry, Modern

systematic studies in African botany. Monographs in Systematic

Botany 25: 259-277.

STEINER, K. E. & WHITEHEAD, V. B. 1990. Pollinator adaptation

to oil-secreting flowers — Rediviva and Diascia. Evolution 44:

1701-1707.

STEINER, K. E. & WHITEHEAD, V. B. 1991. Oil flowers and oil bees:

further evidence for pollinator adaptation. Evolution 45: 1493 —

1501.

VOGEL, S. 1974. Olblumen und olsammelnde Bienen. Tropische und

subtropische Pflanzenwelt 7: 283—547.

WHITEHEAD, V. B. & STEINER, K. E. 1985. Oil-collecting bees in

South Africa. African Wildlife 39: 144—147.

Bothalia 22,1: 19-22 (1992)

Studies in the genus Riccia (Marchantiales) from southern Africa. 24.

R. moenkemeyeri , subgenus Ricciella: new records

S. M. PEROLD*

Keywords: Marchantiales, Riccia moenkemeyeri , southern Africa, subgenus Ricciella

ABSTRACT

Riccia moenkemeyeri was twice described by Stephani (1887, 1891), the second time as R. abnormis. Amell (1952) described

it as R. undulata. It is clearly a plastic species (Jones 1957) and is widely distributed in tropical Africa, from Sierra Leone

(as R. undulata), Nigeria, Cameroon and into the Congo Basin. Until recently, Sim’s specimens from the Matopos in Zimbabwe,

were the most southerly records known, but the species has now also been collected in southern Africa, just east of Pretoria

and at Kransberg, in the western Transvaal.

UITTREKSEL

Riccia moenkemeyeri is twee keer deur Stephani (1887, 1891) beskryf, die tweede keer as R. abnormis. Amell (1952) het

dit as R. undulata beskryf. Dit is duidelik dat dit ’n varierende spesie is (Jones 1957), wydverspreid in tropiese Afrika,

vanaf Sierra Leone (as R. undulata), Nigerie, Kameroen en tot in die Kongo-bekken. Tot onlangs, was Sim se eksemplare

afkomstig van die Matopo-heuwels in Zimbabwe, die mees suidelike rekords bekend, maar die spesie is nou ook in Suider-

Afrika net oos van Pretoria en by Kransberg, in Wes-Transvaal, versamel.

Riccia moenkemeyeri Steph. in Botanische Jahr-

biicher fur Systematik, Pflanzengeschichte und Pflanzen-

geographie 8: 95 (1886 ‘1887’); Steph. : 372 (1898); Jones:

211 (1957); Vanden Berghen: 189 (1972). Type: Niger

Gebiet, Alt Calabar in terra, leg. Moenkemeyer N3,

11-10-1884 (G024384, holo.!; S, iso.)

R. abnormis Steph.: 213 (1891); ibid.: 364 (1898). Type: Kamerun,

Bateki, leg. P Dusen 125, 17 Oct. 1890 (G).

?R. chevalieri Steph.: 116 (1912); ibid.: 1 (1917). Type: Central African

Republic, Haut-Oubangui, Plateau des Ungourras, 650 m, Nov. 1902,

leg. Chevalier, ex Herb. Corbiere.

Ricciella undulata S. Amell: 105 (1952). Type: Africa occidentals.

Sierra Leone, Freetown, stream-side above Calabar Point, S. Amell 2252

(S).

Thallus monoicous, annual, in crowded and often over-

lapping, gregarious patches; glaucous green, turning white

over older parts and along undulating margins, sometimes

with purple-red band on inner side; medium-sized to fairly

large; branches once to several times furcate, closely to

moderately divergent (Figures 1A; 2A), up to 9— 10(— 12)

mm long, segments 2.0— 5.0 x 1.6-2. 5 mm, 0.8 mm thick

medianly but thinner toward margins, ± 2—3 times wider

than thick in section (Figure IE), oblong to ovate, apex

rounded to subacute, emarginate, dorsally deeply grooved

distally (Figure 2B), becoming shallowly grooved to flat

or concave proximally; thallus margins rapidly thinning,

acute, winged and attenuate, ultimately consisting of a

single row of echlorophyllose cells; flanks green to purple-

red below, and rather steep, then abruptly sloping obli-

quely upward and outward (Figure IE), becoming white

toward margin; ventrally rounded, green, sometimes api-

cally with 1 or 2 rows of vestigial red scales (Figure IB);

when dry (Figure 1C) concave dorsally, margins apically

incurved to inflexed or somewhat recurved.

* National Botanical Institute, Private Bag X101, Pretoria 0001.

MS. received: 1991-09-06.

Anatomy: dorsal epidermis chlorophyllose, forming

numerous small, slightly domed to flat areas, generally

enlarging toward margins and proximally, sometimes

rupturing and partly exposing the air chambers below, cells

polygonal, 42-55 x 20-37 /an, surrounding a central,

4- or 5-sided air pore (Figure ID; 2D), mostly only ±

12.5 pm wide toward apex, pores enclosed by smaller com-

panion cells, + 15 x 10 jttm; assimilation tissue 350—400

pm thick, nearly '/2 the thickness of thallus, air chambers

about 24 across width of thallus, centrally narrow and

vertical, 50-60 pm wide, somewhat wider laterally and

sloping obliquely, uniseriate, but in transverse section

(Figure IE) often appearing to be secondarily partitioned

due to forward or lateral inclination, enclosing cellular

unistratose plates, cells irregular in shape and size, 32—87

x 25—37 pm (Figure IF); storage tissue 400—450 am

thick, slightly more than the thickness of thallus, cells

averaging 50 pm in width, containing angular, closely

packed starch granules, but with small spaces wedged in

between; rhizoids smooth or tuberculate, ± 15 pm wide.

Scales mostly quite firmly attached to flanks and difficult

to detach, dark wine-red and shiny or hyaline, not

extending to thallus margins (Figure 2C), spaced, cells

polygonal, 50—75 x 30—45 pm.

Antheridia in a row along groove (Figure 1A), hyaline

necks emerging from small depressions, 200— 295(— 375)

am long. Archegonia median, deeply imbedded, obliquely

orientated, necks sloping toward the apex of the thallus,

up to ± 300 pm long, upper part hyaline and basally

purple, difficult to detect from above. Sporangia oblique

and protruding ventrally (Figure 1G), single or 2 adjacent

or serially arranged, subspherical, ± 500 pm wide, con-

taining 145-190 spores each. Spores (65 — )68 — 75( — 85)

am in diameter, triangular-globular; polar, light tan to

yellowish brown, semitransparent; wing ± 5 am wide,

slightly wider at generally perforated marginal angles,

margin finely crenulate; ornamentation reticulate, but com-

pletely dissimilar on 2 spore faces: distal face (Figure 2E)

20

Bothalia 22,1 (1992)

with (8)9 or 10 areolae across diameter, 8 — 10( — 12) pm

wide, occasionally incompletely separated, walls low,

covered with fine granules and slightly raised into papillae

at nodes; proximal face lacking a triradiate mark, but each

of 3 facets with up to ± 100 tiny, mostly less than 2.5 /xm

wide, shallow, but clearly defined areolae, the walls form-

ing a fine network (Figure 2F). Chromosome number.

n = 9 (Figure 1H) (Bornefeld on S.M. Perold 2603 pers.

comm.); n = 8 (Jovet-Ast 1969).

Riccia moenkemeyeri is a tropical African species,

known from Sierra Leone (as R. undulata ), Ghana,

Nigeria, Cameroon, the Congo Basin, (Region du Lac

Moero (Vanden Berghen 1972)), Angola, Zimbabwe,

Malawi, and now also with outliers into the Transvaal,

southern Africa (Figure 3). Its presence on Fernando Po,

as reported by Stephani (1887) for the type specimen,

Moenkemeyer 3, must have been a mistake, as the locality

on the label states that it is from Calabar, Niger, and in

Species hepaticarum (Stephani 1898) it also does so.

The species grows in damp places, on rich loamy soil,

mostly near streams and in association with other Riccia

species (in southern Africa) such as R. stricta (Lindenb.)

Perold, R. atropurpurea Sim and with Exormotheca

pustulosa Mitt.

Riccia moenkemeyeri is characterized by a more or less

persistent dorsal epidermis which is marked out into small

areolae, each with a small central air pore; by numerous,

narrow air chambers, appearing to be in more than one

layer; by undulating thallus margins which terminate in

a single row of hyaline cells and by a highly distinctive

spore ornamentation with 8—10 large areolae on the distal

face and numerous tiny areolae on the proximal face which

lacks a triradiate mark.

According to the classification used in previous papers

in this series, R. moenkemeyeri is placed in subgenus

Ricciella, section Spongodes, on account of the presence

of air chambers in the assimilation tissue. Although not

truly growing in rosettes, it would be more properly placed

in the informal group ‘Crystallina’, together with R. crys-

tallina L. emend. Raddi, R. cavernosa Hoffm. emend.

Raddi and R. cupulifera A.V. Duthie, than in group

‘Vesiculosa’ with R. bullosa Link ex Lindenb. , R. garsidei

Sim, R. volkii S. Arnell and R. rubricollis Garside &

Duthie ex Perold, which mostly have rather ‘swollen’ thalli

with large, inflated air chambers. The oblique orientation

of the ventrally protruding sporangia is a character which

it apparently shares only with R. stricta (Lindenb.) Perold,

but the latter species has long, narrow, ribbon-like

branches and is placed in subgenus Ricciella, section

Ricciella.

The specimen, S.M. Perold 2603, collected in March

1990, appeared to consist of male plants only, but on serial

and longitudinal sections of several branches, it was found

to also have young, deeply imbedded archegonia with long

necks that are, however, not visible from above. Mature

antheridia have necks up to 375 jam long, but in young an-

theridia they are considerably shorter. Dr E.W. Jones

FIGURE 1 . — Riccia moenkemeyeri. Morphology and anatomy. A— C,

thallus: A, dorsal face, turgid, with rows of antheridial necks;

B, ventral face; C, dry. D, air pore (crosshatched) dorsal, sub-

dorsal cells (stippled lines) enclosing air chamber. E, t.s. of

branch; F, t.s. of part of thallus showing air chambers; G, l.s.

of sporangium with forward sloping neck; H, chromosomes.

A-H, S.M. Perold 2603. Drawings by G. Condy; karyotype by

T. Bornefeld. Scale bars on A— C, F, G = 1 mm; D, F = 50

/im; H = 1 pm.

Bothalia 22,1 (1992)

21

FIGURE 2. — Riccia moenkemeyeri. Morphology and spores. A, thallus; B, apex with groove and scales; C, apical scales seen from the side;

D, air pore; E, distal face of spore; F, proximal face of spore. A— D, S.M. Perold 2603; E, F, T.R. Sim 9072. Scale bars on A— C =

1 mm; D— F = 50 /mi. SEM micrographs by S.M. Perold.

kindly examined part of my collection and commented on

the antheridial necks being shorter than 100 gm. He also

found (pers. comm.) that West African specimens of R.

moenkemeyeri are only exceptionally without spores, the

branches are longer and less divaricate and the epidermis

is more persistent. The Condy 22 and 23 collections were

gathered in April 1991, (slightly later in the following

season than my specimen) and both had sporangia with

mature spores. Seventeen months after collection, a sample

of S. M. Perold 2603 was kept damp for a few days in a

Tupperware dish and it soon resumed growth.

Jones (1957) reported R. moenkemeyeri to be a very

plastic species, a wide range of forms occurring in a single

site. He regarded R. chevalieri Steph. as closely resem-

bling R. moenkemeyeri in vegetative features; the type

specimen, Chevalier 88, however, only had ‘a few male

inflorescences but no female’, and its identification could

thus not be confirmed by spore ornamentation.

Riccia undulata S. Amell was placed in synonymy under

R. moenkemeyeri by Jones (Jones & Harrington 1983),

although Arnell (1952) reported the fronds to be up to 7

mm wide, which is much wider than the measurements

(2.0— 2.5 mm) given by Jones (Jones & Harrington 1983)

for the type specimen. The spores of the two species are

identical, however.

Jovet-Ast (1975) reported on spore germination and

development of the protonema in R. moenkemeyeri, con-

cluding that the various stages (quadrant, plate and column

formation) were similar to those in R. cavernosa.

The reason for the discrepancy in the chromosome

counts of R. moenkemeyeri as reported by Bomefeld

(pers. comm.) and Joyet-Ast (1969), has not been ascer-

tained. Jovet-Ast maintains that in Riccia, n always equals

8 or multiples of 8. Jovet-Ast (pers. comm.) expressed

surprise that Bomefeld’s (1989) counts do not agree with

hers and suggests that Bornefeld’s counts be verified.

Bomefeld (1984) postulates that the different chromosomes

of the basic set in Riccia, which he identifies as A, BB,

CC, DD and E, can multiply heterogeneously (for which

he has coined the term ‘nothopolyploidy’); this would

explain how aberrant numbers could arise.

SPECIMENS EXAMINED

TRANSVAAL.— 2528 (Pretoria): 18 km NE of Cullinan, north of Little

Eden Resort, nr ‘Die Grotte’, Malanspruit, next to river path, on soil

below overhanging rock, (—DA), S.M. Perold 2603 (PRE). 2427

(Thabazimbi): Kransbeig, at seasonally flowing stream, south-lacing bank

under rock, (— BC), Condy 22 (PRE); Kransberg, near rondavel, below

stream crossing, at water’s edge among ferns, (-BC), Condy 23 (PRE).

ZIMBABWE. — [previously misidentified and reported as R. albomar-

ginata (Best 1990)] 2028 (Bulawayo): stream at Bulawayo, (-?AA), T.R.

Sim 9069 ( PRE-CH1015 ) (PRE); Matopos, (— CA), T.R. Sim 9068

FIGURE 3. — Map showing distribution of R. moenkemeyeri in southern

Africa.

22

Bothalia 22,1 (1992)

(CH 1014) (BOL, PRE). 2030 (Masvingo): (-BB), T.R. Sim 9070 (PRE-

CH1012), 9072 (PRE-CH1013) (PRE).

MALAWI. — 1434: 31 km E of Lilongwe, on road to Zomba, on left

side of road, beyond legume patch, on damp soil of footpath leading

to flat rocks, (— AA), S.M. Perold 2690 (PRE). 1535: at roadside in

Zomba, damp earth wall nr river, (—AD), S.M. Perold 2655 (PRE).

ANGOLA. — Dist. Pungo Andongo, hab. gregaria ad cavemas rup.

editiorum in Pedra de Cazella ipsius Praesidii, Welwitsch 229 (as R.

abnormis ) (BM); Dist. Golungo Alto, habit ad rupes limosas rivuli,

Carenghe in Alto Queta, Welwitsch 309a (BM); Dist. Golungo Alto, ad

terrain humidam juxta Rivuli de Quarengue in Queta, Welwitsch 309b

(BM).

BELGIAN CONGO. — Leonard 11894, U895 (BR).

CAMEROON.— Cap Debunsch, J.S. Jungner 1891 (NY); Bipinde,

Urwaldgebiet, G. Zenker 2431h (as R. abnormis) (BM, E); P. Dusen

s.n. (as R. abnormis ) (BM).

NIGERIA.— Ibadan University Bot. Garden, on moist sand under large

trees near the river, E.W. Jones 1187-, Calabar, sandy roadside ditch near

the harbour, E.W. Jones 209; Sanga River Forest Res., Kurmi Kadar,

on very wet heavy red loam on bank of small stream, still flowing at

end of dry season, E.W. Jones 927; Abuja, on earth of rocky bank in

shade of trees by the Rest House, E.W. Jones 893 (all at Herb. Jones).

ACKNOWLEDGEMENTS

I sincerely thank Dr E.W. Jones for his examination of,

and comments on, the specimens I sent him, as well as

the loan of his own collections. I also wish to express my

gratitude to the Curators of BM, BOL, BR, G and NY

for the loan of specimens. Sincere thanks to Dr T.

Bornefeld, Wurzburg, for the chromosome count and the

karyotype; the resident artist at PRE, Ms G. Condy for

the illustrations and for collecting specimens of R.

moenkemeyeri; to Mrs J. Mulvenna for typing the text and

to Mrs A. Romanowski for developing and printing

photographs.

REFERENCES

ARNELL, S. 1952. Hepaticae collected in South and West Africa (1951).

New and little known species. Botaniska Notiser 105 : 307—315.

BEST, E.B. 1990. The Bryophyta of Zimbabwe — an annotated check-

list. Kirkia 13: 293-318.

BORNEFELD, T. 1984. Chromosomenanalyse der Gattung Riccia L.

aus S- und SW-Afrika und allgemeine Bemerkungen zur Zytogene-

tik der Lebermoose. Nova Hedwigia 40: 313-328.

BORNEFELD, T. 1989. The Riccia of S and SW Africa. Chromosome

numbers and composition of the chromosome sets. Nova Hedwigia

48 : 371-382.

GEISSLER, P. & BISCHLER, H. (eds). 1990. Index hepaticarum.

Racemigemma to Zoopsis. Vol. 12. Cramer, Berlin, Stuttgart.

JONES, E.W. 1957. African hepatics XIII. The Ricciaceae in tropical

Africa. Transactions of the British Bryological Society 3:

208-227.

JONES, E.W. & HARRINGTON, A.J. 1983. The hepatics of Sierra Leone

and Ghana. Bulletin of the British Museum (Natural History),

(Botany) 11: 215—289.

JOVET-AST, S. 1969. Le caryotype des Ricciaceae. Revue bryologique

et lichenologique 36: 673—689.

JOVET-AST, S. 1975. Germination et phase protonemique chez quelques

especes du genre Riccia. Revue bryologique et lichenologique

41: 263 —276.

STEPHANI, F. 1887. Hepaticae africanae. Botanische Jahrbiicher fur

Systematik, fflanzengeschichte und Pflanzengeographie 8: 79—96.

STEPHANI, F. 1891. Hepaticae africanae. Hedwigia 30: 201—217.

STEPHANI, F. 1898. Species hepaticarum. Bulletin de l 'Herbier Boissier

6: 309-411.

VANDEN BERGHEN, C. 1972. Hepatiques et Anthocerotees. . Resul-

tats scientifiques de l 'exploration hydrobiologique du Bassin Lac

Bangweolo & Luapula: 8: 1—202.

Bothalia 22,1: 23-35 (1992)

Aspidonepsis (Asclepiadaceae), a new southern African genus

A. NICHOLAS* and D.J. GOYDER**

Keywords: Asclepias, Asclepiadaceae, Aspidonepsis , new genus, new species, southern Africa, taxonomy, Unguilobium

ABSTRACT

Aspidonepsis, an endemic southern African genus, is described and compared to the closely allied genus Aspidoglossum.

This newly described genus is composed of two subgenera, Aspidonepsis and Unguilobium, consisting of three and two

species respectively. Asclepias diploglossa, A. flava, A. cognata and A. reneensis are transferred to Aspidonepsis, and A.

shebae is newly described. All species are discussed, illustrated and a key is given to aid in their identification.

UITTREKSEL

Aspidonepsis, ’n genus endemies in suidelike Afrika, word beskryf en met die naverwante genus Aspidoglossum vergelyk.

Die nuut beskrewe genus bestaan uit twee subgenusse Aspidonepsis en Unguilobium, met drie en twee spesies onderskeidelik.

Asclepias diploglossa, A. flava, A. cognata en A. reneensis word na Aspidonepsis oorgeplaas, terwyl A. shebae nuut beskryf

word. A1 die spesies word bespreek, geillustreer en 'n sleutel om te help met hul identifikasie, word gegee.

INTRODUCTION

A. A. Bullock’s work on the family Asclepiadaceae (1952

to 1967) has received wide acceptance in Africa north of

the Limpopo River. In southern Africa, however, his

generic concepts and names have seldom been applied.

This is explained partly by the feet that his research seldom

included southern African plants and partly by the

rejection of his work by Dyer (1975).

Unfortunately, three elements detract from Bullock’s

work: 1, he admitted that his delimitation of genera was

only tentative (1952); 2, when resurrecting or expanding

existing genera he seldom gave new descriptions for these

taxa. As a result, the generic circumscriptions and exact

application of some of these names is still unclear; 3, his

species concepts were often very broad and there is now

growing consensus that some species will need to be

re-split.

Most southern African herbaria therefore still follow

N.E. Brown’s treatment of the Asclepiadaceae as outlined

in the Flora capensis (1907—1909). However, workers like

N.E. Brown had followed the tradition of their time and

separated genera using floral differences only. They even

separated some genera on the basis of a single character.

Phenomena like convergent evolution were seldom taken

into account, and workers were unaware that the evolu-

tion of analogous floral morphologies had taken place

within the family. Bullock (1952) was the first to realize

that such convergent evolution had taken place and that

many genera in the family not only contained a number

of unrelated entities, but that these entities could only be

identified in terms of consistently produced correlated

character combinations. He was the first taxonomist to

attempt a phylogenetically based classification for the

African members of the tribe Asclepiadeae.

* National Botanical Institute, Private Bag X101, Pretoria 0001.

Formerly: South African Liaison Botanist, Royal Botanic Gardens, Kew,

Richmond, Surrey TW9 3AE, England, UK.

** The Herbarium, Royal Botanic Gardens, Kew, Richmond, Surrey TW9

3AE, England, UK.

MS. received: 1990-10-24.

What Bullock has done at the generic level, N.E. Brown

has accomplished at the specific level. Consequently the

work of N.E. Brown (species delimitation) and Bullock

(generic delimitation) should be seen as complimentary

rather than antagonistic.

Recent investigations concerning the southern African

members of the genus Asclepias sensu N.E. Brown have

shown that Bullock’s generic concepts should be redefined

and extended to embrace the taxa of this subcontinent

(Nicholas 1981). Bearing in mind that the type species of

the genus Asclepias L. is A. syriaca L. , the authors agree

with Bullock in the exclusion of Asclepias from Africa

except as an adventive. The process of moving the southern

African taxa of Asclepias sensu N.E. Brown to their

correct generic position has already begun (Nicholas &

Goyder 1990). The authors understand the desirability of

giving a brief generic synopsis of the subtribe Ascle-

piadineae in Africa at this early stage of their work.

However, as a number of genera still need to be: 1,

resurrected from synonomy; 2, newly described; 3,

extensively redefined; they feel that it is at present unwise

to publish information that may change as their research

progresses.

Aspidonepsis diploglossa (Turcz.) A. Nicholas & D.J.

Goyder, A. flava (N.E. Br.) A. Nicholas & D.J. Goyder,

A. cognata (N.E. Br.) A. Nicholas & D.J. Goyder,

A. reenensis (N.E. Br.) A. Nicholas & D.J. Goyder and

A. shebae A. Nicholas & D.J. Goyder form a phylogenetic

unit quite distinct from the rest of Asclepias sensu N.E.

Brown and can be distinguished from other genera in the

tribe Asclepiadeae by the following set of consistently

present correlated characteristics:

1, a globose, fusiform or napiform tuber just below the

soil surface;

2, a single erect stem (rarely up to 3 in A. flava);

3, spreading to ascending linear to narrowly elliptic leaves

which are ranked up the stem;

4, inflorescences gathered together at the top of the flower-

ing stem, even if nodally produced;

24

Bothalia 22,1 (1992)

5, persistent inflorescence bracts, often grading with the

leaf system;

6, cucullate corona lobes which are produced 0.5 to 1.8

mm above the insertion of the corolla;

7, a saccate corona lobe cavity;

8, wishbone-shaped pollinaria, with semicircular to hemi-

ovoid pollinia.

Aspidonepsis is confined to high altitude grasslands of

the Drakensberg and its foothills, although outlying popu-

lations of some species may be found in mountainous

situations as far south as Grahamstown and on montane

‘islands’ nearer the Transkei-Natal coast. The northern

limit of distribution of this endemic southern African genus

is the eastern Transvaal. Species are usually, but not

always, found in situations subject to annual burning. Popu-

lations are intermittent in the wild and usually consist of

few widely dispersed individuals. Occasionally up to three

tubers are produced in a connected series, possibly

representing subsequent year’s growths.

Aspidonepsis bears a number of similarities to the genus

Aspidoglossum (Table 1), and it is the authors’ opinion that

the two genera may have originated from the same distant,

ancestral stock. However, if this is the case, then the two

taxa have since evolved along very different lines, for a

number of major disjunctions in morphology now exist,

such as the aggregation of inflorescences near the stem

apex and the central cavities in the corona lobes of

Aspidonepsis. In contrast Aspidoglossum bears inflores-

cences that are produced along the length of the stem and

there is no corona lobe cavity.

The affinity of these two genera can be clearly seen in

the corona lobe and pollinarium morphology of

Aspidoglossum delagoense (Schltr.) Kupicha, which is

very similar to Aspidonepsis (Figure 1). However, all other

features of this species place it clearly within Aspidoglos-

sum, of which A. biflorum E. Mey. is not only the type

species but also typical of the genus as a whole (Kupicha

1984). Aspidoglossum has more species and is morpho-

logically more diverse than Aspidonepsis.

The five species recognized in Aspidonepsis fell into two

well-defined groups that require recognition at subgeneric

level. The first group is characterised by spreading or

ascending corolla lobes and cup- or dish-shaped corona

lobes with a tooth-like appendage projecting from the floor

of the corona lobe cavity. The second group has reflexed

corolla lobes and corona lobes with a more angled outer

margin and no tooth-like structure projecting from the

floor of the corona lobe cavity.

A total of 187 pressed specimens were examined during

the course of this study from the following herbaria: BOL,

CPF, GRA, J, K, NBG, NH, NU, PRE, SAM and TCD*.

Additional data were obtained from spirit collections and

supplemented by observations in the field.

* Herbarium abbreviations are taken from Holmgren et al. (1990).

TAXONOMY

Aspidonepsis A. Nicholas & D.J. Goyder, gen. nov.,

Aspidoglosso affinis sed sinu coronae lobis prominenti et

appendice distali coronae lobis non filiformi nec ornata

differt.

Herba perennis. Caudex: tuber globosum, fusiforme vel

napiforme. Caulis unicus (raro duo vel tres), erectus,

gracilis, usque 625 mm tantum longus. Folia expansa,

anguste elliptica vel linearia in subgenere Aspidonepse,

sed ascendentia, linearia vel nonnunquam lanceolata,

margine manifeste revoluta in subgenere Unguilobio.

Inflorescentia umbellata, terminalis subterminalisve vel ad

nodos disposita, 2—17-flora (in subgenere Aspidonepse),

4—11-flora (in subgenere Unguilobio)', bracteae ad anthesin

persistentes. Coronae lobi partibus inferioribus ad colum-

nam staminalem connatis; 0.5— 1.8 mm supra corollam

producti, cucullati; sinus profundus appendice linguiformi

centrali ornatus in subgenere Aspidonepse. Appendix

proximalis ad apicem deltato-falcata et apicem stylii

aequans vel superans impendensque; extremum distale

coronae appendice parva ornatum vel appendice carente;

sinus profundus rimiformis in subgenere Unguilobio.

Appendix antherae reniformis vel pescapriformis profunde

apicaliter fissa.

TYPUS. — Aspidonepsis diploglossa (Turcz.) A. Nicho-

las & D.J. Goyder, vide infra.

Perennial geophytic herb. Rootstock a globose, fusiform

or napiform tuber. Stems 1 (rarely as many as three

in A. flava only), erect, never more than 650 mm tall.

Leaves spreading to ascending, linear, lanceolate to nar-

rowly elliptic, older leaves shorter and broader; petiole

TABLE 1.— A comparison of Aspidonepsis and Aspidoglossum.

important differences between the two taxa.

Bothalia 22,1 (1992)

25

FIGURE 1 . — Comparison of 1, Aspidonepsis flava, Coleman 813, (PRE); 2, Aspidoglossum delagoense, Barbosa & De Lemos 7958, (K); 3,

Aspidoglossum biflorum, Bolus 238, (K). A, flower with part of corolla removed: LA, x 11; 2A, X 9.5; 3A, x 7. B, corona lobe, side

view: IB, x 28; 2B, X 15; 3B, x 11. C, pollinarium: 1C, X 48; 2C, X 22; 3C, x 55.

0—5 mm long. Inflorescences umbelliform; terminal, sub-

terminal and nodal; bracts present at anthesis. Corolla

catilliform or reflexed with lobe apices ascending. Coro-

na with lower parts fused to the staminal column; lobes

produced 0.5 — 1.8 mm above the corolla, cucullate; cavity

saccate with appendage (subgenus Aspidonepsis) or

without appendage (subgenus Unguilobium) . Anther ap-

pendages reniform to pescapriform, with a deep apical

cleft, or rectangular. Style: head swollen; apex truncated.

Pollinaria wishbone-shaped; corpusculum fusiform;

26

Bothalia 22,1 (1992)

translator apparatus articulated and winged; pollinia

semicircular to hemiovoid or clavate. Habitat : high alti-

tude mountain grasslands. Distribution : southern African

Drakensberg. Etymology : Aspidonepsis — Aspidoglossum’s

cousin. Aspid(os), (Greek for shield) but used here to

indicate the genus Aspidoglossum, and anepsia (Greek for

cousin).

Key to subgenera and species

la Corolla not fully reflexed when mature. Corona lobe cavity

with a tongue-like appendage (Subgenus Aspidonepsis:

2a Proximal corona lobe appendages projecting over the style

apex A. cognata

2b Proximal corona lobe appendages not projecting over the

style apex:

3a Corona lobes with arm-like proximal appendages that cross

over each other and reflex back into the coronal cavity.

These appendages are below the style apex and level with

the anther wings A. flava

3b Corona lobes without true proximal appendages, instead,

the proximal ends are produced into dentate or obtuse

(but not protruding) shoulders that are level with the style

apex A. diploglossa

lb Corolla reflexed when mature. Corona lobe cavity without

an appendage (Subgenus Unguilobium):

4a Distal corona lobe appendage present (if somewhat short).

Transkei, Natal and southern Transvaal bordering Natal

A. reenensis

4b Distal corona lobe appendage absent. Eastern Transvaal

only A. shebae

ENUMERATION OF THE SUBGENERA AND SPECIES

A. Subgenus Aspidonepsis

Inflorescences 2— 17-flowered. Flowers yellow, green,

brown and purple or these in combination. Corolla catil-

liform with lobe apices curving upwards or spreading,

lobes with abaxial surface glabrous. Corona lobes : upper

proximal margin various, distal margin obtusely rounded

or truncate and raised above the proximal appendages

(except A. diploglossa)-, cavity saccate with a centrally

produced laterally flattened tongue-shaped or botuliform

appendage. Anther appendages reniform or pescapriform

with a deep apical cleft. Pollinia narrowing proximally;

translator arms in two distinct parts, winged (Table 2).

This subgenus is composed of three species: Aspidonep-

sis diploglossa, A. flava and A. cognata. For a number

of years these three species were considered conspecific,

and lumped together under the oldest name, viz. A.

diploglossa. However, although all three species are

vegetatively similar, close examination shows that they are

distinct entities with very different floral morphologies

(Nicholas 1987). They are usually found in annually burnt

or grazed, high to medium altitude, montane grasslands.

They are found along the Natal-Transkei Drakensberg, and

on scattered island mountain ranges in the eastern Cape

and Natal midlands. The flowers of this subgenus are

predominantly yellow or yellow-green, although occa-

sionally flowers with brown or purple markings can be

found.

1. Aspidonepsis diploglossa (Turcz.) A. Nicholas &

D.J. Goyder, comb. nov. Type: South Africa, Cape

Province, peaks of the Winterberg, Ecklon 23 (KW holo.,

photo!; PRE!, iso.).

Gomphocarpus diploglossus Turcz. : 258 (1848). Asclepias diploglos-

sa (Turcz.) Druce: 605 (1917).

Asdepias schizoglossoides Schltr.: 32 (1894); Schltr.: 451 (1896); N.E.

Br.: 688 (1908); Wood: 461 (1910); Phillips: 194 (1917). Type: South

Africa, eastern Cape, Mrs Barber s.n. (K! , neo., here designated).

Rootstock 1 or several tubers connected in series, 9-35

x 6—12 mm. Stems 1, erect, 170— 400(— 500) mm tall,

bifariously pubescent. Leaves ascending to spreading,

narrowly lanceolate, occasionally falcate, rarely linear or

narrowly elliptic, 5-84(-130) x (0.25— )0. 5 — 7.0 mm; apex

acuminate or occasionally acute; base petiolate to cune-

ate; apetiolate or petiole up to 4 mm long. Inflorescences

1—3 per plant, 4— 16-flowered, bracts present at anthesis;

peduncles up to 9.5 mm long or occasionally inflores-

cences apedunculate. Flowers 4-9 x 6—13 mm, yellow

TABLE 2. — A comparison of the two subgenera Aspidonepsis and Unguilobium. All measurements in mm

* characters forming discontinuities between the two taxa.

Bothalia 22,1 (1992)

27

or yellow-brown; pedicel 6—16 mm long. Calyx : lobes

lanceolate, occasionally triangular or narrowly ovate,

2. 5-4.6 x 1.0— 1.5 mm, apex acuminate, pubescent to

tomentose. Corolla : lobes ovate or occasionally elliptic,

free to the base, 4 — 6( — 7) x 2.4— 4.1 mm; inside yellow,

occasionally tinted with purple or lilac, outside yellow,

brown or purple, these often in combination; abaxial

surface with a few sericeous hairs. Corona lobes produced

+ 0.5 mm above corolla, cucullate-cyathiform, 4— 6(— 7)

x 2.4— 4.1 mm, upper proximal ends forming 2 rounded

shoulders, occasionally extended into short pointed

appendages, level with or projecting (slightly) onto style

apex, distal end obtuse or rounded without a distinct

appendage and level with or lower than style apex, sac-

cate cavity with a tongue-like or deltoid-oblong append-

age 0.2— 0.8 mm wide, projecting 0.2— 0.7 mm above

upper lobe margin, colour yellow to bright yellow. Stami-

nal column 2.0— 2.8 mm long; anther wings shallowly

concave in upper two thirds, rounded at base, 0.75—1.1 x

0.3— 0.5 mm; anther appendages pescapriform or ovate

with a deep apical cleft, membranous, 0.3— 0.6 x 0.6— 0.9

mm, decumbent on style apex. Style apex truncated, with

thickened undulating margins, concave in centre, 1.1— 2.1

mm diameter, bright green to white. Pollinaria: corpus-

culum (0.22— )0. 2 8— 0.32 x 0.08—0.16 mm; translator

arms 0.2— 0.32(— 0.36) mm long, thin with small trans-

parent hook-like wings, pollinia clavate, 0.68— 0.80(— 0.84)

x 0.24—0.36 mm. Fruits and seeds not seen. Specific

epithet etymology, from the Greek words diplo- (two)

and glosso- (tongue); probably in reference to the corona

lobe and the appendage in its central cavity. (Figure 2.1).

Aspidonepsis diploglossa is found in annually burnt

montane grasslands, normally on south- or east-facing hill-

side slopes or mountain plateaux. Usually, but not always,

occurring in wettish areas. Collectors often report it as

rare, although a great many collections exist. It is usually

found growing at altitudes ranging from 1 500 to 2 400

m, but occasionally also at lower altitudes. Plants flower

from October to January. The tubers of this plant lie just

below the soil surface, and when sectioned reveal white,

woody flesh that oozes sticky, milky latex.

A. diploglossa , a mountain-loving species, exhibits a

rather strange distribution. It may be found at high altitudes

around Grahamstown and Hogsback in the eastern Cape,

then there is a gap in the Transkei Drakensberg (which

may be an artifact caused by poor collection in this area)

and then it occurs abundantly along the Natal Drakens-

berg and its foothills as far as Van Reenen’s Pass. After

yet another gap it is found again in the Wakkerstroom area.

A. diploglossa may also inhabit mountain islands in the

Natal midlands at places such as Inanda, Greytown and

Weenen. However, it occurs in the most unlikely place near

the southern Natal coast at the Umtamvuna Nature

Reserve, where it grows at an altitude of only 350 m. This

nature reserve is well known scientifically because it lies

within the narrow belt of Natal Group sandstone in the

coastal region between Port Shepstone and Port St Johns.

Its rich flora includes a number of rare plants and en-

demic species. However, the occurrence of Aspidonepsis

diploglossa at such a low altitude and so near the sea, is

surprising and inexplicable (Figure 3).

Unfortunately, when R. Schlechter described Asclepias

schizoglossoides in 1894 he not only failed to cite the

specimens he examined, but was also unaware that he was

dealing with an already described taxon. Turczaninow had

named this species Gomphocarpus diploglossus in 1848,

citing Ecklon 23 as the type. N.E. Brown picked up

these two errors when preparing the Asclepiadaceae for

Flora capensis, and in correspondence with Schlechter

discovered that the latter taxonomist had based the

name Asclepias schizoglossoides on a Barber specimen

‘probably collected in British Kaffraria’. As a result, N.E.

Brown (1908) suspected that the specimen may be part of

Mrs Barber’s gathering numbered 35. N.E. Brown’s

selection of Barber 35 as the type of the name Asclepias

schizoglossoides for Flora capensis was probably correct.

However, due to the destruction of Schlechter’s ascle-

piadaceous collections housed at Berlin herbarium during

the Second World War, we cannot confirm this. In this

paper we have, therefore, chosen Barber 35 (K) as the

neotype of the name Asclepias schizoglossoides .

W.H. Harvey has written (in pencil) on two Trinity

College Dublin herbarium (TCD) sheets of this species,

the name Gomphocarpus luteus (var.) (3 heterophyllus.

This name was never validly published, and must be con-

sidered nothing more than a manuscript name.

Aspidonepsis diploglossa differs from A. flava and A.

cognata in possessing longer (occasionally narrower)

leaves, a deeply cleft anther appendage, yellow to yellow-

brown flowers and a simple cup-shaped corona lobe, the

upper proximal ends of which are no more than blunt

rounded shoulders level with the style apex. See Table 3.

NATAL. — 2730 (Vryheid): Altemooi, (—AD), Thode A1173 (NH,

PRE). 2731 (Louwsburg): near Ngome, (-CD), Schrire 7037 (NH). 2828

(Bethlehem): Royal Natal National Park, (-DB), Trauseld 122 (PRE);

Mont Aux Sources, ( — DD), Schweickerdt 779 (PRE). 2829 (Harrismith):

Van Reenen, (-AD), Jacobsz 1656 (PRE); Klawervlei, (— CA), Blom

287 (PRE); Cathedral Peak State Forest, (-CC), Killick 1016 (CPF, PRE).

2830 (Dundee): Weenen, (— CC), Rogers 28436 (K). 2929 (Underberg):

Giant’s Castle, (-AB), Stewart 2070 (K, NU); Tabamhlope Mountain,

(—BA), West 1383 (NH, PRE); Highmoor State Forest, (— BC), Killick

& Vahrmeijer 3583 (K, NH, PRE); Restmount area, (— CB), Hilliard

& Burn 15557 ( K); Bushman’s Nek area, (— CC), Hilliard & Burtt 17436

(K, PRE); Garden Castle Nature Reserve, (—CD), Hilliard & Burtt 7866

(K, NU); Runnymeade, (-DB), Moll 1480 (NU); near Maiwaga, (—DC),

Rennie 235 (NU); Glengariff, (— DD), Rennie 488 (NU). 2930 (Pieter-

maritzburg): near Pietermaritzburg, (—AC), Ram s.n. (NU); Caversham,

(—AD), Mogg 2471 (PRE); Greytown, (—BA), Wylie s.n. (K, NH 21644,

PRE ex Transvaal Museum 34205); Dargle, (— CA), Fannin 39 (K, TCD);

near Richmond, (—CD), Wood 10819 (NH); Inanda, (— DB), Groom s.n.

(K ex Wood 1408, NH 4106).

TRANSKEI. — 3028 (Matatiele): near Ramatseliso, (— BB), Board-

man All (PRE). 3029 (Kokstad): Ensikeni, (—BA), Haygarth s.n. (NH

ex Wood 12049). 3130 (Port Edward): Umtamvuna Nature Reserve,

(— AA), Abbott 2868 (NH).

CAPE.— 3227 (Stutterheim): near Fort Cunynghame, (—AD), Sim s.n.

(BOL); Hogsback, (— CA), Rattray s.n. (BOL 15767); Dohne Hill,

(— CB), Sim 1237 ( BOL, NU, PRE, SAM). 3326 (Grahamstown): Cold-

spring, (-AD), Glass 276 (K, PRE, SAM); Howison’s Poort, (—AD),

Hutton s.n. (TCD); Grahamstown, (— BC), MacOwan 850 (K).

WITHOUT PRECISE LOCALITY. —Eastern Cape, Barber 35. s.n.

(K); Cape, (Mrs Barber records it as being collected at the Winterberg,

but its occurrence there is highly improbable. Possibly she meant the

Winterhoek Mountains near Uitenhage or the Klein Winterhoek near the

Zuurberg. where its occurrence is much more likely) Barber 84 ( K, TCD).

2. Aspidonepsis flava (N.E. Br.) A. Nicholas & D.J.

Goyder, comb. nov. Type: Transkei, Malowe Mountain,

Tyson 1086 (K! lecto., here designated; BOL!, SAM!,

isolecto.)

Asclepias flava N.E. Br.: 687 (1908); Wood: 460 (1910).

Bothalia 22,1 (1992)

FIGURE 2. — 1, Aspidonepsis diploglossa ; 2, A. Jlava. A, whole plant with flowers: 1A, X 0.7; 2A, X 0.4. B, flower with part of corolla removed:

IB, x 10; 2B, x 9. C, corona lobe: 1C & 2C1, side view, x 14 & x 27; 2C2, angled view to show crossed, inwardly flexed proximal

appendages, X 30. D, gynostegium excluding corona: ID, X 16; 2D, X 19. E, abaxial surface of anther: IE, X 24; 2E, X 30. F, pollinarium:

IF, X 51; 2F, x 65. 1G, translator apparatus showing winged spur, X 89. 1A, Ruddock 136 (CPF); IB, ID, IE, IF, Boardman All (PRE);

IC, 1G, Boardman 186 (PRE); 2A-2C1, 2D-2F, Coleman 813 (PRE); 2C2, Wood 4249 (NH).

Bothalia 22,1 (1992)

29

TABLE 3. — A comparison of the three species of subgenus Aspidonepsis. All measurements in mm

# Discontinuities between A. diploglossa and the other two taxa; • discontinuities between A. flava and the other two taxa; + discontinuities

between A. cognata and the other two taxa; i discontinuities between A. flava and A. cognata: * discontinuities between all three taxa.

Rootstock a globose tuber, occasionally several connec-

ted in series, 6—10 x 4—9 mm. Stems usually 1, rarely

as many as 3, slender, erect, 180—475 mm long. Leaves

spreading to erect, lanceolate, linear to narrow-elliptic,

7—83 x 0.5— 6.0(— 7.0) mm; apex acuminate or rarely

acute; base petiolate to cuneate. Inflorescences 1 — 3( — 6)

per plant, 1—3 per stem, 4 -18 (—2 4) -flowered; bracts

present at anthesis, 2.6— 5.3(— 7.5) x 0.15—0.5 mm; pedun-

cles (4 — )10 — 175 mm long. Flowers 3— 5(— 6) X 5—8 mm;

pedicel 5-11 mm long. Calyx : lobes lanceolate,

2.0— 3.6(— 4.0) x 0.7— 1.2 mm. Corolla: lobes ovate, oc-

casionally elliptic, free to the base, 3.5 -5.0 x 2. 0-3.2

mm, inside greenish yellow or yellow, outside yellowish

green, pale yellow or yellow with a purple apex, margins

occasionally slightly revolute. Corona lobes produced

0.5— 0.8 mm above corolla, cucullate-crateriform, in side

view boxing glove-shaped, 1.0— 1.6 mm long, upper

proximal ends extending into 2 short (0.25—0.70 mm),

subulate or arm-like appendages that meet and are then

reflexed back to point to distal end of lobe, lower than style

apex, distal end dilated and bowl-shaped with upper

margin overtopping the style apex (even if only slightly);

FIGURE 3. — Distribution of Aspidonepsis diploglossa.

cavity crateriform, 0.4 -0.7 mm deep with a central

sausage-shaped appendage projecting 0.4— 0.7 mm above

cavity margin; orange-yellow, golden yellow, yellow-green

or yellow. Staminal column 1.0— 1.5 mm long; anther wings

shallowly concave in upper two thirds, rounded in lower

third, truncate basally, 0.5— 0.7 x 0.2—0.45 mm; anther

appendages reniform, membranous, (0.2— )0.3— 0.4 (—0.5)

x 0.5— 0.8 mm, decumbent on sides and top of style apex.

