ISSN 0006 8241 = Both;

Bothalia

A JOURNAL OF BOTANICAL RESEARCH

Vol. 33,2

Oct. 2003

TECHNICAL PUBLICATIONS OF THE NATIONAL BOTANICAL INSTITUTE,

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.

Three booklets of the contents (a) to Vols 1-20, (b) to Vols 21-25 and (c) to Vols 26-30, are available.

STRELITZIA

A series of occasional publications on southern African flora and vegetation, replacing Memoirs of

the Botanical Survey of South Africa and Annals of Kirstenbosch Botanic Gardens.

MEMOIRS OF THE BOTANICAL SURVEY OF SOUTH AFRICA

The memoirs are individual treatises usually of an ecological nature, but sometimes dealing with

taxonomy or economic botany. Published: Nos 1-63 (many out of print). Discontinued after No. 63.

ANNALS OF KIRSTENBOSCH BOTANIC GARDENS

A series devoted to the publication of monographs and major works on southern African flora.

Published: Vols 14-19 (earlier volumes published as Supplementary volumes to the Journal of

South African Botany). Discontinued after Vol. 19.

FLOWERING PLANTS OF AFRICA (FPA)

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 botanical artists have contributed to

the series, such as Fay Anderson, Peter Bally, Auriol Batten. Gillian Condy, Betty Connell, Stella

Gower, Rosemary Holcroft, Kathleen Lansdell, Cythna Letty (over 700 plates), Claire Linder-

Smith and Ellaphie Ward-Hilhorst. The Editor is pleased to receive living plants of general interest

or of economic value for illustration.

From Vol

An index

FLORA i

A taxonoi

and Botsv

to genera

and ecoloj

Contributi

vaziland, Namibia

specific taxa, keys

well as taxonomic

PALAEOi

A palaeofh ach of the informa-

tion is pret populations. Now

available:

Molter

i

Molten

<!

Prodroi

I

F

Towarc

b

J.M. & H.M.

1989, by J.M.

'85, by J.M. &

laremont 7735,

.tinction, 1999,

BOTHALIA

A JOURNAL OF BOTANICAL RESEARCH

Volume 33,2

Scientific Editor: G. Germishuizen

Technical Editor: B.A. Momberg

NATIONAL

Botanical

INSTITUTE

2 Cussonia Avenue, Brummeria, Pretoria

Private Bag X101, Pretoria 0001

ISSN 0006 8241

Oct. 2003

FTo o n ^NTEMP

; 1 2 AW *Wj In celebration

THE CENTENARY OF THE NATIONAL HERBARIUM (PRE)‘ IN 2003 AND

ITS IMMEASURABLE CONTRIBUTIONS TO BOTANY IN SOUTH AFRICA AND FURTHER AFIELD

Editorial Board

D.F. Cutler

B.J. Huntley

P.H. Raven

J.P. Rourke

M.J.A. 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, The Netherlands

Acknowledgements to referees

Archer, Mrs C. National Botanical Institute, Pretoria, RSA.

Archer, Dr R. National Botanical Institute, Pretoria, RSA.

Bouman, Dr F. Hugo de Vries Laboratory, Amsterdam, The Netherlands.

Burgoyne, Ms P. National Botanical Institute, Pretoria, RSA.

Deall, G.B. Environmental Consultant, ECOREX, Box 1848, White River, RSA.

Glen, Dr H.F. National Botanical Institute, Pretoria, RSA.

Herman, P.P.J. National Botanical Institute, Pretoria, RSA.

Hilliard, Dr O.M. Royal Botanic Garden, Edinburgh, Scotland, UK.

Killick, Dr D.J.B. 465 Sappers Contour, Lynnwood, 0081 Pretoria, RSA.

Kreisel, Prof. Dr H. Ernst-Moritz-Arndt-Universitat, Greifswald, Germany.

Leistner, Dr O.A. National Botanical Institute, Pretoria, RSA.

Long, Dr D.G. Royal Botanic Garden, Edinburgh, UK.

Nordal, Prof. 1. University of Oslo, Blindern, Norway.

Oliver, Dr E.G.H. National Botanical Institute, Cape Town, RSA.

Ortiz, Dr S. Universidade de Santiago, Santiago de Compostela, Spain.

Radcliffe-Smith, Dr A. Royal Botanic Gardens, Kew, UK.

Robbrecht, Prof. Dr E. National Botanic Garden, Meise, Belgium.