Style apex truncate, margin undulate, apex concave with

a small central pore, 1.1— 1.6 mm wide. Pollinaria :

corpusculum 0.16— 0.20(— 0.26) x (0.60— )0.08— 0.10 mm;

translator arms (0.20— )0.18— 0.28 mm long; pollinia

dilated distally, narrowing proximally, (0.48— )0.52— 0.64

(— 0.68) x 0.16— 0.24 mm. Fruits: mature follicles not seen,

immature follicles narrowly fusiform with an attenuate

apex, not echinate. Seeds not seen. Specific epithet

etymology: from the Latin word flav(us) meaning pale yel-

low. This is in reference to the pale yellow flowers of this

species. (Figure 2.2).

Aspidonepsis flava is usually found growing in annually

burnt montane grasslands. Colonies are usually scattered

and occur at altitudes between 600 and 2 000 m, rarely

at altitudes as low as 450 m. Distributed from Grahams-

town in the eastern Cape through Transkei to Natal. This

species is commonly found in the Drakensberg or its

foothills, although it can be found in the midland and

coastal belts if mountainous areas provide it with a suitable

refuge (Figure 4). A. flava flowers in the midsummer

months between November and January, although there

is one record of a plant flowering in October.

This taxon was first described by N.E. Brown in Flora

capensis (1908), and is abundant in southern Natal and

the Transkei interior. The limits of its southern distribu-

tion is near Grahamstown where a few specimens have

been collected. Plants grow in small colonies in annually

burnt grasslands, and usually occur on hillside slopes

amongst scattered rocks where they receive some protec-

tion from grazing animals and fire. Plants may have up

to three tubers connected in series, each probably

representing a previous year’s growth. Like A. diploglossa

30

Bothalia 22,1 (1992)

these globose tubers are found just below the soil surface,

and have white, latex-filled flesh.

A. flava is distinguished from the other two species in

subgenus Aspidonepsis by its longer peduncles, smaller,

paler coloured flowers, smaller pollinaria, bowl-shaped

corona lobes with arm-like proximal appendages that cross

one another and are then reflexed into the corona lobe

cavity and the sausage-shaped appendage projecting from

the floor of the corona lobe cavity. Table 3.

NATAL. — 2929 (Underberg): Cobham State Forest, (— CB), Cowan

124 (NU); near Underberg, (—CD), Dyer 3744 (K, NH); Mawahqua Mtn

area, (—DA), Rennie 275 (NU); Mpendle, (— DB), Huntley 625 (NH);

Nkonzo State Forest, (— DD), Nicholas <£ Norris 1159 (CPF, NH, PRE).

2930 (Pietermaritzburg): Howick, (—AC), Hutton 408 (BM, K, PRE);

Benvie, Karkloof, (—AD), Hilliard & Bum 13491 (NU); Winterskloof,

(— CB), Sim s.n. (PRE); near Byrne Village, (— CC), Stewart 2023 (K,

NU); Weza State Forest, (—DA), Nicholas 2080 (NH); Fort Donald,

(—DC), Tyson 1660 (SAM). 3030 (Port Shepstone): Ixopo, (— AA),

Shirley s.n. (NU).

TRANSKEI. — 3028 (Matatiele): near Eland’s Height, (—CD), Stewart

1908 (NU). 3029 (Kokstad): near Mt Currie, (—AD), Hutchinson 1823

(K), Tyson 1686 (BOL, PRE, SAM); Ensikeni, (—BA), Haygarth s.n.

ex Wood 12049 (NH 18644, SAM); Malowe, (— BD), Tyson 2723 (K,

SAM); Vaal Bank, (— CB), Haygarth s.n. ex Wood 4230 (K, NH). 3127

(Lady Frere): Mount Kwenkwe, (—DA), Bolus 10215 (BOL), Engcobo,

(— DB), Bolus 10216 (BOL). 3128 (Umtata): Mhlahlane, (— BC), Hutch-

ings 1387 (KEI); Bazija, (-CB), Baur 556 (K, SAM).

CAPE. — 3326 (Grahamstown): Grahamstown, (— BC), Glass 1503 (K,

NBG).

WITHOUT PRECISE LOCALITY.— Natal (Liddesdale), Wood 4249

(K, NH); Gerrard 1315 (BM, K).

3. Aspidonepsis cognata (N.E. Br.) A. Nicholas &

D.J. Goyder, comb. nov. Type: Transkei, Mount Insizwa,

Schlechter 6496 (K!, holo.; BOL!, NH!, PRE!, iso.)

Asclepias cognata N.E. Br. : 687 (1908).

Rootstock a tuber, ±7x7 mm. Stems 1, erect, 180—550

mm tall. Leaves spreading to ascending, linear, occasion-

ally lanceolate, (7 — )11 — 68 x (0.3— >0.7 — 4.0(— 6.0) mm;

apex acuminate, base shortly petiolate, occasionally

cuneate. Inflorescences occasionally subtended by leaves,

1—2 per plant, 1— 7(— 9)-flowered; bracts not fugaceous,

grading with leaves; peduncles 3 — 76(— 92) mm long.

Flowers 5—12 x 7—17 mm; pedicels 6—12 mm long.

Calyx : lobes lanceolate, 3.0— 5.0 x 1.0— 1.8 mm, apex

acuminate. Corolla glabrous; lobes elliptic, occasionally

narrow-elliptic to ovate, (5.8 — )7.6 — 10.5 x 2.6-5.8 mm,

apex acute, inside yellow, yellow-purple, brown-purple,

yellow and lilac, outside pale greenish yellow sometimes

suffused purple, or mustard yellow, or greenish brown,

or base yellow and apex purple, or base mauve and apex

yellow to dark brown, or yellow-brown with purple veins.

Corona lobes produced 1.5— 1.8 mm above corolla, cucul-

late, bonnet-shaped, 3.0— 4.8(— 5.3) [oblique measurement]

X 1.3— 2.5 mm, upper proximal ends extended into 2 short

(0.6 — )0.8 — 1.2 mm, subulate or arm-like appendages

sometimes projecting over style apex, dilated distal end

overtopping style apex by 0.6— 1.0 mm and truncated along

its upper margin; cavity 0.8 -1.3 mm deep with a yellow

tongue-like central appendage projecting 0.8— 1.3 mm

above lip of corona lobe (i.e. almost level with the upper

margin of the distal end); colour dull yellow-green,

mustard yellow, or yellow and purple, with red or brown

along the margin. Staminal column 1.5— 2.6 mm long;

anther wings 0.8— 1.4 x 0.4— 0.6 mm; anther appendages

reniform, membranous, 0.3— 0.6 x 0.8— 1.3 mm, decum-

bent on the sides of the style head. Style apex truncated

with thickened undulate margins, concave with a small

pore in the centre, 1.5— 2.8 mm wide. Pollinaria: corpus-

culum 0.2— 0.3 x 0.1—0.12 mm; translator arms 0.32—0.56

mm long; pollinia semi-circular to semi-ovate with a short

narrow proximal end, 0.72—0.96 X 0.24—0.32 mm. Fruits:

mature follicles not seen, young follicles tomentose (but

not echinate). Seeds not seen. Specific epithet etymology:

from the Latin word cognat(us) meaning related. Unfor-

tunately, N.E. Brown did not explain the sense in which

he applied this name. (Figure 5).

Aspidonepsis cognata may be found scattered in annually

burnt (but not always) montane grassland, usually occur-

ring in river valleys or near streams where the soil is quite

damp. This graceful species flowers between November

and December (although there is one record for October),

and occurs at altitudes between 1 200 and 2 100 m, rarely

lower. A. cognata is confined to a small area in the south-

ern Natal and northern Transkei Drakensberg (Figure 6).

It is unfortunate that N.E. Brown (1908) chose Schlech-

ter 6469 as the type of Asclepias cognata, because this

collection is not typical of the species as a whole. However,

all specimens of Schlechter 6469 examined, although not

typical, clearly belong to this species. In appearance

Hilliard & Burtt 7855 is more representative of the species.

Aspidonepsis cognata can be distinguished from the

other species in subgenus Aspidonepsis by its larger

flowers, larger corona lobes which are broadly helmet-

shaped, wider anther appendages and its longer translator

arms and pollinia. (See Table 3). The corona lobe shape

is highly diagnostic, in particular the subulate or arm-like

proximal appendages which may project over the style

apex, and the raised distal end which is usually truncated

along its upper margin and overtops the style head.

NATAL.— 2929 (Underberg): Fort Nottingham Commonage, (-BD),

Wright 2241 (NU); Gxalingenwa Valley, (— CB), Hilliard & Burn 17090

(K, PRE); Garden Castle State Forest, (— CC), Hilliard & Burtt 13767

(K, NU); Umzimkulu headwaters, (—CD), Hilliard & Bum 7855 (K,

NU); Mpendle, (— DB), Hilliard <6 Bum 13856 (NU).

TRANSKEI. — 3029 (Kokstad): Ensikeni, (-BA), Haygarth s.n. ex

Wood 12045 (K, NH 13661); Mount Insizwa, (—CD), Schlechter 6496

Bothalia 22,1 (1992)

31

FIGURE 5.—Aspidonepsis cognata.

A, whole plant with flowers, X

0.5; B, flower with part of

corolla removed, X 4.5; C,

corona lobe side view, x II; D,

gynostegium excluding corona,

x 9; E, abaxial surface of

anther, x 19.5; F, pollinarium,

X 36. A— F, Hilliard & Burn

9056 (NU).

(BOL, K, NH, PRE); Weza State Forest, (-DA), Nicholas 2081 (NH,

MO).

B. Subgenus Unguilobium

Unguilobium A. Nicholas & D.J. Goyder, subgen.

nov.

Folia ascendentia, margine manifeste revoluta. Inflores-

centia 4— 11-flora. Corolla reflexa; pagina abaxialis pubes-

centia. Coronae lohi ad columnam staminalem circa 1 mm

super insertionem corollae conjuncti, cucullati; appendix

proximalis ad apicem deltato-falcata et apicem styli

aequans vel superans impendensque; extremum distale

coronae appendice parva ornatum (A. reenensis) vel ap-

pendice carente (A. shebae)-, sinus profundus rimiformis.

TYPUS. — Aspidonepsis reenensis (N.E. Br.) A. Nicho-

las & D.J. Goyder vide infra.

Stems 1, erect, thin, up to 625 mm tall. Leaves ascend-

ing, linear, occasionally lanceolate, older leaves shorter

and broader, margins noticeably revolute. Inflorescences

4— 11-flowered, bracts present at anthesis and grading in

size and shape with leaf system. Flowers purple, brown,

lilac and yellow. Corolla reflexed, lobe apices ascending,

abaxial surface pubescent. Corona produced high on

staminal column, + 1 mm above corolla; lobes with

proximal appendages deltoid-falcate with obtuse apex level

with or projecting over style apex, distal end of corona

with arm-like appendage reflexed into corona lobe cavity

(A. reenensis ) or without appendage (A. shebae). Staminal

column : anther wings ear-like in outline; anther append-

ages pescapriform, deeply cleft at apex (A. shebae), or

ovate to rectangular and occasionally cleft at apex (A.

reenensis). Style apex with slightly thickened, undulate

margins. Pollinia : distal end noticeably dilated and

narrowed towards proximal end. Etymology, from the Latin

words ungu(is) (claw) and lob (us) lobe, in reference to the

claw-shaped corona lobes of this subgenus (Table 2).

There are two species in subgenus Unguilobium, viz.

A. reenensis (the type species) and A. shebae. Both are

32

Bothalia 22,1 (1992)

FIGURE 6. —Distribution of Aspidonepsis cognata , A; A. reenensis,

# ; and A. shebae , A .

confined to mountainous areas of the southern African

Drakensberg. A. reenensis is found in the southern regions

of this mountain system (namely Natal), whereas A. shebae

is found in the northeastern region (the eastern Transvaal).

As such, these species are quite widely separated geo-

graphically (Figure 6). Although probably related (even

if somewhat distantly), they can be easily told apart using

corona lobe and anther appendage shape.

4. Aspidonepsis reenensis (N.E. Br. ) A Nicholas &

D.J. Goyder, comb. nov. Type: South Africa, Natal, Van

Reenen, Wood 8635 (K! holo.; GRA!, NH!, PRE!, SAM!,

iso.).

Rootstock a tuber, 17 — 25( — 41) x 7—14 mm. Stems 1,

erect, 240— 520(— 625) mm long, scabrous. Leaves linear,

10.0—56.0 x 0.7— 2. 5(— 4.0) mm, apex acuminate, base

cuneate; usually apetiolate or petiole up to 1 mm long.

Inflorescences occasionally a number massed towards the

stem apex, 1 — 3( — 4) per plant, (l-)4— 8-flowered; bracts

2.50-5.90 x 0.25-0.50 mm; peduncles (9— )12— 65(— 75)

mm long. Flowers (4— )5— 7 x 7—11 mm; pedicel 9—15

(—21) mm long. Calyx reflexed, lobes lanceolate, apex

acuminate, 2.7— 4.5 x 1.0 — 1.7( — 2.5) mm. Corolla : lobes

narrow-elliptic to ovate, 5.5— 6.5 x 2.5— 3.8 mm, colour

(inside and out) dark reddish brown, dark brown, brown,

dull reddish purple or purple, margins light yellow or

FIGURE 1. —Aspidonepsis reenensis.

A, whole plant with flowers, X

0.4; B, flower with part of

corolla removed, x 7; C,

corona lobe, side view, x 12;

D, gynostegium excluding

corona, x 8.5; E, anther, X

27. F, anther appendage: FI,

uncleft, x 12; F2, cleft, x 12;

G, pollinarium, X 40. A-E,

F2, G, Kitlick 1205 (PRE); FI,

Trauseld 1042 (PRE).

Bothalia 22,1 (1992)

33

purple to white, abaxial surface puberulent to villous,

especially in centre and towards the base. Corona lobes

produced from staminal column 0.8— 1.0 mm above corolla,

cucullate, almost cyathiform, (1.6— )2. 2 — 2.6 x 1.3— 1.8

mm; upper proximal ends forming 2 short, falcate, arm-

like appendages with rounded or broad and frilly apices,

(0. 4 — )0.7 — 1 . 3 x 0.4— 1.0 mm, projecting over or (at least)

raised above the style apex; distal appendage short (±0.5

mm), broad and arm-like, reflexed into the cavity (some-

times totally hidden by sides of lobe), appendage below

style apex and almost level with corpusculum. Staminal

column ± 3 mm tall, slightly inflated in lower portion be-

low each corona lobe; anther wings shaped like an

elongated ear lobe, 0.8— 1.1 x (0.3— )0.4— 0.5 mm; anther

appendages ovate to rectangular, appearing wrinkled,

white, membranous, occasionally cleft at apex, 0.8— 1.5 x

1.0— 1.3 mm, decumbent on style apex. Style apex truncate

with undulate margins, concave with a small central pore,

1.8-2. 4 mm wide. Pollinarium : corpusculum 0.28-0.34

(—0.40) x 0.12—0.18 mm; translator arms 0.44— 0.64 mm

long, thin, transparent; pollinia dilated distally with a long

narrow proximal arm-like section, 0.84—1.00 x 0.22—0.28

mm. Fruits and seed not seen. Specific epithet etymology.

a latinization of Reenen from Van Reenen's Pass, the type

locality of this species (Figure 7).

FIGURE 8. —Corona lobe variation in Aspidonepsis reenensis. A, Wood

8635 (PRE); B, Stewart 2110 (NU); C, Hilliard & Bum 7796

(NU); D, Killick & Vahrmeijer 3654 (PRE); E, Hilliard & Bum

9423 (NU); F, Killick 1205 (NU); G, Franks s.n. (NH 12112);

H, Hilliard & Bum 9481 (NU); I, Wood 8635 (NH); J, Trauseld

1042 (PRE); K, Hilliard & Bum 7796 (NU); L, Rennie 1109

(NU); M, Wood 8635 (SAM).

A. reenensis grows in dry mountain grasslands, often

in sandy situations on top of the Cave Sandstone zone of

the Little Berg. It also occurs in Themeda triandra veld,

which is indicative of a fire climax community (Killick

1963). This species, said by collectors to be frequent to

rare, is found in the Natal Drakensberg, from Bushman’s

Nek in the south to Van Reenen’s Pass in the north (Figure

6). It occurs at altitudes varying from 1 500 to 2 100 m,

and flowers in the midsummer months, December and

January, with one record from November.

Corona lobe structure in the tribe Asclepiadeae is very

species-specific and usually uniform within a species

(Nicholas 1987). There are however certain exceptions, A.

reenensis being one of them. The corona lobe structure

of this species is extremely variable, although one can still

see an underlying, and therefore unifying, corona lobe

pattern (Figure 8).

NATAL.— 2829: (Harrismith): Van Reenen, (—AD), Franks s.n. ex

Wood 12112 (NH); Hilliard & Bum 9481 (NU); Wood 8635 (GRA, K,

NH. PRE, SAM); Mount Manyanyeza, (-AD), Stewart 2110 (NU);

Cathedral Peak State Forest, (— CC), Killick 1205 (CPF, K, NH, PRE).

2929 (Underberg); Giant’s Castle Nature Reserve, (—AD), Trauseld 1042

(PRE); Highmoor State Forest, (— BC), Killick & Vahrmeijer 3654 (K,

PRE); Cobham State Forest, (— CC), Hilliard & Bum 9423 (NU); Garden

Castle State Forest, (—CD), Hilliard & Bum 7796 (NU); Mawahqua

Mtn area, (—DC), Rennie 1109 (NU).

5. Aspidonepsis shebae A. Nicholas & D.J. Goyder,

sp. nov., A. reenensi (N.E. Br.) A. Nicholas et D.J. Goyder

affinis sed coronae lobis unguiformibus nec cyathi-

formibus, appendice proximali brevi falcataque nec rotun-

data vel fimbriata, appendice distali brachiformi carente

differt.

TYPE. — Transvaal, 2430 (Pilgrims Rest): (—DC), Mt

Sheba Nature Reserve, Forrester & Gooyer 216 (PRE!,

holo.).

Rootstock a tuber, ± 15 x ± 7 mm. Stems 1, erect, -

190—340 mm long. Leaves linear or occasionally lanceo-

late, 7—44 x 1—4 mm, older leaves smaller and broader,

apex acuminate; usually apetiolate, rarely with petiole up

to 0.5 mm long. Inflorescence 1—2 per plant, (2— )4— 11-

flowered; peduncle (5— )19— 90 mm long. Flowers 4.0— 6.5

x 6.0— 8.0 mm; pedicel 10—15 mm long. Calyx : lobes

lanceolate, 3.4— 3.6 x 1.1— 1.3 mm. Corolla : lobes ovate

or rarely elliptic, 5.1— 5.8 x 3.0— 3.6 mm; inside: base pale

yellow with a lilac apex, or base lilac with a dark purple

apex; outside: base green-yellow with a purple or dark

purple apex; margins pale yellow to white; abaxial surface

pubescent. Corona lobes produced from staminal column

± 1 mm above corolla, claw-like (unguiform), 1.8— 3.0

x 2.0— 2.1 mm; upper proximal ends extended into 2

short, falcate, subulate appendages projecting over style

apex; distal end a square, blunt shoulder which is ± level

with style apex; cavity a shallow, central channel ±0.9

mm deep; yellow in dried specimens. Staminal column

±2.5 mm tall; anther wings ear-shaped, ±0.7 x 0.4—0.45

mm; anther appendages pescapriform, deeply cleft at

apex, membranous, ±0.5 x ±0.7 mm, decumbent on style

apex. Gynoecium: style apex truncate, concave with a

small central pore, 1.6— 1.8 mm wide; ovaries noticeably

pubescent. Pollinarium : corpusculum 0.20— 0.26 x 0.10—

0.12 mm; translator arms 0.28—0.40 mm long; pollinia

34

Bothalia 22,1 (1992)

FIGURE 9. — Aspidonepsis shebae. A,

whole plant with flowers, X

0.5; B, flower with part of

corolla removed, x 7.5; C,

corona lobe, side view, X 14;

D, gynostegium excluding

corona, X 11; E, abaxial

surface of anther, x 24; F, pol-

linarium, x 53. A, B, D-F,

Smuts & Gillett 2326 (PRE);

C, Forrester <6 Gooyer 216

(PRE).

clavate, 0.68 —0.76 x 0.32 —0.36 mm. Fruits and seeds not

seen. Specific epithet etymology, a latinization of Sheba

from Mt Sheba, the type locality. (Figure 9).

A. shebae probably occurs in montane grasslands, and

is restricted to high altitude areas (1 400 to 2 100 m) of

the Pilgrim’s Rest region of the eastern Transvaal (Figure

6). Plants flower in December-January, and according to

one set of collections is said to be frequent.

Vegetatively A. shebae is very similar to A. reenensis,

and it is probably closely related to this species (Table

4). In floral morphology, however, these two species

differ greatly, especially in corona lobe structure (Figure

10).

TRANSVAAL.— 2430 (Pilgrim's Rest): Mt Sheba Nature Reserve,

(—DC), Forrester & Gooyer 216 (PRE); Mauchsberg, (—DC), Smuts

& Gillett 2326 (PRE). 2530 (Lydenburg): Mount Anderson, (-BA),

Smuts & Gillett 2370 (PRE).

ACKNOWLEDGEMENTS

The authors would like to thank the directors and staff

of the Royal Botanic Gardens, Kew and the National

Botanical Institute, South Africa for the opportunity,

resources and help given in the preparation of this paper.

They are also indebted to those institutions who kindly

loaned herbarium specimens. The referees are thanked for

their valued advice on the manuscript. Lastly, Ms M.

Wilmot-Dear is thanked for help with the Latin diagnoses

and Mr J.M. Fothergill for the artwork.

Bothalia 22.1 (1992)

35

TABLE 4. — A comparison of the two species in subgenus Unguilobium. All measurements in mm

* characters forming discontinuities between the two taxa.

FIGURE 10. — Corona lobe variation in Aspidonepsis shebae. A, horrester

& Gooyer 216 (PRE), X 16; B, Smuts & Gillett 2326 (PRE),

x 15; C, Smuts & Gillett 2326 (PRE), x 15; D, Smuts & Gillett

2370 (PRE), x 18.5.

REFERENCES

BROWN, N.E. 1907-1909. Asclepiadeae. In W.T. Thiselton-Dyer, Flora

capensis 4,1: 518-1036, 1129—1133. Lovell Reeves, London.

BULLOCK, A. A. 1952. Notes on African Asclepiadaceae I. Kew Bulletin

1952: 405-426.

BULLOCK. A. A. 1953a. Notes on African Asclepiadaceae II. Kew Bulle-

tin 1953: 51-67.

BULLOCK, A. A. 1953b. Notes on African Asclepiadaceae III. Kew

Bulletin 1953: 329-362.

BULLOCK, A. A. 1954a. Notes on African Asclepiadaceae IV. Kew

Bulletin 1954 : 349 - 373.

BULLOCK, A. A. 1954b. Notes on African Asclepiadaceae V. Kew Bulle-

tin 1954 : 579 - 594.

BULLOCK. A. A. 1955a. Notes on African Asclepiadaceae VI. Kew

Bulletin 1955: 265-292.

BULLOCK, A. A. 1955b. Notes on African Asclepiadaceae VII. Kew

Bulletin 1955: 611-626.

BULLOCK, A. A. 1956. Notes on African Asclepiadaceae VIII. Kew

Bulletin 1956: 503-522.

BULLOCK, A. A. 1961a. Notes on African Asclepiadaceae IX. Kew

Bulletin 15: 193—206.

BULLOCK, A. A. 1961b. An overlooked synonym of Glossonema Decne.

(Asclepiadaceae). Kew Bulletin 15: 248.

BULLOCK, A. A. 1963. Notes on African Asclepiadaceae X. Kew Bulle-

tin 17: 183-196.

BULLOCK, A. A. 1967. Nomenclatural notes : XVII. Vincetoxicum

again! Kew Bulletin 21: 351, 352.

DRUCE, G.C. 1917. Nomenclatural notes: chiefly African and Australian.

Report of the Botanical Society Exchange Club of the British Isles

4,6 Supplement 2: 601—653.

DYER, R.A. 1975. The genera of southern African flowering plants 1:

470—499. Government Printers, Pretoria.

HOLMGREN, P.K., HOLMGREN, N.H. & BARNETT, L.C. (eds).

1990. Index herbariorum. Part. I: the herbaria of the World. 8th

edn. Regnum Vegetabile 120: 1—693. I.A.P.T. & New York Bo-

tanic Gardens, New York.

KILLICK, D.J.B. 1963. An account of the plant ecology of the Cathedral

Peak area of the Natal Drakensberg. Memoirs of the Botanical

Survey of South Africa No. 34: 1—178.

KUPICHA, F. 1984. Studies on African Asclepiadaceae. Kew Bulletin

38 : 599 -672.

NICHOLAS, A. 1981. Taxonomic studies in Asclepias L. ( Asclepiadeae )

with particular reference to the narrow-leaved species in southern

Africa. M.Sc. thesis. University of Natal, Pietermaritzburg.

NICHOLAS, A. 1987. Notes on Asclepias diploglossa, A. cognata and

A. flava (Asclepiadaceae). Bothalia 17: 29-32.

NICHOLAS, A. & GOYDER, D.J. 1990. Corona lobe variation and the

generic position of Asclepias macro (Asclepiadaceae). Bothalia

20: 87-90.

PHILLIPS, E.P. 1917. A contribution to the flora of the Leribe plateau

and environs : with a discussion on the relationships of the floras

of Basutoland, the Kalahari and south eastern regions. Annals

of the South African Museum 16: 189—198.

SCHLECHTER. R. 1894. Beitrage zur Kenntnis siidafrikanischer

Asclepiadaceen. Botanische Jahrbiicher 18: 1—40.

SCHLECHTER, R. 1896. Revision of extra-tropical South African

Asclepiadaceae. Journal of Botany, British and Foreign, London

34: 311-315, 417-421, 449-458

TURCZANINOW, N. 1848. Asclepiadeae. Aliquae indescriptae. Bulle-

tin de la Societe des Naturalistes de Moscou 1: 250—262.

WOOD, J.M. 1910. Revised list of the flora of Natal. Transactions of

the Royal Society of South Africa 1: 459—461.

Bothalia 22,1: 37-51 (1992)

Notes on African plants

VARIOUS AUTHORS

BRYOPHYTA

NEW AND INTERESTING RECORDS OF MOSSES IN THE FLORA OF SOUTHERN AFRICA AREA:

2. GIGASPERMACEAE-BARTRAMIACEAE

New records identified from geographical regions

referred to in the 2nd fascicle of the moss Flora of southern

Africa (Magill 1987) are reported here. The records are

listed in the same taxonomic order as in the 2nd fascicle

of the moss flora and in the same format as in the first

paper in the series (Van Rooy & Perold 1990).

GIGASPERMACEAE

Gigaspermum repens (Hook.) Lindb. in Ofvers. (Magill 1987: 299)

Southern Cape (3421 AB: Van Zanten et al. 7608363 ).

FUNARIACEAE

Goniomitrium africanum (C. Mull.) Broth. (Magill 1987: 313)

Western Transvaal (2526 CA: Van Rooy 680).

Physcomitrium spathulatum (Homsch.) C. Mull. (Magill 1987: 318)

Northern Transvaal (2329 DD: Brenan M3267) and eastern Trans-

vaal (2430 DB: Vorster 649, 1553).

Funaria

bergiana (Homsch.) Broth. (Magill 1987: 323)

Lesotho (2928 BD: Van Rooy 3180, 3183, 3205. 2929 AC: Van Rooy

3260, 3289, 3343, 3369. 2929 CB: Van Rooy 3564), central Cape

(3224 BC: MacLea sub Rehmann 523, 523B) and the eastern Cape

(3027 CB: Van Rooy 2703).

urceolata (Mitt.) Magill (Magill 1987: 326)

Orange Free State (3026 BB: Van Rooy 2456 ) and the eastern Cape

(3027 CC: Van Rooy 2662).

spathulata Schimp. ex C. Mull. (Magill 1987: 329)

Eastern Cape (3027 CA: Van Rooy 2633. 3023 CB: Van Rooy 2690.

3027 DC: Van Rooy 2760).

BRYACEAE

Orthodontium lineare Schwaegr. (Magill 1987: 336)

Southern Cape (3322 CD: Van Zanten et al. 7609412a).

Mielichhoferia bryoides (Harv.) Wijk & Marg. (Magill 1987: 338)

Central Cape (3124 ‘DB: MacLea sub Rehmann 548. 3224 AD:

MacLea sub Rehmann 543b).

Brachymenium

acuminatum Harv. in Hook. (Magill 1987: 343)

Eastern Cape (3027 CB: Van Rooy 2714).

pulchrum Hook. (Magill 1987: 345)

Southwestern Cape (3318 CD: Stephens PRE-CH10132).

Pohlia

baronii Wijk & Marg. (Magill 1987: 349)

Northern Transvaal (2329 BB: Hardy 5268. 2329 DD: Brenan

M3246).

elongata Hedw. (Magill 1987: 351)

Southern Transvaal (2627 BB: Moss PRE-CH9775).

nutans (Hedw.) Lindb. (Magill 1987: 35 3)

Northwestern Cape (3017 BB: Van der Westhuizen & Deetlefs 44).

Leptobryum pyriforme (Hedw.) Wils. (Magill 1987: 357)

Central Transvaal, cultivated in a hothouse (2528 CA: Clarke

PRE-CH13558) and Lesotho (2828 DC: Deall & Killick 81).

Bryum

nitens Hook. (Magill 1987: 371)

Orange Free State (2827 DD: Perold 1311).

capillare Hedw. (Magill 1987: 372)

Orange Free State (2729 AC: Perold 1247, 1248).

torquescens Bruch ex De Not. (Magill 1987: 373)

Lesotho (2928 BB: Van Rooy 3221. 2929 AC: Van Rooy 3318).

pseudotriquetrum (Hedw.) G.M.S. (Magill 1987: 378)

Northern Transvaal (2329 DB: Magill 6512. 2329 DD: Perold

2459).

andicola Hook. (Magill 1987: 384)

Eastern Cape (3027 CD: Van Rooy 2819).

Rhodobryum roseum (Hedw. ) Limpr. (Magill 1987: 389)

Lesotho (2929 AC: Van Rooy 3280, 3286).

BARTRAMIACEAE

Anacolia breutelii (C. Mull.) Magill (Magill 1987: 411)

var. breutelii

Eastern Cape (3027 CA: Van Rooy 2640. 3027 CD: Van Roov 2688,

2786, 2809, 2818. 3027 DC: Van Rooy 2729, 2731, 2738).

var. squarrifolia (Sim) Magill

Lesotho (2929 CC: Magill 4347).

REFERENCES

MAGILL, R.E. 1987. In O.A. Leistner, Flora of southern Africa. Part

1. Mosses, Fascicle 2. Gigaspermaceae-Bartramiaceae. National

Botanical Institute, Republic of South Africa.

VAN ROOY, J. & PEROLD, S.M. 1990. New and interesting records

of mosses in the Flora of southern Africa area: 1. Sphagnaceae-

Grimmiaceae. Bothalia 20: 211—213.

J. VAN ROOY* and S.M. PEROLD*

* National Botanical Institute, Private Bag X101, Pretoria 0001.

MS. received: 1991-09-24.

EUPHORBIACEAE

NOTES ON EUPHORBIA SPECIES FROM THE NORTHWESTERN CAPE

Preparation of an account of the Euphorbiaceae for

the Cape Department of Nature Conservation project

‘Flora of Namaqualand’ has highlighted the chaotic

taxonomy of the genus Euphorbia in southern Africa.

The monumental publication of Alain White, R. Allen

Dyer & Boyd L. Sloane (1941) provides a wealth of in-

formation in the form of illustrations, but it does little to

explain the extent of variation that occurs in the field.

The keys provided are often unreliable and many of the

species are extremely difficult to distinguish from one

another.

Over the past 30 years many new species of Euphorbia

have been described from southern Africa. Far too narrow

a species concept seems to have been applied which some-

times bears little relevance to the position in nature.

38

Bothalia 22,1 (1992)

During recent investigations it was found that several

recently described taxa were impossible to separate from

earlier, in some cases very little-known ones. Re-collection

and cultivation of some of these has allowed a more

in-depth comparison with newer taxa and makes some

synonymy necessary.

E. celata R.A. Dyer in Bothalia 11: 278 (1974). Type:

Hall 4272 (PRE, holo.).

E. miscella Leach: 341 (1984a). Type: Leach & Williamson 16545

(NBG, holo.!; PRE, iso.!).

H. Hall collected E. celata at two widely separated

localities: north of Vredendal and along the bank of the

Groen River west of Garies, about 120 km further to the

north. More recent collecting has proved it to be fairly

plentiful, if localized, at both of these localities and has

in addition revealed its occurrence along the Swartlintjies

River some 80 km further northwards as well as near

Komaggas, yet further north near Lekkersing, the type

locality of E. miscella, and near Alexander Bay at the

mouth of the Orange River. The species thus seems to be

of scattered occurrence over most of the low-lying parts

of Namaqualand.

Both the localities Lekkersing and Alexander Bay were

mentioned by White, Dyer & Sloane (1941) and Wilman

(1946) for Euphorbia wilmaniae Marloth. Leach (1984a)

was the first to point out that this material (at least from

Lekkersing) represented a quite different species to

E. wilmaniae. However, in sorting this out, he described

the new species E. miscella. This was unnecessary in my

opinion. Leach compared the new species only to E.

namuskluftensis Leach and no mention is made of E.

celata.

Euphorbia celata (Figure 1) is generally an insignificant

plant with only short portions of stem protruding above

the soil. Plants are mostly found in very exposed places,

usually on or near the summit of low hills or on small

outcrops in low-lying areas. They grow in crevices in rock

outcrops or in flat patches of coarse quartz, granite or shale

gravel shallowly overlying bedrock.

In E. celata the stems may become densely tufted (as

in E. namuskluftensis), but this is not usual and they tend

rather to form small clusters connected by rhizomes to the

central tuber, making subsidiary roots and tubers along

the way. Plants with a neat, central, + turnip-shaped tuber

such as in Dyer (1974: fig. 2) have been dug out in several

localities, but more usually excavation yields a dense mass

of slender, interwoven rhizomes rooting sporadically along

their length with several subsidiary tubers and it is difficult

to locate the central tuber. This is as recorded by Leach

for E. miscella (Leach 1984a: fig. 1).

In the southern part of its distribution range (north of

Vredendal and along the Groen River) E. celata has rela-

tively few tubercles on the stems and they are essentially

arranged into three angles. Along the Swartlintjies River,

FIGURE 1.— Euphorbia celata. A, portion of young stem of male plant; B, portion of young stem of female plant; C, side view of male cyathium;

D, side view of female cyathium; E, face view of male cyathium; F, male flowers with bracteole; G, dissection of female cyathium showing

involucral gland, part of lobe, female flower, bracteole and bracteole primordia; H, rudimentary female floret in male cyathium. All drawn

from Bruyns 3704, north of Vredendal. Scale bar: A, B, 3 mm; C— H, 1 mm.

Bothalia 22,1 (1992)

39

on low hills south of Komaggas and in the numerous popu-

lations around Lekkersing, die stems are generally shorter

and thicker with more densely arranged, more compressed

tubercles. However, in these localities and in the hills just

north of Komaggas (where plants were seen relatively

sheltered among and inside bushes) any specimens that

were even slightly protected showed new growth very

similar to that from the southern portion of the distribu-

tion range.

North of Vredendal and along the Groen River the leaves

of E. celata are usually 3—6 mm long. North of this they

are often much smaller (as short as 1.5 mm) and their size

is dependent on the situation of the plant (in sheltered

branches they are still 3—4 mm long). The leaf is about

as long as or longer than the tubercle to which it is affixed

and in this respect is exactly as in E. miscella.

In E. celata (as in E. miscella ) a terminal cyathium is

produced, subtended by 2—3 leaf-like bracts in the axils

of which further cyathia may develop. This is different to

the situation in E. namuskluftensis where the cyathia are

solitary and are borne near the apex of the branch in the

axils of the tubercles. In E. namuskluftensis each cyathium

is ‘borne on a short, glabrous, bracteate peduncle’ (Leach

1983: 190). This is lacking in both E. celata and E.

miscella.

For E. miscella and E. celata the dimensions of the

cyathia are summarized below (measurements based on

Figure 1 for E. celata are given in brackets after Dyer’s

data; measurements based on Bruyns 4637 are given in

brackets after Leach’s data for E. miscella.

E. miscella (Leach 1984a)

Male cyathium: up to 4.5 x 3.5 (4.0 x 2.5— 3.0 mm),

stipe 2—4 mm long (1-2 mm long), pedicels glabrous.

Female cyathium: 3.5 x 4.5 (2.0-2. 5 x 2.0 mm), style

±2.5 mm long, capsule 7 mm diam.

E. celata (Dyer 1974)

Male cyathium: 4x4 mm (3.5 x 4.5 mm), peduncle 4

mm (1 mm) (pedicels sparsely pubescent towards apex).

Female cyathium: 4.0 x 1.5 (3.0 x 2.5 mm), style 2.5

mm long (4 mm long, part protruding from involucre 2.5

mm long), capsule 5.5— 6.0 mm diam.

The material on which Figure 1 is based was collected

at the type locality of E. celata but differs from the

flowering material seen by Dyer (1974). Bruyns 4637 was

collected from near the type locality of E. miscella. It is

highly unlikely that more than one very similar species

occurs at these localities and thus one is compelled to

accept that variability occurs in E. celata, as is to be

expected in most species. The small differences in the

measurements between the material from north of

Vredendal and that from Lekkersing are not considered

to be significant and the remarkable similarity between

Leach’s E. miscella and E. celata forces one to the

conclusion that these two represent the same species.

Specimens examined

CAPE. — 2816 (Oranjemund): Kortdoringberg, (—DA), Van Jaarsveld

5421A (NBG). 2817 (Vioolsdrift): 10 km N of Lekkersing, (-CC), Bruyns

4638 (BOL); Lekkersing, (-CC), Marloth 12441 (STE). 2917 (Spring-

bok): south of Lekkersing, (— AA), 18 July 1970, Wisura 1622 (NBG);

Williamson 3323 (BOL), 3206 (NBG); Leach 16545 (NBG); N of

Gemsbokvlei, (— AA), Bruyns 4637 (BOL); hills N of Komaggas, (—CD),

Bruyns 4617 (BOL); S of Komaggas, (—CD), Bruyns 4595 (BOL). 3017

(Hondeklip Bay): Swartlintjies River, (— AB), Bruyns 3556 (BOL); Groen

River, (-DD), Hall 4282 (NBG); Bruyns 1733 (BOL). 3118 (Vanrhyns-

dorp): north of Hoi R. Station, (-AD), Hall 4722 (PRE); Bruyns 3704

(BOL).

E. quadrata Nel in Jahrbuch der Deutschen Kakteen-

gesellschaft: 42 (1935). Type: Herre sub SUG 6519 (STE,

holo.!; BOL, iso.!).

E. francescae Leach: 563 (1984b). Type: Williamson 3248 (NBG!).

Euphorbia quadrata has been remarkably rarely

collected. Originally discovered by Hans Herre in 1930

in the Stinkfontein Mountains, it was rediscovered by

Oliver, Tolken & Venter in 1977 and since then only three

collections seem to have been made. This is partly due

to the remoteness of the areas in which it occurs and also

to the inconspicuousness of the plants. On the summit of

the Cornellsberg they grow fully exposed, forming small,

densely branched, rounded shrubs with somewhat of an

‘alpine shrublet’ habit. However, lower down they were

found to reach 1 m in height but were easily overlooked

due to their nondescript appearance with large numbers

of slender, scarcely succulent twigs and few leaves. At

another locality further to the west they were common on

the sheltered upper southwestern slope of a peak, growing

inside other bushes and among rocks and were again

inconspicuous despite being up to 0.5 m tall. It appeared

that any branches projecting from the sheltering bushes

were soon eaten back, and exposed plants were found

grazed right back to the main stem.

In E. quadrata a complex system of thickened, brown,

root tubers develops. The main stem may be up to 25 mm

thick and this branches under the soil surface into more

slender roots which are swollen at intervals into sausage-

like tubers up to 30 mm thick. The aerial stems have a

peculiar colouration: the youngest tissue usually to just

below the leaves is a striking red-purple after which it is

striped lengthwise through the splitting of this reddish

bark. The reddish hue disappears after one season and

older stems are covered with a uniformly greyish bark and

have a somewhat rubbery consistency, not becoming truly

woody (Figure 2).

As is usual in bisexual Euphorbia cyathia, the female

flower appears first and the free parts of the styles are fully

divergent after 3-5 days (Figure 2B, G). The female

flower in E. quadrata remains erect for about 10 days and

then bends downwards through the space left by the

missing involucral gland just as the male flowers begin

to appear (Figure 2C, D). If no pollination takes place,

the style shrivels up within two days of its becoming fully

recurved. If pollination does take place, the style gradually

becomes erect and the ovary begins to swell. It has been

found that an interval of several months may occur between

pollination and the onset of development of the ovary.

Colours observed in the flowers are unusual: the male

pedicel is pale translucent green, the filament bright

pinkish red and the anther is yellow lined with red along

the pore; the female pedicel is greyish green except for

red just beneath the ovary (the rudimentary calyx), the

ovary is dark green and the styles are pink.

40

Bothalia 22,1 (1992)

FIGURE 2. — Euphorbia quadrata. A, portion of young stem; B, G, side views of young cyathium with female floret still erect; C, side view

of cyathium with female flower recurved; D, H, face views of cyathium; E, I, dissection of cyathium showing involucral lobe, male flowers,

female flower and some bracteoles; F, cluster of male flowers. B— F drawn from Bruyns 4044 (Cornellsberg summit); A, G, H, I from

Bruyns 3936 (Vandersterrberg). Scale bar: A-D, G, H, 3 mm; E, F, I, 1 mm.

Euphorbia quadrata and E. francescae both come from

the summit of the Stinkfontein Mountains. E. francescae

is given as differing from the former by 1, its much smaller

plants with very short, tuberculate branches (± 30 mm

long); 2, longer male flowers; 3, subglobose ovary; 4,

very much shorter styles which are free almost to the

base.

I have visited the summit of the Cornellsberg to

investigate Euphorbia francescae. Compared to the type

specimen, which is a remarkably minute plant, others at

the same locality formed shrublets up to 200 mm tall and

the same in diameter (as observed also by Oliver et al.

on their specimen). All specimens seen were much stunt-

ed by grazing and by the heavy winds which sweep across

this very exposed mountain top and this clearly explains

the fact that they were much smaller than either the plants

seen further down the mountain or the specimen cultivated

at Stellenbosch by Herre (the type of E. quadrata). Their

branches were therefore unusually short. Nevertheless, the

30 mm given by Leach for the length of the branches was

observed to be on the small side and 30—100 mm would

be more accurate. Leach considered Nel’s species to be

without small tubercles on the stems and Nel (1935) does

not mention them in his description, but they are clearly

present on the isotype specimen at BOL.

Nel does not mention the size of the male flowers. Leach

(1984b: 567) gives the filaments in E. francescae as +

1.5 mm long and states that they are longer than in E.

quadrata. In Figure 2C the protruding male flower has

a filament 2 mm long whereas in Figure 2E the filament

is slightly less than 1.5 mm long. On the isotype of E.

quadrata (BOL) the filaments are 2 mm long. Therefore

I see no basis for Leach’s statement about the relative sizes

of the male flowers.

Bothalia 22,1 (1992)

41

The styles of E. quadrata were given by Nel as 9 mm

long. Leach gives them for E. francescae as ‘± 1.5 mm

long, free almost to the base’. In Leach’s illustrations

(1984b: fig. 2.2) the styles are a little over 2 mm long with

fused and free parts almost exactly equal in length. The

somewhat thickened base that he shows (which he

obviously did not include when measuring the style) has

not been found in other material from the same site and

is probably due to the styles already having begun to dry

out (as pointed out above, this happens quickly after the

female flower becomes recurved, if pollination has not oc-

curred). In my collections the lengths of the styles were

3.5 mm on the Comellsberg and 8 mm on the Vandersterr-

berg. These measurements are from very few specimens

(one plant in the first case and three in the second) but

even from these and including Nel’s data, it is clear that

E. quadrata varies considerably in this character so that

this distinction is meaningless.

I conclude that it is not possible to distinguish E.

francescae from E. quadrata with any certainty. The name

E. francescae is therefore placed into synonymy.

Specimens examined

CAPE.— 2816 (Oranjemund): Vandersterrberg, (— BD), Bruyns 3936

(BOL). 2817 (Vioolsdrif): Cornellsberg, (— CA), Oliver, Tolken &

Venter 697 (PRE); Williamson 3248 (NBG); Bruyns 4044 (BOL); near

summit of Stinkfontein Mountains, Herre sub SUG 6519 (STE, holo.;

BOL, iso.).

E. exilis Leach in Leach & Williamson, South African

Journal of Botany 56: (1990). Type: Leach & Bayer 17129

(NBG!).

E. glandularis Leach & Williamson: 75 (1990). Type: Leach & Hihon-

Taylor 17019 (NBG!).

Leach & Williamson (1990: 77) gave E. exilis (Figure

3) as occurring mainly on the Knersvlakte and E. glan-

dularis in a small area to the northwest of Steinkopf (Leach

& Williamson 1990: 76). More recently similar plants have

been collected near Nigramoep (northwest of Springbok)

and north of the Groen River (west of Garies). Since

the plants are not readily distinguishable, except by an

experienced collector, from the much commoner E.

ephedroides E. Mey. ex Boiss. and members of the E.

decussata complex, the species will probably be found in

much of the intervening territory in due course.

Leach & Williamson distinguished E. exilis from E.

glandularis by the ‘smaller stature (300-400 mm as op-

posed to 0.6— 1.0 m), its densely branched, clump-forming

FIGURE 3. — Euphorbia exilis. A— C, portion of young stem; D, older male cyathium with two developing males on axillary branches; E, bisexual

cyathia; F, female cyathia with one capsule nearing maturity; G, female cyathium with capsule near maturity; H, dissection of male cyathium

showing involucral gland, part of lobe, male flowers and rudimentary female flower (separated); I, male and female flowers from separate

(unisexual) cyathia. A, C, D, G drawn from Bruyns 3214 (Nigramoep); B, H, from Bruyns 3835 (Groen River); E, F, I from Bruyns 3225

(N of Vredendal). Scale bar: A, 3 mm; B, C, H, I, 1 mm; D-G, 2 mm.

42

Bothalia 22,1 (1992)

habit with soft, more slender branches, much smaller

stipular glands, its frequently pink involucres and maroon

bracts, its smaller capsule which is not ribbed in the

sinuses and is borne on a longer pedicel, styles which are

not appressed to the top of the capsule, and less

prominently keeled seeds which are only + 1.2 mm thick

as opposed to 1.6 mm’ in E. glandularis.

Plants collected near Nigramoep were 300—400 mm tall

and thus, although geographically closest to E. glan-

dularis, were actually more similar in stature to E. exilis.

They were also extremely densely branched, as were those

along the Groen River, far more so than plants seen at

Hoi River (Knersvlakte) but very similar to those seen at

Quaggaskop on the Knersvlakte. Plants from the Groen

River had slightly smaller stipular glands than those from

Nigramoep (compare Figure 3B & 3C). Stipular glands

of this size are also found on specimens from the Kners-

vlakte [Hall 3734 (NBG) and others] and the minute

stipular glands shown by Leach & Williamson (1990: fig.

4.9) had perhaps already dried out when drawn and were

thus a fraction of their fresh size (Figure 3).

Recent collections suggest that the length of the female

pedicel is an unreliable character. In particular, in E. glan-

dularis the capsule is shown to squash the involucre on

development (Leach & Williamson 1990: fig. 3.10) where-

as in E. exilis this is not the case. In the material figured

here from near Nigramoep (Figure 3G) the pedicel is

slightly longer than that of a specimen from the Kners-

vlakte (Figure 3F). As can be seen in Figure 3E, F & I,

the styles vary significantly in length at a single locality

(Knersvlakte again), being quite close to the ovary (more

like E. glandularis) in 3F & I and more erect in 3E (more

like E. exilis). In the plant from Nigramoep (Figure 3G)

they are much closer to the ovary, though not as closely

adpressed as shown by Leach & Williamson for E. glan-

dularis. The difference in the length of the united part of

the styles (± 0.5 mm) and the 0.4 mm difference in thick-

ness of the seeds given by Leach, are considered to be

of questionable significance. Thus one of these names is

unnecessary and E. glandularis is reduced to synonymy.

Specimens examined

CAPE. — 2917 (Springbok): Klipfontein, (—BA), Williamson 3773,

3789; Leach & Hilton-Taylor 17019 (all NBG); NE of Nigramoep, (— BC),

Bruyns 3214 (NBG). 3017 (Hondeklip Bay): 5 km NW Baievlei, (— DB),

Bruyns 4590 (BOL); 2 km towards Soutfontein, (-DB), Bruyns 4588

(BOL); north of Groen River, (-DB), Bruyns 3835 (BOL). 3018 (Kamies-

berg): Kamagab, (—CD), Bruyns 4586 (BOL). 3118 (Vanrhynsdorp): north

of Vredendal, (—AD), Hall 3734 (NBG); Bruyns 3225 (NBG); Quaggas-

kop, (— BC), Leach & Hilton-Tayor 16994 (NBG); Bruyns 4035 (BOL);

Hoi River, (-CB), Leach & Hall 14180 (NBG); Aties, (-DA), Leach

& Bayer 17129 (NBG).

REFERENCES

DYER, R.A. 1974. A new species of Euphorbia (Euphorbiaceae).

Bothalia 11: 278-280.

LEACH, L.C. 1983. A new Euphorbia from South West Africa (Namibia).

Journal of South African Botany 49: 189-192.

LEACH, L.C. 1984a. A new Euphorbia from South Africa. Journal of

South African Botany 50: 341—345.

LEACH, L.C. 1984b. A new Euphorbia from the Richtersveld. Journal

of South African Botany 50: 563—568.

LEACH, L.C. & WILLIAMSON, G. 1990. The identities of two con-

fused species of Euphorbia (Euphorbiaceae) with descriptions

of two closely related new species from Namaqualand. South

African Journal of Botany 56: 71—78.

NEL, G.C. 1935. Euphorbia quadrata Nel nov. sp. Jahrbuch der Deut-

schen Kakteengesellschafr. 42, 43.

WHITE, A., DYER, R.A. & SLOANE, B.L. 1941. The succulent

Euphorbieae. Abbey Garden Press, Pasadena.

WILMAN, M. 1946. Preliminary checklist of the flowering plants and

ferns of Griqualand West. Deighton Bell, Cambridge.

P. BRUYNS*

* Bolus Herbarium, University of Cape Town, Private Bag, Rondebosch

7700.

MS. received: 1991-03-18.

FABACEAE

THE IDENTITY OF ARGYROLOBIUM OBSOLETUM AND THE CORRECT NAMES FOR SOME SPECIES OF POLHILUA (CROTALARIEAE)

The genus Polhillia C.H. Stirton comprises seven rare

and highly localised species, all of which are endemic to

the southwestern Cape Province of South Africa. The

generic circumscription of Stirton (1986a) was slightly

broadened by Van Wyk & Schutte (1989) to include two

species that were previously misplaced in Argyrolobium

Eckl. & Zeyh. and Melolobium Eckl. & Zeyh. The species

of Polhillia all have the bilabiate calyx of Argyrolobium

and Melolobium and also have a similar combination of

quinolizidine alkaloids (Van Wyk et al. 1988; Van Wyk

& Verdoorn 1990), but can easily be distinguished from

the two genera by the absence of true peduncles (Van Wyk

& Schutte 1989). The intemode directly below the inflores-

cence is elongated to function as a peduncle, a condition

that can be recognised by the presence of a foliage leaf

at the apex of the pseudo-peduncle. Other diagnostic

characters are the woody habit, conduplicate leaflets,

virtual absence of bracts and bracteoles, sheathing stipules,

short calyx, pubescent corolla, imbricate keel petals and

a chromosome number of 2n = 32 (Stirton 1986a; Van

Wyk & Schutte 1989).

In considering various species for inclusion in the new

genus, Stirton (1986a) overlooked Argyrolobium obsole-

tum Harv. Harvey (1862) queried the correct taxonomic

position of this species by inserting a question mark behind

the generic name. The holotype in the Thunberg collec-

tion in Uppsala clearly belongs to the type species of

Polhillia, hitherto known as P. waltersii (C.H. Stirton)

C.H. Stirton. A name change is therefore unavoidable. I

am also using this opportunity to change the gender of

some specific epithets that were not given in the correct

form when species were transferred from Argyrolobium

and Melolobium by Stirton (1986a) and Van Wyk & Schutte

(1989). To avoid confusion, all known species of Polhillia

are included in the following synonymy.

Polhillia C.H. Stirton in South African Journal of

Botany 52: 167 (1986a); Van Wyk & Schutte: 397 (1989);

Van Wyk: 265-288 (1991).

Lebeckia Thunb. subgenus Plecolobium C.H. Stirton:

318 (1981). Type: Polhillia waltersii (C.H. Stirton) C.H.

Bothalia 22,1 (1992)

43

Stirton. [now P obsoleta (Harv.) B-E. van Wyk, see

below] .

1. P. brevicalyx (C.H. Stirton ) Van Wyk & Schutte

in Kew Bulletin 43: 420 (1989).

Argyrolobium brevicalyx C.H. Stirton: 443 (1984). Type: Burgers 3188

(K!, holo. ; STE!, iso.).

2. P. canescens C.H. Stirton in South African Journal

of Botany 52: 174 (1986a). Type: Bayer 3104 (NBG!,

holo.).

3. P. connata (Harv. ) C. H. Stirton in South African

Journal of Botany 52: 174 (1986a), as 'P. connatum. Type:

Thom 37 (K!, holo.).

4. P. involucrata (Thunb.) Van Wyk & Schutte in Kew

Bulletin 43: 420 (1989), as ‘ P. involucratum .

Psoralea involucrata Thunb.: 607 (1823). Argyrolobium involucratum

(Thunb.) Harv.: 75 (1862). Melolobium involucratum (Thunb.) C.H.

Stirton: 355 (1986b), as ‘ M . involucratum (Harv.) C.H. Stirton’. Type:

Thunberg s.n. sub UPS-THUNB 17575 (UPS!, lecto., chosen by Stirton

1986b).

5. P. obsoleta (Harv) B-E. van Wyk , comb. nov.

Argyrolobium obsoletum Harv., Flora capensis 2 : 70 (1862). Type:

Cape, without precise locality, Thunberg s.n. sub UPS-THUNB 16504

(UPS!, holo.).

Lebeckia waltersii C.H. Stirton: 318 (1981). Polhillia waltersii (C.H.

Stirton) C.H. Stirton: 173 (1986a). Type: Rourke 1484 (K! , holo.; NBG!,

STE!, iso.).

Aspalathus sericea sensu Thunb.: 574 (1823) non DC.

P. obsoleta is known only from one locality at Worcester,

where the first recent collection was made in 1977 (Stirton

1986a). A specimen not seen by Stirton (1986a) however,

indicates that the species may have been much more widely

distributed in an area that is now largely under wheat

cultivation. This specimen, Edwards s.n. sub BOL 13438

(BOL), was collected at Porterville (3318 BB Cape Town)

and sent to the Bolus Herbarium in 1909.

6. P. pallens C.H. Stirton in South African Journal

of Botany 52: 171 (1986). Type: Burgers 2633 (STE!, holo.;

K!, STE!, iso.).

7. Polhillia sp. A. [see Stirton: 178 (1986a)].

Stirton (1986a) expressed uncertainty about the identity

of an anomalous specimen, Hutchison 253 (K), which was

previously included by Stirton (1981) under L. waltersii

(P. obsoleta).

ACKNOWLEDGEMENTS

I thank Dr C.H. Stirton for verifying the identity of

Edwards s.n. in the Bolus Herbarium and the curators of

BOL, K, NBG, PRE, STE and UPS for allowing me to

study the material of Polhillia.

REFERENCES

HARVEY, W.H. 1862. Leguminosae. In W.H. Harvey & O.W. Sonder,

Flora capensis 2 : 67—76. Hodges & Smith, Dublin.

STIRTON, C.H. 1981. Studies in Leguminosae-Papilionoideae of southern

Africa. Bothalia 13: 317—325.

STIRTON, C.H. 1984. A new species of Argyrolobium (Fabaceae) from

the southern Cape. Journal of South African Botany 50: 443-448.

STIRTON, C.H. 1986a. Polhillia, a new genus of papilionoid legumes

endemic to South Africa. South African Journal of Botany 52:

167-180.

STIRTON, C.H. 1986b. Melolobium involucratum (Fabaceae), a new

combination for South Africa. South African Journal of Botany

52: 354-356.

THUNBERG, C.P. 1823. Flora capensis : 574, 706. Schultes, Stuttgardt.

VAN WYK, B-E. 1991. A review of the tribe Crotalarieae (Fabaceae).

Contributions from the Bolus Herbarium 13: 265—288.

VAN WYK, B-E. & SCHUTTE, A.L. 1989. Taxonomic relationships

amongst some genera of Leguminosae tribe Crotalarieae and

Argyrolobium (Genisteae). Kew Bulletin 44: 397—423.

VAN WYK, B-E. & VERDOORN, G.H. 1990. Alkaloids as taxonomic

characters in the tribe Crotalarieae (Fabaceae). Biochemical

Systematics and Ecology 18: 503—515.

VAN WYK, B-E., VERDOORN, G.H. & GREINWALD, R. 1988. The

major alkaloids of the genus Polhillia. South African Journal of

Botany 54: 389-391.

B-E. VAN WYK*

* Department of Botany, Rand Afrikaans University, P.O. Box 524, Jo-

hannesburg 2000.

MS. received: 1991-06-11.

PROTEACEAE

THE CORRECT AUTHOR CITATION FOR PARANOMUS REFLEXUS

In the most recent revision of Paranomus (Levy ns 1970),

the authority for Paranomus reflexus is given as P. reflexus

(Phill. & Hutch.) N.E. Br. in Transactions of the Royal

Society of South Africa 21: 263 (1933). This citation is also

used in Gibbs Russell et al. (1987). However, an earlier

combination made by Fourcade in 1932 has been over-

looked. Curiously, Fourcade’s combination was published

in the same volume of the same journal as N.E. Brown’s

was, namely Transactions of the Royal Society of South

Africa vol. 21, but in part 1 published in December 1932,

whereas N.E. Brown’s combination appeared in part 3

published in November 1933.

The correct author citation for this species is therefore

Paranomus reflexus (Phill. & Hutch.) Fourcade in Trans-

actions of the Royal Society of South Africa 21: 97 (1932).

REFERENCES

BROWN, N.E. 1933. Nivenia Vent, and Nivenia R. Br. Transactions of

the Royal Society of South Africa 21: 259—270.

FOURCADE, H.G. 1932. Contributions to the flora of the Knysna and

neighbouring divisions. Transactions of the Royal Society of South

Africa 21: 75—102.

GIBBS RUSSELL, G.E. et al. 1987. List of species of southern African

plants edn 2, part 2. Memoirs of the Botanical Survey of South

Africa No. 58.

LEVYNS, M.R. 1970. A revision of the genus Paranomus (Proteaceae).

Contributions from the Bolus Herbarium 2: 37.

J.P. ROURKE*

* National Botanical Institute, Kirstenbosch , Private Bag X7, Claremont

7735.

MS. received: 1991-10-29.

44 Bothalia 22,1 (1992)

GOMPHILLACEAE (LICHENES)

A NEW SPECIES OF BULLATINA FROM THE TRANSKEI WILD COAST

Bullatina viridis Brusse, sp. nov.

Thallus crustosus, foliicola, viridis, usque ad 7 mm

diam., laevis, trichomatibus hyphophorisque instructus;

prothallus argenteo-albidus. Trichomata (Figurae 4 & 5)

albida, simplicia, arcuata, usque ad 1.2 mm longa, basi

circa 50 /urn crassa, ad apicem acuta. Hyphophora (Figurae

4 & 5) piliformia, albida, erecta, 0.22—0.30 mm alta, basi

20—30 pm crassa, ad apicem obtusa; ex apicibus fascicu-

lus lacrimiformis conidiophororum et catenarum

conidiorum cellulis algarum pendens. Conidiophora

(Figura 6) hyalina, leviter ramosa, septata. Cellulae coni-

diogenae nonnihil inflatae, terminales. Conidia concatena-

ta (Figura 6), holoblastica, acropeta, hyalina, simplicia,

ellipsoidea vel clavata, 6.0— 8.5 x 1.3— 2.0 ^m. Cortex

(superior) monostratus, 3—5 pm crassus, paraplecten-

chymatus, cellulis 3—5 pm diam. Stratum gonidiale 15—20

pm crassum. Algae coccoideae, virides, 4.5-11.0 fim

diam. Apothecia viridia, sessilia, usque ad 0.4 mm diam.

(Figura 4). Excipulum thallinum 25—30 pm crassum,

cellulas algarum continens, crystallis destitutum. Excipu-

lum proprium reductum, hyalinum, prosoplectenchyma-

tum, 10-15 pm crassum. Hypothecium destitutum.

Hymenium hyalinum, J— , 55—80 pm altum. Paraphyses

leviter ramosae anastomosaeque vel fere simplices, paralle-

lae, septatae, ecapitatae, bene gelatinosae, luminibus circa

FIGURE 4. — Bullatina viridis Brusse, habit. F. Brusse 5864 , holotype.

Scale in mm.

FIGURE 5. — Bullatina viridis

Brusse, camera lucida

drawing of a trichome and

hyphophores, showing the

teardrop- shaped conidial

mass. F. Brusse 5864 ,

holotype. Bar = 0.1 mm.

0.8 pm crassis. Epihymenium 5—8 pm crassum, cum

gonocystibus, 2.5— 5.0 pm diam., crystallis destitutum.

Asci late clavati, parietibus J— hyalinis, apici leviter in-

crassatis, ascoplasmate J vinoso-rubescente. Ascosporae

singulae, 30—60 x 10—22 gm, hyalinae, muriformes,

11—18 x 2— 5-loculares, halonatae, ovales, extremis

obtusis, J— , parietibus tenuibus, cyanophilis, interdum in

pycnidia mutantes, vel in spermatia, hyalina, 3-4 x 0.7

/urn, fatiscentes. Pycnidia propria non visa.

TYPUS. — Transkei, 3228 (Butterworth): Dwesa Nature

Reserve. About 1 km from campsite on road to mPume

gate. On living fronds of the cycad Encephalartos villosus

Lem., in understorey vegetation of coastal forest. Alt. 40

m (— BD). F. Brusse 5864, 1991-01-05 (PRE, holo.; BM,

iso.). Figurae 4, 5 & 6.

Thallus crustose, foliicolous, green, up to 7 mm across,

smooth, with trichomes and hyphophores; prothallus

whitish with a silvery sheen. Trichomes (Figures 4 & 5)

completely whitish, simple, arcuate, up to 1.2 mm long,

about 50 pm thick at base, apex acute. Hyphophores

(Figures 4 & 5) piliform, whitish, erect, 0.22—0.30 mm

tall, 20—30 pm thick at base, apex obtuse, a teardrop-

shaped bundle (when wet) of conidiophores and chains

of conidia hang down from apex, which contains algal cells

in its core. Conidiophores (Figure 6) pendant, hyaline,

sparsely branched, septate. Conidiogenous celts somewhat

swollen, terminal. Conidia concatenate (Figure 6), holo-

blastic, acropetal, hyaline, simple, ellipsoid to clavate,

6.0— 8.5 x 1.3— 2.0 pm. Upper cortex single-layered, 3—5

pm thick, paraplectenchymatous, cells 3—5 pm diam. Al-

gal layer 15—20 pm thick. Algae coccoid, green, 4.5—11.0

pm diam. Apothecia green, sessile, up to 0.4 mm diam.

(Figure 4). Thalline exciple 25—30 pm thick, containing

algal cells, crystals (of calcium oxalate) absent. Proper

exciple reduced, hyaline, prosoplectenchymatous, 10—15

pm thick. Hypothecium absent. Hymenium hyaline, J— ,

Bothalia 22,1 (1992)

45

55—80 /xm high. Paraphyses lightly branched and anas-

tomosed to nearly simple, parallel, septate, ecapitate,

strongly gelatinized, lumens about 0.8 /xm thick. Epihy-

menium 5-8 /xm thick, with gonocysts, 2.5— 5.0 /xm

diam., crystals (of calcium oxalate) absent. Asci broadly

clavate, 1-spored, wall J- hyaline, somewhat thickened

towards apex, ascoplasma J wine-red. Ascospores 30-60

x 10—22 /xm, hyaline, muriform, 11— 18-locular x

2-5-locellate, halonate, oval, ends obtuse, J— , walls thin,

cyanophilic (the walls stain deep blue in lactophenol Cotton

Blue), sometimes changing into pycnidia, or disintegrating

into hyaline spermatia, 3-4 x 0.7 /xm. Proper pycnidia

not seen.

The thallus bears whitish trichomes and much smaller

whitish hyphophores, quite abundantly. The hyphophores,

which represent the anamorph of this lichen, are of the

hanging teardrop type, which is also found in Bullatina

aspidota (Vain.) V&zda & Poelt, until now the only

species of Bullatina (V&zda 1979: fig. 9; Vezda & Poelt

1987). However, the hyphophores and trichomes are

similar in size in B. aspidota (0.5— 0.6 mm long), whereas

in B. viridis they are clearly dimorphic — the hyphophores

only reaching 0.3 mm high and being stubble-like, whereas

the trichomes are much longer, becoming 1.2 mm long

in some cases.

This new species is a very distinct species from a macro-

scopic habit point of view. The whole lichen is green in

colour, and the proper exciple does not crack away from

the thalline exciple except very rarely in old specimens.

The colour may be due to the fact that the whole lichen

lacks calcium oxalate crystals, which are common in the

greyer species.

FIGURE 6. — Bullatina viridis Brusse, camera lucida drawing of the

conidiophores and chains of conidia. F. Brusse 5864 , holotype.

Bar = 10 n m.

The genus Gyalectidium Mull. Arg., closely related to

the genus Bullatina V£zda & Poelt, produces a com-

pletely different anamorph in the form of a small erect

scale, which may be variously ciliate or ragged along the

upper edge, with the conidial mass in the axil of this scale

(Serusiaux & De Sloover 1986; V&zda 1979, 1983).

This new species is not likely to be confused with the

only other Bullatina species, B. aspidota due to the latter’s

relatively thick thallus which is white due to calcium

oxalate crystal encrustation. The apothecia are deeply

sunken into the thallus as well (Santesson 1952: fig. 64)

and the ascospores are larger (42—80 x 20—32 /mi) than

those of B. viridis (30—60 x 10-22 /xm).

Superficially B. viridis may be mistaken for a Tricharia

with white trichomes, especially one with epithecial

gonocysts present, such as in T. vulgaris (Mull. Arg.) R.

Sant. T vulgaris has, in fact, got ascospores of a very

similar size range to B. viridis, but T. vulgaris lacks

trichomes and possesses a stalked-auriculoid hyphophore

(Kalb & Vezda 1988; fig. 2), rather than the teardrop

hyphophore of B. viridis, and many other Tricharia species

(Kalb & Vezda 1988; Vezda 1979). T. vulgaris, like all

other Tricharia species, has a well -developed proper

exciple which is paraplectenchymatous and 20—30 /xm

thick under the hymenium, and thicker on the flanks. This

degree of exciple development is lacking in B. viridis. On

the other hand, the apothecia of Tricharia lack a thalline

exciple and stand clear of any thalline tissue, whereas those

of Bullatina possess a thalline exciple, usually referred to

as being immersed in the thallus (cryptolecanorine).

However, whatever the situation is, the flanks of the

hymenium have tissue containing algae in B. viridis, which

is not the case in all Tricharia species, including T.

vulgaris.

Thus far, this new species has only been collected at

the type locality, Dwesa Forest on the Transkei Wild Coast,

but will undoubtedly be found at other forested localities

in warm subtropical and tropical areas.

ACKNOWLEDGEMENTS

The author would like to thank Prof. R. Santesson for

his views on this lichen, and Dr A. Vezda for kindly

reviewing this manuscript. Prof. Colin H. Dickinson

introduced the author to this important locality. Thanks

are extended to the Transkei Forestry Department for

permits to collect at Dwesa.

46

Bothalia 22,1 (1992)

REFERENCES

KALB, K. & v£zDA, A. 1988. Neue oder bemerkenswerte Arten der

Flechtenfamilie Gomphillaceae in der Neotropis. Bibliotheca

Lichenologica 29: 1-80.

SANTESSON, R. 1952. Foliicolous lichens 1. A revision of the taxonomy

of the obligately foliicolous, lichenized Fungi. Symbolae Botanicae

Upsalienses 12,1: 1-590.

SERUSIAUX, E. & DE SLOOVER, J.R. 1986. Taxonomtcal and eco-

logical observations on foliicolous lichens in northern Argentina,

with notes on the hyphophores of Asterothyriaceae. Veroffent-

lichungen des Geobotanischen Institutes der Eidgenossischen

Technischen Hochschule, Stiftung Riibel, in Zurich 91: 260—292.

vfiZDA, A. 1979. Flechtensystematische Studien XI. Beitrage zur

Kenntnis der Familie Asterothyriaceae (Discolichenes). Folia

Geobotanica et Phytotaxonomica 14: 43—94.

V^ZDA, A. 1983. Foliicole Flechten aus der Kolchis (West-

Transkaukasien, UdSSR). Folia Geobotanica et Phytotaxonomica

18 : 45 -70.

vfeZDA, A. & POELT, J. 1987. Flechtensystematische Studien XII. Die

Familie Gomphillaceae und ihre Gliederung. Folia Geobotanica

et Phytotaxonomica 22: 179—198.

F. BRUSSE*.

* National Botanical Institute, Private Bag X101, Pretoria 0001.

MS. received: 1991-07-15.

MORACEAE

NEW RECORDS OF FICUS SPECIES AND THEIR POLLINATORS ON GRAND COMORE

The Comoros consist of four major islands of volcanic

origin, the largest and geologically youngest of which is

Grand Comore (Ngazidja). The fig trees ( Ficus spp.,

Moraceae) of the Islands were revised by Berg (1986), who

recognised nine Ficus species, five of which were record-

ed from Grand Comore. The nine Comoran Ficus spe-

cies include two species that are also present on the African

mainland and six species shared with Madagascar. There

is also one endemic species, and one of the species shared

with Madagascar is represented by an endemic sub-

species.

Fig trees are pollinated exclusively by fig wasps

(Hymenoptera: Agaonidae). Knowledge of the fig wasps

of the Comoro Islands is minimal, with just a single record

from Mayotte of Allotriozoon heterandromorphum, the

pollinator of F. lutea (Wiebes 1974). This report describes

the results of a survey of Ficus species and their pollina-

tors carried out on Grand Comore in July 1990. Specimens

of eight Ficus spp. were collected, of which four were new

to the island. The pollinators of four of these Ficus species

were also collected. These included the pollinator of one

of the taxa endemic to the islands, which had not been

recorded previously. Plant collection codes (CH) and the

herbarium where the material is deposited are indicated.

The wasps are retained by the author.

F. pachyclada Bak. subsp. pachyclada

North of Magoudjou, near South African Research Farm, CHI (RUH),

10 m, free-standing among rocks in remnant woodland, 25-vii-90.

Previously recorded from Madagascar. Additional uncertain records,

based on juvenile specimens, from Grand Comore. Previous records cited

in Berg (1986).

F. bojeri Bak.

Near Dzamadjou, 6 km north of Moroni, CHI (RUH), 4 m, free-

standing tree along roadside, 22-vii-90; near Maoueni, CH3 (RUH), 9

m, strangler along roadside, figs in leaf axils and projecting from the

main trunk, 24-vii-90; near Maoueni, CH4 (RUH), 13 m, large buttressed

tree along roadside, 24-vii-90. Previously recorded from Madagascar,

Seychelles and other islands in the Comoros. Previous records cited by

Berg (1986). New records for Grand Comore.

F. sycomorus L.

Hadjambou, in the northeast of the island, CH5 (RUH), 6 m buttressed

tree in pasture, 23-vii-90. Pollinating wasp Ceratosolen arabicus Mayr,

as in mainland Africa. The non-pollinating agaonid Ceratosolen galili

also present. Widespread in Africa, Madagascar and other islands in

the Comoros. Previous records cited in Berg (1986) . New record for Grand

Comore.

F. tiliifolia Bak.

Mount Karthala, above Moroni, CH6 (RUH), at side of contour path

in disturbed forest, one of a group of six 3—4 m saplings, 26-vii-90.

Previously recorded from Madagascar and other islands in the Comoros.

Previous records cited by Berg (1986). New record for Grand Comore.

F. lutea Vahl

Maoueni, 5 km north of Moroni, CH7 (RUH), small strangler in

remnant woodland, 22-vii-90; Maoneni-Grill forest, CH8 (RUH), large

free-standing tree in cleared pasture near forest, 24-vii-90. Pollinating

wasp Allotriozoon heterandromorphum Grandi, as in mainland Africa.

Widespread in Africa, Madagascar and other Indian Ocean islands.

Previous records cited by Berg (1986).

F. rubra Vahl

Maoueni, 5 km north of Moroni, CH9 (RUH), small strangler,

22-vii-90; Maoneni-Grill forest, CH10 (RUH), small strangler, figs on

pedicels in leaf axils and more or less sessile from major branches,

24-vii-90. Pollinating wasp Nigeriella avicola Wiebes, as on Aldabra (Wie-

bes 1975); southeast coast near Chamou Beach, CHII (RUH), large stran-

gler on baobab, 27-vii-90. Recorded from Madagascar, the Comoros and

other Indian Ocean Islands. Previous records cited by Berg (1986).

F. antandronarum (H. Perrier) C.C. Berg subsp. bemardii

C.C. Berg

Maoneni-Grill forest, CH12 (RUH), small strangler, mature figs yellow.

Pollinating wasp Elisabethiella sp. indesc., 24-vii-90; Maoneni-Grill

forest, CH13 (RUH), small strangler, 24-vii-90. This subspecies is

restricted to the Comoro Islands. Previous records cited by Berg (1986).

New pollinator record for the subspecies and species (Wiebes & Compton

1990).

F. reflexa Thunb. subsp. aldabrensis (Bak.) C.C. Berg

Road between Moroni and Mt Karthala, CHI4 (RUH), scrub/planta-

tion, at roadside, small strangler of Jack fruit, 26-vii-90; South Coast

road, near Sima Ambonii, CH15 (RUH), roadside, small strangler on

baobab, 25-vii-90. This subspecies previously recorded from Aldabra

and other islands in the Comoros. Previous records cited by Berg (1986).

New record for Grand Comore.

Eight of the nine Ficus species known from the Comoro

Islands were collected. The remaining species, F. kartha-

Bothalia 22,1 (1992)

47

lensis C.C. Berg, was originally described from Grand

Comore. Consequently, all the fig species known from the

Comoros have now been recorded from Grand Comore.

The natural vegetation of Grand Comore consists of

coastal and upland forest zones and a small area of heath-

land at the summit of Mt Karthala (Bijnens et al. 1987).

In the lowlands, most of the original forest cover on the

island has been converted to agriculture. The extent of

native tree removal at lower altitudes is nonetheless

variable, and F. sycomorus is not uncommon, either in

remnant patches of disturbed forest or growing in pastures.

The smaller strangler figs also persist at lower altitudes,

if suitable host trees remain. At higher altitudes, forest

cover is extensive on the active volcano Mt Karthala and

there is a remnant Maoeni-Grill forest. F. tiliifolia and F.

antandronarum subsp. bemardii were detected only in

these areas.

ACKNOWLEDGEMENTS

The author would like to thank Vanessa Rashbrook for

her assistance in the field, Marco Boni (South African

Consul) for providing logistical assistance and Prof. C.C.

Berg for identifying the plant material.

REFERENCES

BERG, C.C. 1986. The Ficus species (Moraceae) of Madagascar and the

Comoro Islands. Bulletin du Museum national d'Histoire naturelle

Section B Adansonia 8: 17—55.

BIJNENS, L., STEVENS, J., JANSSENS, L. & LOUETTE, M. 1987.

Community structure of Grand Comoro land birds with special

reference to the ecology of congeneric species. Revue de Zoologie

africaine 101: 221—232.

WIEBES, J.T. 1974. Species of Agaon Dalman and Allotriozoon Grandi

from Africa and Malagasy (Hymenoptera: Chalcidoidea, Agaoni-

dae). Zoologische Mededelingen 48: 123—143.

WIEBES, J.T. 1975. Fig insects from Aldabra (Hymenoptera: Chal-

cidoidea). Zoologische Mededelingen 49: 225—239.

WIEBES, J.T. & COMPTON, S.G. 1990. Agaonidae (Hymenoptera:

Chalcidoidea) and Ficus (Moraceae): fig wasps and their figs,

VI (Africa concluded). Proceedings Koninklijke Nederlandse

Akademie van Wetenschappen 93: 203—222.

S.G. COMPTON*

* Department of Zoology and Entomology, Rhodes University,

Grahamstown 6140.

MS. received: 1991-06-24.

FABACEAE

A NEW SPECIES OF PRIESTLEYA FROM THE SOUTHWESTERN CAPE

Priestleya boucheri Oliver & Fellingham, sp. nov.,

in genere singularis propter florescentias occultas foliis

subinvolucratis, bracteas aurantiacas, flores grandes,

calycem subinaequalem, vexillum elongatum reflexum.

TYPE. — Cape, 3418 (Simonstown): Grabouw area,

Kogelberg Reserve, Five Beacon Ridge, summit of ridge,

1 160 m, 20 May 1989, Boucher & Oliver 5531 (STE,

holo. ; BOL, K, MO, PRE, S, iso.).

A low compact woody single- stemmed shrub up to 500

mm tall. Stem and branches leafy only in the ultimate

80—100 mm, the younger long villous, the older glabrous

with prominent leaf scars, distinctly 3-ridged below the

leaf scars with the main ridge below the leaf scar and the

two smaller lateral ridges below the stipules, the bark

yellowish brown and corky. Leaves spirally arranged,

imbricate, erect, incurved, subsessile, 26-50 x 12-17

mm, linear-elliptic to narrowly obovate, acute rarely

subobtuse, green but slightly glaucous, the younger silky

villous all over, noticeably shaggy-edged with the hairs

all pointing to the apex and longer on the adaxial surface,

soon becoming almost glabrous with some adpressed old

dark-brown hairs, margins yellow, veins yellowish, mucro

minute and reddish brown, venation pinnate, stomata

numerous and visible on both surfaces; stipules 2, minute,

enlarging alongside the leaf scars.

Inflorescences 2-flowered fascicles on lateral absolute

brachy blasts 1.0— 1.5 mm long, mostly aggregated into 4-

or 6-flowered synflorescences enclosed within the upper

leaves at the ends of the main branches, flowers erect and

arranged in a circle; pedicel 3 mm long, pubescent; bract

7.0 -8.0 x 6.5 mm, very broadly ovate, shortly acuminate,

at first creamy-green soon turning papery and yellow-

brown, long ciliate, abaxially sparsely villous, adaxially

villous, clasping the base of the calyx and the pedicel.

Calyx : tube 7.5—10.0 x 5.0 mm, pale creamy-green,

becoming papery and yellow-brown; lobes 5.0— 5.5 mm

long, the lowest subequal to or 0.5 mm longer than the

others, the upper two lobes fused more than the others,

free distally for 1 mm, darker in colour than the tube,

villous, long ciliate. Petals : standard 25 x 6 mm, ovate-

elliptic, reflexed over the fused calyx lobes in the mature

flower, claw 4x2 mm with 2 basal lateral lobes 3 mm

long, separated abaxially by 2 ridges and a median

channel; alae 24 x 4 mm, oblong, falcate, obtuse, claw

3.5 mm long, lobe with an internal thickening in the upper

part; keel 25.0 x 4.5 mm, navicular, acute, claw 2 mm

long, lobes connate above for 2/3 and below for '/2 their

length. Stamens diadelphous, vexillary filament free, 26

mm long, the others connate for about 12 mm into a tube

thickened at the base and with 2 knobs adjoining the free

filament, the longest connate filament 28 mm and the

shortest 21 mm long; anthers ± 1 mm long. Ovary + 6

mm long, obliquely narrowly ovate, long, silky, villous,

with forward pointing hairs; style 21 mm long, glabrous;

stigma simple. Fruit 25 x 9 x 5 mm, including the

remnant style base, 6-seeded, villous with appressed hairs

pointing towards the apex, golden brown; seeds 4.5 x 2.0

mm, compressed, olive-brown with a white aril. Figure 7.

Diagnostic characters : P. boucheri is very distinct in

the genus on account of the hidden inflorescences, the large

flowers, the longer lower calyx lobe, the calyx being

yellow-brown at maturity and the large but narrow standard

which is reflexed over the calyx at maturity.

Discussion : this species was brought to our attention by

C. Boucher who has made a special study of the

48

Bothalia 22,1 (1992)

FIGURE 7. — Priestleya boucheri. A, flowering branch; B, terminal portion of A with the upper leaves opened outwards to reveal the synflores-

cence; C, single florescence, with the bracts removed, on an absolute brachyblast; D, flower; E, calyx laid out (upper lobes to the right);

F, base of flower with the calyx removed; G, standard laid out flat with the break indicating the reflexion zone; H, wing, lateral outside

view; I, keel, adaxial & lateral views; J, androecium; K, gynoecium; L, fruit; M, seed, x 6. All drawn from the type, Boucher & Oliver

5531. A, B, x 1; C-E, X 3; F, x 6: G-L, x 3; M, x 6.

Bothalia 22,1 (1992)

49

vegetation and flora of the Kogelberg Reserve (Boucher

1977) and is currently involved with sensitive conserva-

tion matters in the area. His collection, made late in the

year, was in fruit and remained unidentifiable, but tenta-

tively placed near the genus Liparia. Subsequently further

material, which was in a young flowering stage, was

collected by D. le Mai tre in April during a survey of rare

species in the Kogelberg area for the Department of

Forestry. This enabled us to ascertain that the collections

represented a remarkable new species quite unlike either

of the two species of Liparia (Bos 1967). Material with

fully open flowers was then obtained in late May to analyse

in detail.

Knowing that the material was collected in the Kogel-

berg area, which had been visited many times by that

veteran plant collector Thomas Stokoe during the period

1920 to 1955, a search was made for a collection of his

in BOL and NBG including SAM. Two fruiting twigs came

to light under the incertae of Priestleya in SAM, one col-

lected in 1944 from just ‘Kogelberg’, the other in 1953

recorded from Five Beacon Ridge. While reading through

the letters of Stokoe to the late W. & M. Cloete of Klein-

mond [in the possession of EGHO], a reference was found,

dated 1953, to an unknown Priestleya which he had not

been able to collect in flower despite several attempts to

do so. He mentioned it as growing on the crest of the Five

Beacon Ridge near Kogelberg and a sketch of the locality

was given. From this sketch it is clear that our type

collection and Stokoe’s note refer exactly to the same

population.

Two distinct populations of this species are known to

exist at present (Figure 8), separated by only 0.5 km. The

type population on Five Beacon Ridge contained eight

plants, two of which were seedlings. The population on

the ridge towards Kogelberg {Le Mai tre 401; Oliver 9139)

FIGURE 8.— Known distribution of Priestleya boucheri.

contained 10 plants, two of which were seedlings. In a

detailed note accompanying his collection, Le Mai tre

notes that the fire history showed that the area had been

burnt in March 1976 and therefore the plants were probably

13 years old in 1989. He further records that R.A. Haynes

and F. J. Kruger, also of the Department of Forestry, found

another population of 10 plants to the southwest in 1981,

but that only one dead plant could be located in 1986 due

to a fire a few years earlier. All the populations were

destroyed in a lightning-induced fire in February 1990 and

will therefore have to regenerate from seed.

The large, flat, hairy leaves give the plants a remarkably

proteoid facies similar to Protea stokoei, P. caespitosa and

Leucadendron gandogeri which all grow in the immediate

vicinity. The flowers are hidden from view from a dis-

tance and are only seen when the shrub is viewed from

above, very much like the condition occurring in many

species of Leucadendron, This is due to the large sub-

terminal leaves which curve over the synflorescences. The

bud and fruit stages are completely hidden. In the bud stage

the flowers are erect with the standard covering the alae

and the keel. In the mature flower the position of the

standard becomes horizontal, i.e. perpendicular to the rest

of the flower. This position is reached after a 180° bend

occurring closely over the fused calyx lobes followed by

a second, distally from and close to the first but 90° in

the opposite direction (Figure ID). With the standards of

all flowers in the synflorescence assuming this position,

the ‘involucral’ leaves are pushed open to expose the

flowers in a cuplike formation (Figure IB). After pollina-

tion, the second bend in the standard is straightened again,

allowing the standard to be totally reflexed over the calyx

lobes (Figure IF). The involucral leaves then close over

the synflorescence and hide the developing fruits com-

pletely.

The pollinating agent was not noted during the visit to

the flowering population. However, the large size and

whole arrangement of the flowers noted above strongly

suggests visitation by a large bee.

The floral characters place the species very close to

Liparia and in particular to L. parva. That genus is charac-

terized by large narrow flowers each in the axil of a large

petaloid bract, with the calyx very unequal, the lowest lobe

being subpetaloid and much longer than the others. In the

new species the calyx lobe is subequal to 0.5 mm longer

than the other lobes. In L. splendens, however, the keel

petals are held together by a most remarkable interlocking

‘catch’ system of the alae. The flower colour and structure

is very similar to the lemon-yellow-flowered L. parx’a

which is endemic in the southern Cape Peninsula, whereas

the more widespread L. splendens has bright orange to

reddish orange flowers. In Priestleya and Xiphotheca the

flowers are generally smaller and have a very broad

rounded standard and usually equal calyx lobes.

In the plant architecture and structure of the inflores-

cence this species is quite unlike Liparia and is identical

to the genus Priestleya sens. str. (= Priestleya section

Priestleya). In Liparia the flowers are borne in the axils

of leaf-like bracts at the ends of main or leafy lateral

branches, forming a simple condensed racemose flores-

cence. This florescence, which in most cases is fairly

heavy due to the number of large flowers, hangs down-

50

Bothalia 22,1 (1992)

wards in a nodding fashion. It is also non-innovating and

so further growth of the axis has to be initiated from an

axillary bud on the upper side of the curved branch just

below the florescence.

In the new species the flowers are borne in pairs at the

ends of lateral absolute brachyblasts, i.e. extremely

shortened and leafless branchlets. These 2-flowered flores-

cences are grouped mostly in pairs or threes at the ends

of the main branches to form a synflorescence which is

enclosed within the involucre-like upper leaves (Figure 9).

The florescences are non-innovating, but this does not

affect the growth pattern of the plant as growth continues

from the apical bud of the main axis in the centre of the

synflorescence. On old branches the remains of some

brachyblasts lower down clearly indicate the position of

the synflorescences of previous years.

A detailed study of inflorescences in Priestleya and

related genera has shown that three basic types can be

distinguished (Schutte & Van Wyk in prep.): 1, simple

terminal or subterminal racemes with a rachis extension,

i.e. a sterile apical portion. This type can be many-flowered

( Liparia ) or few-flowered ( Priestleya hirsuta and related

species); 2, axillary 2- or 4-flowered fascicles, also with

a sterile apical portion. The flowers are arranged in op-

posite pairs (decussate). This inflorescence type occurs

in Priestleya calycina and related species and also P.

boucheri (see Figure 7C where the small terminal sterile

part can be distinguished); 3, axillary 2-flowered fascicles

without a sterile apical portion, i.e. a simple determinate

reduced inflorescence. This type is superficially similar

to the previous type, but can easily be distinguished by

the absence of a terminal sterile part. Section Anisothea

of Priestleya has this type of inflorescence and, together

with the non-intrusive base of the calyx and characteristic

combination of alkaloids, these provide convincing

supportive evidence for excluding the section from

Priestleya. This was done by Ecklon & Zeyher (1836) who

proposed the new generic name Xiphotheca.

Several species of Priestleya occur in the area around

Kogelberg Peak and they are very striking plants. The two

tall, almost tree-like species, P. calycina and P tomentosa

(X. villosa) have silvery leaves and conspicuous heads of

bright yellow flowers. An as-yet-undescribed species

occurs on the southern slopes of Five Beacon Ridge in

Spinnekopsneskloof and was also collected by Boucher

{Boucher 1812).

Chemical analyses of samples from both populations on

the Kogelberg clearly indicate that this species contains

alkaloids characteristic of Priestleya sens. str. and Liparia

parva and none of the unique compounds found in

Xiphotheca (Van Wyk et al. 1991). The new species has

large amounts of quinolizidine alkaloids such as sparteine,

11,12-dehydrosparteine, lupanine, isolupanine and 13-

hydroxylupanine. The relative quantities of these alkaloids

are closely similar to the combinations found in other

species of Priestleya (virtually identical to that found in

P. latifolia, for example). Liparia splendens differs from

L. parva in the much higher proportion of ammodendrine,

but otherwise the alkaloids of Liparia are similar to those

found in Priestleya. In contrast, Priestleya section

Anisothea {— Xiphotheca) has a unique combination of

alkaloids not found in Priestleya sens, str., i.e. anabasine

i

FIGURE 9. — Diagram of the structure of a synflorescence in Priestleya

boucheri consisting of three two-flowered florescences on lateral,

non-innovating brachyblasts— dotted lines represent extended axes.