Smithies, Mrs S. National Botanical Institute, Pretoria, RSA.

Ronse De Craene, Dr L.P Katholieke Universiteit Leuven, Belgium.

Tancin, Ms C. Hunt Institute for Botanical Documentation, Carnegie Mellon University, Pittsburgh,USA.

Van Wyk, Prof. A.E. University of Pretoria, RSA.

Weberling, Prof. Dr F. Universitat Ulm, Germany.

Wigginton, M.J. 36 Big Green, Warmington, Peterborough, UK.

CONTENTS

Volume 33,2

1. Jamesbrittenia bergae (Scrophulariaceae), a distinctive new species from Limpopo, South Africa. P.

LEMMER 141

2. A new Cyrtanthus species ( Amaryllidaceae: Cyrtantheae) endemic to the Albany Centre, Eastern Cape,

South Africa. D.A. SNIJMAN 145

3. Understanding Erica xwillmorei , a nineteenth century English garden hybrid. E. CHARLES NELSON

and E.G.H. OLIVER 149

4. Notes on African plants:

Asteraceae. A new combination in the genus Macledium (Mutiseae). N.C. NETNOU and B.-E.

VAN WYK 158

Cleveaceae-Marchantiales. Sauteria nyikaensis , a new liverwort species from Malawi. S.M. PEROLD 165

Euphorbiaceae. Excoecaria madagascariensis', a first record for the Flora of southern Africa region.

J.E. BURROWS, S.M. BURROWS, P. LOFFLER and L. LOFFLER 155

Fabaceae. The correct name for Acacia montana. P.P. SWARTZ 164

Lycoperdaceae-Gasteromycetes. Calvatia sect. Macrocalvatia redefined and a new combination in

the genus Calvatia. J.C. COETZEE and A.E. VAN WYK 156

Lycoperdaceae-Gasteromycetes. Author citation and publication date of the name Calvatia cranii-

fonnis. J.C. COETZEE and A.E. VAN WYK 160

Mesembryanthemaceae. Brianhuntleya, a new genus endemic to the Worcester-Robertson Karoo,

South Africa. P. CHESSELET, S. HAMMER and I. OLIVER 160

5. Inflorescences of Cliffortia L. (Rosaceae) and related vegetative branching patterns. A.C. FELLINGHAM

and H.P. LINDER 173

6. More accurate publication dates for H.C. Andrews’ The Heathery , particularly volumes 5 and 6. R.J.

CLEEVELY. E.C. NELSON and E.G.H. OLIVER 195

7. Embryology and systematic relationships of Kiggelaria (Flacourtiaceae). E.M.A. STEYN, A.E. VAN

WYK and G.F. SMITH 199

8. Vegetation of the rock habitats of the Sekhukhuneland Centre of Plant Endemism, South Africa. S.J.

SI EBERT, A.E. VAN WYK, G.J. BREDENKAMP and F. SIEBERT 207

9. National Botanical Institute South Africa: administration and research staff 31 March 2003, publications

1 April 2002-31 March 2003. Compiler: B.A. Momberg 229

10. Guide for authors to Bothalia 245

New genus, species, name and combinations in Bothalia 33,2 (2003)

Acacia theronii P.P. Swartz, nom. nov., 164

Brianhuntleya Chesselet, S. A. Hammer & I. Oliver, gen. nov., 161

Brianhuntleya intrusa (Kensit) Chesselet, S. A. Hammer & I. Oliver, comb, nov., 161

Calvatia capensis (Lloyd) J.C.Coetzee, Eicker & A.E.van Wyk, comb, nov., 156

Cyrtanthus macmasteri Snijman, sp. nov., 145

Jamesbrittenia bergae P.Lemmer, sp. nov., 141

Macledium zeyheri (Sond.) S. Ortiz subsp. thyrsiflorum (Klatt) N.C.Netnou, comb, nov., 158

Sauteria nyikaensis Perold, sp. nov., 167

iv

Bothalia 33,2: 141-143 (2003)

Jamesbrittenia bergae (Scrophulariaceae), a distinctive new species from

Limpopo, South Africa

P. LEMMER*

Keywords: Jamesbrittenia Kuntze. Limpopo [Northern Province], Manuleae. new species, Scrophulariaceae, Thabazimbi

ABSTRACT

Jamesbrittenia bergae P.Lemmer is a new species from the Farm Brakvallei near Thabazimbi in Limpopo [Northern

Province], Large, medium-textured, bright scarlet flowers with yellow throats are borne singly in upper leaf axils; the flowering

branches, although terminal, do not form elongated racemes; leaves are greyish green, coarsely serrated. It grows in full sun

on ferricrete outcrops.