(a bipiperidyl alkaloid) and lupinine (a bicyclic quinolizi-

dine).

The position of this new species is thus anomalous

on account of its liparioid flowers and it shows the con-

nection between the two genera. Indeed, the status of

Liparia as a genus distinct from Priestleya sens. str. can

be seriously questioned. Generic delimitation in the tribe

Liparieae is currently under critical revision (Schutte &

Van Wyk in prep.) because the present system does

not reflect major discontinuities in intergeneric relation-

ships.

Specimens examined

CAPE.— 3418 (Simonstown): Kogelberg Reserve, Five Beacon Ridge,

1 160 m, 8-10-1980, mature fruiting, (— BB), Boucher 4975 (STE); ibid.,

20-5-1989, flowering, Boucher & Oliver 5531 (BOL, K, MO, PRE, S,

STE); ridge between Kogelberg and Five Beacon Ridge, 1 120 m,

22-4-1986, in bud, (-BB), Le Maine 401 (PRE, STE); ibid., 1 130 m,

25-5-1989, flowering, Oliver 9139 (PRE, STE): Kogelberg, 11-1944,

mature fruiting, (— BB), Stokoe in SAM 56330 (SAM); Five Beacon

Ridge near Kogelberg, 9-1953, young fruit, (-BB), Stokoe in SAM 65718

(SAM).

REFERENCES

BOS, J.J. 1967. The genus Liparia L. (PAP). Journal of South African

Botany 33: 269-292.

Bothalia 22,1 (1992)

51

BOUCHER, C. 1977. Provisional checklist of flowering plants and ferns

in the Cape Hangklip area. Journal of South African Botany 43:

57-80.

ECKLON, C.F. & ZEYHER, C. 1836. Enumeratio plantarum afri-

cae australis extratropicae 2: 166. Perthes & Besser, Ham-

burg.

SCHUTTE, A.L. & VAN WYK, B.-E. in prep. The re-instatement of

the genus Xiphotheca Eckl. & Zeyh. (Fabaceae). Taxon.

VAN WYK, B.-E., VERDOORN, G.H., GREINWALD, R. & BACH-

MANN, P. 1991. Taxonomic significance of major alkaloids in

the genus Priestleya. Biochemical Systematics and Ecology 19:

595-598.

E.G.H. OLIVER* A.C. FELLINGHAM* and B.-E. VAN WYK**

* Stellenbosch Herbarium, National Botanical Institute, P.O. Box 471,

Stellenbosch 7601.

** Department of Botany, Rand Afrikaans University, P.O. Box 524, Jo-

hannesburg 2000.

MS. received: 1991-08-27.

•-

Bothalia 22,1: 53-58 (1992)

The occurrence in southern Africa of the hepatic, Symphyogyna

brasiliensis (Pallaviciniaceae)

S.M. PEROLD*

Keywords: hepatic, Metzgeriales, Pallaviciniaceae, southern Africa, Symphyogyna brasiliensis, S. lehmanniana

ABSTRACT

In Magill & Schelpe (1979) Symphyogyna lehmanniana is confirmed as occurring in southern Africa. Subsequently, however,

Grolle (1980) reported that this name, which has been applied to a liverwort widespread in Africa, is a synonym of S. brasiliensis.

To draw attention to this synonymy, a description of S. brasiliensis , illustrated with photographs, is given here, as well as

an account of its distinguishing features and its distribution.

UITTREKSEL

In Magill & Schelpe (1979) word die teenwoordigheid van Symphyogyna lehmanniana in suidelike Afrika bevestig. Daarna

berig Grolle (1980) egter dat hierdie naam, wat vir ’n wydverspreide lewermos in Afrika gebruik is, ’n sinoniem is van

S. brasiliensis. Om die aandag op hierdie sinonimie te vestig, word ’n beskrywing van S. brasiliensis , geillustreer met foto’s,

sowel as inligting oor sy onderskeidende kenmerke en verspreiding hier gegee.

INTRODUCTION

In Magill and Schelpe’s (1979) checklist, Symphyogyna

lehmanniana is listed as one of the species of Symphyo-

gyna occurring in southern Africa.

Subsequently, however, Grolle (1980) concluded that this

name, applied to a liverwort reported to be widespread

in Africa, is a synonym of S. brasiliensis. During the

course of his investigations Grolle had compared speci-

mens assigned to both species, their types, as well as

spores, and could find no differences. He had also

examined a number of other specimens, previously

assigned to several so-called different species of Symphyo-

gyna and Pallavicinia, all of which he eventually placed

in synonymy under S. brasiliensis. Three of these species,

namely S. lehmanniana, S. valida and Pallavicinia

capensis, have types which were collected in southern

Africa. The following description of 5. brasiliensis, which

is illustrated with photographs, is given to draw attention

to the above information.

The two Metzgerialean genera, Symphyogyna and

Pallavicinia were previously placed in the family

Dilaenaceae (Dum.) Warnst. by Arnell (1963) and by

Schuster (1964, 1982), but this name is illegitimate (Grolle

1972). They are now classified by Grolle (1983) in the

family Pallaviciniaceae Migula and are placed in different

subfamilies, the Symphyogynoideae (Trev.) Schust. and

the Pallavicinioideae (Migula) Grolle respectively. They

are frondose liverworts, characterized by thinly winged

thalli with a thicker midrib, containing a median con-

ducting strand of elongate, tapering cells with thickened,

perforated walls. The thalli are often stipitate and are either

procumbent or erect; their margins are entire, dentate,

undulate or lobate; the epidermal cells are ± rectangular

and lack nodular thickenings and the capsules are

* National Botanical Institute, Private Bag X101, Pretoria 0001.

MS received: 1991-11-11.

elongated. The two genera are separated by the type of pro-

tection provided for the archegonia and young sporophyte:

in Pallavicinia the archegonia, and after fertilization, the

pseudoperianth and capsule, are surrounded by a short

tubular or annular involucre; in Symphyogyna the arche-

gonia are subtended by a laciniate, scale-like involucre with

the margins of the latter free and directed forward, no

pseudoperianth is developed after fertilization and the

young sporophyte is enclosed only by the shoot-calyptra,

which has a cluster of unfertilized archegonia at its tip.

Symphyogyna brasiliensis Nees & Mont, in Annales

des sciences naturelles, Botanique ser. II. 5: 67 (1836).

Type: Brazil, Est. Minas Geraes, Sao Joao Batista, Martius

s.n. (STR, lecto.).

S. lehmanniana (Mont. & Nees) Gottsche et al.: 483 (1846). Type:

Cape Province, Table Mountain, ‘in Promontorio Bonae Spei in vertice

ad latus boreale montis Tabularis locis umbrosis’, Ecklon s.n. (STR,

lecto.).

S. tenuicostata Steph.: 306 (1895). Type: Tanzania, Usambara, Holst

688 (JE, W, iso.).

S. valida Steph.: 69 (1917); syn. fide S. Arnell: 111 (1963). Type:

Zululand, Eshowe, Haakon s.n. (JE, M, UPS, iso.).

Pallavicinia capensis S. Arnell: 177 (1954); syn. fide S. Arnell: 111

(1963). Type: Cape Province, Knysna, Gouna (not Guona) Forest, S.

Arnell 1769 (PRE, holo. ! ; UPS, iso.).

Terricolous, growing on damp soil; thallose, prostrate

and creeping (Figures LA; 2B), in crowded, overlying

mats, bright green when fresh, rarely developing a purplish

or reddish tinge along margins and over costa, linear,

simple or dichotomously branched, sometimes with

ventral intercalary branching, medium-sized to large,

10—20 x 2— 3 mm, 280 gm thick over ventrally bulging

costa, from which arise pale brownish, translucent

rhizoids, smooth and mostly ±12.5 pm wide; apex entire

or with a shallow notch (Figure IB), the two halves very

slightly overlapping in centre, bearing numerous 2-celled

slime papillae, ± 65 x 20 pm (Figure IB), these also

present ventrally near the apex; margins entire, undulate,

without slime papillae. Wings translucent, unistratose,

54

Bothalia 22,1 (1992)

FIGURE 1 .—Symphyogyna brasiliensis. A, male thallus with androecia; B, apical notch of thallus with slime papillae (indicated by an arrow);

C, cross section of thallus showing costa with central conducting strand and unistratose lateral wings; D, marginal and laminal cells seen

from above; E, antheridium; F, oil bodies and chloroplasts visible inside cells; G, scale-like involucre which partly covers antheridium.

A-G, Glen 2687. A, x 22; B, F, x 350; C, E, G, x 87; D, x 175.

Bothalia 22.1 (1992)

55

FIGURE 2. — Symphyogyna brasiliensis. A, female thallus with cluster of archegonia; B— E, various stages in maturation of capsule: B, C, young

capsule enclosed in calyptra; D, capsule and seta emerged from calyptra; E, capsule with ripe spores and long, tenuous seta; F, cross

section of seta; G, cross section through middle of shoot-calyptra. A-G, Glen 2687. A, B, E, x 24; C, x 34; D, x 30; F, x 110; G, x 40.

56

Bothalia 22,1 (1992)

FIGURE 3. — Symphyogyna brasiliensis. A, laciniate scale-like involucre which subtends each archegonium; B, cells in capsule wall; C, J, elaters;

D, cross section of capsule wall, only outer cells shown; E— I, spores; . A— J, Glen 2687. A, x 35; B, D, x 175; C, x 58; E, F, x

1750; G, H, x 1630; I, x 1000; J, x 550. C, E-H, SEM micrographs; A, B, D, I, J, LM photographs, all taken by the author.

Bothalia 22,1 (1992)

57

medianly 2 or 3 cells thick, rather abruptly grading into

the roughly triangular costa; outer row of cells along

margin generally rectangular, ± 50.0 x 27.5 /xm, other-

wise cells 5- or 6-sided, 50—75 x 30.0—37.5 /xm (Figure

ID) and ± 40 /xm thick in section, containing numerous

chloroplasts and fusiform or ± triangular oil bodies

(Figure IF). Costa with a central conducting strand (Figure

1C), ± 45 /xm wide, brownish to purple, formed of

strongly elongated, small cells, ± 10.0 x 7.5 /xm, with

straight, thickened walls, the enclosing cells up to 10 or

11 rows deep, hyaline, much larger, mostly 50.0 x 27.5

/xm, irregular in shape, walls thin and wavy.

Dioicous. Androecia usually in 2 crowded (Figure 1A),

more or less parallel rows over the costa and lateral to it

on either side, each short-stalked antheridium (Figure IE)

partially covered by an irregularly shaped, dentate or en-

tire, forwardly-directed, scale-like involucre (Figure 1G),

± 800 x 450 /xm, its cells mostly 67.5 x 45.0 /xm, with

age antheridia turn yellow and cell walls of scales dar-

ken. Gynoecia generally one to several per frond, in

acropetal succession, situated dorsally over the costa and

containing several archegonia in a cluster (Figure 1A),

which is subtended by a posteriorly inserted, scale-like

involucre (Figure 2B), sometimes partly double toward the

base, generally deeply laciniate (Figure 3A), 850-110 x

500—750 /xm, its cells + 72.5 x 50.0 /xm. Calyptra thick-

ening and enlarging into a fleshy shoot-calyptra, up to 3

mm long and as much as 10 cell rows or 260 /xm thick

in cross section (Figure 2G), with several unfertilized

archegonia at the tip (Figure 2C), before it is perforated

by the capsule (Figure 2D). Capsule cylindrical (Figure

2E), 2250 x 650 /xm, opening with 4 valves, wall brown,

several cell layers thick, outer cells irregularly elongate

(Figure 3B), ± 200 x 20-30 /xm, with cell walls evenly

thickened (Figure 3D), inner cells thin-walled. Seta when

young, erect, ± 290 /xm in diameter, with ± 18 cortical

cell rows and ± 16 medullary cells (Figure 2F), eventually

becoming long and tenuous (Figure 2E). Spores light

brown, + globular, 25—30 /xm in diameter, ornamenta-

tion with low, irregularly branched, short or long, curly

ridges (Figure 3E, G, H), on proximal face a discrete

round area with the ornamentation much more densely

arranged (Figure 3F). Elaters bright brown, tapering to

the ends, up to 195 x 7.5 /xm wide in the centre, 2-spiral

(Figure 3C, J).

Symphyogyna brasiliensis is widespread throughout

Africa, occurring in Sierra Leone, Ghana (Jones & Har-

rington 1983), the Cameroons, Zaire, Rwanda, Burundi,

Tanzania, Angola, Zimbabwe, as well as on the islands

of Madagascar, Mauritius, Reunion, Saint-Benoit, St

Helena and Ascension (Grolle 1980; Vana et al. 1979,

reported as S. lehmanniana) . In the Neotropics it is known

from Mexico to Bolivia, Uruguay and Brazil (Grolle 1980).

Gradstein etal. (1983) reported it to be widely distributed

in the tropical mountains of the two continents, Africa and

South America, at altitudes between 1 500 and 3 000 m,

whereas in subtropical Brazil, the Cape Province and on

the Galapagos Islands it descends to near sea level.

In southern Africa it has been collected in northern,

eastern, central and southern Transvaal, in Swaziland,

Natal and Zululand, as well as in the southwestern and

southern Cape (Figure 4).

FIGURE 4. — The distribution of Symphyogyna brasiliensis in southern

Africa.

Symphyogyna brasiliensis is terrestrial, growing on

damp streambanks, along footpaths or in forested areas.

It differs from the other species in the genus by its procum-

bent habit, by the entire margins of the thallus, by its

slightly larger spore size and by its somewhat finer spore

ornamentation. S. filicum Nadeaud from Cameroon Moun-

tain is similar to S. brasiliensis, except that the small cells

in the central strand are thin-walled (Jones 1990) and the

spore sculpture is different. It was recently segregated as

a separate genus, Symphyogynopsis , by Grolle in Grolle

& Piippo (1986). Symphyogyna volkensii Steph. is more

robust (Vanden Berghen 1965) than 5. brasiliensis and has

scattered slime papillae along the wing margins and

different spore ornamentation (Grolle pers. comm.).

As vegetative propagation is unknown, dispersal of 5.

brasiliensis must be by long range aerial transport of its

spores. Its presence on young volcanic islands would

support this assumption.

SPECIMENS EXAMINED

TRANSVAAL. — 2229 (Waterpoort): Soutpansberg, Lokorhela 793

Farm, + 500 m upstream of waterfall, streambank in forest, 1 320 m

altitude, (— DD), Glen 2687 (PRE). 2329 (Petersburg): The Downs,

Petersburg, (—CD), Junod 4011b (PRE); Woodbush For. Res., Magoebas-

kloof, on roadside embankment, (— DD), H. Anderson CH13496 (PRE);

Houtboschdorp, (— DD), D.R.J. Van Vuuren 1478a (PRE). 2430 (Pilgrim’s

Rest): Mariepskop, immediately W of dam in Klaserie River, on wet

earth bank, (— DB), Vorster 511, 575 (PRE); Mariepskop, near forestry

station in riverine montane forest, on shale embankment, (— DB), Vorster

783 (PRE). 2527 (Rustenburg): Buffelspoort, Krom River, (—CD),

Bottomley & Doidge CH3603 (PRE); Bokfontein, Farm Jacksonstuin,

(—DA), Mogg CH13173 (PRE). 2528 (Pretoria): Magaliesberg, Boeken-

houts kloof, (-CB), Mogg CH1562, CH2858 (PRE). 2529 (Witbank):

Dist. Verena, 24 km E of Bronkhorstspruit, on road to Susterstroom,

Farm Klipfontein, in gulley, (— CA), S.M. Perold 452 (PRE); Klipfon-

tein no. 87, in open glades at streamside, (-CC), Mogg 12561 (PRE).

2530 (Lydenburg): 10 km E of Lydenburg, in road stone quarry, off

Sabie/Lydenburg road, (— AB), Rankin 53 (PRE); Rooiwal, in ravine,

(— BD), Bosnian 3178 (PRE); Rosehaugh, (— BD), T.R. Sim 7585 (PRE);

Kaapsche Hoop, (— DB), V. A. Wager 65 (PRE). 2531 (Komatipoort):

Barberton, (-CC), Hendry 2 (PRE). 2627 (Potchefstroom): Witpoortjie,

near Johannesburg, (-BB), Moss CH1465 (PRE); Witpoortjie kloof, on

damp bank, (— BB), C.S. & M. Moss CH1479 (PRE). 2628 (Johannes-

burg): Johannesburg, (— AA), Edwards CH1461 (PRE). 2730 (Vryheid):

Wakkerstroom Dist., Farm Oshoek, (-AD), Glen 1676 (PRE).

SWAZILAND.— 2531(Komatipoort): Horo Forest, (— CB), V A. Wager

92 (PRE). 2631 (Mbabane): Mbabane, (—AC), Rodin CH3975 (PRE);

Mbabane, (—AC), Edwards CH1447 (PRE).

58

NATAL. — 2730 (Vryheid): Scheepers’ Nek, (—DC), T.R. Sim 8231

(PRE); Vryheid, (— DD), T.R. Sim CH1452 (PRE). 2831 (Nkandla):

Eshowe, Signal Hill, (—CD), Van der Plank CH1446 (PRE). 2929

(Underberg): 0.5 km N of Tabamhlope Police Sta., towards Draycott,

on vertical to overhanging streambank, (—BA), Glen 1693 (PRE); Rosetta,

(-BD), T.R. Sim CH1470( PRE); 1 km beyond Sani Pass Hotel, streamlet

at roadside, (-CB), S. M. Perold 2501 (PRE). 2930 (Pietermaritzburg):

Buccleuch, (—AD), T.R. Sim CH1471 (PRE); Ndwedwe, Zwatini Kloof

forest, on moist cliff, (-BD), Strey 7757 (PRE); Pietermaritzburg, Chase

Valley, (— CB), Bews CH4466 (PRE); Zwaartkop, (— CB), T.R. Sim

CH1482 (PRE); Hilton Road, (— CB), T.R. Sim CH1454, CH1469,

CH1483 (PRE); Sweetwaters, (— CB), T.R. Sim CH1466, CH1476 (PRE);

Inchanga, at stream, (—DA), T.R. Sim CH1451, CH1468, CH1484, CH1486

(PRE); New Germany, (-DD), Moonsammy 21 (PRE); Van der Byl 21

CH1460 (PRE). 3029 (Kokstad): Kingston, (— AA), T.R. Sim CH1485

(PRE).

CAPE. — 3318 (Cape Town): Table Mtn, Disa Ravine, alt 2500 ft.,

(-CD), T.R. Sim CH4414 (PRE); Stellenbosch, (-DD), Duthie CH1459

(PRE). 3319 (Worcester): Kloof off Bain’s Kloof, (— CA), Primes CH1475

(PRE); Groot Drakenstein Mts, (— CC), Primes CH4471 (PRE); Du Toit’s

River bridge, at rock face next to road, (— CC), S.M. Perold 1151 (PRE);

4 km N of Villiersdorp, Elands River road, Du Toitsberge, near water-

fall, (—CD), S.M. Perold 624 (PRE). 3320 (Montagu): warm bath at

Uitvlugt, near Barrydale, (—DC), Muir CH4103 (PRE). 3321 (Ladismith):

Garcia’s Pass, (— CC), Muir CH1467 (PRE). 3322 (Oudtshoom): George,

(-CD), H.A. Wager CH1449 (PRE), Wager s.n., 41 (PRE). 3323

(Willowmore): Keurbooms Rivier, (-CD), Burn Davy 17033 (PRE).

3418 (Simonstown): Kalk Bay, (— AB), Potts CH1473 (PRE); Gordon’s

Bay, Felswande siidlich dem Dorfe, (— BB), Cholnoky 386 (PRE); Kogel-

berg near Gordon’s Bay, (— BB), Mogg CH938 (PRE). 3419 (Caledon):

Hermanus Dist., Riviera Kloof, (-AC), Louwrens CH2893 (PRE);

Voelgat, (—AC), Louwrens CH3715 (PRE); Mossel River, (—AD), Potts

CH4475 (PRE); Greyton Kloof, on earth bank, (—BA), S.M. Perold 1164

(PRE). 3423 (Knysna): Knysna, Gouna (not Guona) Forest, (-AA),

Amell 1769 (type of Pallavicinia capensis ), 1790 (PRE); Knysna, (— AA),

Amell 1476 (PRE); Gouna For. Res., N of Knysna, on damp earth bank

at stream, (—A A), S.M. Perold 904 (PRE).

ACKNOWLEDGEMENTS

I wish to express my gratitude to Dr E.O. Campbell,

New Zealand, and Dr R. Grolle, Germany, for critically

reading the manuscript and for their helpful suggestions;

also to Dr H.F. Glen, NBI, for collecting fresh material

that could be photographed. Thanks also to Mrs A.

Romanowski, photographer, for developing and printing

Bothalia 22,1 (1992)

the photographs and Mrs J. Mulvenna, typist, for her

valued contribution.

REFERENCES

ARNELL, S. 1954. A new species of Hepaticae from South Africa. Revue

bryologique et lichenologique 23: 173-179.

ARNELL, S. 1963. Hepaticae of South Africa, pp. 411. Swedish Natural

Science Council, Stockholm.

GOTTSCHE, C.M., LINDENBERG, J.B.G. & NEES ab ESENBECK,

C.G. 1844-1847. Synopsis hepaticarum. pp. 835. Hamburg,

Meissner. Reprinted 1967, Cramer, Lehre.

GRADSTEIN, S.R., POCS, T. & VA&\, J. 1983. Disjunct hepaticae

in tropical America and Africa. Acta Botanica Hungarica 29:

127-171.

GROLLE, R. 1972. Die Namen der Familien und Unterfamilien der

Lebermoose (Hepaticopsida). Journal of Bryology 7: 201—236.

GROLLE, R. 1980. Miscellanea hepaticologica 201—210. Journal of

Bryology 11: 325—334.

GROLLE, R. 1983. Nomina generica hepaticarum; references, types and

synonymies. Acta botanica Fennica 121: 1—62.

GROLLE, R. & PIIPPO, S. 1986. Bryophyte flora of the Huon Penin-

sula, Papua New Guinea. XVI. Acta botanica Fennica 133: 59 -79.

JONES, E.W. 1990. African hepatics XL. An artificial key to the genera

of African hepatics. Journal of Bryology 16: 9-40.

JONES, E.W. & HARRINGTON, A.J. 1983. The hepatics of Sierra Leone

and Ghana. Bulletin of the British Museum (Natural History),

(Botany) 11: 215-289.

MAGILL, R.E. & SCHELPE, E.A. 1979. The bryophytes of southern

Africa. An annotated checklist. Memoirs of the Botanical Survey

of South Africa No. 43: 1—39.

NEES AB ESENBECK, C.G. & MONTAGNE, C. 1836. Jungerman-

niearum herbarii Montagneani species exposuerunt C.G. Nees

ab Esenbeck et C. Montagne. Annales des sciences naturelles,

Botanique ser. II. 5 : 52 -72.

SCHUSTER, R.M. 1964. Studies on antipodal Hepaticae. IV. Metz-

geriales. Journal of the Hattori Botanical Laboratory 27: 183 —216.

SCHUSTER, R.M. 1982. Studies on Hepaticae. LIX. On Sandeothal-

lus Schust., gen. n. and the classification of the Metzgeriales.

Nova Hedwigia 36: 1—16.

STEPHANI, F. 1895. Hepaticae africanae. Botanische Jahrbiicher 20:

299-321.

STEPHANI, F. 1917. Species hepaticarum 6: 1-693.

VA&k, J., POCS,T. & DE SLOOVER, J.L. 1979. Hepatiques d’Afrique

tropicale. Lejeunia 98: 1—23.

VANDEN BERGHEN, C. 1965. Hepatiques recoltees par le Dr J.-J.

Symoens dans la region peri-tanganyikaise. Bulletin Societe r.

de botanique de Belgique, Bruxelles 98: 129—174.

Bothalia 22,1: 59-75 (1992)

Morphology, evolution and taxonomy of Wachendorfia (Haemodoraceae)

N.A. HELME* and H.P. LINDER*

Keywords: Dilatris, evolution, Haemodoraceae, morphology, phylogeny, taxonomy, Wachendorfia

ABSTRACT

Wachendorfia Burm. is a small genus endemic to the Cape Floral Region. Four species are recognised in this study. Two

species were originally described by Burman in 1757 and these were followed by numerous other descriptions of what is

essentially one very variable species (W. paniculata Burm.). This variation is discussed and reasons are given as to why

the recognition of formal infraspecific taxa is inappropriate. Formal taxonomic descriptions, distribution maps and a key

to the species are provided. Rhizome morphology, leaf anatomy and pollen and seed coat structures were investigated and

illustrations are provided. A cladogram was inferred and this is consistent with an ecological speciation model for the genus.

The two species with the most restricted distribution ( W. brachyandra W.F. Barker and W parviflora W.F. Barker) are considered

to be the most recently evolved. Features of systematic and ecological interest (e.g. floral enantiomorphy) are discussed.

UITTREKSEL

Wachendorfia Burm. is ’n klein genus endemies aan die Kaapse Floristiese Streek. Vier spesies word in hierdie ondersoek

erken. Aanvanklik is twee spesies in 1757 deur Burman beskryf, en hulle is gevolg deur talle ander beskrywings van wat

in wese een baie varierende spesie ( W. paniculata Burm.) is. Hierdie variasie word bespreek en redes word verstrek waarom

die formele erkenning van infraspesifieke taksons onvanpas is. Formele taksonomiese beskrywings, verspreidingskaarte en

’n sleutel tot die spesies word verskaf. Risoommorfologie, blaaranatomie en stuifmeel- en saadhuidstrukture is ondersoek

en word gel llustreer. ’n Kladogram is afgelei en is in ooreenstemming met ’n ekologiese spesiasiemodel vir die genus. Die

twee spesies met die mees beperkte verspreiding (W brachyandra W.F. Barker en W parviflora W.F. Barker) word beskou

as die waarvan die evolusionere ontwikkeling mees onlangs plaasgevind het. Kenmerke van sistematiese en ekologiese belang

(bv. enansiomorfie by die blom) word bespreek.

INTRODUCTION

Wachendorfia Burm. is a small genus (four spp.) of the

Haemodoraceae (tribe Haemodoreae), restricted to the

Cape Floral Region of southern Africa ( sensu Bond &

Goldblatt 1984).

The Haemodoraceae is divided into two tribes: the

Haemodoreae and the Conostyleae (Geerinck 1969;

Hutchinson 1973). The southern African Haemodoraceae

{Barberetta Harv. , Dilatris Berg, and Wachendorfia) are

all members of the tribe Haemodoreae, which is defined

by the presence of two perianth whorls, a short or absent

tube and three or six stamens. The Conostyleae is

represented in Australia and North America and is charac-

terised by one perianth whorl, long, often hairy tubes, and

six stamens.

The presence of arylphenalenone pigments was hypothe-

sized by Simpson (1990) to be the most reliable synapo-

morphy defining the Haemodoraceae. These pigments

provide the rhizomes, and probably the flowers of Wachen-

dorfia and related genera, with their red component. In

addition, he suggested that 1, bifurcate cymes; 2, an

inferior ovary position; and 3, discoid or ovoid-globose

seeds may be further synapomorphies. However, Barberet-

ta does not possess a cyme (interpreted as an autapo-

morphy by Simpson (1990)) and a number of genera

including Wachendorfia have superior ovaries, a reversal

according to Simpson (1990). Almost all of the species

(63 out of 78) and the genera (12 out of 17) are found

* Bolus Herbarium, University of Cape Town, Rondebosch 7700.

MS. received: 1991-07-09.

in South America, southern Africa and Australia. The

family may therefore be interpreted as being essentially

Gondwanan.

Wachendorfia species have attractive yellow flowers,

with a form of enantiomorphy that manifests itself in the

production of both ‘left- and ‘right-handed’ floral forms

within a population. This form of enantiomorphy is

restricted to Wachendorfia and Barberetta (Ornduff &

Dulberger 1978). Floral enantiomorphy has been inter-

preted as a mechanism to increase intermorph (and thus

interplant) pollinations, ensuring effective outcrossing

(Wilson 1887; Ornduff & Dulberger 1978), as a means of

reducing pollen wastage or of preventing damage to the

gynoecium during pollinator visits. There is, however, still

much to be learnt concerning the evolutionary and

ecological significance of this trait. The fact that this

extremely unusual feature is most common in a small,

endemic fynbos genus makes the systematic study of this

group potentially fascinating. A sound taxonomic base is

essential if one wishes to use the genus for exploring some

of the intriguing evolutionary questions posed by the group,

yet the taxonomy is at present confused, with the species

limits of W paniculata vague, and several possible new

species unpublished. In this study we hope to come to an

understanding of species delimitation within the genus, and

increase our knowledge of the ecology and possible

evolutionary history of the species, with a view to high-

lighting areas of potentially rewarding systematic study.

METHODS

Plant morphology

Gross morphology was determined by field observation

and examination of extensive material lodged in BOL,

60

Bothalia 22,1 (1992)

NBG, PRE, SAM and STE. These specimens have all been

provided with determinavit labels. Rhizome and culm

(annual flowering stem) morphology was investigated using

fresh plants at different stages of development. Details

were observed with a Wild stereo dissecting microscope.

Thin sections were hand-cut with a scalpel and then stained

with iodine to investigate the presence of starch.

Pollen morphology

Pollen was taken from either herbarium specimens or

from live material collected in the field. For both light

microscopy (LM) and scanning electron microscopy

(SEM), pollen was acetolysed according to the method of

Erdtman (1960). Half the acetolysed pollen was mounted

in glycerine on a slide and sealed with wax for light micro-

scopy. The other half was mounted on aluminium stubs

TABLE 1. — The characters, character states and coding used in the

cladistic analysis. The first state is coded 0, the second, 1 and

the third, 2

Gross plant morphology:

1 Plant height greater than 1 m/less than 1 m

2 *Corm length greater than 30 mm/less than 30 mm

3 Leaf length greater than 500 mm/less than 500 mm

4 Leaf width greater than 50 mm/less than 50 mm

5 Leaves glabrous/with short and dense indumentum

6 Bracts scarious/herbaceous

7 Max. peduncle length greater than 100 mm/less than 100 mm

8 Cilia on the petal margins present/absent

9 Ratio stamen to tepal length: greater than 0.6/less than 0.6

10 Average minimum tepal width greater than or equal to 6 mm/

less than 6 mm

11 Max. tepal length greater than 20 mm/less than 20 mm

Leaf anatomical characters:

12 *Leaf chlorenchyma pallisade two cells wide/one cell wide/

absent

13 Sclerenchyma cap on the vascular bundles well developed/

poorly developed or absent

14 Mucilage canals along leaf margins/scattered in mesophyll

15 Cuticle thick/thin

16 Lignification of epidermis light/heavy

17 Guard cell with one lip/two lips

18 *Subsidiary cell kidney-shaped/rectangular

19 Cortical cells unlobed/lobed

Pollen characters:

20 Sulcus to verrucae distance greater than 10 ^m/less than 10

^im

21 Exine sculpturing verrucate/reticulated

Seed characters:

22 *Seed shape spherical/ovate/disk-shaped

23 Seed diameter greater than 4 mm/less than 4 mm

24 Epidermal cell shape rectangular/irregular

Chemical characters:

25 Colour of pressed flowers purple/orange

* cladistically informative characters.

TABLE 2.— Cladistic character codes for each species. Characters

arranged according to Table 1, missing data codes as 9

Characters

and sputter coated with Au/Pd and examined in a

Cambridge S200 SEM at lOkV. LM photographs using

differential interference contrast and oil immersion were

taken on a Zeiss Axioskop photomicroscope using Ilford

FP4 film. Pollen measurements were made using engraved

stage micrometer units in conjunction with an eyepiece

scale.

Pollen sources, vouchers in BOL

Wachendorfia brachyandra : Cape Point, Salter 8718.

W. graminifolia : Constantia Nek, Barker 170.

W. paniculata: Hermanus, Burman 936: Mamre, Esterhuysen 5244:

Waaihoek Peak, Esterhuysen 8283.

W. parviflora : Clanwilliam, Le Roux 2626.

W. thyrsiflora: Humansdorp, Fourcade 2436.

Dilatris corymbosa : Cape Point, Pillans 4589.

D. pillansii: Cape Point, Salter 7902.

Seed morphology

Seeds were taken from herbarium material or from

material collected in the field. Seeds were air dried,

mounted on aluminium stubs, and sputter coated with

Au/Pd and examined in a Cambridge S200 SEM at 5kV.

Seed sources, vouchers in BOL

Wachendorfia brachyandra'. Kirstenbosch, NBG s.n.

W. paniculata : Mamre, Esterhuysen 5244 ; Kommetjie, Helme 7;

Boulders beach, Leighton 34', Tulbagh, Leighton 37.

W. parviflora'. Kirstenbosch, NBG s.n.

W. thyrsiflora: Kirstenbosch, NBG s.n.

Dilatris ixioides: Pakhuis Pass, Esterhuysen 3165

D. viscosa: Tulbagh, Leighton 40.

Leaf anatomy

Fresh material used for the anatomical study was

collected from plants in the field, and fixed and preserved

in Kew Cocktail (16:16:1:1 = distilled water :EtOH: forma-

lin: glycerine). Fresh material was favoured because

cellular constitution is more easily observed and is less

likely to show distortion than dried and reconstituted

material. Sections were sledge microtomed to a thickness

of 25 micrometers and stained in Alcian Blue and Safranin

(lignin stain and counterstain) for 30 minutes (Tolivia &

Tolivia 1987). Sections were then washed in distilled water

and dehydrated in an alcohol series before mounting in

DPX.

Epidermal leaf scrapes were mounted in Hoyer’s

solution (Anderson 1954), which results in cleared,

permanent mounts. These slides were then used for the

examination of leaf surface features such as stomata, cell

shape and hair structure.

Cladistic analysis

A cladistic analysis was performed on the four species

of Wachendorfia and the outgroup, Dilatris. Twenty-five

characters were investigated for use in the analysis, but

only four proved to be cladistically informative (Table 1),

because their evolutionary polarity could be determined

and they occur in more than one species. A list of charac-

ter codes is given in Table 2. The most parsimonious

cladogram, that is the diagram that requires the least

Bothalia 22,1 (1992)

61

number of character changes, was determined using

Hennig86 (Farris 1988), and PAUP vers. 2.4 (Swofford

1985) was used to calculate the patristic distances for each

species and to optimize the characters on the cladogram.

Dilatris was used as the outgroup because it has many

morphological features in common with Wachendorfia .

The only other possible outgroup is Barberetta, but no

good material was available of this genus. Ornduff (1979)

suggested that, on the basis of chromosome numbers,

Wachendorfia and Barberetta (both with n=15) are more

closely related to each other than either is to Dilatris

(n = ±19—21). De Vos (1956), however, suggested that

Wachendorfia and Dilatris are closely related, based on

embryological studies, but she unfortunately did not

investigate Barberetta which possesses an unusual floral

structure for the Haemodoraceae, namely, a simple raceme,

unlike Wachendorfia or Dilatris , which have helicoid

cymes. Evidence therefore exists both for and against

choosing Dilatris as the outgroup for Wachendorfia,

although the case for choosing Barberetta could be

stronger. Simpson (1990) showed that on available infor-

mation, Barberetta is the sister taxon of Wachendorfia and

that these two together are the sister group of a clade

consisting of the New World genera Schlekia and

Pyrrorhiza : Dilatris was viewed as more distantly related.

However, as the outgroup need not be the sister taxon of

the study group (Watrous & Wheeler 1981; Maddison

et al. 1984), we used Dilatris as the outgroup. It should

be borne in mind that this may result in a poorer resolu-

tion of the phylogeny than when the sister taxon is used

as the outgroup.

TAXONOMY

Historical background

The earliest description of a Wachendorfia was by

Plukenet in 1700 and he named these specimens from the

Cape of Good Hope Erythrobulbus . These descriptions

were probably based on material collected by Oldenland.

Forty years later Breyne (1739) illustrated and described

what was clearly a Wachendorfia under the name Asphode-

lus. Hendrik Oldenland, the Superintendent of Simon van

der Stel’s Company Garden in Cape Town during the

period 1691—1699, made some of the earliest collections

in the vicinity of Cape Town. This collection of notes and

plants was neglected for many years, until Johannes

Burman published the first post-Linnean descriptions of

Wachendorfia in his monograph of the genus (Burman

1757). Two species, W thyrsiflora Burm. and W. panicu-

lata Burm., were accurately described. In 1758, Loefler

unwittingly applied the name Wachendorfia to what is now

known as Callisia Loefl. (Commelinaceae). The name

Wachendorfia commemorates E.J. van Wachendorff, who

was Professor of Botany and Chemistry at Utrecht during

the 18th Century (Jackson 1987).

Thunberg visited the Cape between 1772 and 1775, and

made extensive collections, which included specimens of

Wachendorfia, some of which were described by Linnaeus

(filius) as W. graminifolia in his Supplementum planta-

rum (1781). Thunberg, in his Prodromus (1794) and Flora

capensis (1811), described W graminea, W. tenella and W.

hirsuta. Salisbury added W. elata, W humilis, W. pallida

and W. brevifolia in 1796, whereas two more names were

proposed in the 19th century. Baker (1896) relegated most

of these above-mentioned names to the rank of variety or

synonym of W thyrsiflora and W. paniculata.

The genus was then virtually ignored for 120 years, un-

til Barker (1949) published two new species, W. parviflora

W.F. Barker and W brachyandra W.F. Barker. Although

these species have a more restricted range and are perhaps

less common than the initial two species, it is surprising

that such striking plants could go undetected for so long.

Wachendorfia Burm., in Wachendorfia: 2 (1757);

Linnaeus: 864 (1759); Bak. 6: 1 (1896); W.F. Barker: 206

(1950); Geerinck: 58 (1969); Dyer: 946 (1976). Type

species: W. thyrsiflora Burm.

Wacchendorfia Burm.f. (sphalm.) (1768).

Wachendorfia Cothen. (orth. var.) (1790).

Pedilonia Presl (1829).

Pre-Linnean synonyms:

Erythrobulbus Pluk.: v 5 (1700).

Asphodelus Breyne: t. 22 (1739).

Perennial, rhizomatous herbs 0.1— 2.5 m tall; leaves

usually annual (one species perennial); annual culms

develop from a rhizome. Rhizome fleshy, bright red,

irregular to ovoid, nodes three, opposite, each with an

axillary bud, apical bud producing the culm; new rhizome

produced annually from an axillary bud, 1—3 old ones

remaining attached; roots thin, adventitious, clustered

around node between old and new rhizomes; old rhizome

leaf bases forming a tunic around rhizome; ramets often

formed. Leaves erect or spreading, linear, lanceolate, or

falcate, firm in texture, always longitudinally plicate,

simple and entire, glabrous to hirsute, green to yellow-

green, 0.1— 0.9 m long; lower leaves sheathing at base of

culm; rhizome leaves 3, stem leaves variable in number,

lowest two stem leaves amplexicaul and opposite, leaves

above spirally arranged and not amplexicaul. Culm annual;

0.1— 2.5 m tall. Inflorescence a lax deltoid to dense cylin-

drical panicle; peduncles herbaceous, terete to angular,

very short to long (up to 0.25 m), near vertical, or

spreading; main axis covered in short hairs, pilose at base,

glandular nearer apex. Bracts erect or recurved, herba-

ceous or scarious, acute, usually hairy, older scarious

bracts often recurved. Flowers with tepals 6, borne in 2

whorls, subequal, spreading, oblong; lower 3 free, upper

3 united at base with small dark and light markings near

base; two open spur-like nectaries produced from the bases

of outer upper tepal and 2 adjacent inner tepals; outer

segments hairy on outside, especially central upper tepal;

yellow to pale apricot. Stamens 3, opposite the inner

tepals; filaments free, filiform, declinate, the upper two

pendulous, the lowest turning sideways opposite style;

anthers ovate-sagittate, two-celled, dehiscing longitudinal-

ly, introrse, usually included, 0.2— 3.0 x 0.5— 1.0 mm;

pollen boat-shaped, monosulcate, heteropolar, with a

verrucate, two-layered exine. Ovary superior, pubescent,

3-locular, 1 ovule per locule, axile placentation; style

filiform, bent sideways (left or right in any one plant), in

same direction as one of the stamens; stigma minute,

capitate. Fruit an acutely 3-lobed capsule, dehiscing locu-

62

Bothalia 22,1 (1992)

licidally, usually glandular, wider than tall, pubescent.

Seeds 1 per locule, spherical to ovoid, densely hairy.

Key features of the genus

Perennial herbs, three of the four species with annual

leaves. Annual flowering stems (culms). Perennial rhizome

with red sap. Leaves lanceolate, linear or falcate, simple,

longitudinally plicate, entire. Inflorescence a panicle.

Flowers zygomorphic, yellow, with right- and left-handed

forms. Ovary 3-locular, locules uni-ovular. Confined to

the fynbos biome.

Key to the species

la Perianth segments narrow, 3—6 mm broad; bracts herbaceous,

all erect; leaves usually as long or longer than culm; plants

usually short (less than 0.25 m tall) 4. W. parviflora

lb Perianth segments broad, 5—18 mm wide; bracts scarious

or submembranous, the upper usually recurved; leaves

usually shorter than culm; plants usually tall (greater than

0.25 m tall):

2a Stamens and style about half the length of the tepals, clustered

3. W. brachyandra

2b Stamens and style from two-thirds to nearly as long as the

tepals, spreading:

3a Plant tall (over 0.6 m); inflorescence dense, cylindrical;

leaves broad (greater than 15 mm wide), glabrous, peren-

nial 1. W. thyrsiflora

3b Plant short (usually below 0.6 m); inflorescence variable,

lax to very dense panicle; leaves narrow (usually less than

20 mm broad), glabrous to hairy, annual .... 2. W. paniculata

1. Wachendorfia thyrsiflora Burm. in Wachen-

dorfia; 3 (1757). Thunb.: 306 (1811); Bak. 6: 1 (1896); W.F.

Barker: 206 (1950). Type: Oldenland s.n. (?G, not seen);

iconotype: Burm.: 13, figs 1 & 2 (1757).

W. elata Salisb., Prodr. 45 (1796).

Pre-Linnean synonym:

Asphodelus latifolius Breyne: t. 22 (1739).

Rhizomatous perennial herb; 0.6— 2.5 m tall. Rhizome

up to 150 mm long, irregular, usually cylindrical, sheathed

by overlapping leaf bases; rooting from nodes. Leaves

perennial, erect, lorate to lanceolate, deeply plicate, firm,

glabrous, usually shorter than the flowering stem, up to

900 x 80 mm; younger plants with leaves less than 50

mm broad. Leaf anatomy: palisade layer not distinct;

cuticle with variable lignification; sclerenchyma caps

poorly developed; subsidiary cells kidney-shaped;

mucilage canals large. Culm up to 2.5 m tall, stout (up

to 15 mm diameter near base), erect, densely pilose at base

(hairs up to 4 mm long) running into short (1 mm long)

glandular hairs near apex. Inflorescence a dense cylindrical

panicle, up to 0.6 x 0.2 m. Peduncles many, short (up

to 100 mm), regularly spaced on distal half of axis.

Younger plants often more laxly branched; seldom more

than 7 pedicels per peduncle; no secondary peduncles.

Bracts lanceolate, scarious, recurved, persistent, 10—40

mm long. Tepals bright yellow, nectar guides on base of

upper 3 tepals light to dark; all slightly spathulate; 12-28

X 8— 14 mm; outer adaxial tepal narrower and shorter than

the rest, slightly recurved; upper 5 tepals overlapping,

leaving lower tepals somewhat isolated; tepals sometimes

fringed with orange cilia. Stamens : filaments 3/4 tepal

length, 15—20 mm long, spreading; anthers 1.2— 2.0 x

0.5— 1.0 mm. Gynoecium: ovary yellowish, 2—3 x 1—2

mm; style 13—18 mm long. Fruit a 3-angled capsule, 10

x 7—10 mm. Seed large, up to 5 mm long, oval or kidney-

shaped; dense microscopic blisters (1.5 pm) on seed hairs.

Chromosome number: n=15 (Ornduff 1979).

Flowering time: beginning of September to end of

November, occasionally to mid-January, with a peak in

September and October.

Distinguishing features: large size; marshy habitat; large

yellow flowers with long spreading stamens; inflorescence

a dense, cylindrical panicle; peduncles short.

Distribution and habitat: W thyrsiflora is a widespread

species, occurring from as far north as the Olifants River

valley between Citrusdal and Clanwilliam, south to the

Cape Peninsula, inland to the Franschhoek Mountains, and

along the south coast and associated mountain ranges as

far east as Humansdorp (Figure 1). This species is more

of a habitat specialist than any of the other species, and

is confined to permanently moist sites, especially along

streams and in seepages. The species has a wide altitudinal

range, varying from about 5 — +1 200 m a.s.l.

Status: locally abundant, often dominant in marshes.

Some populations are very small and may be declining

due to afforestation or weed infestation. The populations

in mountain fynbos are generally stable.

Vouchers: Leighton 1353 (BOL); Levyns 3368 (BOL); Liebenberg 7923

(PRE); Pillans 8086 (BOL).

2. Wachendorfia paniculata Burm. in Wachendorfia:

11 (1757). Sims: t. 616 (1803); Thunb.: 307 (1811); Bak.

6: 1 (1896); W.F Barker: 206 (1950). Type: Oldenland

s.n. (G?, not seen); iconotype: Burm.: 15, fig. 3 (1757).

W graminifolia L.f. : 101 (1781). W. graminea Thunb.: 309 (1811) nom.

meg. Type: near river at Drakenstein. Thunberg 1242 (UPS, lecto. —BOL,

microfiche!).

W. hirsuta Thunb.: 308 (1811); Sims: t. 614 (1803). Type: sand fields

between Swartland and Saldanha Bay. Thunberg 1243 (UPS, holo.— BOL,

microfiche!).

W. tenella Thunb.: 308 (1811). Type: between Langevlei and Heeren-

logenment. Thunberg 1246 (UPS, holo. —BOL, microfiche!).

W. brevifolia Solander ex Ker-Gawl. (1809). Type: in the Banksian

Herbarium (BM).

W. herbertii Sweet: 400 (1826), based on W. paniculata var. fl Herbert

(1826). Specimen from the Cape of Good Hope, flowering at Spofforth

in July. Iconotype: Botanical Magazine 53: t. 2610 (1826), here designated.

Rhizomatous perennial herb; 0.1— 0.9 m tall. Rhizome

ovoid, up to 50 mm in diameter, appearing tunicate when

dry. Leaves annual; narrowly lanceolate or linear to

broadly falcate, erect or spreading, dull green to yellow-

green, glabrescent to hairy, 3-nerved; 0.1— 0.7 x

5— (20)— 35 mm. Leaf anatomy: palisade layer one cell

thick; cuticle with intermediate lignification; sclerenchyma

cap development variable; subsidiary cells rectangular;

mucilage canals small. Culm occasionally up to 1 m tall,

slender to robust (3—15 mm diameter), erect, covered with

short, dense simple hairs. Inflorescence a lax to dense

racemose panicle, composed of 5—20 scorpioid cymes,

each cyme bearing 1—7 flowers. Peduncles variable in

number and length, longer at the base (may be branched

again to form a secondary peduncle), up to 0.4 m long,

slender to robust; pedicels short (less than 100 mm). Bracts

Bothalia 22,1 (1992)

63

FIGURE 1. — Distribution of W. thyrsiflora.

scarious (often submembranous when fresh), often

recurved, especially when old, 5—50 mm long, veined,

long-acuminate, pilose, partly sheathing. Tepals apricot,

yellow, or orange; slightly scented; 13.0— (21.0)— 31.0 x

4.0— (10.0)— 16.0 mm; inner abaxial tepal often slightly

broader than the rest; outer adaxial tepal shorter, narrow-

er, recurved; margins sometimes fringed with short cilia.

Stamens from two-thirds to three-quarters the tepal length,

spreading; anthers 2— 3 x 0.8— 1.0 mm. Gynoecium: style

15—22 mm long, as long as shortest tepal. Fruit an acutely

three-lobed capsule; 10 x 5 mm. Seed spherical, coarsely

pilose, brown, 2 mm in diameter. Chromosome number :

n=15 (Ornduff 1979). Figure 2.

Flowering time: August to December, very rarely to

early February (high altitude), with most records from

September to November.

Common name : rooikanol or spinnekopblom.

Distinguishing features: plants from 0.15—0.80 m tall;

bracts scarious when mature; stamens and style spreading,

from two-thirds to three-quarters the length of the tepals;

tepals broad; leaves usually shorter than the flowering stem.

Nomenclatural notes: according to Savage (1945) there

is a specimen of Wachendorfia in the herbarium of the

Linnean Society, annotated by Burman as ‘ Wachendorfia

paniculata'. As the Burman herbarium is generally thought

to be in the Delessert Herbarium at Geneva (Stafleu &

Cowan 1976) further investigation would be required to

determine where the type of the name is housed.

The types of Linnaeus the Younger are mostly housed

in the herbarium of the Linnean Society. However, there

FIGURE 2. — W. paniculata flower structure. A, left-handed flower (stigma deflected to left); B, right-handed flower (stigma deflected

to right); C, side view showing large perianth-aperture (arrowed).

64

Bothalia 22,1 (1992)

FIGURE 3. — Distribution of W. paniculata.

is no material of W. graminifolia (Savage 1945), whereas

there is a specimen annotated as such by Thunberg at

Uppsala. This may well be the type material, but can in

the absence of further information only be designated as

a lectotype.

Distribution and habitat : this is the most widespread

species, ranging from Nieuwoudtville to Port Elizabeth

(Figure 3). It is ubiquitous in the fynbos biome, and is

found mainly on soils derived from Table Mountain Sand-

stone, although it has also been recorded from areas with

granitic soils, alluvial sands, and clayey soils derived from

Malmesbury shales. The species is found from sea level

to about 1 700 m.

W. paniculata grows in mesic and dry mountain fynbos,

lowland fynbos on acid or alkaline soils, renosterveld, and

strandveld. It is most common in younger vegetation (less

than 10 years since last fire), probably because it favours

less dense vegetation. The species does not require per-

manent moisture and can be found in areas ranging from

seasonally dry sands to permanently moist shales. This

wide ecological tolerance is certain to be one of the major

reasons for the success of the species.

Status : locally abundant, but usually varies from sparse

to common. This species is unlikely to become endangered

in the near future due to its wide distribution, varied

habitats, and ability to thrive in disturbed (even ploughed)

sites.

Variation within the species : W paniculata is excep-

tionally variable with respect to plant size, leaf hairiness

and shape, and flower size and structure. This has resulted

in the large number of synonyms for this species.

Barker (1950) resurrected W. graminifolia L.f. as a

segregate of W paniculata. However, there is a continuum

of variation in all traits mentioned by Barker as reliable

discriminators, for example, perianth length, plant robust-

ness, leaf hairiness and width. In addition, specimens show

combinations of characters which make the separation of

two species, based on these characters, futile, e.g. a plant

may be robust and have tepals 25—30 mm long (both W.

graminifolia characteristics), as well as narrow, hairy

leaves (both W. paniculata characteristics). The various

combinations are endless. For these reasons we do not

think that W graminifolia can be separated from W. pani-

culata .

An extensive review of all available herbarium specimens

showed that there is neither temporal nor geographical

separation between the ‘varieties’. In addition, the chromo-

some number is constant for specimens from a wide geo-

graphical range, and exhibiting a wide range of morpho-

logical variation (Ornduff 1979). This evidence supports

the idea that W. paniculata and all plants closely resembling

it represent a single species. This variation in W. paniculata

should somehow be recognised. However, the variation

is too continuous to recognize formal taxa, and Rosendahl’s

(1949) comment probably applies here: ‘some authors

when dealing with highly polymorphous groups seem to

feel that it is necessary to take account of all variants that

can be distinguished and fit them into a formal scheme.

The trouble with this procedure is that in attempts to set

up a series of units of descending rank, a point of diminish-

ing returns is soon reached, beyond which confusion rather

than clarification results. Such schemes may have some-

thing to commend them in theory but not in practice’. The

alternative is to recognize informal forms.

Stuessy (1990) defines ‘form’ as ‘specimens with small

genetic variations, not geographically correlated, and often

growing with more ‘typical’ plants’. Subspecies and

varieties apply only to allopatric taxa, and the variation

within W paniculata should thus be described as forms.

Form 1: this form has been recorded from Kleinmond

in the east, north to Mamre, and south to the Cape Penin-

sula, and is characterised by large tepals fringed with dark

cilia. There is continuous variation for this and other

characters in the plants, and the distinctness of the morph

as a whole is doubted.

Bothalia 22,1 (1992)

65

FIGURE 4. — W. brachyandra flower

structure. A, right-handed

flower illustrating clustered

stamens and style characteris-

tic of species; B, side view

showing small perianth aper-

ture (arrowed).

Form 2: specimens from the mountains between Ceres

and Pakhuis Pass are frequently very tall, with extremely

lax panicles and long, thin leaves. This form does have

a degree of environmental and geographical correlation,

which suggests that it might be viewed as a subspecies (al-

lopatric, genetically similar). However, the form is not

strictly allopatric as the typical form of W. paniculata is

known to occur in the area and for this reason we would

not recommend the use of a formal subspecific rank for

this taxon.

Form 3: dwarf specimens with very narrow, hairy leaves

can often be found growing together with more ‘typical’

forms, for example in the hills above Glencairn and

Simonstown.

Vouchers: Barker 169, 1097 (NBG); Lewis in BOL 22263 (BOL); Pillans

9138 (BOL); Stokoe 818 (BOL); Wolley Dod 526 (BOL).

3. Wachendorfia brachyandra W.F. Barker in

Journal of South African Botany 15: 41 (1949); W.F.

Barker: 207 (1950). Type: Cape Peninsula, Kirstenbosch,

Barker 1096 (NBG, holo.!).

Rhizomatous perennial herb, 0.10—0.65 m tall. Rhizome

small, globose to ovate-oblong, 5—20 mm diameter.

Leaves annual, erect or spreading, linear to lanceolate,

often falcate, glabrous, dark green to yellow green, up to

700 x 35 mm. Leaf anatomy, palisade layer poorly

defined, consisting of one or two cells; cuticle lightly

lignified; sclerenchyma cap development variable; subsi-

diary cells rectangular; mucilage canals small. Culm about

3 mm in diameter, covered in short glandular hairs, seldom

branched to form secondary peduncles; usually less than

0.4 m long. Inflorescence a lax panicle with 6—17 flowers

per peduncle; peduncles and pedicels slender; peduncles

short near tip of axis, a few much longer at the bottom

(up to 0.2 m long). Bracts mostly scarious, oblong ovate

acuminate, almost sheathing, not recurved lower down,

up to 80 mm long; densely pilose. Tepals light apricot

yellow, the markings dark; 12—20 x 4—14 mm; outer

adaxial tepal smaller than the others, only slightly

recurved; cilia seldom present on tepal edges. Stamens

clustered, not spreading, half as long as tepals, 6-14 mm

long; anthers 2.0— 2.5 x 0.5— 1.0 mm. Style: short, not

much curved sideways, 7—12 mm long. Fruit a 3-lobed,

dry capsule broader than long, 5 x 8—10 mm, carpels

obtuse. Seeds spherical, coarsely hairy, 2 mm diameter.

Figure 4.

Flowering time: late August to early December, with

a peak in September and October.

Distinguishing features: stamens and style short, half

the length of the tepals, clustered; inflorescence lax; tepals

pale yellow-apricot.

Distribution and habitat: this species has the most

restricted range of all the species, and is confined to the

extreme southwestern Cape (Figure 5). It is undoubtedly

undercollected, and is known from altitudes ranging from

50— ±600 m on either sandstone or granite derived soils.

The species seems to favour damp sites, often growing in

partial shade in forest margins, or in seeps and drainage

lines in fynbos. It is also commonly found in recently burnt

fynbos areas, and will persist for many years in a fairly

open habitat (e.g. herbaceous margins of forests).

66

Bothalia 22,1 (1992)

C, rear view.

Status : locally common, but with a sparse, scattered dis-

tribution. Due to the nature of its habitat, W brachyandra

may well become endangered, as swampy, moist areas are

frequently drained for agriculture, building, etc.

Vouchers: Compton 16359 (NBG); Salter 8718 (NBG), 9046 (BOL).

4. Wachendorfia parviflora W.F Barker in Journal

of South African Botany 15: 39 (1949); W.F. Barker: 207

(1950). Type: Cape Peninsula, Camps Bay, Salter 7457

(NBG, holo. !).

Rhizomatous perennial herb, 0.1— 0.4 m tall, usually

dwarf, 0.1— 0.2 m. Rhizome small globose-ovoid, oblong,

5—25 mm in diameter. Leaves annual, erect or spreading,

linear to lanceolate, usually falcate and longer than in-

florescence, blue-green, softly hairy, up to 360 x 25 mm.

Leaf anatomy: palisade layer one cell thick; cuticle with

intermediate lignification; sclerenchyma caps well de-

veloped; subsidiary cells rectangular; mucilage canals

small. Culm short (usually less than 0.2 m); culm,

peduncles and pedicels covered in short glandular hairs.

Inflorescence a very short, dense panicle; peduncles very

short, densely clustered on axis. Bracts herbaceous, green,

erect, often produced beyond the flowers. Tepals dull

yellow, fading brownish purple, segments narrow, 15—25

x 3— 6 mm, upper segments usually broader than the rest,

lowermost tepal isolated. Stamens: two thirds length of

tepals, 10-12 mm long; anthers 1.5— 2.0 x 0.5 mm. Style:

two-thirds length of tepals, 11 mm long. Fruit a dry

capsule, broader than long, 6-7 x 11-14 mm; carpels

obtuse, covered in glandular hairs. Seeds globose, coarsely

hairy, 2-4 mm in diameter. Figure 6.

Flowering time: early August to late September, with

a peak in early September.

Distinguishing features: dwarf habit; erect herbaceous

bracts; tepals very narrow.

Distribution and habitat: W parviflora is essentially a

species of the western Cape, ranging from Nieuwoudtville

to the Cape Peninsula and McGregor (Figure 7). There

is a record from near Soebatsfontein and another from 15

km east of Hondeklipbaai (3017BB), which suggests that

this species probably occurs all the way up the west coast

as least as far north as Hondeklipbaai. This area is under-

collected (Gibbs Russell et al. 1984) and it may be more

common and widespread in the area.

W parviflora grows on both sandstone and granite

derived soils, and there are a few records from shale areas.

The species is found at altitudes ranging from 15 -±500

m a.s.l. in habitats ranging from dry, sandy hollows in

coastal fynbos to moist, rocky ledges in thick mountain

fynbos. It is often found in association with W paniculata.

Status: uncertain, probably uncommon and sparse

throughout its range. Appears to be replaced by W pani-

culata in many apparently suitable areas.

Vouchers: Barker 4600 (NBG), W.F. Barker 9-8-1935 (BOL);

Hanekom 1168 (PRE); Johnson 236 (NBG); Salter 7457 (NBG).

DETAILED MORPHOLOGICAL OBSERVATIONS

Rhizome

Wachendorfia possesses a distinctive rhizome, contain-

ing a red fluid rich in arylphenalenone pigments. The

air-dried rhizome varies in diameter from 5—50 mm,

although the fresh rhizomes may be twice this size. The

shape is usually spherical to ovoid, although irregular

shapes may occur. An air-dried rhizome is usually covered

by a papery tunic composed of the dry leaf bases.

FIGURE 7. — Distribution of W. parviflora.

Bothalia 22,1 (1992)

67

FIGURE 8. — A, W. paniculata : rhizome morphology, showing the three previous years shrivelled rhizomes, separated in

each case by a cluster of adventitious roots, a, active rhizome; d, three year old rhizome. B, W. parviflora: rhizome

and culm with leaves removed. Culm nodes 1 and 2 with amplexicaul leaves; node 3 and onwards with spiral arrangement

of leaves; nodes 2 and 3 with small sterile buds visible. C, W. paniculata: rhizome with all rhizome leaves removed;

large node ringing middle of rhizome is node 2, axillary bud visible (large arrow); node 3 visible near base of culm

(small arrow). D, E, W. parviflora: D, close up of young culm, nodes 2 & 3 visible, the sterile bud of node 2 arrowed;

E, pileate epidermal hair, note four enlarged basal cells and multicellular hair construction. F, W. paniculata: light

photomicrograph of pollen grain, illustrating central

A new rhizome is produced every year, with the previous

year’s rhizome remaining attached to the present year’s

rhizome. The old rhizome shrivels, until all that remains

is the extensive system of vascular traces. Occasionally

up to three old rhizomes may be found attached in

sequence to the current rhizome. A cluster of thin, short

adventitious roots sprouts from between each rhizome

(Figure 8A, B).

Each rhizome has three nodes, each with a ‘rhizome

leaf and an axillary bud (Figure 8C). The axillary buds

are arranged distichously. Each axillary bud is capable of

producing a new rhizome, although only one is actually

produced. The new rhizome may be produced below or

sulcus, proximal verrucae and micropore-pitted border.

to the side of the parent rhizome. This means that next

year’s rhizome will be exploring new soil, either deeper or

some horizontal distance from the parent. This may be an

important function given the absence of tap roots. The an-

nual flowering shoot is formed by the apical bud, resulting

in a sympodial growth form. W. thyrsiflora may produce

stoloniferous lateral outgrowths from the main rhizome,

some of which may ultimately extend several metres. These

extended rhizomes are then capable of producing new

ramets. Vegetative reproduction is common in monocots,

and seems to be an important feature of Wachendorfia.

The rhizomes stain positive for starch. In W panicula-

ta and W parviflora the amyloplasts are concentrated

68

Bothalia 22,1 (1992)

FIGURE 9. — Leaf anatomy of Wachendorfia: scale bar, 250 /mi. A, B, W. thyrsiflora : A, leaf section illustrating kidney-shaped subsidiary

cells and poorly defined palisade layer; B, vascular bundle with small sclerenchyma caps. C, D, W. paniculata : C, vascular bundle

with large sclerenchyma caps; D, stoma with twin epidermal lips and single palisade layer. E, W. brachyandra: variable palisade

width and bundle with large cap. F, W. parviflora : single palisade layer and trichome base with four swollen epidermal cells (arrowed).

G, Dilatris pillansii: leaf edge with double palisade layer: c, mucilage canal; vb, horizontally aligned vascular bundle; spaces between

densely packed cortical cells are the result of lobing. H, D. pillansii: stoma with twin epidermal lips and kidney-shaped subsidiary cells.

within the vascular stele, with almost none in the cortex.

In W thyrsiflora amyloplasts are equally common in the

cortex and the stele. The reasons for this variation are

unknown, but may be related to the persistence of the

rhizome of W. thyrsiflora, resulting in an extremely long

underground organ.

The xylem elements have an amphivasal arrangement,

which is similar to that reported for the Restionaceae by

Linder (1990).

Bothalia 22,1 (1992)

Leaf morphology

Although the plicate leaves of Wachendorfia are distinc-

tive, they often show considerable intraspecific variation.

The colour varies from dark green through to light yellow-

green, and is often a function of leaf age. The shape may

vary within a species, some populations of W paniculata

having almost needle-like leaves, and others having broad,

falcate leaves. W. thyrsiflora usually has large ensiform

leaves, whereas W brachyandra has lanceolate or falcate

leaves. W. parviflora is normally characterised by falcate

leaves, but may frequently have lanceolate leaves. W. thyrsi-

flora has distinctly longer and broader leaves than any of

the other species (up to 900 x 80 mm). The other three

species show much interplant variation in leaf size, but

are all within approximately the same range, up to 400

x 40 mm. There seems to be some altitudinal variation

in leaf size and shape in W. paniculata, with specimens

growing at high altitudes tending to have extremely long,

narrow, linear leaves.

Wachendorfia leaves originate from both the rhizome

and the culm. There are usually three large rhizome leaves,

arising from the three rhizome nodes (Figure 8C). These

leaves sheath the culm at the base. The culm leaves fall

into two types: the lowest two are amplexicaul and

opposite, whereas those further up the stem are spirally

arranged (Figure 8D). The lower leaves thus give the

distinct impression of being two-ranked. The leaves are

always firm in texture, longitudinally plicate and entire.

The leaves of W. thyrsiflora and W. brachyandra are always

glabrous or very nearly so, while those of W parviflora

are distinct in having a dense coating of numerous short

hairs (Figure 8E). W paniculata leaves vary from being

nearly glabrous to densely hairy with long white hairs.

Within-plant variation in leaf hairiness is small, although

within-population variation may be great in this species.

There is no obvious ecological reason for the difference

in leaf hairiness within the species, as it does not seem

to follow a gradient of rainfall, altitude or any other single

factor.

The dense leaf hairs in W. parviflora are of two types.

Pointed, unicellular hairs are by far the most common,

outnumbering the pileate, tricellular hairs by about 5:1.

Adaxial and abaxial leaf surfaces are equally hairy. Four

or occasionally five large epidermal cells support the base

of each hair. The unicellular hairs taper to a point and

appear to have hollow bases, whereas the tricellular hairs

have small terminal cap cells that may be glandular (Figure

8E).

The two species confined to permanently damp habitats

(W thyrsiflora and W. brachyandra ) have glabrous leaves,

whereas the two dryland species ( W. paniculata and W.

parviflora) usually have hairy leaves.

Leaf anatomy

The leaf anatomy of Dilatris pillansii W.F. Barker, D.

corymbosa Berg., and the four species of Wachendorfia,

was examined. Wachendorfia leaves are plicate, and both

thinner and wider than the narrow, rigid leaves of Dilatris,

which have a number of special associated features.

69

The palisade layer in Dilatris is two cells wide (Figure

9H), which is a similar arrangement to the twin layer of

columnar palisade cells in Conostylis R. Br. (Green 1959).

Lachnanthes Ell. lacks a palisade layer altogether (Simpson

& Dickison 1981). There is thus a range of palisade

structures within the family. The Cape genera reflect this

variation well, with Dilatris having a two cell layer, W.

thyrsiflora without a distinct layer (Figure 9A), and the

other Wachendorfia species with either a one or a two cell

layer (Figures 9C— F). Cortical air spaces are absent in

Dilatris and the cortical cells are lobed (Figure 9H). All

Wachendorfia species have air spaces between the isodia-

metric cortical cells (Figures 9A-F).

The epidermal cuticle is thick in Dilatris, W. thyrsiflora

and W. paniculata, but is noticeably thinner in W parvi-

flora and W brachyandra. Lignification of the epidermal

layer is light in all taxa bar W parviflora, in which exten-

sive lignification is evident (Figure 9A— H).

Stomata in Dilatris and Wachendorfia species are para-

cytic. The epidermal lip has a well-developed upper

component and a lower lip of about half the length of the

upper. This double epidermal lip is also recorded in Lach-

nanthes (Simpson & Dickison 1981). Green (1959) made

no mention of these epidermal lips in his study of the

Australian genus Conostylis, although his drawings show

slight spurs on the outer edges of the stomata. Dilatris has

a kidney-shaped subsidiary cell (Figure 9H), a feature

which it shares with W. thyrsiflora. All other Wachendorfia

species have rectangular subsidiary cells. Stomata appear

to be equally common on both sides of the leaf and are

very slightly sunken relative to the epidermal layer. Large

substomatal cavities are present in all taxa.

In Wachendorfia, the large first order vascular bundle

within the extreme edge of the leaf closest to the culm,

lies horizontally (i.e. tangential to the culm axis), whereas

all others in the leaf are vertically orientated (i.e. radial

to the culm axis). This feature is also found in Conostylis

(Green 1959). This may be the result of the leaf sheathing

around the rhizome and the culm, thus flattening one edge

of the leaf and distorting the apparent bundle position.

Sclerenchyma cap development in the vascular bundles is

very variable. W. thyrsiflora has poorly developed caps,

whereas in Dilatris and W. parviflora they are well-

developed. This feature is variable in both W. brachyandra

and W. paniculata, and in the latter the bundles may be

exceptionally well developed.

W. thyrsiflora has larger leaf cortex mucilage canals than

the other species in the genus. This may be an allometric

feature associated with the overall large size of the species

or the result of ecological or phylogenetic factors. We

suggest that it does have ecological relevance, as the

various species’ leaf sections were taken from leaves of

the same size (thus reducing any allometric effects), and

W. thyrsiflora is unique in being confined to permanently

damp sites. Large quantities of mucilage can be seen in

cut sections of W. thyrsiflora leaves and most of this

appears to come from the large canals in the leaf. Dilatris

has a single large canal within the leaf (Figure 9G), and

this is always located in the side of the leaf closest to the

aerial stem. There are a number of other smaller canals

scattered throughout the leaf cortex which may act as

mucilage ducts.

70

Bothalia 22,1 (1992)

Floral morphology

Considerable interplant and interpopulation variation

occurs in the length of the inflorescence, the number of

cymes produced, the number of flowers in a cyme, the

size and colour of the flowers, the degree of perianth

spreading, and the size of the nectaries.

Wachendorfia flowers are enantiomorphic. The style of

some flowers is sharply directed to the right, whereas in

other plants it is deflected to the left (Figure 2). In both

cases, one of the three stamens is borne close to the style,

the other two are deflected in the opposite direction.

Dilatris is also enantiomorphic, but differs in that both

left- and right-handed flowers may be found on the same

inflorescence.

The upper three tepals in Wachendorfia have pale mark-

ings surrounded by dark rings which may act as nectar

guides. There is much interplant variation in the contrast,

size, and shape of these markings. Large ‘semi-extrafloral’

nectaries are formed by the bases of the outer upper and

the two adjacent inner tepals, one being present on each

side of the flower. These are elongated into spurlike

structures clearly visible on the outside edges of the

flowers. Significant quantities of nectar are produced,

which may persist even after the flower itself has withered.

Nectar is also produced long before the flower is open.

Sugar concentration in an open flower is about 20%,

rising to 50% in a withered flower (Ornduff & Dulberger

1978).

The genus is essentially spring flowering, although there

are some intra- and interspecific differences worth noting.

High altitude forms of W. paniculata are noted for their

late flowering. The species has an extremely long flower-

ing period, and this may be attributed to the varied habitats

in which the species is found. W. thyrsiflora is often found

flowering in late summer. This might be a reflection of

its moist habitat, as severe moisture stress is unlikely,

perhaps allowing this species to flower much later than

the other species which favour drier habitats. W parviflora

is interesting in that it flowers early, with no flowering

recorded later than September. It is often the only species

of the genus in flower in early August.

The three dryland species can be separated on flower

structure. In Figure 10 three floral measurements are

plotted for W. parviflora, W brachyandra, and W. panicu-

lata. On the y-axis a ratio of stamen (= style) length over

maximum tepal length for each flower is plotted, (demon-

strating the distinctness of W. brachyandra) , and on the

x-axis the maximum tepal width recorded for each flower

is given (demonstrating the distinctness of W parviflora).

W. thyrsiflora is a very distinct species in many different

ways (habitat, size, anatomy), and for this reason is not

included in this analysis. The figure illustrates how the

three species can be separated by a combination of floral

characters, but it should be noted that it does not demon-

strate the existence of clearcut phenetic groups.

Pollen

Erdtman (1966) recognized three pollen types within the

Haemodoraceae. Six genera, including both Dilatris and

Wachendorfia, are characterized by monosulcate pollen

(20—90 pm long) with a ‘usually not very distinct’ exine

stratification and sexine pattern. Barber etta was not studied

by Erdtman (1966). Simpson (1983) published a systematic

palynological survey of the Haemodoraceae, in which the

pollen was investigated by light microscopy, and both scan-

ning and transmission electron microscopy, thus allowing

a detailed characterisation of the wall structure. He includ-

ed two species of Dilatris, Wachendorfia thyrsiflora and

Barberetta in his study, and his observations provide a use-

ful descriptive base for the present study which included

the other three species of Wachendorfia.

Dilatris pillansii and D. corymbosa possess very similar

pollen grains, are monosulcate and heteropolar, with

verrucate to baculate non-apertural sculpturing and

gemmate to psilate apertural sculpturing (Figure 11A).

Grains are approximately 50 pm long (polar) and 25 pm

wide (equatorial). The exine surface appears to be quite

4 8 12 FIGURE 10. — Scatter diagram of the

Maximum tepal width (mm) maximum tepal width against

the ratio of stamen/petal length

for W. parviflora, W. brachyan-

dra and W. paniculata, show-

ing how these species can be

separated morphologically.

■ W. parviflora * W. brachyandra * W. paniculata

Bothalia 22,1 (1992)

71

FIGURE 11. — Scanning electron

micrographs of pollen grains.

A, Dilatris corymbosa : reti-

culate exine. B, W. thyrsiflora:

broad micropore-pitted border.

C, W. parviflora : verrucae pat-

terning. D, IV. brachyandra: v,

verrucae; m, micropore-pitted

border; s, sulcus. E, F, W.

paniculata: E, heteropolar,

boat-shaped grain; F, close-up

of verrucae and granular

secondary exine sculpturing

beneath.

different from the exine of Wachendorfia pollen, as the

verrucae seem to have coalesced to form a more continuous

bumpy pattern. Dilatris also has a far less obvious

micropore-pitted aperture border.

Wachendorfia pollen is monosulcate and heteropolar,

with a convex aperture wall consisting of widely separated,

two-layered baculate exine elements, over a relatively thick

two-layered, fibrillar intine (Simpson 1983). The concave

non-apertural wall is proximally verrucate, with a granular

appearance between the verrucae (Figure 11F). The

verrucae thus appear to ‘saddle’ the pollen grain (Figures

8F; 11B — F). The aperture wall is encircled by a smooth

border pitted with micropores (Figure 11B— F). The

‘coarsely granular, distinctly convex operculum’ noted by

Erdtman (1966) is an aperture wall with an outer layer of

closely spaced exine structural elements, a point noted by

Simpson (1983). These apertures often disintegrate during

acetolysis, probably because of the lack of continuous

exine material between the elements. The grains are

distinctly hemispheric in shape. All species have grains

40—50 x 20—25 p m. In all species the size of the exine

verrucae is relatively constant, about 1 pm in diameter,

and all species have the granular substance visible between

the verrucae. This granular layer is the lower layer of the

two-layer exine. W. thyrsiflora (Figure 11B) differs from

the rest of the genus in that the distance between the sulcus

and the start of the verrucae, i.e. the width of the micro-

pore-pitted border, is twice as great as in the other spe-

cies. There is thus no reliable way to identify Wachendor-

fia pollen, with the exception of that of W. thyrsiflora , to

species level.

The pollen of Barberetta is virtually identical to that of

Wachendorfia. This suggests that Barberetta is more close-

ly related to Wachendorfia than it is to Dilatris.

Seeds

The seeds of Wachendorfia are all ovoid or spherical,

about 2 mm in diameter, and covered in numerous

short hairs (Figure 12B-F). There is little interspecific

variation, and they cannot be reliably used for species

identification. Hair length and width shows very little

variation. All the species have curious verrucae on the

hairs (Figure 12E, F), and their function and origin is not

known.

W thyrsiflora has the most distinctive seed, as they are

large and somewhat kidney-shaped. When viewed in a

72

Bothalia 22,1 (1992)

SEM, the hairs on the seed also have a greater density

of verrucae than the hairs of other species. The seed dis-

persal biology is unknown, vith hydrochory and ane-

mochory possible. The seed of W. thyrsiflora has been

observed floating on streams (pers. obs.), and as this

species is generally associated with streams, hydrochory

may be important here.

Pollination biology

The few studies on the evolutionary significance of

enantiomorphy have suggested that it is probably a feature

which promotes outcrossing, that is, increases the level

of intermorph pollination (Wilson 1887; Ornduff & Dul-

berger 1978). This conclusion is based on the findings that

W. paniculata has a weakly developed self-incompatibility

system, and that intermorph crosses produce more seeds

than self-pollinations or intramorph pollinations. Selec-

tion for outcrossing cannot be the driving force behind

the evolution of the Dilatris flower structure, as left- and

right-handed flowers are found on the same plant. The

FIGURE 12. — Scanning electron micrographs of seed. A, Dilatris viscosa L.f.: disc-like seed with a central funicle and reticulate patterning.

B, W. brachyandra: seed hairs often have relatively few verrucae. C, W. paniculata : typical raised funicle and hairy, oval seed. D,

W. brachyandra : spherical seed with the usual central funicle(f). E, W. parviflora : seed hairs and verrucae. F, W. thyrsiflora : dense

verrucae on its seed hairs.

Bothalia 22.1 (1992)

73

DILATRIS

W.thyrsiflora

W.paniculata

W.brachyandra

W.parviflora

FIGURE 13.— Cladogram for the species in Wachendorfia. All characters are indicated; those that are phylogenetically informative are indicated

by thicker lines. The length of the components indicates the patristic distances on the diagram. Character codes are given in Table 1, and

the distribution of the characters in Table 2.

evolutionary history behind such a strange system is com-

pletely unknown (Ornduff & Dulberger 1978).

The seemingly wasteful production of large quantities

of ‘extrafloral’ nectar, often prior to anthesis, is once again

difficult to explain in evolutionary terms, and deserves

further study. The nectaries open to the outside of the

flower and nectar stealing is therefore very easy, as the

Thief does not have to get anywhere near the pollen or

stigmatic surfaces. Ants are seldom seen on the plants due

to the glandular hairs on the stem, so the ‘extrafloral’

nectar does not appear to have any function in feeding ‘ant

guards’ (see Faegri & Van der Pijl 1966).

The pollinators are unknown. This would be a rewarding

field of study, as the unusual floral morphology may reflect

interesting evolutionary/ecological interactions. One of the

problems with the pollination syndrome is that in all

species, except W. brachyandra, the stigma and anthers

seem too far apart for most insects to touch when visiting

the flower. We might predict that seed set would be greater

in W. brachyandra than in the other species (ease of polli-

nation), but preliminary observations do not support this

prediction. There must therefore be some insects which

can pollinate the other species. These insects would have

to be significantly larger than the common honeybee. Apis

mellifera , as this species is not able to reach the anthers.

The carpenter bee, Xylocopa caffra, is large enough but

it is very seldom seen on low growing Wachendorfia , as

it tends to favour tall shrubs and small trees, e.g. Virgilia

oroboides. The first author has seen tabanid flies (family

Tabanidae) taking nectar from W. parviflora and W

paniculata , and in the process, the wings of this large-

bodied fly touched the anthers. It is possible that this group

of flies may be important pollinators of the genus due to

their size and anthophilous behaviour, but more observa-

tions are needed. Other potential pollinators include

numerous small beetles that appear to feed on the pollen.

These beetles clamber over the anthers and may effect a

degree of ‘mess and soil’ pollination.

Phytogeny and speciation

The most parsimonious cladogram is given in Figure

13. The patristic distances (i.e. the degree of divergence

from its ancestor) of each component is indicated by the

relative length of that component. All the characters,

including the autapomorphies, have been indicated on the

diagram. Phylogenetically informative characters are

indicated by thicker lines.

W thyrsiflora is basal in the genus and has many more

autapomorphies than any of the other species, indicated

by the long patristic distance from its basal node. This sug-

gests that W. thyrsiflora has diverged morphologically from

the rest of the genus. It also occupies the most peculiar

habitat, in perennially wet marshes and along streams,

whereas the other species are all in habitats that are at least

seasonally dry, and the morphological divergence may

reflect this habitat specialisation.

The remaining three species occupy relatively similar

habitats, and also show relatively low patristic differences

between them. W. paniculata has a wide ecological and

geographical range, but is restricted to well-drained

habitats. W. brachyandra is restricted to the southwestern

Cape, where it occurs in seasonally wet places and forest

margins. There is no evidence to date of it occurring with

any of the other species, but the stamen-petal ratio may

reflect a pollination isolation mechanism.

W parviflora is often found occurring with W. paniculata

and appears to be ecologically very similar. However, there

is a temporal separation in the flowering time, with W.

parviflora flowering earlier than W. paniculata. This sug-

gests that W parviflora may be a neotenous form of W

paniculata. Morphologically, the hairy leaves and narrow

tepals may be juvenile structures (although this has not

been substantiated with ontogenetic studies on W. panicu-

lata and W. brachyandra). The speciation mechanism that

may lead to such a neotenous form is obscure. Unfortu-

nately this species has not been investigated cytologically,

so the possibility of cytological reorganization underlying

this evolution is not excluded.

The short patristic distance between the ancestral node

and W. paniculata suggests that the ancestral habitat of

Wachendorfia was well-drained. Conversely, the large

patristic distance to W thyrsiflora suggests that its peren-

nially wet habitat is a derived habitat. It is interesting that

the diversification in the genus occurred in the dryland

clade, rather than the marshland clade. Although the total

patristic distances of the W. thyrsiflora clade and the W

paniculata clade are the same, the W paniculata clade,

because it contains three species, has a much wider range

of variation. This is expressed morphologically by the

different taxa showing different patristic values, pheno-

logically by the much wider range of flowering times and

ecologically by the range of substrates and moisture

regimes. The total geographical ranges of the two clades

are the same. These results would caution against auto-

74

Bothalia 22,1 (1992)

matically assuming that because the wetland taxon is

taxonomically isolated, it represents the ancestral habitat.

CONCLUSIONS

Wachendorfia consists of three well-defined species and

a fourth extremely variable species. This latter species,

W paniculata, has been the cause of much confusion in

the past: at least seven specific names were given to the

various forms. This variation is continuous, making the

recognition of intraspecific taxa totally arbitrary. There

is never any real geographic separation of these forms,

so the use of subspecific or varietal rank would be inap-

propriate (Stuessy 1990). However, there is a form that

shows a type of geographical separation along with a

distinctively lax morphology (W. paniculata form 2).

Although this form is confined to the mountains between

Ceres and Pakhuis Pass, it is not the only form in the

area, so subspecific rank would not be appropriate (no

allopatry).

Linder (1990) found that ‘using only macro-morphologi-

cal structures at specific level in the Restionaceae only

reflects a portion of the available information and reliance

on such a small portion of the available data set may lead

to mistakes when assessing the relationships among

species’. The present study supports this view and we feel

that the reliance on macro-morphological features is often

the reason for inadequate, confusing species delimitations

within Wachendorfia. The present study assessed not only

macro-morphological variation within Wachendorfia, but

also pollen and seed coat features, rhizome storage

products, and basic features of leaf anatomy. It is hoped

that this broader data base strengthens the proposed clas-

sification of Wachendorfia.

The cladistic analysis suffered from a lack of informa-

tive characters and the resulting cladogram should be

interpreted with caution. However, the diagram of patristic

distances (Figure 11) illustrates the difference between the

habitat specialist (W. thyrsiflora) and the habitat generalists,

the former having many more derived traits, suggesting

that ecological specialization has occurred. W. parvifiora

is notable for the number of derived characters that support

the idea of it being a neotenous species.

The reproductive biology is not sufficiently well known

to understand the evolution of the peculiar form of floral

enantiomorphy in Wachendorfia. Although there is a

phylogeny at generic level (Simpson 1990) the functional

purpose of enantiomorphy and extrafloral nectaries is not

understood. This may well be linked closely to the polli-

nation biology in the genus.

ACKNOWLEDGEMENTS

We wish to thank the Electron Microscope Unit of the

University of Cape Town for assistance with the survey

of seed and pollen morphology, and the Directors of PRE,

NBG, SAM and STE for the loan of herbarium material.

Wendy Hitchcock made the drawings and Mike Baumgart-

ner helped with computing. A. Nicholas provided ready

assistance at Kew and G. Duncan kindly looked after live

plants at Kirstenbosch. We wish to thank the Foundation

for Research Development for funding this research.

REFERENCES

ANDERSON, L.E. 1954. Hoyer’s solution as a rapid permanent mounting

medium for bryophytes. The Bryologist 57: 242—244.

BAKER, J.G. 1896. Haemodoraceae. In W.T. Thiselton-Dyer, Flora

capensis, Vol. 6. Reeve, London.

BARKER, W.F. 1949. Plantae novae africanae, Series XXIX. Journal

of South African Botany 15: 39—42.

BARKER, W.F. 1950. Haemodoraceae. In R.S. Adamson & T.M. Salter,

Flora of the Cape Peninsula. Juta, Cape Town.

BOND, P. & GOLDBLATT, P. 1984. Plants of the Cape Flora. Journal

of South African Botany Supp. Vol. 13.

BREYNE, J. 1739. Prodromus fasciculi rariorum plantarum. Breyne,

Danzig.

BURMAN, J. 1757. Wachendorfia. Amsterdam.

BURMAN, N.L. 1768. Flora indica. Cornelius Haak, Leiden.

COTHENIUS, C.A. VON 1790. Dispositio vegetabilium methodica.

Berlin.

DE VOS, M.P. 1956. Studies on the embryology and relationships of

South African genera of the Haemodoraceae: Dilatris Berg, and

Wachendorfia Burm. Journal of South African Botany 22: 41—63.

DYER, R.A. 1976. The genera of southern African flowering plants. Vol.

2. Dept, of Agricultural Technical Services, Pretoria.

ERDTMAN, G. 1960. The acetolysis method. A revised description.

Svensk Botanisk Tidskrift 54: 561—564.

ERDTMAN, G. 1966. Plant morphology and plant taxonomy: angio-

sperms. Hafner, New York.

FAEGRI, K. & VAN DER PIJL, L. 1966. The principles of pollination

ecology. Pergamon Press, Oxford.

FARRIS, J.S. 1988. Hennig86 Reference, Version 1.5. Privately published.

GEERINCK, D. 1969. Genera des Haemodoraceae et des Hypoxidaceae.

Bulletin du Jardin Botanique National de Belgique 39: 47—82.

GIBBS RUSSELL, G.E., RETIEF, E. & SMOOK, L. 1984. Intensity

of plant collecting in southern Africa. Bothalia 15: 131—138.

GREEN, J.W. 1959. The genus Conostylis R. Br. I. Leaf anatomy.

Proceedings of the Linnean Society of New South Wales 84:

194-206.

HERBERT, W. 1826. W. paniculata var. j3. Curtis 's Botanical Magazine

53: t. 2610.

HUTCHINSON, J. 1973. The families of flowering plants. 3rd edn.

Clarendon Press, Oxford.

JACKSON, W.P.U. 1987. Origins and meanings of names of South African

plant genera. Ecolab, University of Cape Town.

KER, J.B. 1809. Wachendorfia brevifolia. Dingy-flowered Wachendorfia.

Curtis’s Botanical Magazine 29: t. 1166.

LINDER, H.P. 1990. A morphological study on the Thamnochortus

erectus complex. South African Journal of Botany 56: 443—450.

LINNAEUS, C. 1759. Flora capensis. Wannman, Uppsala.

LINNAEUS, C. fil. 1781. Supplementum plantarum. Braunschweig,

Uppsala.

LOEFLER, P. 1758. Iter hispanicum. Kostnad, Stockholm.

MADDISON, W.P., DONOGHUE, M.J. & MADDISON, D.R. 1984.

Outgroup analysis and parsimony. Systematic Zoology 33: 83—103.

ORNDUFF, R. 1979. Chromosome numbers and relationships of certain

African and American genera of Haemodoraceae. Annals of the

Missouri Botanical Garden 66: 577—580.

ORNDUFF, R. & DULBERGER, R. 1978. Floral enantiomorphy and

the reproductive system of Wachendorfia paniculata (Haemo-

doraceae). New Phytologist 80: 427—434.

PLUKENET, L. 1700. Almagesti botanici mantissa. London.

PRESL, K.B. 1829. Pedilonia, novum plantarum genus. Pragae.

ROSENDAHL, C.O. 1949. The problem of subspecific categories.

American Journal of Botany 36: 24—27.

SALISBURY, R.A. 1796. Prodromus stirpium in horto ad Chapel Allerton

vigentium. London.

SAVAGE, S. 1945. A catalogue of the Linnean Herbarium. Linnean

Society, London.

SIMPSON, M.G. 1983. Pollen ultrastructure of the Haemodoraceae and

its taxonomic significance. Grana 22: 79—103.

SIMPSON, M.G. 1990. Phylogeny and classification of the Haemo-

doraceae. Annals of the Missouri Botanical Garden 77: 722 —784.

SIMPSON, M.G. & DICKISON W.C. 1981. Comparative anatomy of

Lachnanthes and Lophiola (Haemodoraceae). Flora 171: 95—113.

SIMS, J. 1803. Wachendorfia hirsuta. Curtis 's Botanical Magazine 17:

t. 614.

SIMS, J. 1803. Wachendorfia paniculata. Curtis ’s Botanical Magazine

17: t. 616.

STAFLEU, F.A. & COWAN, R.S. 1976. Taxonomic literature, Vol. 1.

Bohn, Scheltema & Holkema, Utrecht.

STUESSY, T.F. 1990. Plant taxonomy. Columbia University Press, New

York.

Bothalia 22,1 (1992)

75

SWEET, R. 1826. Hortus brittanicus. 1st edn. Ridgway, London

SWOFFORD, RL. 1985. Phylogenetic analysis using parsimony. Ver.2.4.

Illinois Natural History Survey, Champaign, Illinois.

THUNBERG, C.R 1794. Prodromus plantarum capensium. Edman,

Uppsala.

THUNBERG, C.R 1807—1913. Flora capensis. Edman, Uppsala.

TOLIVIA, D. & TOLIVIA, J. 1987. Fasga: a new phytochromatic for

simultaneous and differential staining of plant tissue. Journal of

Microscopy 148: 113-117.

WATROUS, L.E. & WHEELER, Q.D. 1981. The outgroup comparison

method of character analysis. Systematic Zoology 30: 1—11.

WILSON, J. 1887. On the dimorphism of the flowers of Wachendorfia

paniculata. Transactions and Proceedings of the Botanical Society

of Edinburgh 17: 73.

Bothalia 22,1: 77-91 (1992)

An overview of Penicillium (Hyphomycetes) and associated teleomorphs

in southern Africa

A.L. SCHUTTE*

Keywords: Eupenicillium , fungi, Hyphomycetes, Penicillium, overview, southern Africa, Talaromyces , taxonomy

ABSTRACT

Literature on the hyphomycete genus Penicillium Link and its teleomorphs, Eupenicillium Ludwig and Talaromyces C.R.

Benjamin, is surveyed in the Republic of South Africa, Lesotho, Mozambique, Namibia, Swaziland and Transkei up to

1990. References are grouped under the headings, general mycology, plant pathology, industrial application, medical importance,

mycotoxins and chemical work. An alphabetical list of the species recorded in southern Africa as well as the host and/or

substrate from which each species has been reported is presented with relevant literature references; specimens in various

culture collections are also incorporated. Although most of the known Penicillium species have already been reported from

southern Africa, in-depth work is still required in all fields of research concerning this genus.

UITTREKSEL

Literatuur aangaande die hifomiseetgenus Penicillium Link en sy teleomorwe Eupenicillium Ludwig en Talaromyces C.R.

Benjamin in die Republiek van Suid-Afrika, Lesotho, Mosambiek, Namibie, Swaziland en Transkei is nagegaan tot 1990.

Die verwysings word gegroepeer onder die opskrifte mikologie, plantpatologie, industriele toepassing, mediese belang,

mikotoksiene en chemiese werk. ’n Alfabetiese lys van die spesies wat in suidelike Afrika aangeteken is, asook die gasheer

en/of substraat waarop elke spesie aangemeld is, word met die toepaslike verwysings gegee; eksemplare in verskeie fungus-

versamelings word ook ingesluit. Alhoewel die meeste van die bekende Penicillium- spesies reeds in suidelike Afrika aangeteken

is, is diepgaande werk op alle navorsingsgebiede random hierdie genus steeds nodig.

INTRODUCTION

‘Species of Penicillium are so abundant and so con-

spicuous in all sorts of stale or decaying organic matter

that they constitute a part of the common conception of

mould, and are loosely referred to as ‘blue’ or ‘green’

mould’ (Raper & Thom 1949). Representatives of this

multi-faceted genus are of ecological importance because

they are abundant and widespread in the environment; they

are fruit deteriorators and contribute greatly to post-harvest

decay; they have industrial applications such as in cheese-

making; and they produce secondary metabolites and

mycotoxins, including the indispensable antibiotics.

The generic name Penicillium (Latin, penicillus = little

brush) was first introduced in 1809 by Link who very

briefly described the genus with three species, namely P.

candidum Link, P. expansum Link and P. glaucum Link.

The true identity of these fungi has been difficult to

determine, but an apple-rotting fungus was linked to P.

expansum by Thom (1910). Although the validity of the

generic name has been questioned over the years,

Hawksworth (1985) concluded that Penicillium Link should

be considered correct and indicated that he had previously

designated a neotype of P expansum Link as the type

species of the genus.

Succeeding the works of Thom (1910, 1930), the manual

by Raper & Thom (1949) has been the standard work on

Penicillium for nearly 30 years. Subsequently, a new era

in Penicillium identification was heralded by Pitt (1973),

who used the ability of isolates to grow at reduced water

* Mycology Unit, Plant Protection Research Institute, Private Bag X134,

Pretoria 0001.

MS. received: 1991-09-26.

activity, correlated with penicillus types, as well as growth

rates at 5°C and 37°C, as differential criteria. This concept

was later fully developed in a monograph (Pitt 1979).

Shortly afterwards, a well-illustrated atlas of penicillia by

Ramirez (1982) was published. However, the value of Pitt’s

(1979) guide to the taxonomy of Penicillium was confirmed

at the First international Penicillium and Aspergillus

workshop (Samson & Pitt 1985), when Pitt’s species

concept and methods were incorporated in the recommen-

dations for future taxonomic practice in this genus.

Previously, the name Penicillium was applied to both

the hyphomycetous and ascomycetous states. However,

separation of the teleomorphic states of Penicillium from

the anamorph, as implemented by Pitt (1979), is in accor-

dance with Art. 59 of the International Code of Botanical

Nomenclature and is of practical value for the taxonomist.

Stolk & Scott (1967) re-introduced the use of the teleo-

morph name Eupenicillium Ludwig for a portion of the

genus Penicillium. Monographic contributions to the genus

Eupenicillium were made by Scott (1968a, b) and Stolk

& Samson (1983). The teleomorphic genus Talaromyces

C.R. Benjamin is separated from Eupenicillium on the

basis of ascocarp morphology. The former genus is charac-

terised by the production of gymnothecia composed of

loosely intertwined hyphae, as opposed to cleistothecia.

Stolk & Samson (1972) as well as Pitt (1979) have

contributed to the taxonomy of this group.

A multidisciplinary approach to the identification of

Penicillium is becoming more prevalent (Bridge et al.

1985). Protein electrophoresis (Bent 1967), the API ZYM

testing system (Bridge & Hawksworth 1984), pyrolysis gas

chromatography (Soderstrom & Frisvad 1984), physio-

logical and biochemical methods (Bridge 1985), enzyme

electrophoresis (Cruickshank & Pitt 1987), studies on

78

Bothalia 22,1 (1992)

thermal denaturation of DNA (Paterson etal. 1990), elec-

tron microscopy (Ramirez 1982; Kozakiewicz 1989) and

the production of secondary metabolites and mycotoxins

(Frisvad & Filtenborg 1983; Frisvad et al. 1990) have

recently been used to supplement traditional methods of

identification.

Members of this genus identified at the Mycology Unit

in recent years were often found to differ somewhat from

the descriptions given by Pitt (1979). This raised the ques-

tion of whether these variations are consistent for all South

African isolates. In addition, preliminary investigations

indicated that Penicillium species are frequently only

briefly mentioned in publications or included in lists of

fungi from surveys. It was therefore considered advan-

tageous to gather this scattered information in order to

compile a list of Penicillium species recorded in southern

Africa, to bring this information in line with modern

taxonomic systems, and to indicate areas requiring further

research.

This paper is an overview of publications dealing with

all aspects of the Penicillium species reported in South

Africa, Lesotho, Mozambique, Namibia, Swaziland and

Transkei up to 1990. Literature is grouped according to

various fields of research and presented in chronological

order. The National Collection of Fungi, including the

dried collection and the culture collection, collections

donated to the Mycology Unit, the collection of the

Medical Research Council as well as catalogues of inter-

national culture collections, served as additional sources

of information. Foreign isolates used for chemical work,

have been mentioned but not listed. No attempt has been

made to verify published data, the identity of Penicillium

isolates, or any other information.

OVERVIEW OF LITERATURE

General mycology

The first published record of the genus Penicillium in

southern Africa appears to be that of P. digitatum (Pers.

ex Fr.) Sacc. on citrus (Pole Evans 1911). In this publica-

tion Pole Evans mentioned that in 1903, the Government

Entomologist for Natal reported great losses to the orange

crop, due to a mould. He noted that he had collected the

causative fungus, P digitatum, from fallen oranges in the

Northern Transvaal five years before (i.e in 1906). Doidge

(1950) listed all Penicillium species recorded up to 1945,

including specimens in the Collection of the Timber

Research Laboratories, Chamber of Mines, Johannesburg,

as well as those mentioned by Thom (1930).

The Penicillium specimen accessioned in the National

Collection of Fungi first was ‘P armeniacum Berk’

(PREM 187 — see checklist), recorded by the Government

Laboratories Johannesburg, on Zea mays on 12 September

1906. This fungus was not a Penicillium, however, but

probably belongs in Monilia (Thom 1930).

The second Penicillium entry, ‘P gratioti Sartory’

(PREM 5587 — see checklist), was recorded by P.A. van

der Bijl from the City Deep Mine in Johannesburg, on

7 December 1912. Thom (1930) provided more data about

this isolate, recording its optimum temperature and utili-

zation of various sugars. Its true identity is not clear,

however, as the name is no longer in use and Raper &

Thom (1949) referred to P gratioti only as: ‘apparently

some member of the P. janthinellum series’.

Numerous penicillia have been reported subsequently

in general surveys of fungi on various substrates. Cohen

(1950) conducted the first survey of soil fungi in South

Africa, comparing the effect of different burning and

grazing treatments, and he recorded nine Penicillium

species. Scott (1968a) described eight new Eupenicillium

species from soil and included these in a more extensive

monograph of the genus (Scott 1968b). Penicillium was

found to be the genus of Fungi Imperfecti with the largest

number of species represented in Zululand soil (Eicker

1969). The same locality yielded P. olsonii Bain. & Sartory

throughout the soil profile, whereas P. javanicum Van

Beyma showed a marked decrease with increasing soil

depth (Eicker 1970). Eicker (1973) found the penicillia to

have an even distribution in different litter layers of

Eucalyptus maculata Hook. f. and later found the genus

to be common on Panicum coloratum L. litter (Eicker

1976). P. cyclopium Wesding was isolated from angora goat

dung, but Mitchel (1970) indicated that it was probably an

aerial contaminant.

High quality stored maize obtained from six localities,

studied by Van der Westhuizen & Bredell (1972) was found

to have a high percentage of Penicillium spp., with P.

oxalicum Currie & Thom often comprising 30% of the

fungi recorded. On stored lucerne seed, species of this

genus did not increase during an increased period of

storage (Marasas & Bredell 1973). The composition and

distribution of soil fungi in the western Transvaal was

studied by Papendorf (1976) and one of his isolates,

described as the new species P striatosporum Stolk (Stolk

1969), was later re-identified by Pitt (1979) as P. restric-

tum Gilman & Abbott. Penicillium spp. were found to be

scarce on leaves and litter of Cenchrus ciliaris L.

(Bezuidenhout 1977), in aerospora of an Eragrostis curvula

(Schrad.) Nees pasture (Van der Merwe et al. 1979) and

in the soil of Kaokoland, Namibia (Eicker et al. 1982).

Many of the above-mentioned species are included in the

checklist and bibliography of South African fungi com-

piled by Gorter (1979) for the period 1947-1977.

Allsop et al. (1987) found a more varied fungal flora

present in the rhizosphere than in the non-rhizosphere area

of a fynbos site; several Penicillium species were reported,

including P. novae-zeelandiae Van Beyma and Eupenicil-

lium pinetorum Stolk, reported in South Africa for the first

time. McLean & Berjak (1987) studied the mycoflora of

maize and indicated P. variabile Sopp as the most frequent

internal contaminant of maize seed, while P. brevicom-

pactum Dierckx was isolated from 15 % of the seedlings.

Wittaker et al. (1989) reported a decline in Penicillium

species after hot water treatment of stored maize seed,

penicillia were found to be present on Eucalyptus (Lund-

quist & Baxter 1985), Pinus in the Transvaal (Lundquist

1986), Pinus in the Cape (Lundquist 1987) and common

on stored seed of indigenous plants (Isaacs & Benic 1990).

P. crustosum Thom and P. purpurescens (Sopp) Biourge

have been indicated as endophytes of grass species (De

Villiers 1989). Ramirez (1990) based the description of P.

krugeri Ramirez on 26 isolates collected from soil at

different localities in the Kruger National Park in 1987.

Bothalia 22,1 (1992)

79

Apparently the type material of this fungus has been lost

(C. Ramirez pers. comm.).

Additional reports of South African isolates may be

found in the monographs on Penicillium by Thom (1930),

Raper & Thom (1949), Pitt (1979) and Stolk & Samson

(1983), as well as in catalogues of international culture

collections.

Plant pathology

During the early 1900’s the deteriorators, P. digitatum,

P. expansum Link and P. italicum Wehmer, became

a major problem for the fruit producing industry by

hampering exports to Europe (Pole Evans 1920). Most of

the South African isolates mentioned by Thom (1930) had

been sent to the USA for identification by V.A. Putterill.

Putterill was in charge of a mycological laboratory in Cape

Town in 1918, and later worked at the fruit inspection

service (Doidge 1950). These first South African Penicil-

lium records probably concerned fruit rot, although they

are listed as having an undetermined host.

P. digitatum on citrus was reported by Pole Evans (1911)

who stressed the importance of good sanitation in orchards

to combat this fungus. To determine the presence of patho-

genic fungal spores at the Cape harbour, Pole Evans (1920)

exposed agar plates in the railway trucks and in cold

storage rooms on the docks and on the ships.

These pathogens were later listed by Verwoerd (1929).

Doidge & Van der Plank (1936) subsequently conducted

a survey on the fungi causing rot of oranges and lemons,

indicating P. digitatum as the most important, with P.

italicum and P verrucosum Dierckx also present. They

(Doidge & Van Der Plank 1936) remarked that although

a large number of additional Penicillium spp. were isolated

during the survey, no attempt was made to identify these

species which were apparently saprophytic and growing

on decaying tissues. Van der Plank (1945) did experimental

work with hypochlorous acid as a bleach and disinfectant

of citrus fruit, finding it effective against P. digitatum

conidia. Martin (1960) listed seven saprophytic Penicillium

species in citrus soil and found five species in adjacent

virgin soil. Other Penicillium species of plant pathologi-

cal interest were mentioned by Doidge et al. (1953), Roth

(1967), Wager (1972) and Gorter (1977). The bulb pathogen

P corymbiferum Westling, isolated by Wager, was de-

posited in the IMI culture collection where it was examined

by Pitt (1979).

Matthee (1968) studied P expansum, the pathogen and

deteriorator of stored pome fruits, and indicated that older

or bruised fruit was more susceptible. Holtzhausen &

Knox-Davies (1974) used this fungus as an experimental

organism in chemical seed treatments. Combrink et al.

(1980) found that a longer exposure time of apples to a

sodium hypochlorite solution had a better fungicidal effect

on P. expansum conidia than a stronger solution. P

funiculosum Thom reportedly caused a core rot of apples

and formed a moist infection (Combrink et al. 1985).

Members of the genus were also isolated from litchi fruit

(Roth 1963), bananas (Roth & Loest 1965) and mangoes

(Wehner et al. 1981). P. pinophilum Hedgcock apparently

enhances disease symptoms of groundnut pods in the

presence of Chalara elegans Nag Raj & Kendrick (Baard

1988). This fungus was able to decompose filter paper as

well as detached groundnut pods. Surface disinfected roots

of Medicago spp. yielded eight different Penicillium spp.

(Lamprecht et al. 1988). P. spinulosum Thom was found

to be pathogenic on onions (Naude & Jooste 1989) and

P. hirsutum Dierckx on bulbs of flowering plants (Schutte

1990).

Unidentified members of the genus were reported on

Japanese radish seed (Holtzhausen 1978), groundnuts

(Ferreira & Lutchman 1989), recalcitrant seed (Berjak et

al. 1989; Mycock & Berjak 1990), barley seed (Liibben

et al. 1989) and maize cultivars (Rheeder et al. 1990).

Industrial applications

Penicillia encountered in industry were first reported

by Van der Bijl (1920) in his study of the deterioration of

cane sugar crystals and solutions in storage. This record

is also of taxonomic interest as two of these Penicillium

isolates had been sent to Thom, whose comments accom-

panying the identifications are included. One of these

isolates was deposited in PREM: 14262 P. luteum-

purpurogenum group.

Davel & Neethling (1930) dealt with fungi in dairies and

mentioned the use of P. camembertii Thom, P glaucum

and P roquefortii Thom in cheese factories, indicating that

members of this group can be troublesome in these

surroundings. Coles (1925) recorded P. glaucum on Stilton

and Wenslydale cheese and Radmore (1986) did a micro-

biological study of air in dairies. Other work done on

penicillia in the dairy industry is discussed under the

heading ‘Mycotoxins’.

An interesting use for Penicillium was found in reducing

the stickiness of molasses meal (Roth 1968), for which

P. notatum Westling was used on a commercial scale.

Although photographs of eight different Penicillium spp.

are included, only the series to which they belong are

given. The wine industry noted various identified and

unidentified Penicillium spp. on grapes (Le Roux et al.

1973), their incidence on healthy grapes being 60% and

on Botrytis infected fruit 70%.

Heat resistant fungi posing problems for apple juice

canners, turned out to be teleomorphs of P. vermiculatum

Dangeard and P. brefeldianum Dodge (Van der Spuy et

al. 1975). This work is referred to world-wide in connec-

tion with heat resistance of fungal spores. The thermophilic

Talaromyces dupontii Griffen & Maublanc, was isolated

during a study of fungi in mushroom compost (Eicker

1977). Penicillium species encountered later when various

casings for mushroom production were tested, were

indicated as potentially harmful (Smit 1984). Martin &

Keen (1978) found P. crustosum to be common in home-

made beer as well as on sorghum malt used for brewing.

A low incidence of Penicillium spp. on commercial and

industrial sorghum malt was reported by Rabie & Liibben

(1984).

Medical importance

Although members of the genus are known to cause

allergies and to produce mycotoxins, Penicillium is men-

tioned infrequently in literature on medical mycology.

80

Bothalia 22,1 (1992)

Fungal allergy was the motivation for three five-year

surveys of aerospora, two done in Johannesburg (Ordman

& Etter 1956; Ordman 1963) and one in Windhoek

(Ordman 1970). Penicillium made up about 10% of the

fungi isolated and showed no seasonal prevalence. Fungal

contamination of food was investigated by Gilman (1972),

in an attempt to correlate diet and liver cancer in man and

a variety of identified penicillia were listed. Antimycotic

and antibacterial activity of soil fungi was studied by

Eicker (1975) who found positive effects against both

organisms, by P. chrysogenum and P cyclopium. Horwitz

& Wehner (1977) warned that the presence of antibiotics

produced by P. chrysogenum Thom used in salami curing

may pose a health hazard for persons sensitive to penicil-

lin. Penicillium was also amongst the fungi present on com

believed to be the cause of oesophageal cancer in Transkei

and in the high rate area of the disease, 43 % of the samples

were infected with this organism (Marasas et al. 1981).

Marasas & Van Rensburg (1986) found this genus most

prevalent on crops in the area where Mseleni joint disease

occurs in Kwazulu . Some of the work mentioned under

the heading ‘Mycotoxins’ also has a medical application.

Mycotoxins

The discovery in the 1960’s of aflatoxin and its carcino-

genic effects created renewed interest in fungal contamina-

tion. In the search for members of the aflatoxin-producing

Aspergillus flavus group, numerous species of the closely

related genus Penicillium were also encountered and details

of their distribution recorded. Scott (1965), the first South

African to test fungi for toxicity by feeding day old

ducklings with infected meal, found P. islandicum Sopp,

P oxalicum, P rubrum Stoll and P. urticae to be acutely

toxic, whereas P. piceum Raper & Fennell had a less severe

effect. This paper subsequently became a citation classic.

The fungal flora of stock feeds, and the incidence of

toxicity, was investigated by Van Warmelo (1967), who

found that Penicillium had a low incidence on these

substrates. Wehner & Rabie (1970) did toxicity tests with

micro-organisms from nuts and dried fruit, including P

frequentans Westling, P notatum and three unidentified

Penicillium spp., none of which turned out to be toxic.

Martin (1974) compiled a table of all information avail-

able on mycotoxin-producing fungi, dividing them into

field and storage fungi. Mutagenicity of Penicillium

mycotoxins to Salmonella typhimurium was studied by

Wehner et al. (1978) and negative results were reported

for griseofulvin, patulin and penicillic acid. In a similar

study, the mycotoxin emodin, produced by P rugulosum

Thom, was found to be a frameshift mutagen (Wehner et

al. 1979). As no local isolates were mentioned in the above-

mentioned work, the species concerned have not been

included in the appended list.

The presence of mycotoxin-producing fungi on cheese

was investigated by Liick et al. (1976) and unidentified

Penicillium spp. were isolated from 33 out of 43 cheese

samples. Some of the isolates tested had a toxic effect on

ducklings. Seven isolates of P. roquefortii, isolated from

blue cheese showed a variation in toxicity, whereas the four

isolates of P. camembertii tested had a less pronounced

effect (Liick et al. 1978). A noteworthy finding of Liick

& Wehner (1979) was that Penicillium isolates grown on

maize were more toxic to ducklings than those grown on

milk curd. Kriek & Wehner (1981) proved the toxicity of

P. italicum, isolated from an orange, to laboratory animals.

The effect of maize meal infected with this fungus was

not as detrimental to ducklings as to rats. The nature of

the lesions observed in rats was similar to those caused

by the toxic P. islandicum. Dutton & Westlake (1985) found

the incidence of Penicillium spp. as well as contamination

by its mycotoxins to be low on cereal and animal feedstuffs.

Kellerman et al. (1988) implicated Penicillium as a

mycotoxin producer but gave no examples. The Medical

Research Council tested various isolates of 30 Penicillium

spp. for toxicity to ducklings and found most to have a

detrimental effect (C.J. Rabie pers. comm.). All isolates

were identified by J.I. Pitt and are listed under the

abbreviation MRC. These authors all studied the relation-

ship between fungi and mycotoxins, but the mycotoxins

themselves called for more detailed chemical studies.

Chemical work

A variety of Penicillium mycotoxins have been extracted

and characterized in South Africa. Steyn (1969) described

a new, rapid and sensitive system for the separation and

detection of eleven different mycotoxins, followed by work

on secalonic acid D, a toxic metabolite of P. oxalicum

(Steyn 1970). The isolation of viridicatum toxin from P.

viridicatum Westling was reported by Hutchison et al.

(1973). Nagel et al. (1972) reported on the production of

the highly toxic citreoviridin and made a study of the

morphological characteristics of various isolates of its

producer, P. pulvillorum Turfitt. Steyn et al. (1982) studied

the biosynthesis of the above-mentioned citreoviridin.

Holzapfel (1968), Steyn et al. (1975), McGrath et al.

(1976) and Neethling & McGrath (1977) studied various

aspects of cyclopiazonic acid (e.g. biosynthesis, structure

and production), a toxic metabolite of P. cyclopium.

However, Frisvad (1989) stated that the isolate used for

all the above-mentioned cyclopiazonic acid work, namely

CSIR 1085, was not P. cyclopium but P griseofulvum

Dierckx. Pitt came to the same conclusion as indicated

by De Jesus et al. (1981). Frisvad (1989) stated that P.

viridicatum (CSIR 1029) used by Hutchison et al. (1973)

had also been misidentified.

Various mycotoxins other than the above-mentioned

were studied locally. Oxalin produced by P. oxalicum

received attention from Nagel et al. (1976), Vleggaar &

Wessels (1980) and Steyn & Vleggaar (1983), while PR

toxin produced by P. roquefortii was studied by Gorst-

Allman & Steyn (1982). Certain isolates of P. crustosum

are able to produce tremorgenic mycotoxins and these were

examined in detail by Maes et al. (1982) and De Jesus et

al. (1983a, b, c). P. janthinellum Biourge, associated with

rye grass staggers was found to produce janthitrems,

tremorgenic mycotoxins studied by De Jesus et al. (1984).

For most of these investigations the authors obtained

authenticated isolates or had their fungal cultures verified,

mostly by Pitt.

In 1985, South Africa hosted the IUPAC Symposium on

mycotoxins and phycotoxins (Steyn & Vleggaar 1986)

where a paper concerning synthesis of the Penicillium

mycotoxins cyclopiazonic acid and viridamine was

presented by Holzapfel (1986).

Bothalia 22,1 (1992)

81

DISCUSSION

The large number of undetermined Penicillium species

in the literature cited is an indication that scientists in South

Africa have a history of not attempting to identify mem-

bers of this genus. Other than that done by Scott (1968a,

b), work published on Penicillium in South Africa is

clearly fragmentary and many of the isolates obtained early

this century were identified overseas. The use of correctly

identified Penicillium isolates in any scientific research

must be stressed. Mistaken identities have been reported

for South African studies (Frisvad 1989); voucher speci-

mens deposited in recognized culture collections will assist

in overcoming this problem and will also make isolates

available to other scientists.

With the exception of P hordei Stolk, P. olivicolor Pitt

and Talaromyces stipitatus (Thom) C. R. Benjamin, all

the Penicillium species listed by Samson & Pitt (1985)

as common, have been recorded in southern Africa.

However, teleomorphic penicillia have been reported

infrequently as they require special isolation techniques

(Scott 1968b). Synnematous members of the genus appear

to be scarce and most representatives in the National

Collection of Fungi, PREM and PPRI are recent acquisi-

tions.

The role that penicillia play in the ecology of natural

ecosystems as well as in cultivated areas, has not been

investigated in this country. Certain Penicillium species

have antimycotic as well as antibacterial activities (Eicker

1975). Others are strongly antagonistic to soil-borne plant

pathogens such as Gaeumannomyces , Pythium and Rhizoc-

tonia, whereas some members of Talaromyces have

antifungal as well as antiprotozoal capacities (Domsch et

al. 1980). Biological control of plant pathogens by Penicil-

lium species deserves attention, as it may well be of

economic importance.

The successful use in Penicillium taxonomy of physio-

logical and various biochemical methods, mycotoxin

profiles and electron microscopy, has been indicated.

However, these techniques have not yet been applied to

this genus in South Africa and may be of value in

determining relationships between species and groups as

well as indicating new species.

Much meaningful work on Penicillium, one of the more

common and economically important genera of fungi, is

therefore still to be done in the fields of taxonomy, ecology,

biological control and chemotaxonomy.

ACKNOWLEDGEMENTS

The encouragement of members of staff and assistance

of Alice Baxter of the Plant Protection Research Institute

in preparing this manuscript, is gratefully acknowledged.

REFERENCES

ALLSOPP, N., OLIVIER, D. L. & MITCHELL, D.T. 1987. Fungal popu-

lations associated with root systems of proteaceous seedlings at

a lowland fynbos site in South Africa. South African Journal of

Botany 53: 365-369.

BAARD, S.W. 1988. Interactions between fungi from blackhulled ground-

nut pods. Phytophylactica 20: 61—64.

BENT, K.J. 1967. Electrophoresis of proteins of 3 Penicillium species

on acrylamide gels. Journal of General Microbiology 49:

195-200.

BERJAK, P, FARRANT, J.M., MYCOCK, D.J. & PAMMENTER,

N.W. 1989. Deterioration and pathology of recalcitrant seeds.

Twenty-seventh annual congress of the South African Society for

Plant Pathology. Phytophylactica 21: 106 (Abstract).

BEZUIDENHOUT, H. 1977. 'n Ondersoek van die Hyphomycetes ge-

assosieer met Cenchrus ciliaris L. M.Sc. thesis. University of

Pretoria.

BRIDGE, P. D. 1985. An evaluation of some physiological and biochem-

ical methods as an aid to the characterization of species of

Penicillium subsection Fasciculata. Journal of General Microbi-

ology 131: 1887-1895.

BRIDGE, P.D. & HAWKSWORTH, D.L. 1984. The API ZYM enzyme

testing system as an aid to the rapid identification of Penicillium

isolates. Microbiological Sciences 1: 232—234.

BRIDGE, P.D. , HAWKSWORTH, D.L., KOZAKIEWICZ, Z.,

ONIONS, A.H.S., PATERSON, R.R.M. & SACKIN, M.J. 1985.

An integrated approach to Penicillium systematics. In R.A.

Samson & J.I.Pitt, Advances in Penicillium and Aspergillus

systematics. Plenum Press, New York.

COHEN, C. 1950. The occurrence of fungi in the soil, after different

burning and grazing treatments of the veld in the Transvaal. South

African Journal of Science 46: 260—264.

COLES, L. 1925. The manufacture of Stilton and Wensleydale cheese.

Union of South Africa Department of Agriculture Reprint No.

20: 1-10.

COMBRINK, J.C., KOTZE, J.M., WEHNER, F.C. & GROBBELAAR,

C.J. 1985. Fungi associated with core rot of Starking apples in

South Africa. Phytophylactica 17: 81—83.

COMBRINK, J.C., MATTHEE, F.N. & KOCK, J.S. 1980. Stability of

sodium hypochlorite solutions and their effectiveness against

conidia of Penicillium expansum Thom. Phytophylactica 12:

85-87.

CRUICKSHANK, R.H. & PITT, J.I. 1987. Identification of species in

Penicillium subgenus Penicillium by enzyme electrophoresis.

Mycologia 79: 614—620.

DAVEL, H.B. & NEETHLING, H.L. 1930. Skimmels en hul bestryding

in die melkery. Union of South Africa Department of Agriculture

Science Bulletin No. 80: 1—18.

DE JESUS, A.E., GORST-ALLMAN, C.P, STEYN, P.S., VAN

HEERDEN, F.R., VLEGGAAR, R. & WESSELS, PL. 1983a.

Tremorgenic mycotoxins from Penicillium crustosum. Biosynthesis

of Penitrem A. Journal of the Chemical Society Perkin Trans 1

8: 1863-1983.

DE JESUS, A.E., STEYN, P.S., VAN HEERDEN, F.R. & VLEG-

GAAR, R. 1984. Structure elucidation of the janthitrems, novel

tremorgenic mycotoxins from Penicillium janthinellum. Journal

of the Chemical Society Perkin Trans 1 4: 697—701.

DEJESUS, A.E., STEYN, P.S., VAN HEERDEN, F.R., VLEGGAAR,

R. & WESSELS, P.L. 1983b. Tremorgenic mycotoxins from

Penicillium crustosum: isolation of penitrems A-F and the structure

elucidation and absolute configuration of penitrem A1 . Journal

of the Chemical Society Perkin Trans 1 8: 1847—1983.

DEJESUS, A.E., STEYN, P.S^, VAN HEERDEN, F.R., VLEGGAAR,

R. & WESSELS, P.L. 1983c. Tremorgenic mycotoxins from

Penicillium crustosum. Structure elucidation and absolute con-

figuration of penitrems B-F1 . Journal of the Chemical Society

Perkin Trans 1 8: 1857—1861.

DE JESUS, A.E., STEYN, P.S., VLEGGAAR, R., KIRBY, G.W.,

VARLEY, M.J. & FERREIRA, N.P. 1981. Biosynthesis of

a-cyclopiazonic acid. Steric course of proton removal during

cyclisation of /3-cyclopiazonic acid in Penicillium griseofulvum.

Journal of the Chemical Society Perkin Trans 1 12: 3292—3294.

DE VILLIERS, J.C. 1989. Endofitiese fungusse teenwoordig in ’n aantal

Suid-Afrikaanse grasspesies. M.Sc. thesis. University of Potchef-

stroom for C.H.E.

DOIDGE, E.M. 1950. The South African fungi and lichens to the end

of 1945. Bothalia 5: 1-1094.

DOIDGE, E.M., BOTTOMLEY, A M., VAN DER PLANK, J.E. &

PAUER, G.D. 1953. A revised list of plant diseases in South Africa.

Union of South Africa Department of Agriculture Science Bulletin

No. 346: 1-125.

DOIDGE, E.M. & VAN DER PLANK, J.E. 1936. The fungi which cause

rots of stored citrus fruits in South Africa. Union of South Africa

Department of Agriculture and Forestry Science Bulletin No. 162:

1-23.

DOMSCH, K.H., GAMS, W. & ANDERSON, T-H. 1980. Compendium

of soil fungi. Academic Press, London.

82

Bothalia 22,1 (1992)

DUTTON, M.F. & WESTLAKE, K. 1985. Occurrence of mycotoxins

in cereals and animal feedstuffs in Natal, South Africa. Journal

of the Association of Official Analytical Chemists 68: 839—842.

EICKER, A. 1969. Microfungi from surface soil of forest communities

in Zululand. Transactions of the British Mycological Society 53:

381-392.

EICKER, A. 1970. Vertical distribution of fungi in Zululand soils.

Transactions of the British Mycological Society 55: 45—57.

EICKER, A. 1973. The mycoflora of Eucalyptus maculata leaf litter. Soil

Biology and Biochemistry 5: 441—448.

EICKER, A. 1975. A survey of the antimycotic and antibacterial activity

of soil microflora from Transvaal. Journal of South African Botany

41: 187-198.

EICKER, A. 1976. Non-parasitic mycoflora of the phylloplane and litter

of Panicum coloratum. Transactions of the British Mycological

Society 67: 275 -281.

EICKER, A. 1977. Thermophilic fungi associated with the cultivation

of Agaricus bisporus (Lange) Singer. Journal of South African

Botany 43: 193 —207.

EICKER, A., THERON, G.K. & GROBBELAAR, N. 1982. ’n Mikro-

biologiese studie van ‘kaal kolle’ in die Giribesvlakte van

Kaokoland, S.W.A.-Namibie. South African Journal of Botany

1: 69-74.

FERREIRA, N.P. & LUTCHMAN, V.S. 1989. Fungal contamination

of groundnuts in the northern Transvaal. Twenty-seventh annual

congress of the South African Society for Plant Pathology.

Phytophy lactic a 21: 107 (Abstract).

FRISVAD, J.C. 1989. The connection between the penicillia and aspergilli

and mycotoxins with special emphasis on misidentified isolates.

Archives of Environmental Contamination and Toxicology 18:

452-467.

FRISVAD, J.C. & FILTENBORG, O. 1983. Classification of terverticillate

penicillia based on profiles of mycotoxins and other secondary

metabolites. Applied and Environmental Microbiology 46:

1301-1310.

FRISVAD, J.C., SAMSON, R.A. & STOLK A.C. 1990. Notes on the

typification of some species of Penicillium. Persoonia 14:

193-202.

GILMAN, G. A. 1972. Report of a survey of food storage and handling

techniques used in eastern Transvaal and Swaziland with special

reference to fungal contamination in the diet and the incidence

of liver cancer in man. Feb. 1968— Dec. 1971 R397: 1—126. Tropical

Products Institute, Ministry of Overseas Development, London.

GORST-ALLMAN, C.P & STEYN, PS. 1982. Biosynthesis of PR toxin

by Penicillium roquefortii. Part 2. Evidence for an hydride shift

from 2H N.M.R. spectroscopy. Tetrahedron Letters 23: 5359—

5362.

GORTER, G.J.M.A. 1977. Index of plant pathogens and the diseases

they cause in cultivated plants in South Africa. Republic of South

Africa, Department of Agricultural Technical Services Science

Bulletin No. 392: 1-177.

GORTER, G.J.M.A. 1979. An annotated checklist and selected biblio-

graphy of South African fungi for the period 1946—1977. Republic

of South Africa, Department of Agricultural Technical Services

Technical Communication No. 163: 1—34.

HAWKSWORTH, D.L. 1985. The typification and citation of the generic

name Penicillium. In R.A. Samson & J.I. Pitt, Advances in

Penicillium and Aspergillus systematics. Plenum Press, New York.

HOLZAPFEL, C.W. 1968. The isolation and structure of cyclopiazonic

acid, a toxic metabolite from Penicillium cyclopium Westling.

Tetrahedron Letters 24: 2101—2119.

HOLZAPFEL, C.W. 1986. Chemical studies relevant to the synthesis

of some mycotoxins: the cyclopiazonic acids and viridamine. In

PS. Steyn & R. Vleggaar, Mycotoxins and phytotoxins. Elsevier,

Amsterdam.

HOLTZHAUSEN, M.A. 1978. Seed-borne fungal pathogens and dis-

eases of Japanese radish and their control in South Africa.

Phytophylactica 10: 107—114.

HOLTZHAUSEN, M.A. & KNOX-DAVIES, PS. 1974. Pathogens of

cruciferous vegetable crops in commercial seed samples in South

Africa. Phytophylactica 6: 289—294.

HORWITZ, C. & WEHNER, F.C. 1977. Antibiotics in mould-cured

salami. South African Medical Journal 52: 669.

HUTCHISON, R.D., STEYN, PS. & VAN RENSBURG, S.J. 1973.

Viridicatumtoxin, a new mycotoxin from Penicillium viridicatum

Westling. Toxicology and Applied Pharmacology 24: 507-509.

ISAACS, J. & BENIC, L.M. 1990. A study of the fungal mycoflora

associated with stored seeds of eight indigenous families. Twenty-

eighth annual congress of the South African Society for Plant

Pathology. Phytophylactica 22: 154 (Abstract).

KELLERMAN, T.S., COETZER, J.A.W., & NAUDE, T.W. 1988. Plant

poisonings and mycotoxicoses of livestock in southern Africa.

Oxford University Press, Cape Town.

KOZAKIEWICZ, Z. 1989. Ornamentation types of conidia and coni-

diogenous structures in fasciculate Penicillium species using

scanning electron microscopy. Botanical Journal of the Linnean

Society 99 : 273 -293.

KRIEK, N.P.J. & WEHNER, F.C. 1981. Toxicity of Penicillium italicum

to laboratory animals. Food and Cosmetics Toxicology 19: 311—315.

LAMPRECHT, S.C., KNOX-DAVIES, PS. & MARASAS, W.F.O. 1988.

Fungi associated with root rot of annual Medicago spp. in South

Africa. Phytophylactica 20: 281-286.

LE ROUX, G., ESCHENBRUCH, R. & DE BRUIN, S.I. 1973. The

microbiology of South African wine-making. Part 8. The

microflora of healthy and Botrytis cinerea infected grapes.

Phytophylactica 5: 51—54.

LUBBEN, A., & RABIE, C.J. & MARASAS, W.F.O. 1989. Die voor-

koms en toksisiteit van fungi op garssaad in suidelike Afrika.

Twenty-seventh annual congress of the South African Society for

Plant Pathology. Phytophylactica 21: 107 (Abstract).

LUCK, H. & WEHNER, F.C. 1979. Mycotoxins in milk and dairy

products. South African Journal of Dairy Technology 11: 169—176.

LUCK, H., WEHNER, F.C., PLOMP, A. & STEYN, m'. 1976. Myco-

toxins in South African cheese. South African Journal of Dairy

Technology 8: 107—110.

LUCK, H., WEHNER, F.C. & STEYN, M. 1978. Toxicity of Penicillium

spp. from mould-ripened cheeses. South African Journal of Dairy

Technology 10: 191, 192.

LUNDQUIST, J.E. 1986. Fungi associated with Pinus in South Africa

Part 1. The Transvaal. South African Forestry Journal 138: 1—14.

LUNDQUIST, J.E. 1987. Fungi associated with Pinus in South Africa.

Part 2. The Cape. South African Forestry Journal 140: 4—15.

LUNDQUIST, J.E. & BAXTER, A.P 1985. Fungi associated with

Eucalyptus in South Africa. South African Forestry Journal 135:

9-19.

MAES, C.M., STEYN, PS. & VAN HEERDEN, F.R. 1982. High per-

formance liquid chromatography and thin-layer chromatography

of penitrems A-F, tremorgenic mycotoxins from Penicillium

crustosum. Journal of Chromatography 234: 489—493.

MARASAS, W.F.O. & BREDELL, I.H. 1973. Mycoflora of South Afri-

can lucerne (Medicago saliva L.) seed. Phytophylactica 5: 89—94.

MARASAS, W.F.O. & VAN RENSBURG, S.J. 1986. Mycotoxicological

investigations on maize and groundnuts from the endemic area

of Mzeleni joint disease in Kwazulu. South African Medical

Journal 69 : 369-374.

MARASAS, W.F.O., WEHNER, F.C., VAN RENSBURG, S.J. & VAN

SCHALKWYK, D.J. 1981. Mycoflora of com produced in human

oesophageal cancer areas in Transkei, Southern Africa. Phyto-

pathology 71: 792-796.

MARTIN, J.P 1960. Fungi and nematodes in South African citrus orchard

soils in relation to the citrus replant problem. Soil Science Society

of America Proceedings 24: 469—472.

MARTIN, P.M.D. 1974. Fungi associated with common crops and crop

products and their significance. South African Medical Journal

48 : 2374 -2378.

MARTIN, P.M.D. & KEEN, P. 1978. The occurrence of zearalanone

in raw and fermented products from Swaziland and Lesotho.

Sabouraudia 16: 15—22.

MATTHEE, F.N. 1968. Skimmelvrot by kernvrugte. Deciduous Fruit

Grower 18: 109-111.

McGrath, r.m., steyn, ps., ferreira, n.p & neethling,

D.C. 1976. Biosynthesis of cyclopiazonic acids in Penicillium

cyclopium : the isolation of dimethylallylpurophosphate: cyclo-

acetoacetyltryptophanyl dimethylallyltransferase. Bio-organic

Chemistry 4: 11—23.

McLEAN, M. & BERJAK, P. 1987. Maize grains and their associated

mycoflora— a micro-ecological consideration. Seed Science and

Technology 15: 831—850.

MITCHELL, D.T. 1970. Fungus succession on dung of South African

ostrich and angora goat. Journal of South African Botany 36:

191-198.

MYCOCK, D.J. & BERJAK, P 1990. Fungal contamination associated

with several homoiohydrous (recalcitrant) seed species. Phyto-

phylactica 22: 413—418.

NAGEL, D.W., PACHLER, K.G.R., STEYN, PS., VLEGGAAR, R.

& WESSELS, PL. 1976. The chemistry and 13C NMR assign-

ments of oxaline, a novel alkaloid from Penicillium oxalicum.

Tetrahedron Letters 32: 2625—2631.

Bothalia 22,1 (1992)

83

NAGEL, D.W., STEYN, P.S. & SCOTT, D.B. 1972. Production of

citreoviridin by Penicillium pulvillorum. Phytochemistry 2 : 627—

630.

NAUDE, S.P & JOOSTE, W.J. 1989. Five fungal pathogens of storage

onions. Twenty- seventh annual congress of South African Society

for Plant Pathology. Phytophylactica 21: 110 (Abstract).

NEETHLING, D.C. & McGRATH, R.M. 1977. Metabolic development

and mitochondrial changes during cyclopiazonic acid production

in Penicillium cyclopium. Canadian Journal of Microbiology 23:

856 - 872.

ORDMAN, D. 1963. The air-borne fungi in Johannesburg. A second

five-year survey: 1955—1959. South African Medical Journal 37:

325-328.

ORDMAN, D. 1970. Seasonal respiratory allergy in Windhoek: the pollen

and fungus factor. South African Medical Journal 44: 250—253.

ORDMAN, D., & ETTER, K.G. 1956. The air-borne fungi in Johan-

nesburg. A five-year survey as a basis for the study of fungus

allergy in South Africa. South African Medical Journal 30:

1054-1058.

PAPENDORF, M.C. 1976. The soil mycoflora of an Acacia karroo com-

munity in the western Transvaal. Bothalia 12: 123—127.

PATERSON, R.R.M., KING, G.J. & BRIDGE, P.D. 1990. High

resolution thermal denaturation studies on DNA from 14 Penicil-

lium strains. Mycological Research 94: 152—156.

PITT, J.I. 1973. An appraisal of identification methods for Penicillium

species: novel taxonomic criteria based on temperature and water

relations. Mycologia 65: 1135—1157.

PITT, J.I. 1979. The genus Penicillium and its teleomorphic states

Eupenicillium and Talaromyces. Academic Press, London.

POLE EVANS, I.B. 1911. The citrus fruit-rot, caused by the blue mould ,

Penicillium digitatum (Fr.), Sacc. Union of South Africa

Department of Agriculture Farmers Bulletin No. 11: 1—7.

POLE EVANS, I.B. 1920. Report on cold storage conditions for export

fruit at Cape Town. Union of South Africa Department of

Agriculture No. 2: 1—11.

RABIE, C.J. & LUBBEN, A. 1984. The mycoflora of sorghum malt.

South African Journal of Botany 3: 251-255.

RADMORE, K. 1986. A microbiological study of the air in dairy

processing and packaging plants. M.Sc. thesis. University of

Pretoria.

RAMIREZ, C. 1982. Manual and atlas of the penicillia. Elsevier,

Amsterdam.

RAMIREZ, C. 1990. A new species of Penicillium isolated from soil

at the Kruger National Park in South Africa. Mycopathologia

110: 23-29.

RAPER, K.B. & THOM, C. 1949. A manual of the penicillia. Baltimore,

Williams & Wilkins.

RHEEDER, J.P, MARASAS, W.F.O., VAN WYK, PS., DU TOIT, W.,

PRETORIUS, A.J. & VAN SCHALKWYK, D.J. 1990. Incidence

of Fusarium and Diplodia species and other fungi in naturally

infected grain of South African maize cultivars. Phytophylactica

22: 97-102.

ROTH, G. 1963. Post harvest decay of litchi fruit. Republic of South

Africa, Department of Agricultural Technical Services Technical

Communication No. 11: 1—16.

ROTH, G. 1967. Citrus fruit decay in South Africa caused by Penicillium

digitatum Sacc. Phytopathologische Zeitschrift 58: 383—396.

ROTH, G. 1968. Biological treatment of molasses meal to reduce

stickiness. Proceedings of the South African Sugar Technologists’

Association, April 1968: 57-64.

ROTH, G. & LOEST, F.C. 1965. Collar rot of banana hands and its

associated micro-organisms. Republic of South Africa, Depart-

ment of Agricultural Technical Services Technical Communica-

tion No. 44: 1—14.

SAMSON, R.A. & PITT, J.I. 1985. Advances in Penicillium and

Aspergillus systematics. Plenum Press, New York.

SCHUTTE, A.L. 1990. A Penicillium bulb pathogen. Twenty-eighth

annual Congress of the South African Society for Plant Pathology.

Phytophylactica 22: 150 (Abstract).

SCOTT, D.B. 1965. Toxigenic fungi isolated from cereal and legume

products. Mycopathologia et Mycologia Applicata 25: 213—222.

SCOTT, D.B. 1968a. Studies on the genus Eupenicillium Ludwig. 4. New

species from soil. Mycopathologia et Mycologia Applicata 36:

1-27.

SCOTT, D.B. 1968b. The genus Eupenicillium Ludwig. CSIR Research

Report No. 272: 1—150. Pretoria.

SEIFERT, K.A. & SAMSON, R.A. 1985. The genus Coremium and the

synnematous penicillia. In R.A Samson & J.I. Pitt, Advances in

Penicillium and Aspergillus systematics. Plenum Press, New

York.

SMIT, M. 1984. n Studie van verskillende deklaagmateriale in die kom-

mersiele verbouing van Agaricus brunnescens Peck. M.Sc. thesis,

University of Pretoria.

SODERSTROM, B. & FRISVAD, J.C. 1984. Separation of closely related

asymmetric penicillia by pyrolysis gas chromatography and

mycotoxin production. Mycologia 76: 408—419.

STEYN, P.S. 1969. The separation and detection of several mycotoxins

by thin-layer chromatography. Journal of Chromatography 45:

473 -475.

STEYN, P.S. 1970. The isolation, structure and absolute configuration

of secalonic acid D, the toxic metabolite of Penicillium oxalicum.

Tetrahedron Letters 26: 51—57.

STEYN, P.S. & VLEGGAAR, R. 1983. Roquefortine, an intermediate

in the biosynthesis of oxaline in cultures of Penicillium oxalicum.

Journal of Chemical Society Chemical Communications 1983:

560, 561.

STEYN, P.S. & VLEGGAAR, R. 1986. Mycotoxins and phycotoxins.

Papers presented at the sixth International IUPAC Symposium.

Elsevier, Amsterdam.

STEYN, PS., VLEGGAAR, R., FERREIRA, N.P, KIRBY, G.W. &

VARLEY, M.J. 1975. Steric course of proton removal during the

cyclisation of /3-cyclopiazonic acid in Penicillium cyclopium.

Journal of the Chemical Society Chemical Communications 459:

465, 466.

STEYN, PS., VLEGGAAR, R., WESSELS, PL. & WOUDENBERG,

M. 1982. Biosynthesis of citreoviridin. A carbon-13 N.M.R. study.

Journal of the Chemical Society Perkin Trans 1 9 : 2175 —2178.

STOLK, A.C. 1969. Four new species of Penicillium. Antonie van

Leeuwenhoek 35: 261—274.

STOLK, A.C. & SAMSON, R.A. 1972. The genus Talaromyces. Studies

on Talaromyces and related genera 2. Studies in Mycology 2:

1-65.

STOLK, A.C. & SAMSON, R.A. 1983. The ascomycete genus Eupenicil-

lium and related Penicillium anamorphs. Studies in Mycology 23:

1-149.

STOLK, A.C. & SCOTT, D.B. 1967. Studies on the genus Eupenicillium

Ludwig. 1. Taxonomy and nomenclature of penicillia in relation

to their sclerotioid ascocarpic states. Persoonia 4: 391-405.

THOM, C. 1910. Cultural studies of species o/ Penicillium. Bulletin for

the Bureau of Animal Industry 118: 1—109. U.S. Department of

Agriculture.

THOM, C. 1930. The Penicillia. Bailliere, Tindall & Cox, London.

VAN DER BUL, PA. 1920. Studies on some fungi and the deterioration

of sugar. Union of South Africa, Department of Agriculture

Science Bulletin No. 18: 1—19.

VAN DER MERWE, W.J.J., EICKER, A., MARASAS, W.F.O. &

KELLERMAN, T.S. 1979. Aerospora of an Eragrostis curvula

pasture in South Africa. Onderstepoort Journal of Veterinary

Research 46: 19—25.

VAN DER PLANK, J.E. 1945. The use of hypochlorous acid and its

salts in citrus packhouses for bleaching sooty blotch and as

disinfectants against mould. Union of South Africa, Department

of Agriculture and Forestry Science Bulletin No. 241: 1—60.

VAN DER SPUY, J.E. , MATTHEE, F.N. & CRAFFORD, D.J. A. 1975.

The heat resistance of moulds Penicillium vermiculatum Dangeard

and Penicillium brefeldianum Dodge in apple juice. Phytophylac-

tica 7: 105-107.

VAN DER WESTHUIZEN, G.C.A. & BREDELL, I.H. 1972. The fungus

flora of high quality stored maize in South Africa, 1966-1972.

Final Report. Republic of South Africa Department of Agricul-

tural Technical Services. Unpublished.

VAN WARMELO, K.T. 1967. The fungus flora of stock feeds in South

Africa. Onderstepoort Journal of Veterinary Research 34: 439—

450.

VERWOERD, L. 1929. A preliminary checklist of diseases of cultivated

plants in the winter rainfall area of the Cape Province. Union

of South Africa, Department of Agricultural Science Bulletin No.

88: 1-18.

VLEGGAAR, R. & WESSELS, PL. 1980. Stereochemistry of the

dehydrogenation of (2S)-histidine in the biosynthesis of roque-

fortine and oxaline. Journal of the Chemical Society Chemical

Communication 4: 160—162.

WAGER, V.A. 1972. Records of diseases of ornamental plants not

previously recorded in South Africa. Republic of South Africa,

Department of Agricultural Technical Services Technical Com-

munication No. 100: 1—14.

WEHNER, F.C., BESTER, S., KOTZE, J.M. & BRODRICK, H.T. 1981.

Fungi associated with post-harvest decay of mangoes in South

Africa. South African Mango Grower’s Association Research

Report 2: 81-88.

84

Bothalia 22,1 (1992)

WEHNER, F.C. & RABIE, C.J. 1970. The micro-organisms in nuts and

dried fruits. Phytophylactica 2: 165—170.

WEHNER, F.C., THIEL, RG. & DU RAND, M. 1979. Mutagenicity

of the mycotoxin emodin in the Salmonella! microsome system.

Applied and Environmental Microbiology 37: 658—660.

WEHNER, F.C., THIEL, P.G., VAN RENSBURG, S.J. & DEMASIUS,

I.P.C. 1978. Mutagenicity to Salmonella typhimurium of some

Aspergillus and Penicillium mycotoxins. Mutation Research 58:

193-203.

WITTAKER, A., MYCOCK, D.J. & BERJAK, P. 1989. Towards

elimination of seed storage mycoflora: wet-heat trails with

maize. Twenty-seventh Annual Congress of the South Afri-

can Society for Plant Pathology. Phytophylactica 21: 110 (Ab-

stract).

CHECKLIST OF PENICILLIUM, EUPENICILLIUM AND TALAROMYCES SPECIES RECORDED IN SOUTHERN AFRICA

Penicillium species recorded in southern Africa up to 1990 are arranged alphabetically and the host and/or substrate from which each species

has been recorded is given with the relevant literature reference. Species names are listed as cited in the original publication in roman type, with

a cross reference to the epithet currently accepted by Pitt (1979) in bold, except in the case of Eupenicillium , where the revision proposed by

Stolk & Samson (1983) has been followed, or where older epithets have been traced (Seifert & Samson 1985). In the past, ascosporic fungi were

included in the genus Penicillium, posing nomenclatural problems ( A ) in designating the anamorph-teleomorph relationship. Consequently, species

known to produce a teleomorphic state have been listed under Penicillium with a cross reference to either Eupenicillium or Talaromyces , which

are listed separately.

The following abbreviations are used in the list:

CBS, South African isolates listed in the 1990 List of Cultures of the Centraalbureau voor Schimmelcultures, Baarn, The Netherlands.

IMI, cultures in the 1988 Catalogue of the Culture Collection of CAB International Mycological Institute, Kew, United Kingdom.

MRC, isolates in the Culture Collection of the Medical Research Council, all identified by Pitt (C.J. Rabie pers. comm.).

PPRI, isolates in the Culture Collection of the National Collection of Fungi. Several of these have been identified or verified by Pitt.

PREM, isolates deposited in the National Collection of Fungi as dried material.

The National Collection of Fungi recently acquired three additional fungal culture collections. Most of these cultures were no longer viable

and had scant accompanying data, but local isolates are listed with numbers under their appropriate abbreviations:

CSIR, isolates listed in a collection obtained from the Council for Scientific and Industrial Research, which included some isolates of Scott (1968a, b).

MCP, the collection of Papendorf (1976), received from the University of Potchefstroom for C.H.E. These isolates are listed under the substrate

soil, but some isolates could have been isolated from Acacia karroo litter.

UCT, a collection obtained from the University of Cape Town which contained isolates of Allsopp et al. (1987).

GENUS PENICILLIUM

acidoferum (see P. canescens)

aculeatum Raper & Fennell

cereal and legume products: Scott (1965)

soil: CSIR 348

adametzii Zaleski

Allium cepa : PREM 44729

soil: Papendorf (1976); Allsopp etal. (1987); MCP 35, 221, 222, 1159

ventilation tubing: Doidge (1950)

Zea mays: Van der Westhuizen & Bredell (1972)

adametzioides Abe ex G. Smith

Zea mays : McLean & Berjak (1987); PREM 47619

alutaceum (see E. terrenum)

arenicola Chalabuda

mushroom casing: Smit (1984)

asperum (see E. crustaceum)

atramentosum Thom

chicken feathers and droppings: PPRI 4086; PREM 48602

dung: PPRI 3703, 4049; PREM 49878, 50682

atrovenetum (see P. melinii)

armeniacum Berk {Monilia, Thom 1930)

Zea mays: PREM 187

aurantiobrunneum (see P. glabrum)

aurantiocandidum (see P. aurantiogriseum)

aurantiogriseum Dierckx

Arachis hypogaea: MRC 330

Aristea major: PPRI 4302

cheese: PREM 49040, 49042

Hordeum vulgare: MRC 2670

Panicum miUaceum: MRC 245

Vigna subterranea: MRC 284

Zea mays: McLean & Berjak (1987); PREM 47622

= aurantiocandidum Dierckx

soil: Eicker (1969, 1973)

= cyclopium Westling

Allium cepa: PREM 44737

Arachis hypogaea: Gilman (1972)

cereal and legume products: Scott (1965)

cheese: Luck & Wehner (1979)

dung: Mitchell (1970)

natural gum: Roth (1968)

soil: Eicker (1975); CSIR 409; MCP 378

Sorghum caffrorum: CSIR 519, 534, 542, 543

Vitis vinifera: Doidge (1950, et al. 1953)

Zea mays: Van der Westhuizen & Bredell (1972) CSIR 258, 303, 358,

403, 461, 462, 543, 659, 719; PREM 43751, 44302, 44303

= johanniolii Zaleski

undetermined host: Thom (1930)

= lanosocoeruleum Thom

Medicago spp. : Van Warmelo (1967)

Vitis spp.: Le Roux et al. (1973)

= martensii Biourge

Arachis hypogaea: Gilman (1972)

Zea mays: Gilman (1972); Van der Westhuizen & Bredell (1972); CSIR

660

= solitum Westling

material: Doidge (1950)

biforme (see P. camembertii)

brefeldianum (see E. javanicum var. javanicum)

brevicompactum Dierckx

aerospora: Roth (1968)

apple puree: MRC 3137

Avena saliva: MRC 2824

brattice cloth: Doidge (1950)

cereal and legume products: Scott (1965)

compost: PPRI 3186

debris: PPRI 4068

fodder: PPRI 3631

Medicago saliva: PREM 44475, 44477, 44519

natural gum: Roth (1968)

Prunus persica var. nucipersica: PPRI 3597

soil: Eicker (1975); CSIR 327; MCP 371

Bothalia 22,1 (1992)

85

Sorghum caffrorum : CSIR 531, 547

Vitis spp. : Le Roux et al. (1973)

Zea mays : McLean & Berjak (1987); Pitt (1979); Van der Westhuizen

& Bredell (1972); CSIR 81, 95, 219, 330, 378, 459, 593, 623, 665,

675; PPRI 3630; PREM 43741, 43742 , 47537, 47831

undetermined host: CBS 287.53 (albino mutant)

= stoloniferum Thom

soil: Cohen (1950)

Zea mays : Van der Westhuizen & Bredell (1972); CSIR 238

camembertii Thom

cheese: Davel & Neethling (1930); Liick et al. (1978); PPRI 3122; PREM

47740

= biforme Thom

Arachis hypogaea: Van Warmelo (1967)

Medicago spp.: Van Warmelo (1967)

canescens Sopp

Barleria obtusa: PPRI 3808

flannel: PREM 33287

Protea cynaroides: PPRI 3786

soil: Papendorf (1976)

Zea mays: Van der Westhuizen & Bredell (1972)

= acidoferum Sopp (near P. canescens, Raper & Thom 1949)

Citrus sinensis: Doidge (1950)

= kapuscinskii Zaleski

soil: MCP 384

swine meal: Van Warmelo (1967)

capsulatum Raper & Fennell

dried fish: Pitt (1979); IMI 140 284

Medicago sativa: PREM 44469

Zea mays: CSIR 181

casei (see P. roquefortii)

charlesii (see P. fellutanum)

chermesinum Biourge

soil: Martin (1960)

chrysogenum Thom

aerospora: Roth (1968)

Arachis hypogaea: Van Warmelo (1967); PPRI 3658; PREM 48261

cereal and legume products: Scott (1965)

fishmoth gut: PREM 49016, 49017

grass: PPRI 4277

Hordeum vulgare: MRC 2807

Medicago spp.: Lamprecht (1988); PREM 48321

molasses meal: Roth (1968)

mushroom casing: Smit (1984)

natural gum: Roth (1968)

nuts and dried fruit: Wehner & Rabie (1970)

soil: Eicker (1975); Martin (1960); PREM 48767

Sorghum caffrorum: CSIR 427; MRC 1682

Zea mays: Gilman (1972); McLean & Berjak (1987); Van der Westhuizen

& Bredell (1972); Van Warmelo (1967); CSIR 436, 453, 477

= notatum Westling

aerospora: Roth (1968)

Allium cepa: PREM 44738

Cenchrus ciliaris: Bezuidenhout (1977)

cereal and legume products: Scott (1965)

Medicago sativa: PREM 44466, 44552

molasses meal: Roth (1968)

natural gum: Roth (1968)

nuts and dried fruit: Wehner & Rabie (1970)

soil: CSIR 317, 318

Sorghum caffrorum: CSIR 285, 286

Vitis spp. Le Roux et al. (1973)

Zea mays: Van der Westhuizen & Bredell (1972); CSIR 265, 302, 328,

428, 434, 644

undetermined host: Doidge (1950)

citreo nigrum Dierckx

mushroom casing: Smit (1984)

= citreoviride Biourge

Zea mays: Van der Westhuizen & Bredell (1972); CBS 239.65; CSIR

138, 505, 568, 590

citreoviride (see P. citreo nigrum)

citrinum Thom

aerospora: Roth (1968)

Allium cepa: PREM 44777

Arachis hypogaea: Gilman (1972); MRC 241, 263, 283, 294, 2109

Avicennia spp.: PREM 47616, 47617

cereal and legume products: Scott (1965)

dried leaves: MRC 320, 333, 334

fruit: Doidge (1950); Thom (1930)

Ipomoea batatas: PPRI 3571

Manihot esculenta: MRC 212, 232, 249

Medicago spp.: Lamprecht (1988); PREM 48312

natural gum: Roth (1968)

Phaseolus spp.: MRC 178, 210, 222 , 304, 3D

soil: Allsopp etal. (1987); Cohen (1950); Eicker (1969, 1970); Papendorf

(1976); CSIR 370, 372 , 373, 374

Sorghum caffrorum: MRC 2332

Vigna subterranea: MRC 224, 280

Zea mays: Gilman (1972); McLean & Berjak (1987); Van der Westhuizen

& Bredell (1972); CSIR 152 , 352 , 393, 394 , 549, 661, 708; MRC

257, 258, 262, 266, 293, 294, 307, 437, 444; PREM 44304, 44305,

47620, 47621

= steckii Zaleski

Arachis hypogaea: Van Warmelo (1967)

cereal and legume products: Scott (1965)

soil: Eicker (1969, 1970); CSIR 346, 381, 382, 384, 385, 387

Zea mays: Van der Westhuizen & Bredell (1972); Van Warmelo (1967);

CSIR 341, 383, 426, 444, 454, 595, 670; PREM 43752

claviforme (see P. vulpinum)

commune (see P. puberulum)

concentricum (see P. coprophilum)

coprophilum (Berk. & Curt.) Seifert & Samson

cubed dogfood: PPRI 3700

debris: PPRI 3725, 3902, 3903; PREM 49881

dung: PPRI 3726, 4107, 4128; PREM 49863, 50683, 50714

grass roots: PREM 47700

Zea mays: CBS 473.75

soil: PPRI 3611, 4280; PREM 47700, 47701

= concentricum Samson, Stolk & Hadlock

Zea mays: Seifert & Samson (1985)

coralligerum (see P. herquei)

corylophilum Dierckx

aerospora: Doidge (1950); Thom (1930)

Asparagus virgatus: PPRI 3785

contaminant: PREM 48560

lime juice: PPRI 4303

Medicago spp.: Lamprecht (1988); PREM 48316

soil: PPRI 4304

Zea mays: PREM 44307

corymbiferum (see P. hirsutum)

crustosum Thom

Arachis hypogaea: PREM 48018

cheese: PPRI 3892

dried fish: MRC 316

fishmoth gut: PREM 49015

Manihot esculenta: MRC 247

meat pie: MRC 1271

Oryza sativa: MRC 285

Phaseolus spp.: MRC 228

Prunus armeniacum: MRC 3015

Prunus persica: PPRI 3587

soil: Eicker (1975)

Sorghum caffrorum: Martin & Keen (1978)

Stipagrostis uniplumis: De Villiers (1989); PPRI 3457

Zea mays: Gilman (1972); Van der Westhuizen & Bredell (1972); PREM

47864

cyclopium (see P. aurantiogriseum)

dangeardii (see T. flavus)

decumbens Thom

Dalbergia obovata: PPRI 3721; PREM 49888

mushroom casing: Smit (1984)

soil: Martin (1960)

Zea mays: Van der Westhuizen & Bredell (1972); CSIR 2

undetermined host: Thom (1930)

dendriticum Pitt

contaminant: PPRI 4002; PREM 48605

86

Bothalia 22,1 (1992)

debris: PPRI 3782

fodder: PPRI 3887, 4225

Protea scolopendriifolia : PPRI 4014; PREM 47704

Watsonia marginata : PPRI 3724

digitatum (Pers. ex Fr.) Sacc.

aerospora: Pole Evans (1920)

Carica papaya : Doidge (1950, et al. 1953)

Citrus aurantium: Doidge et al. (1953)

citrus fruit: Pole Evans (1911); Roth (1967)

Citrus limonia: Doidge (1950, etal. 1953); CSIR 562, 563; PPRI 3740

Citrus nobilis var. deliciosa : Doidge (1950, et al. 1953)

Citrus parodist: PPRI 3319; PREM 48908

Citrus sinensis: Doidge (1950, et al. 1953); Doidge & Van der Plank

(1936); Van der Plank (1945); Verwoerd (1929); CSIR 558, 561;

PPRI 3737

soil: Eicker (1969, 1973)

= digitatum Sacc. var. califomicum Thom

Physalis peruviana: Doidge et al. (1953)

Citrus sinensis: Doidge (1950); Doidge & Van der Plank (1936); PREM

30659

digitatum var. califomicum (see P. digitatum)

divaricatum Thom (Scopulariopsis, Raper & Thom 1949)

sugar: Van der Bijl (1920)

diversum Raper & Fennell

Eucalyptus cloeziana: PPRI 3731; PREM 49865

Medicago sativa: PREM 44517

duclauxii Delacr.

aerospora: Roth (1968)

Asparagus officinalis: PPRI 4083

grass roots: PPRI 3130; PREM 47754

mine timber: Doidge (1950); Pitt (1979); Raper & Thom (1949); IMI

200 309

molasses meal: Roth (1968)

natural gum: Roth (1968)

soil: PPRI 3983, 4305; PREM 48938

dupontii (see T. thermophilus)

echinulatum Raper & Thom ex Fassatiova

granadilla juice: PPRI 3585

elongatum (see P. expansum)

erubescens (see E. terrenum)

expansum Link

aerospora: Pole Evans (1920)

Arachis hypogaea: Pitt (1979); MRC 199; PREM 48381

cereal and legume products: Scott (1965)

granadilla juice: PPRI 3584; PREM 49415

Malus sylvestris: Combrink et al. (1980); Doidge (1950, et al. 1953);

PPRI 4215

molasses meal: Roth (1968)

natural gum: Roth (1968)

pome fruit: Matthee (1968)

Psidium guajava: PREM 48383

Strelitzia reginae: PPRI 4278

soil: CSIR 398, 410; PPRI 4279

Vigna subterranea: MRC 174

Vitis vinifera: Doidge (1950, etal. 1953); Le Roux et al. (1973); MRC

1131

Zea mays: Van der Westhuizen & Bredell (1972); CSIR 15, 71, 226, 326,

404, 443, 483, 527, 659, 717; MRC 177; PREM 47512

undetermined host: Holtzhausen & Knox-Davies (1974)

= elongatum Dierckx

Vitis vinifera: Doidge (1950, et al. 1953)

fellutanum Biourge

face cream: PPRI 4306

Protea spp. : PPRI 3980

soil: MCP 390, 391

Zea mays: Pitt (1979); CBS 268.65; IMI 162 083, 162 114; CSIR 284

- charlesii G. Smith

cereal and legume products: Scott (1965)

soil: Papendorf (1976); MCP 48, 117

Zea mays: Gilman (1972); Van der Westhuizen & Bredell (1972); CSIR

284, 389, 401, 476

flavidorsum (see P. glabrum)

frequentans (see P. glabrum)

funiculosum Thom

Ananas comosus: Doidge (1950, et al. 1953); PPRI 4307

Arachis hypogaea: Baard (1988); Gilman (1972); Pitt (1979); PPRI 3634;

PREM 48015

cereal and legume products: Scott (1965)

Cyperaceae spp.: PPRI 3632; PREM 48604

Eucalyptus maculata: Eicker (1973)

Malus sylvestris: Combrink et al. (1985)

Medicago sativa: PREM 44513

Phaseolus spp.: MRC 281

soil: Allsopp et al. (1987); Doidge (1950); Eicker (1969, 1973) Martin

(1960); Papendorf (1976); CSIR 141, 362, 365, 367, 368, 369;

MCP 189, 336; PPRI 3504; UCT

Zea mays: Gilman (1972); McLean & Berjak (1987); Van der Westhuizen

& Bredell (1972); CSIR 23, 82, 83, 92, 93, 221, 242, 300, 613;

PPRI 3633; PREM 43754 , 43755, 43756, 43757, 47637

undetermined host: Thom (1930); Raper & Thom (1949)

= varians G. Smith

Zea mays: Van Warmelo (1967)

fuscum (Sopp) Biourge (application uncertain, Pitt 1979)

Medicago sativa: PREM 44401

glabrum (Wehmer) Westling

dung: PPRI 4308

Medicago sativa: PREM 44535, 44550

Melianthus comosus: PPRI 3807

soil: Allsopp et al. (1987)

wine bottle cork: PPRI 3637; PREM 48406

= aurantiobrunneum Dierckx

soil: Cohen (1950)

= flavidorsum Biourge

soil: Cohen (1950)

= frequentans Westling

Allium cepa: PREM 44767

Arachis hypogaea: Gilman (1972)

cereal and legume products: Scott (1965)

nuts and dried fruit: Wehner & Rabie (1970)

soil: Papendorf (1976); MCP 122, 185, 190

Sorghum caffrorum: CSIR 546

Zea mays: Gilman (1972); Van der Westhuizen & Bredell (1972); Van

Warmelo (1967); PREM 44300

gladioli (see E. crustaceum)

glandicola (Oud.) Seifert & Samson

debris: PPRI 3705; PREM 49879

fodder: PREM 48588

grass roots: PPRI 3123

= granulatum Bain,

aerospora: Roth (1968)

molasses meal: Roth (1968)

Triticum aestivum: MRC 1135

Zea mays: Van der Westhuizen & Bredell (1972); PREM 43750

glaucum Link (nomen confusum, Pitt 1979)

cheese: Coles (1925); Davel & Neethling (1930)

Corylus avellana: PREM 23651

nuts: Doidge (1950)

granulatum (see P. glandicola)

gratioti Sartory (indeterminate, Pitt 1979)

underground, gold mine: Doidge (1950); PREM 5587

griseofulvum Dierckx

birdseed: PPRI 3701

cereal and legume products: Scott (1965)

cubed dogfood: PPRI 3306, 3679

Dalbergia obovata: PPRI 3702; PREM 49887

fishmoth gut: PPRI 3123

Manihot esculenta: Pitt (1979); MRC 270, 273

Medicago spp.: Lamprecht (1988); PREM 48317, 48318

silage: CBS 315.63

soil: De Jesus et al. (1981); Cohen (1950); PPRI 4281

Vigna subterranea: MRC 312

Watsonia marginata: PPRI 3809

Zea mays: MRC 214

= urticae Bain.

cereal and legume products: Scott (1965)

soil: CSIR 391

Bothalia 22,1 (1992)

87

Zea mays'. PREM 44308

griseoroseum Dierckx

= roseocitreum Biourge

aerospora: Doidge (1950); Thom (1930)

herquei Bain & Sartory

cereal and legume products: Scott (1965)

debris: PPRI 3904

soil: Eicker (1975); CSIR 359, 360, 361, 363; PPRI 4218; PREM 48559

Zea mays: Gilman (1972); Van der Westhuizen & Bredell (1972); Van

Warmelo (1967); CSIR 364, 402 , 474, 538, 731

= coralligerum Nicot & Pionnat

soil: CSIR 1072

hirayamae (see E. euglaucum)

hirsutum Dierckx

Allium sativum: PPRI 3792; PREM 47862

Asparagus officinalis: PPRI 4219

Omithogalum spp. : Pitt (1979); Schutte (1990); CBS 502.75; PPRI

3795

Gladiolus spp.: PPRI 3598, 3600, 3601, 3602; PREM 49414

= corymbiferum Westling

Omithogalum spp.: Wager (1972); IMI 068 414

humuli Van Beyma

soil: Eicker (1969, 1973)

Eucalyptus maculata: Eicker (1973)

implicatum Biourge

Allium cepa: PREM 44779

Arachis hypogaea: Van Warmelo (1967)

cereal and legume products: Scott (1965)

Zea mays: Van der Westhuizen & Bredell (1972); CSIR 139, 355, 506

inflatum Stolk & Malla

soil: PPRI 3206; PREM 49071

intricatum (see P. jensenii)

inusitatum (see E. inusitatum)

islandicum Sopp

Arachis hypogaea: Gilman (1972)

cereal and legume products: Scott (1965)

contaminant: PPRI 3124, 3714; PREM 47753, 49869

Sorghum caffrorum: Rabie & Liibben (1984)

Zea mays: Gilman (1972); Van der Westhuizen & Bredell (1972)

undetermined host: Raper & Thom (1949); Thom (1930); CBS 176.68

italicum Wehmer

aerospora: Pole Evans (1920)

Citrus sinensis: Doidge et al. (1953); Doidge & Van der Plank (1936);

Verwoerd (1929)

Citrus limonia: Doidge (1950, et al. 1953)

Citrus maxima: Doidge (1950, et al. 1953)

Citrus nobilis var. deliciosa: Doidge (1950); PPRI 3723; PREM 48607

Citrus sinensis: Doidge (1950); Kriek & Wehner (1981); IMI 78 681;

PPRI 4309; PREM 48606

fodder: PREM 48386, 48389

mushroom casing: Smit (1984)

Prunus persica: Doidge (1950)

Prunus salicina: Doidge (1950)

janczewskii Zaleski

Barleria obtusa: PREM 49890

contaminant: PREM 47702

Encephalartos laevifolius: PPRI 3179

Medicago spp.: Lamprecht (1988); PREM 48320

Pirns elliottii: PREM 48907

soil: Allsopp et al. (1987); Eicker (1969); PPRI 3586

undetermined host: CBS 384.67

= nigricans Bain.

cereal and legume products: Scott (1965)

Medicago sativa: PREM 44523

soil: Martin (1960); CSIR 325

Sorghum caffrorum: MRC 1552

Zea mays: Van der Westhuizen & Bredell (1972); Van Warmelo (1967)

janthinellum Biourge

Arachis hypogaea: PREM 48262

brattice cloth: Doidge (1950)

cereal and legume products: Scott (1965)

Medicago spp.: Lamprecht (1988); PREM 48319

soil: CSIR 319, 320, 321, 322, 340, 342; MCP 365; PREM 48013, 48014,

48905

Zea mays: McLean & Berjak (1987); Van der Westhuizen & Bredell

(1972); PREM 47545

javanicum (see E. javanicum var. javanicum)

jensenii Zaleski

Allium cepa: PREM 44761

mushroom casing : Smit (1984)

soil: Eicker (1969, 1970); PREM 44256

Zea mays: Van der Westhuizen & Bredell (1972), PREM 43740, 43753

= intricatum Thom

soil: Cohen (1950)

flannel: Doidge (1950)

johannioli (see P. aurantiogriseum)

kapuscinskii (see P. canescens)

krugeri Ramirez

soil: Ramirez (1990)

lanosocoeruleum (see P. aurantiogriseum)

lanosum (see P. puberulum)

lapidosum (see E. lapidosum)

lilacinum Thom (Paecilomyces lilacinus, Pitt 1979)

soil: Martin (1960); Papendorf (1976)

swine meal: Van Warmelo (1967)

Zea mays: Van der Westhuizen & Bredell (1972)

lividum Westling

debris: PPRI 3707; PREM 49886

soil: PPRI 4043

luteum (see T. luteus)

luteoviride Biourge (indeterminate, Pitt 1979)

aerospora: Doidge (1950); Thom (1930)

martensii (see P. aurantiogriseum)

megasporum Orput & Fennell

Encephalartos laevifolius: PREM 49069

melinii Thom

debris: PREM 47699

Encephalartos laevifolius: PPRI 3178; PREM 49070

mouse nest material: PPRI 4223

mushroom casing: Smit (1984)

soil: Allsopp et al. (1987); Pitt (1979); PPRI 4042; PREM 47699

= atrovenetum G. Smith

Arachis hypogaea: CBS 240.65

Zea mays: Van der Westhuizen & Bredell (1972); CSIR 323, 324

meridianum (See E. meridianum)

miczynskii Zaleski

debris: PPRI 3710; PREM 49882

soil: Allsopp et al. (1987); PPRI 4040

Zea mays: MRC 426

undetermined host: Pitt (1979)

= pedemontanum Mosca & Fontana

soil: Papendorf (1976); MCP 127

Zea mays: PREM 44301

= soppii Zaleski

Eucalyptus maculata: Eicker (1973)

soil: Eicker (1969, 1970, 1973); PREM 44281

undetermined host: CSIR 1398

minioluteum Dierckx

Dianthus caryophyllus: PPRI 3982

Hordeum vulgare: MRC 1756

paper: PPRI 3659; PREM 49874

soil: PREM 48586, 48587

Zea mays: McLean & Berjak (1987); PPRI 3984, 4020; PREM 47533,

47538, 47539, 47544, 47618

montanense Christensen & Bakus

soil: PPRI 4041

multicolor (see P. sclerotiorum)

nigricans (see P. janczewskii)

Bothalia 22,1 (1992)

notatum (see P. chrysogenum)

novae-zeelandiae Van Beyma

Protea spp. : PPRI 3978

soil: Allsopp et al. (1987); PPRI 4222

ochrochloron Biourge

soil: Eicker (1969)

ochrosalmoneum (see E. ochrosalmoneum)

olivinoviride (see P. viridicatum)

olsonii Bain. & Sartory

debris: PPRI 4038

Eucalyptus maculata : Eicker (1973)

Gloxinia spp. PPRI 3706

mushroom casing: Smit (1984)

soil: Eicker (1969, 1970, 1973)

Tribulus terrestris : PPRI 3308; PREM 49207

Zea mays'. PREM 47861, 47863

oxalicum Currie & Thom

Aloe asperifolia: MCP 351

Arachis hypogaea: PREM 48260, 48567

cereal and legume products: Scott (1965)

debris: PPRI 4039

dried fish: MRC 322

face cream: PPRI 3272

fodder: PREM 48584, 48585

soil: CSIR 331, 332, 333, 335, 338

Sorghum caffrorum : CSIR 296, 522, 523

Zea mays'. Doidge (1950); McLean & Berjak (1987); Nagel et al. (1976);

Steyn (1970); Van der Westhuizen & Bredell (1972); CSIR 210,

293, 368, 504, 555, 589, 615, 620, 643, 650, 676; PREM 47542

palitans (see P. viridicatum)

paraherquei (see P. simpiicissimum)

paxilli Bain.

Encephalartos laevifolius: PPRI 3183, 3184

mouse nest material: PPRI 4220

pedemontanum (see P. miczynskii)

petchii Sartory & Bain, (indeterminate, Pitt 1979)

Ananas comosus : Doidge (1950)

piceum Raper & Fennell

cereal and legume products: Scott (1965)

debris: PPRI 4019; PREM 49864

soil: CSIR 345

pinetorum (see E. pinetorum)

pinophilum Hedgcock

Ananas comosus: Doidge (1950)

Arachis hypogaea: Baard (1988); PPRI 3661; PREM 48033, 48384,

48385

compost: PPRI 3166; PREM 49030

Ehretia rigida: PPRI 4310

Sorghum caffrorum: MRC 1587

Zea mays: PREM 47638

undetermined host: Thom (1930)

= purpurogenum var. rubisclerotium Thom

cereal and legume products: Scott (1965)

Zea mays: CSIR 72, 90, 100, 207, 233, Til , 329, 366, 524

piscarium (see P. simpiicissimum)

primulinum Pitt

Eucalyptus cloeziana: PPRI 3730; PREM 49866

puberulum Bain.

Allium cepa: PREM 44765

Arachis hypogaea: Pitt (1979); MRC 335

Encephalartos laevifolius: PPRI 3205

soil: PPRI 3204

Zea mays: Van der Westhuizen & Bredell (1972)

= commune Thom

aerospora: Roth (1968)

Cenchrus ciliaris: Bezuidenhout (1977)

flannel: PREM 33289

molasses meal: Roth (1968)

Zea mays: Van der Westhuizen & Bredell (1972)

= lanosum Westling

brattice cloth: Doidge (1950)

cheese: Luck & Wehner (1979)

soil: Papendorf (1976); MCP 163

Zea mays: Gilman (1972)

pulvillorum (see P. simpiicissimum)

purpurescens (Sopp) Biourge

Fingerhuthia africana: De Villiers (1989); PREM 49278

mushroom casing: Smit (1984)

Protea spp.: PPRI 4284

stored foods: IMI 141 658

Vitis vinifera: PPRI 3574

purpurogenum Stoll

aerospora: Roth (1968)

Agave sisalana: PREM 48893

Allium cepa: PPRI 4224; PREM 44773

Arachis hypogaea: Gilman (1972); CSIR 13

Manihot esculenta: Pitt (1979); MRC 181

molasses meal: Roth (1968)

natural gum: Roth (1968)

nuts and dried fruit: Wehner & Rabie (1970)

Phaseolus spp.: MRC 182

soil: Allsopp et al. (1987); CSIR 350, 351

Sorghum caffrorum: MRC 2501

sugar: Doidge (1950); Van der Bijl (1920)

Vitis spp.: Le Roux et al. (1973)

Zea mays: Gilman (1972); Van der Westhuizen & Bredell (1972); MRC

315; PPRI 3120, 3783; PREM 49018

= rubrum Stoll

Arachis hypogaea: CSIR 13

cereal and legume products: Scott (1965)

cheese: Luck & Wehner (1979)

Medicago sativa: PREM 44370

Zea mays: Gilman (1972); Van der Westhuizen & Bredell (1972); PREM

43747

purpurogenum var. rubisclerotium (see P. pinophilum)

pusillum (see E. cinnamopurpureum)

putterillii Thom ( Geosmithia putterillii, Pitt 1979)

aerospora: Doidge (1950); Thom (1930)

raciborskii Zaleski

aerospora: PPRI 3712; PREM 49885

Arachis hypogaea: PPRI 3664

soil: Stolk & Samson (1983); PPRI 4217

Watsonia marginata: PPRI 3722

= raistrickii G. Smith

Arachis hypogaea: Pitt (1979); MRC 197

cereal and legume products: Scott (1965)

mushroom casing: Smit (1984)

soil: Allsopp et al. (1987); CSIR 388

Sorghum caffrorum: CSIR 526, 528, 529, 545

Zea mays: McLean & Berjak (1987); Van der Westhuizen & Bredell

(1972); CSIR 4; PREM 47636

raistrickii (see P. raciborskii)

restrictum Gilman & Abbott

Acacia karroo: Pitt (1979)

Helianthus annuus: PREM 47856

soil: Allsopp et al. (1987); Papendorf (1976); MCP 23

= striatisporum Stolk

Acacia karroo: Stolk (1969); IMI 151 749

soil: Stolk (1969); Papendorf (1976); MCP 116, 213

roquefortii Thom

cheese: Davel & Neethling (1930); Doidge (1950); Liick et al. (1978);

CSIR 390, 392, 423, 447, 450, 455, 493, 497, 498, 499, 502, 503,

507, 509, 510, 512; PREM 49041, 49050; PPRI 3167, 3190, 3889;

UCT

Medicago spp.: Lamprecht (1988); PPRI 3125; PREM 48315

Vitis spp.: Le Roux et al. (1973)

Zea mays: Van der Westhuizen & Bredell (1972); Van Warmelo (1967)

= casei Staub

soil: Eicker (1969, 1970, 1973); PREM 44280

roseocitreum (see P. griseoroseum)

roseopurpureum Dierckx

soil: Papendorf (1976)

Bothalia 22,1 (1992)

89

Zea mays : CSIR 400

rubrum (see P. purpurogenum)

rugulosum Thom

aerospora: Roth (1968)

Allium cepa: PREM 44775

Arachis hypogaea : PREM 48388

coconut matting: Doidge (1950)

Gladiolus spp. : PPRI 3596; PREM 49413

natural gum: Roth (1968)

Zea mays'. Van der Westhuizen & Bredell (1972); PREM 43739

= tardum Thom

aerospora: Doidge (1950); Thom (1930)

Arachis hypogaea'. Gilman (1972)

soil: CSIR 344; MCP 372

timber: Doidge (1950)

Zea mays'. Gilman (1972); Van der Westhuizen & Bredell (1972)

sclerotiorum Van Beyma

fodder: PREM 48876

Kniphofia spp.: PREM 48877

soil: Pitt (1979); Raper & Thom (1949); PREM 48571; PPRI 3901, 4069,

4139

Zea mays: Stolk & Samson (1983); MRC 425

= multicolor Grigorieva-Manoilova & Poradielova (application un-

certain, Pitt 1979)

cereal and legume products: Scott (1965)

Eucalyptus maculata: Eicker (1973)

soil: Eicker (1969, 1973, 1975); Papendorf (1976)

Zea mays: Van der Westhuizen & Bredell (1972); CSIR 208, 397; MCP,

PREM 43748

senticosum (see E. senticosum)

simplicissimum (Oudem.) Thom

Arachis hypogaea: PREM 48032, 48564

cereal and legume products: Scott (1965)

dung: PPRI 3214; PREM 49084

flannel bag: Raper & Thom (1949); Pitt (1979); IMI 039 816

Medicago spp.: Lamprecht et al. (1988); PREM 48313. 48314

soil: Eicker (1969); Papendorf (1976); CSIR 339; MCP 178, 179; PPRI

4067; PREM 48902. 48903, 48904

Sorghum caffrorum: MRC 2206

ventilation tubing: Doidge (1950)

Zea mays: Van der Westhuizen & Bredell (1972); CSIR 77

= paraherquei Abe

soil: Papendorf (1976); MCP 39, 105, 188

= piscarium Westling

soil: Papendorf (1976); MCP 187

= pulvillorum Turfitt

cereal and legume products: Scott (1965)

Eucalyptus maculata: Eicker (1973)

soil: Eicker (1969, 1973); PREM 44287

Zea mays: Nagel & Steyn (1972); CSIR 1405, 1406

solitum (see P. aurantiogriseum)

soppii (see P. miczynskii)

spiculisporum (see T. trachyspermus)

spinulosum Thom

Allium cepa: Naude & Jooste (1989); PREM 48012, 48561, 48871, 48872,

48873

Pinus spp.: PPRI 3505

soil: Cohen (1950); Papendorf (1976); MCP 165; UCT

Zea mays: Van der Westhuizen & Bredell (1972); CSIR 200; PREM

43749

undetermined host: Thom (1930)

= terlikowskii Zaleski

Zea mays: CSIR 411, 412, 475, 684, 685

= trzebinskii Zaleski

soil: Cohen (1950)

steckii (see P. citrinum)

stolkiae (see E. stolkiae)

stoloniferum (see P. brevicompactum)

striatisporum (see P. restrictum)

sublateritium Biourge

soil: Cohen (1950)

tardum (see P. rugulosum)

terlikowskii (see P. spinulosum)

terrenum (see E. terrenum)

terrestre Raper & Thom (application uncertain, Pitt 1979)

Zea mays: Van der Westhuizen & Bredell (1972)

thomii Maire

aerospora: Roth (1968)

cereal and legume products: Scott (1965)

Cussonia paniculata: PPRI 3784

molasses meal: Roth (1968)

soil: Eicker (1975); Papendorf (1976); CSIR 752; MCP 38; PPRI 3237,

4044; UCT

Zea mays: Van der Westhuizen & Bredell (1972)

trzebinskii (see P. spinulosum)

urticae (see P. griseofulvum)

variabile Sopp

Allium cepa: PREM 44757, 44764

Arachis hypogaea: Gilman (1972)

cereal and legume products: Scott (1965)

coconut matting: Raper & Thom (1949); IMI 040 040

Hordeum vulgare: MRC 1755

leaves: MRC 319

Medicago sativa: PREM 44547, 44548

paper: PPRI 3657; PREM 49876

soil: CSIR 206, 353, 356, 467

Vitis spp.: Le Roux et al. (1973)

Zea mays: Gilman (1972); McLean & Berjak (1987); Van der Westhuizen

& Bredell (1972); Van Warmelo (1967); CSIR 70, 73, 232 , 269,

296, 395, 464, 472, 548; PREM 43743, 43744, 43745, 43746,

47540, 47541, 47543

varians (see P. funiculosum)

velutinum Van Beyma

Eucalyptus maculata: Eicker (1973)

Medicago sativa: PREM 44522, 44549

soil: Eicker (1969, 1970, 1973); PREM 44260

Zea mays: Van der Westhuizen & Bredell (1972) PREM 44306, 47514

vermiculatum (see T. flavus)

verrucosum Dierckx

cheddar cheese: Pitt (1979)

Citrus sinensis: Doidge (1950, et al. 1953); Doidge & Van der Plank

(1936)

debris: PPRI 3121

Phaseolus spp: MRC 220

soil: PPRI 3575

verruculosum Peyronel

Casuarina spp.: PREM 47707

Eucalyptus maculata: Eicker (1973)

Oryza sativa: MRC 171

soil: Allsopp et al. (1987); Eicker (1969, 1970, 1973); CSIR 347; PPRI

3501; PREM 44265, 48017; UCT

Zea mays: PPRI 3837

viridicatum Westling

aerospora: Thom (1930); Doidge (1950)

Arachis hypogaea: MRC 292

birdseed: PREM 4221

cereal and legume products: Scott (1965)

cheese: MRC 1132

Hordeum vulgare: MRC 1761, 2669, 2830

material: Doidge (1950)

natural gum: Roth (1968)

soil: CSIR 15, 405, 407; PREM 48906

Zea mays: Hutchison et al. (1973); Pitt (1979); Van der Westhuizen &

Bredell (1972); CSIR 255, 349, 354 , 396, 413, 425, 430. 460, 570,

663, 724; MRC 422

= olivinoviride Biourge

Allium cepa: PREM 44769

= palitans Westling

Allium cepa: PREM 44758

flannel: Doidge (1950)

Zea mays: Van der Westhuizen & Bredell (1972)

vulpinum (Cooke & Massee) Siefert & Samson

dung: PPRI 3727; PREM 49880

= claviforme Bain.

90

Bothalia 22,1 (1992)

molasses meal: Roth (1968)

soil: CSIR 1088, 1089

waksmanii Zaleski

Barleria obtusa: PPRI 3704; PRJEM 49884

Encephalartos spp.: PPRI 4283

Medicago sativa : PREM 44474

mushroom casing: Smit (1984)

soil: Eicker (1975); Papendorf (1976); MCP 40

Zea mays : Van der Westhuizen & Bredell (1972); MRC 203

wortmannii (see T. wortmannii)

Penicillium species undetermined

aerospora: Ordman (1963, 1970); Ordman & Etter (1956); Radmore

(1986); Van der Merwe et al. (1979)

Arachis hypogaea : Ferreira & Lutchman (1989); Marasas & Van

Rensburg (1986); Van Warmelo (1967)

Cenchrus ciliaris : Bezuidenhout (1977)

cheese: Luck et al. (1976); Luck & Wehner (1979)

Citrus sinensis: Doidge (1950); Doidge & Van der Plank (1936); Roth

(1967); Verwoerd (1929)

Crucifera spp.: Holtzhausen & Knox-Davies (1974)

Cucumis melo: Doidge et al. (1953)

Eucalyptus spp. : Lundquist & Baxter (1985)

fodder: Dutton & Westlake (1985)

foodstuff: Martin & Keen (1978)

Hordeum vulgare : Liibben & Rabie (1989)

indigenous seed: Isaacs & Benic (1990)

Iris spp. : Doidge et al. (1953)

Litchi chinensis: Doidge et al. (1953); Roth (1963)

Lupinus spp.: Van Warmelo (1967)

Mangifera indica: Wehner et al. (1981)

Malus sylvestris: Doidge et al. (1953); Verwoerd (1929)

Medicago sativa: Lamprecht (1988); Marasas & Bredell (1973); Van

Warmelo (1967); PREM 44530, 44551, 44554

molasses meal: Roth (1968)

Musa spp.: Roth & Loest (1965)

mushroom casing: Smit (1984)

Narcissus spp. : Doidge et al. (1953)

nuts and dried fruit: Wehner & Rabie (1970)

Panicum coloratum: Eicker (1976)

Pinus spp.: Lundquist (1987)

Prunus persica: Doidge et al. (1953)

Prunus salicina: Doidge et al. (1953)

recalcitrant seed: Berjak et al. (1989); Mycock & Berjak (1990)

Raphanus sativus var. longipinnatus: Holtzhausen (1978)

Saccharum officinarum: Doidge (1950)

smoked shrimps: Gilman (1972)

soil: Allsopp et al. (1987); Cohen (1950); Eicker (1975, et al. 1982);

Papendorf (1976); CSIR 312

Sorghum cajfrorum: Rabie & Liibben (1984); CSIR 316, 521, 533

Tulipa spp.: Doidge et al. (1953)

Vitis spp.: Le Roux et al. (1973); Verwoerd (1929)

Zea mays: Gilman (1972); Marasas et al. (1981); Marasas & Van

Rensburg (1986); McLean & Berjak (1987); Van Warmelo (1967);

Wittaker et al. (1989); CSIR 218, 264, 414, 415

GENUS EUPENICILLIUM

alucateum (see E. terrenum)

anatolicum (see E. euglaucum)

baarnense (Van Beyma) Stolk & Scott

Acacia mollesjuna: CBS 339.61

soil: Scott (1968b), CSIR 1059, 1070, 1071, 1090, 1106, 1107, 1130; PPRI

3259

brefeldianum (see E. javanicum var. javanicum)

catenatum Scott

soil: Scott (1968a); Stolk & Samson (1983); CBS 325.67; CSIR 1097;

PREM 48556

cinnamopurpureum Scott & Stolk

Pinus spp.: CBS 492 £6; CSIR 946

soil: Scott (1968b); Stolk & Samson (1983); CBS 490.66, 491.66; CSIR

942, 943, 945, 946, 1126; PREM 48558

Zea mays: Stolk & Samson (1983)

undetermined host: Stolk & Samson (1983)

=£ P. pusillum G. Smith

Zea mays: CSIR 606

crustaceum Ludwig

soil: Scott (1968b), CBS 214.71, 215.71, 216,71; CSIR 1026, 1027, 1057,

1102, 1105, 1124; PREM 48551

=£ P. asperum (Shear) Raper & Thom

Eucalyptus maculata: Eicker (1973)

soil: Eicker (1969, 1970, 1973); PREM 44264

=£ P. gladioli McCulloch & Thom

Gladiolus spp.: Doidge (1950, et al. 1953); PREM 30706

ehrlichii (see E. javanicum var. javanicum)

erubescens (see E. terrenum)

euglaucum (Van Beyma) Stolk & Samson

soil: Stolk & Samson (1983); CBS 467.67

Zea mays: Stolk & Samson (1983); CBS 238.65

= anatolicum Stolk

soil: Scott (1968b); Stolk & Samson (1983) CSIR 1095, 1113

= hirayamae Scott & Stolk

soil: Allsopp et al. (1987); Scott (1968b); CSIR 1112; PPRI 3264; PREM

49212

Zea mays: CBS 238.65; CSIR 445

=£ P. hirayamae Scott & Stolk

Zea mays: CSIR 487, 554; IMI 136 205

hirayamae (see E. euglaucum)

inusitatum Scott

soil: Scott (1968a); Stolk & Samson (1983); CBS 351.67; CSIR 1096;

PREM 48570

=£ P. inusitatum Scott

soil: IMI 136 214

javanicum (Van Beyma) Stolk & Scott var. javanicum

apple juice: Stolk & Samson (1983)

soil: Stolk & Samson (1983); CBS 211.71

undetermined host: Stolk & Samson (1983)

= brefeldianum (B. Dodge) Stolk & Scott

apple juice: CBS 291.62

soil: Scott (1968b); Stolk & Samson (1984), CBS, CSIR 1002, 1010,

1011, 1012, 1013; 1028, 1029, 1030, 1068, 1069, 1108, 1109; PPRI

3260; PREM 48555

undetermined host: Stolk & Samson (1983)

=£ P. brefeldianum B. Dodge

apple juice: Van der Spuy et al. (1975)

= ehrlichii (Klebahn) Stolk & Scott

soil: Scott (1968b), CSIR 1025, 1026, 1027; MCP, ; PPRI 3262, 3695;

PREM 49195, 49362

= javanicum (Van Beyma) Stolk & Scott

Arachis hypogaea: CSIR 416, 417, 419, 420, 421, 424; PREM 48259

soil: Scott (1968b), CSIR 1004, 1005, 1006, 1007, 1008, 1009, 1015, 1018,

1019, 1025, 1026, 1027, 1110; 48382, 48550;

=£ P javanicum Van Beyma

soil: Eicker (1969, 1970, 1973); Martin (1960); Papendorf (1976); MCP

123

Zea mays: Van der Westhuizen & Bredell (1972)

lapidosum Scott & Stolk

soil: Scott (1968b); CBS 318.66, CSIR 1035; PREM 48880

Zea mays: CSIR 1093

unrecorded host: Stolk & Samson (1983)

=£ P. lapidosum Raper & Fennell

soil: IMI 113 748; PREM 48880; UCT

meridianum Scott

soil: Scott (1968a, b); Stolk & Samson (1983); CBS 314.67, 217.71, 219.71;

CSIR 1052, 1037, 1036, 1103; PREM 48884

=£ P meridianum Scott

soil: IMI 136 209

ochrosalmoneum Scott & Stolk

soil: Scott (1968b), Stolk & Samson (1983); CBS 515.67; CSIR 1094;

PREM 48886

Zea mays: Stolk & Samson (1983); CBS 489.66; CSIR 145

=£ P. ochrosalmoneum Udagawa

Zea mays: IMI 116 248

parvum (Raper & Fennell) Stolk & Scott

soil: Scott (1968b), CSIR 973, 1054, 1058; MCP, PPRI 3263; PREM

48557, 48881, 48887, 49194

pinetorum Stolk

soil: Allsopp etal. (1987); Scott (1968b); CBS 328.71; CSIR 1092, 1125;

PPRI 3490; PREM 48883; UCT

Bothalia 22,1 (1992)

91

=£ P. pinetorum Stolk

soil: CSIR 1092

senticosum Scott

soil: Scott (1968a, b); Stolk & Samson (1983); CBS 313.67, 329.71; CSIR

1042, 1104; IMI 216 905; PREM 48882

A P. senticosum Scott

soil: IMI 216 905

shearii Stolk & Scott

Medicago spp. : Lamprecht (1988); PPRI 4017; PREM 48322

soil: Scott (1968b); CSIR 1003, 1016, 1017; PREM 48549

Zea mays: CSIR 722

stolkiae Scott

soil: Scott (1968a, b); Stolk & Samson (1983); CBS 315.67, 330.71, 331.71;

CSIR 1003, 1041, 1074; PREM 48552

^ P stolkiae Scott

soil: IMI 136 210

terrenum Scott

soil: Scott (1968a, b), Stolk & Samson (1983); CBS 313.67, 212.71, 213.71,

220.71, 327.71; CSIR 972, 1020, 1021, 1022, 1023, 1024; PPRI

3266

= alutaceum Scott

soil: Scott (1968a); Stolk & Samson (1983); CBS 317.67; CSIR 1039,

1056, 1091, 1100, 1101; PPRI 3488; PREM 48885

= erubescens Scott

soil; Scott (1968a, b); Stolk & Samson (1983); CBS 318.67, 319.67; CSIR

944, 974, 1040, 1032, 1034, 1038, 1040, 1061; PPRI 3261; PREM

48554, 49199

qfc P. alutaceum Scott

soil: IMI 136 243

=£ P. erubescens Scott

soil: IMI 136 404

=£ P. terrenum Scott

soil: IMI 136 208

Eupenicillium species undetermined

mushroom casing: Smit (1984)

soil: CSIR 1127, 1128, 1129; PPRI; UCT

GENUS TALAROMYCES

avellaneus (Thom & Turesson) C.R. Benjamin (anamorph: Merimbla

ingelheimense, Pitt 1979)

soil: CSIR 958, 959

bacillosporus (Swift) C.R Benjamin (anamorph: Geosmithia swiftii, Pitt

1979)

soil: CSIR 961

flavus (Klocker) Stolk & Samson

apple juice: Pitt (1979)

contaminant: PPRI 3790; PREM 48577

Encephalartos laevifolius: PPRI 3213; PREM 49074

wine bottle cork: MCP 27

= flavus var. macrosporus Stolk & Samson

fruit: Stolk & Samson (1972); CBS 317.63; IMI 197 487

soil: Stolk & Samson (1972); CBS 226.72; PPRI 3791

= vermiculatus (Dang.) C.R. Benjamin

soil: CSIR 960, 962, 963

A P. dangeardii Pitt

apple juice: IMI 197 478

A P. vermiculatum Dang,

apple juice: Van der Spuy et al. (1975)

soil: Eicker (1969, 1973)

luteus (Zukal) C.R. Benjamin

P. luteum Zukal

aerospora: Roth (1968)

Citrus sinensis : Doidge (1950)

flannel: Doidge (1950); PREM 33288

molasses meal: Roth (1968)

natural gum: Roth (1968)

spiculisporus (see T. trachyspermus)

thermophilus Stolk

Celtis africana litter: Pitt (1979); Stolk & Samson (1972); CBS

116.72

P. dupontii Griffin & Maubl.

apple juice; Van der Spuy et al. (1975)

compost: IMI 197 483

mushroom compost: Eicker (1977)

trachyspermus (Shear) Stolk & Samson

Manihot esculenta: Pitt (1979); MRC 724

contaminant: PPRI 3885

= spiculisporus (Lehman) C.R. Benjamin

soil: CSIR 955, 956

=£ P. spiculisporum Lehman

undetermined host: MCP 1125

vermiculatus (see T. flavus)

wortmannii (Klocker) C.R. Benjamin

Arachis hypogaea : MRC 332

Oryza sativa : Pitt (1979)

soil: Allsopp et al. (1987); MCP 1134; Stolk & Samson (1972); CBS

293.53; CSIR 954, 957, 964, 965

Watsonia marginata : PPRI 3675

=£ P. wortmannii Klocker

Oryza sativa: MRC 243

Bothalia 22,1: 93-109 (1992)

Vegetation and checklist of Inaccessible Island, central South Atlantic

Ocean, with notes on Nightingale Island

J.P. ROUX* RG. RYAN**, S.J. MILTON*** and C.L. MOLONEY**

Keywords: Inaccessible Island, Nightingale Island, south Atlantic ocean, vegetation

ABSTRACT

The physiography and climate of Inaccessible and Nightingale Islands are briefly discussed. The vegetation and the major

plant associations are described. Notes are given on the ecology and distribution of each taxon. Taxa newly recorded for

Inaccessible Island include Agrostis goughensis, A.holgateana, A. wacei, Calamagrostis deschampsiiformis, Carex thouarsii

var. recurvata, Conyza albida, Elaphoglossum campylolepium and Uncinia meridensis. One species, C. albida, is alien to

the Tristan group. Two native ferns Asplenium platybasis var. subnudum and Blechnum australe were found on Nightingale

Island for the first time, and the presence of introduced Malus domestica orchards was recorded. Two unidentified taxa

were found that may represent new species: Elaphoglossum sp. at Inaccessible Island and Apium sp. at both Inaccessible

and Nightingale Islands.

The total number of vascular plant species recorded at Inaccessible and Nightingale Islands now stands at 98 and 43,

respectively, of which 26 (28%) and seven (16%) are introduced species. Only Atriple x plebeja and two species of Cotula

occur at Nightingale Island but are absent from Inaccessible Island.

UITTREKSEL

Die fisiografie, klimaat en plantgemeenskappe van Inaccessible- en Nightingale-eilande, word kortliks beskryf. Inligting

oor die ekologie en verspreiding van elke takson word ook verskaf. Agt taksons, Agrostis goughensis, A. holgateana, A.

wacei, Calamagrostis deschampsiiformis, Carex thouarsii var. recurvata, Conyza albida, Elaphoglossum campylolepium

en Uncinia meridensis word almal die eerste keer op Inaccessible-eiland aangeteken. C. albida is ’n indringer in die Tristan-

groep. Nuwe verspreidings vir Nightingale-eiland sluit onder meer die twee varingspesies Asplenium platybasis var. subnudum

en Blechnum australe en gevestigde appelboorde, Malus domestica , in. Twee ongeidentifiseerde taksons wat moontlik nuwe

spesies mag wees, is versamel: Elaphoglossum sp. op Inaccessible-eiland en Apium sp. op beide Inaccessible- en Nightingale-

eilande.

Die totale aantal vaatplantspesies wat op Inaccessible- en Nightingale-eilande aangeteken is, staan nou onderskeidelik op

98 en 43, waarvan 26 (28%) en sewe (16%) indringers is. Atriplex plebeja en twee Cofw/a-spesies is die enigste plante wat

op Nightingale-eiland voorkom maar van Inaccessible-eiland afwesig is.

INTRODUCTION

Inaccessible and Nightingale Islands are uninhabited

islands in the Tristan da Cunha group, central South

Atlantic Ocean. Situated at 37°S, they are among the

temperate oceanic islands least disturbed by human

activities (Wace & Holdgate 1976). Several botanical

collections have been made on the islands despite the

hazardous landing conditions, but most collections were

scant, resulting from short visits only (Groves 1981). The

most recent floristic account of the floras at Inaccessible

and Nightingale Islands is that of Wace & Dickson (1965).

We provide a more complete and up-to-date account of

the vegetation of the two islands as a result of a summer-

long stay on Inaccessible Island between October 1989 and

March 1990. Nightingale Island was visited on one

day in October, two days in November and one day in

December.

GEOLOGY, PHYSIOGRAPHY AND CLIMATE

Inaccessible and Nightingale Islands are of volcanic

origin, associated with the Mid-Atlantic Ridge. Bathy-

* National Botanical Institute, Compton Herbarium, Private Bag X7,

Claremont 7735.

** Percy FitzPatrick Institute for African Ornithology, University of Cape

Town, Rondebosch 7700.

*** Botany Department, University of Cape Town, Rondebosch 7700.

MS. received: 1991-03-22.

metric surveys of the coastal waters suggest that the islands

are remnants of once much larger islands (Baker et al.

1964). Lavas from Inaccessible Island have been dated at

approximately three million years old, whereas Nigh-

tingale Island originated approximately 18 million years

ago (Gass 1967; McDougall & Ollier 1982). The geology

of the islands has been described by Baker et al. (1964).

Inaccessible Island consists of thin basaltic lava flows

interbedded with ash and cinders, with intrusive trachyte

dykes, plugs and domes. Nightingale Island, being older,

is more extensively eroded, and consists almost entirely

of trachytes.

Inaccessible Island has a planar area of approximately

12 km2 (Preece et al. 1986, contra Siddall 1985). It is

surrounded by sheer cliffs which are 500 m high in the

west and 200 m in the east (Figure 1). The plateau is

undulating, with three main drainage basins and a few

small hills, the highest being Swale’s Fell, 511 m (Figure

2). Nightingale Island has a planar area of approximately

3 km2 and is lower-lying than Inaccessible Island, the

highest peak being approximately 370 m (Wace & Hold-

gate 1976). There are no permanent streams on Nigh-

tingale, but there are three bogs (The Ponds) in the central

part of the island (Figure 3).

The climate of Inaccessible and Nightingale Islands is

cool temperate oceanic (Wace & Holdgate 1976). There

are few meteorological observations from the islands, but

94

Bothalia 22,1 (1992)

FIGURE 1. — A view of Inaccessible Island from the north. The island is characterised by its steep cliffs.

INACCESSIBLE ISLAND

N

FIGURE 2. — Map of Inaccessible Island showing the major collecting sites. (Adapted from Fraser et al. 1983.).

Bothalia 22,1 (1992)

95

FIGURE 3. — Map of Nightingale

Island showing the major col-

lecting sites. (Adapted from

Groves 1981).

the climate near sea level is probably similar to that record-

ed at Tristan da Cunha, 40 km away (see Wace & Holdgate

1976). Mean daily minimum and maximum temperatures

measured during October 1989 and December to March

1989—90 at Blenden Hall, + 10 m above sea level at

Inaccessible Island, were 14.0°C and 21.6°C (range

7.5— 28°C). Temperatures on the plateau are consider-

ably cooler than those at the coast. Relative humidity at

Blenden Hall varied between 52% and 100% (mean

82.5%). The prevailing winds are from the west, and there

is often a marked altitudinal variation in wind strength;

winds are stronger on the plateau than along the coast.

Rain is typically associated with the passage of frontal

systems and occurs throughout the year, with a winter

maximum (Wace & Holdgate 1976). The mean rainfall at

Blenden Hall, Inaccessible Island was 151 mm per month

during summer, with at least some rain on 73% of days.

Rainfall on the plateau was approximately one-third higher

than that recorded near sea level, and precipitation on the

plateau is higher still due to the frequent formation of

orographic clouds (typically above 350m). Orographic

cloud covered the top of Inaccessible Island on 50.4% of

days during summer 1989-90. Nightingale Island, 22 km

distant from Inaccessible Island, presumably has a similar

climate to that at Inaccessible. However, being a lower is-

land, Nightingale has less marked altitudinal variation in

climate, and has orographic cloud cover less frequently

than does Inaccessible Island.

VEGETATION

The origin of the floras of isolated, geologically young

islands by means of long distance dispersal has been

discussed extensively (e.g. Tryon 1966; Carlquist 1980;

Huntley 1967; Wace 1960). The main dispersal vectors of

propagules to the Tristan islands are transport by wind,

birds and ocean currents (Wace & Dickson 1965). The

origin of the Tristan flora is primarily South American

or southern circumpolar (Wace & Dickson 1965; Tryon

1966). Preece et al. (1986) discuss the Quaternary paleo-

botany of Inaccessible Island.

The vegetation types of the various islands in the Tristan

group are broadly similar, and we have adopted the

terminology used by Wace & Holdgate (1958) and Wace

96

Bothalia 22,1 (1992)

& Dickson (1965). We recognize four physiognomically

distinct types of vegetation or formations (Figures 4 & 5),

which can be divided into several communities charac-

terised by different dominant species. The communities

are used as a unit of description and can be grouped in

their formations as follows: 1, tussock grassland: Spartina

arundinacea tussock; Blechnum penna-marina heath. 2,

fern bush: Blechnum palmiforme heath; Phylica arborea

bush. 3, wet heath. 4, bogs: Sphagnum bog; Scirpus sulca-

tus bog.

MIDDLE ISLAND

N

Tussock grassland

Tlissock grassland covers most of Nightingale Island and

occurs along the steep cliffs from sea level to ± 500 m

on Inaccessible Island. The formation consists of large

Spartina arundinacea tussocks which are up to 3.5 m tall.

On drier, more exposed ridges S. arundinacea tussock is

replaced by Blechnum penna-marina heath on the west

coast of Inaccessible Island.

Spartina arundinacea tussock

On Inaccessible Island this community forms extensive

stands on the small areas of flat ground at sea level and

on all the steep sea-facing cliffs up to + 500 m (Figure

6). It penetrates the plateau only in the low-lying river val-

leys above the Waterfall and Cave Rock. In areas of tall,

dense growth, the closely spaced tussocks exclude all other

vascular plants.

On drier, better drained ridges and in marshy areas the

tussocks are more widely spaced, allowing multi-species

communities to form. The species more commonly found

on ridges and slopes include Blechnum australe, B. penna-

marina, Elaphoglossum laurifolium, E. succisifolium ,

Rumohra adiantiformis , Empetrum rubrum and Nertera

depressa. In marshy areas Amauropelta bergiana var.

tristanensis , Carex insularis, Mariscus congestus, Holcus

INACCESSIBLE ISLAND

N

FIGURE 4. — Map of Inaccessible Island showing the distribution of the

major vegetation types.

FIGURE 5. —Map of Nightingale Island showing the distribution of the

major vegetation types.

lanatus, Hydrocotyle capitata, Rumex frutescens and Scir-

pus sulcatus var. sulcatus are commonly found. Calystegia

sepium subsp. americana and C. tuguriorum climb over

Spartina tussock in some areas.

Small streams and seepages run down the steep slopes

onto the boulder beaches with Azolla filiculoides, Plan-

tago major and Rumex obtusifolius subsp. obtusifolius

commonly occurring on the streambanks. Apium australe

and Sonchus oleraceus are common on the eroded talus

slopes above the beach between Dirleton Point and

Warren’s Cliff. Rockhopper penguins ( Eudyptes chryso-

come) erode paths in their colonies under the tussock,

resulting in some tussocks standing on a root pillar up to

300 mm high.

Soil slips are common on the steep coastal scarps and

alien species are among the first plants to become

established. Introduced species recorded on slips on Inac-

cessible Island include Conyza albida, Holcus lanatus,

Pseudognaphalium luteo-album, Plantago major, Sonchus

oleraceus and Veronica serpyllifolia, whereas Apium

australe, Pelargonium grossularioides, Scirpus bicolor var.

virens and various mosses are native species that colonize

slips.

On Nightingale Island Spartina arundinacea tussock

extends over most of the island. The tussock bases are

more widely spaced, with the open areas between tussocks

extensively burrowed into by great shearwaters ( Puffinus

gravis) for nests. Species more commonly found among

the tussock include Hypolepis rugosula var. villoso-viscida,

Histiopteris incisa var. carmichaeliana and Scirpus bicolor

var. bicolor. Asplenium obtusatum var. crassum and A.

platybasis var. subnudum were found less commonly.

Bothalia 22,1 (1992)

97

FIGURE 6. — Tussock grassland:

Spartina arundinacea on the

low-lying plain at West Point,

Inaccessible Island.

Disturbed open areas, especially along the path between

the huts and the Ponds, have largely been taken over by

introduced species such as Holcus lanatus, Poa annua,

Rumex obtusifolius and Sonchus oleraceus. Native species

found here are Apium sp. , Cotula australis, C. moseleyi

and Scirpus bicolor var. bicolor.

Blechnum penna-marina heath

This community is largely confined to well-drained

ridges and steep slopes within tussock grassland on the

west-facing coastal scarp of Inaccessible Island. Blechnum

penna-marina generally is dominant, but species such as

Blechnum australe, Elaphoglossum succisifolium, E. lauri-

folium, Lycopodium diaphanum, Acaena sarmentosa,

Empetrum rubrum, Holcus lanatus, Nertera depressa,

Uncinia brevicaulis, Vulpia bromoides and various mosses

also occur. S. arundinacea and Ctenitis aquilina are com-

monly associated with boulder strewn streambeds, gullies

and cliff bases. This association does not occur on Nigh-

tingale Island, where Blechnum penna-marina is scarce.

Fern bush

This formation covers most of the plateau on Inaccess-

ible Island, but is restricted to the region around the Ponds

at Nightingale Island. Two major associations can be iden-

tified.

Blechnum palmiforme heath

This association covers much of the high, western half

of the plateau on Inaccessible Island (Figures 4 & 7). It

extends from ± 250 m to the highest part of the island

at Swale’s Fell. Typical Blechnum palmiforme heath is

absent from Nightingale Island, although small stands of

B. palmiforme occur at First and Second Ponds.

FIGURE 7. — Fern bush: Blechnum

palmiforme heath on the

plateau, Inaccessible Island. B.

palmiforme and Phylica arbo-

rea are dominant in this com-

munity. The stunted growth of

the plants may be ascribed to

the prevailing westerly wind.

98

Bothalia 22,1 (1992)

Blechnum palmiforme is the dominant species and the

procumbent caudices form a confused tangle with only

the apical part turning upwards, bearing a crown of

coriaceous fronds 200—400 mm above the ground. Several

species of these ferns (e.g. Elaphoglossum hybridum, E.

laurifolium, E. succisifolium, Grammitis magellanicum

subsp. magellanicum, Hymenophyllum aeruginosum, H.

peltatum, Lagenophora nudicaulis, Nertera assurgens and

N. depressa) are epiphytic on the caudices. Eriosorus

cheilanthoides , Huperzia insularis, Lycopodium diapha-

num, Apium austral e, Calamagrostis deschampsiiformis,

Carex thouarsii, Scirpus bicolor var. bicolor and Uncinia

meridensis are frequent non-epiphytes in this association.

Watercourses support a number of other taxa including

Callitriche christensenii, Glyceria insularis, Scirpus sul-

catus var. sulcatus, and the introduced species Holcus

lanatus and Rumex obtusifolius . Stunted, procumbent

Phylica arborea scrub occurs on some slopes and ridges.

These plants never flower and usually adopt the height of

the surrounding vegetation as a result of the exposed, wind-

swept habitat.

Phylica arborea bush

This association is largely confined to the more sheltered

eastern part of Inaccessible Island at elevations of ± 150 —

250 m. Moving from B. palmiforme heath, the initially

procumbent Phylica arborea scrub becomes progressively

taller, until it eventually forms a closed canopy up to 5

m high in well sheltered localitites (Figure 8). P arborea

FIGURE 8. — Fern bush: Phylica arborea bush on the plateau, Inacces-

sible Island. In sheltered localities the trees may attain a hight

of up to 5 m.

branches support dense growths of epiphytic lichens and

some ferns such as Hymenophyllum aeruginosum. The

undergrowth consists largely of dense stands of pterido-

phytes with Blechnum palmiforme, Ctenitis aquilina and

Histiopteris incisa var. carmichae liana the dominant

species, although Elaphoglossum laurifolium and Asple-

nium obtusatum var. crassum form monospecific stands

in places. In some areas with dense canopy cover, there

is little understorey vegetation, and the ground is exten-

sively burrowed by breeding great shearwaters.

The Serengeti, a flat, relatively dry region in the centre

of Inaccessible Island, consists of open P arborea wood-

land. B. palmiforme , which attains a height of up to 2 m,

forms a subcanopy between the trees. The ground storey

is sparse, with Eriosorus cheilanthoides, Carex thouarsii

var. thouarsii, Empetrum rubrum, Nertera depressa, Scirpus

bicolor var. bicolor and various mosses the most abundant

species. Several taxa are very scarce or absent in this area

including all grasses, Acaena sarmentosa, Ctenitis aqui-

lina, Lagenophora nudicaulis and Gnaphalium thouarsii.

Phylica arborea trees also occur in tussock grassland

on the coastal slopes of Inaccessible Island, and are absent

only between Dirleton Point and South Hill. Trees occur

singly, in small groups, and occasionally in larger groups

with closed canopies (such as Wilkins’ Copse, east of Skua

Bog at the West Point of Inaccessible Island). These trees

differ markedly from those on the island plateau. There

are very few epiphytic lichens on the branches (presumably

due to the infrequency of precipitation from mist and

clouds at lower altitudes) and the understorey consists

either mainly of Spartina arundinacea or has a composi-

tion similar to Blechnum penna-marina heath.

On Nightingale Island, closed-canopy Phylica arborea

bush is confined to the vicinity of The Ponds. In damp

areas the undergrowth consists largely of Blechnum

palmiforme and Scirpus bicolor var. bicolor , but the drier

slopes support Ctenitis aquilina, Hypolepis rugosula var.

villoso-viscida, Histiopteris incisa var. carmichaeliana,

Carex thouarsii var. thouarsii and Acaena sarmentosa.

Scattered P. arborea also occurs in tussock grassland,

particularly along drainage lines on the eastern peak. In

open areas around the Ponds, Scirpus bicolor var. bicolor

forms dominant stands or hummocky meadows. Each

tussock has a pachycaul habit.

Wet heath

Wet heath is restricted to the highest part of Inaccessible

Island, in a narrow strip along the western edge of the

plateau. It is absent from Nightingale Island. Occurring

on the highest part of Inaccessible Island, it is exposed

to very strong winds, and the vegetation is dense and low-

growing (typically 200—300 mm). The formation is

geographically limited, and could be considered to be a

transitional form between tussock grassland and B.

palmiforme heath. However, it has several characteristic

species, and the large number of breeding birds concen-

trated in this vegetation type result in a very different

proportional composition of the vegetation.

The main community extends along the western rim of

the plateau between Swale’s Fell and the upper part of

Ringeye Valley. This is a diverse community, and the

Bothalia 22,1 (1992)

99

following species are common: Amauropelta bergiana var.

tristanensis, Ctenitis aquilina, Elaphoglossum succisi-

folium, Hypolepis rugosula var. villoso-viscida, Acaena

sarmentosa, Apium australe, Carex insularis, C. thouarsii

var. thouarsii, Holcus lanatus, Hydrocotyle capitata,

Nertera assurgens, Scirpus bicolor var. bicolor, S. sulcatus

var. sulcatus, Spartina arundinacea, Uncinia brevicaulis

and U. meridensis . This is the only area where Glyceria

insularis is found away from watercourses, and Cardamine

glacialis, Deschampsia mejlandii and Ranunculus

mauricatus occur nowhere else on Inaccessible Island.

Disturbed areas in the immediate vicinity of albatross nests

and petrel burrows are colonised by invasive species such

as the introduced Cerastium fontanum, Holcus lanatus,

Poa annua and Rumex obtusifolius, as well as native

species such as Gnaphalium thouarsii.

Southeast of Swale’s Fell and in the northern part of

Ringeye Valley, typical wet heath is absent, and tussock

grassland merges almost directly into Blechnum

palmiforme heath. However, there is a peculiar mixture

of the two formations at Dune Hills, with sparse Spartina

arundinacea growing in amongst B. palmiforme heath.

Farther northwest, towards Molly Bog, S. arundinacea is

replaced by abundant tussocks of Calamagrostis

deschampsiifonnis .

Exposed rocks and cliffs provide a microhabitat utilised

by several species. Species characteristic of damp, shaded

crevices include Asplenium erectum, Elaphoglossum

obtusatum, Grammitis magellanica subsp. magellanica,

Agrostis sp., Lagenophora nudicaulis, Nertera depressa

and Uncinia compacta, whereas sunny, north and west-

facing cliffs support species such as Asplenium obtusatum

var. crassum and Chenopodium ambrosioides var. tomen-

tosum.

Bogs

We consider bogs as vegetated areas with impeded

drainage. Two bogs with differing vegetation types are

recognised.

Sphagnum bog

There are several relatively small bogs dominated by

the moss Sphagnum sp. on the eastern part of the plateau

on Inaccessible Island. Most are located at the head of

small streams. Dick’s Bog in Ringeye Valley has been

described in detail by Preece et al. (1986). Associated

species are Histiopteris incisa var. carmichae liana, Carex

insularis, C. thouarsii, Scirpus sulcatus var. sulcatus and

Spartina arundinacea. This type of bog is not found on

Nightingale Island.

Scirpus sulcatus bog

Skua Bog, at West Point, Inaccessible Island, is an

extensive marshy area at approximately sea level, domi-

nated by Scirpus sulcatus var. sulcatus with a small area

of open water along the coastal edge (Preece et al. 1986)

(Figure 9). Azolla filiculoides, Carex insularis, Holcus

lanatus and Rumex frutescens are the only other species

growing in the bog, which is surrounded by tussock grass-

land. A smaller patch of bog occurs behind the beach to

the west of Blenden Hall, which is invaded by the alien

species Mariscus congestus and Plantago major.

Small patches of Scirpus sulcatus bog are also found

in watercourses on the plateau at Inaccessible Island (e.g.

Molly Bog). These bogs are often invaded by Holcus

lanatus, and may support some Carex insularis and C.

thouarsii var. thouarsii. Many bogs on the plateau are

associated with breeding white-chinned petrels ( Procel -

laria aequinoctialis conspicillata) . This species only

breeds in wet areas on the plateau of Inaccessible Island,

and their burrows are characterised by entrance moats. The

mud and water around the burrow entrance support

Hypolepis rugosula var. villoso-viscida, Callitriche christen-

senii and Scirpus bicolor var. bicolor. In some areas, large

numbers of white-chinned petrels breed together at the

upslope edge of a bog, and there is a specific pattern of

bog colonization. Immediately in front of the nests is an

area of open water and bare mud, and at some distance

from the focus of bird disturbance this is colonized by

FIGURE 9. — Bogs: Scirpus sulcatus

bog. Skua Bog at West Point,

Inaccessible Island, dominated

by Scirpus sulcatus var. sulca-

tus.

100

Bothalia 22,1 (1992)

FIGURE 10. — First Pond, Nightingale

Island. Vegetation in this bog

consists of an outer zone domi-

nated by Scirpus sulcatus var.

sulcatus followed by a zone

of Blechnum palmiforme. The

central part of the bog is domi-

nated by Phylica arborea. The

nesting bird is a yellow-nosed

albatross, Diomedea chloro-

rhynchos.

Callitriche christensenii. Still farther from the nests, there

is the usual stand of Scirpus sulcatus var. sulcatus.

The Ponds on Nightingale Island also support floating

mats of Scirpus sulcatus var. sulcatus, with little open

water (Figure 10). Carex insularis and C. thouarsii also

occur, and Callitriche christensenii is found in the shal-

low water along the edge of The Ponds.

DISCUSSION AND CONCLUSIONS

Seven species and one variety were recorded new from

Inaccessible Island, whereas two species were found to

be new for Nightingale Island. These figures exclude two

taxa that may prove to be new species ( Apium sp. and

Elaphoglossum sp.). Our observations bring the total

TABLE 1. — Numbers of species of vascular plants on Inaccessible and

Nightingale Islands. Endemic species are those entirely restricted

to the Tristan-Gough group of islands, and does not include

endemic varieties or subspecies

* includes two species which may be extinct ( Raphanus sativus and

Physalis peruviana ), but excludes Centella asiatica, for which there is

no collected material.

number of species recorded from the islands to 98 and

43 for Inaccessible and Nightingale, respectively (Table

1). Only three taxa are found on Nightingale that are absent

from Inaccessible, the two species of Cotula (one endem-

ic and one introduced) and the endemic Atriplex plebeja.

A total of 28 introduced vascular plants occur on the is-

lands, all spermatophytes. The proportion of alien plants

is greater on Inaccessible Island (27.7%) than on Nightin-

gale Island (16.3%).

The current survey increased the known species rich-

ness at the two islands by almost 10% (cf. Groves 1981).

Only one new taxon was an introduced species ( Conyza

albida on Inaccessible Island), suggesting that the rate of

transfer from the adjacent inhabited island of Tristan is

relatively low. Tristan has more than 100 introduced

species (Groves 1981). The low proportion of new

introduced species also indicates that further collecting

of the native flora is warranted, and that the number of

vascular plants probably is larger than that reported here.

Some taxa require systematic review, notably Nertera,

Uncinia, some of the grasses (Agrostis, Calamagrostis and

Deschampsia ), and some of the ferns ( Elaphoglossum and

Asplenium).

ACKNOWLEDGEMENTS

We thank the Administrator, Island Council and people

of Tristan for permission to work on the Tristan Islands,

and for their friendship and support. We are grateful to

Dr H.P. Linder for identifying the grasses and to Mr A.

Nicholas (South African Botanical Liason Officer at Kew)

for verifying several Dean and Ryan collections and for

providing literature not available in South Africa. Our visit

to the Tristan group was supported by the South African

Department of Environment Affairs through the South

African Committee for Antarctic Research, the South

African Nature Foundation, and the Wildlife Society of

Southern Africa.

Bothalia 22,1 (1992)

101

REFERENCES

AELLEN, P. 1968. Flowering plants from Tristan da Cunha. In E.

Christophersen, Results of the Norwegian scientific expedition

to Tristan da Cunha 55: 1—29.

AGARDH, J.G. 1839. Recensio specierum generis Pteridis. Lund.

AITON, W.T. 1789. Hortus kewensis. London.

BAKER, P.E., GASS, I.G., HARRIS, P.G. & LE MAITRE, R.W. 1964.

The volcanological report of the Royal Society Expedition to

Tristan da Cunha, 1962. Philosophical Transactions of the Royal

Society, London, Series A, 256 : 439 — 575.

BEAUVOIS, A.M.F.J. PALLISOT DE. 1805. Prodrome des cinquieme

famille de I’Aeteogamie. Paris.

BENTHAM, G. & HOOKER, J.D. 1862—1883. Genera plantarum.

Lovell Reeve, London.

BLAKE, S.F. 1925. New names for five South American Asteraceae.

Proceedings of the Biological Society of Washington 38: 85.

BOCKELER, J.O. 1875. Diagnosen neuer Cyperaceen. Flora 58: 262.

BORKHAUSEN, M.B. 1803. Theoretisch-praktisches Handbuch der

Forstbotanik und Forsttechnologie. Darmstadt.

BORY DE SAINT-VINCENT, J.B.G.M. 1804. Voyage dans les quatre

principals les des Mers d'Afrique : 3. Buisson, Paris.

BRACKENRIDGE, W.D. 1854. Filices. United States Exploring Expe-

dition 16. Sherman, Philadelphia.

BROWN, R. 1810. Prodromus florae novae hollandiae. Johnson, Paris.

BROWN, R.N.R. 1905. The botany of Gough Island. Journal of the

Linnean Society, Botany. 37: 238 —267.

BRUMMI1T, R.K. 1965. New combinations in North American Calyste-

gia. Annals of the Missouri Botanic Garden 52: 214—216.

CARLQUIST, S. 1980. Hawaii, a natural history, edn 2. Pacific Tropical

Botanical Garden, Honolulu.

CARMICHAEL, D. 1818. Some account of the island of Tristan da Cunha

and of its natural productions. The Transactions of the Linnean

Society of London : 12.

CHING, R.C. 1934. A revision of the compound-leaved Polysticha and

other related species in the continental Asia including lapan and

Formosa. Sinensia 5: 70.

CHRISTENSEN, C.F.A. 1905. Index filicum. Hagerup, Kopenhagen.

CHRISTENSEN, C.F.A. 1937. Pteridophyta. In E. Christophersen,

Scientific results of the Norwegian Antarctic expedition 1927, 1928,

16: 12-15.

CHRISTENSEN, C.F.A. 1940. The pteridophytes of Tristan da Cunha.

In E. Christophersen, Results of the Norwegian scientific expe-

dition to Tristan da Cunha, 1937, 1938 6: 1—25.

CHRISTOPHERSEN, E. 1934. Plants of Gough Island (Diego Alvarez).

Scientific results of the Norwegian Antarctic expedition 13: 1—6.

CHRISTOPHERSEN, E. 1944. New phanerogams from Tristan da

Cunha. Results of the Norwegian scientific expedition to Tristan

da Cunha: 11.

CLARKE, C.B. 1883. On Hemicarex Benth. and its allies. Journal of

the Linnean Society, Botany 20: 395.

CLARKE, C.B. 1897. Distribution of three sedges. Journal of Botany

35 : 72.

DE CANDOLLE, A.P. 1821. Regni vegetabilis systema naturale. Paris.

DESVEUX, A.N. 1827. De la famille des fougeres. Memoires de la

Societe Linneenne de Paris 6: 333.

DUSEN, P. 1900. Die Gefasspflanzen der Magellanslander nebst einem

Beitrage zur Flora der Ostkiiste von Patagonien. In Svenska

Expeditionen till Magellandema. Wissenschaftliche Ergebnisse

der Schwedischen Expedition nach den Magellansldndem

1895-1897 3 (Botany): 98.

FORSTER, I.G.A. 1786. Florulae insularum australium prodromus. Got-

tingen.

FORSTER, J.G. A. 1789. Fasciculus plantarum magellanicarum. Com-

mentarii societatis regiae scientiarum gottingensis. 9: 13—45.

FORSTER, J.R. & FORSTER, J.G. A. 1776. Characteres gene rum plan-

tarum. London.

FRASER, M. et al. 1983. The Denstone expedition to Inaccessible Island.

Denstonian Supplement: 1-60. Denstone College, Uttoxeter.

GAERTNER, J. 1788. De fructibus et seminibus plantarum. Stuttgart.

GASS, I.G. 1967. Geochronology of the Tristan da Cunha group of Islands.

Geological Magazine 104: 160—170.

GRAY, S.F. 1821. A natural arrangement of British plants. London.

GROVES, E.W. 1981. Vascular plant collections from the Tristan da Cunha

group of islands. Bulletin of the British Museum (Natural History),

Botany 8: 333—420.

GUEDES, M. & JOVET, P. 1975. Conyza albida Willd. ex Sprengel,

the correct name for ‘C. altissima' Naudin ex Debeaux (C.

naudinii Bonnet). Taxon 24: 393, 394.

HEMSLEY, W. B. 1884. Report of the scientific results of the voyage

of H.M.S. Challenger. London.

HILLIARD, O. & BURTT, B.L. 1981. Some generic concepts in

Compositae-Gnaphaliinae. Botanical Journal of the Linnean So-

ciety 82: 181-232.

HOLTTUM, R.E. 1974. Thelypteridaceae of Africa and adjacent islands.

Journal of South African Botany 40: 133, 134.

HOOKER, J.D. 1852. Flora novae-zelandiae. London.

HOOPER, S.S. 1968. Cyperaceae from Tristan da Cunha. In E.

Christophersen, Results of the Norwegian scientific expedition

to Tristan da Cunha 1937, 1938 54: 1—9.

HUBBARD, C.E. 1981. Gramineae. In E.W. Groves, Vascular plant col-

lections from the Tristan da Cunha group of islands. Bulletin of

the British Museum ( Natural History), Botany 8: 333—420.

HUNTLE Y, B.J. 1967. A preliminary account of the vegetation of Marion

and Prince Edward Island. South African Journal of Science 63:

235-241.

JALAS, J. 1963. Notes on Cerastium L. subsect. Perennia Fenzl

(Caryophyllaceae). Archivum societatis zoologicae botanicae fen-

nicae 'Vanamo' 18,1: 63.

KUHN, F.A.M. 1868. Filices africanae. Engelmann, Leipzig.

LABILLARDIERE, J.J.H. DE 1807. Novae hollandiae plantarum

specimen. Paris.

LAMARCK, J.B.A.P.M. DE 1783. Encyclopedic methodique. botanique.

Paris.