INTRODUCTION

This distinctive new species was discovered by Mr

A.S. Berga, nurseryman and keen indigenous plant

grower, on the Farm Brakvallei near Thabazimbi in

February 2002 (Figure 1). Photographs and specimens

were taken for identification purposes. Literature studies

and specimens examined at the C.E. Moss and Bolus

herbaria pointed us to the genus Jamesbrittenia. The use

of the Jamesbrittenia key in The Manuleae. A tribe of

Scrophulariaceae (Hilliard 1994), however, could not

aid identification.

A pressed flowering specimen of one of the plants

grown from seed collected in the field, was sent for com-

ment to Dr O.M. Hilliard. Royal Botanic Garden.

Edinburgh, who is the authority on the taxonomy of the

genus. She replied that the plant was unknown to her, but

that it might be a hybrid, provided that possible parents

occurred in the area. A thorough search at the original

locality, however, revealed only Jamesbrittenia

burkeana (Benth.) Hilliard. In view of the lack of puta-

tive parents and the fact that seeds of the plants germi-

nated prolifically with the progeny exactly resembling

the original plants, I maintain that this is. indeed, a new species.

Jamesbrittenia bergae P.Lemmer , sp. nov., species

distincta statim floribus pro ratione magnis, splendide coc-

cineis recognita; fortasse J. accrescenti (Hiem) Hilliard

affinis, sed ab ilia colore dissimile florum. tubo corollae

12-18 mm longo (non 18-25 mm longo ut in J. accres-

cente), textura loborum corollae media, apicibus

plerumque retusis margineque non colore dilution (tex-

tura J. accrescentis crassa, marginibus pallidis tenuisque,

apicibus ± rotundatis), ovario glabro (in J. accrescente

minute glanduloso-puberulo), capsulis 4-7 mm x 2-3

mm (non 7-10 mm x 3. 5-4.5 mm ut in J. accrescente ),

et distributione geographico (in ditione Thabazimbi, non

montibus Zoutpansberg dictis vel terra alta provinciae

Mpumalanga dictae), distinguitur.

TYPE. — Limpopo, 2427 (Thabazimbi): Farm Brakvallei,

southern foothills of Berg van Winde, on ferricrete out-

* P.O. Box 912878, 0127 Silverton, South Africa.

MS. received: 2003-04-01.

crops, southern aspect. 1 060 m, (-CB), S.E. Strauss 808

(PRE, holo.!; E!, PRU!).

Multistemmed shrublet up to 200 mm high, with

perennial rootstock; stems ± erect or spreading; most

parts strongly and pleasantly herb-scented when crushed;

plants very floriferous. Leaves alternate, occasionally

opposite towards base of stems, spreading, greyish

green; lamina elliptic to obovate, (5— )8— 1 5(— 25) x (3-)

6— 1 0(— 15) mm. tapering towards base, apex ± obtuse,

margin coarsely serrate to irregularly lobed, both sur-

faces with glandular-hispid hairs, ± 0.3 mm long, and

with evenly distributed, glistening glands, veins

impressed above, raised below; petiole V4— '/3 total leaf

length. Flowers solitary in upper leaf axils, terminal

flowers being the youngest. Pedicels usually 10-20 mm

long. Calyx divided almost to base; tube 0. 5-1.0 mm

long; lobes 5. 5-8.0 x 0.8-1 .5 mm, oblong, apex subacute

to rounded, densely covered with glandular hairs ± 0.3

mm long and with scattered glistening glands. Corolla:

tube 12-18 x 0.8-1. 2 mm. measured midway between

base and limb, but ± cylindrical, abruptly expanding and

geniculate near apex, with multicellular glandular hairs,

0. 7-1.0 mm long, and scattered glistening cells (Figures

2A; 3); well-bearded on outside of expanded part; mouth

rounded, limb nearly regular, 16.5-25.0 mm across later-

al lobes; lobes spreading, broadly ovate, 8.2-1 1 .5 x 8-1 1

mm, base cuneate, apex distinctly retuse, rarely ± round-

ed, outside with few glandular hairs and scattered glis-

tening glands, especially towards base, inside glabrous;

lobes medium-textured, brilliant scarlet above, dull

whitish below, without a thin-textured, pale margin;

throat 3-4 mm diam., bright yellow, with transverse, V-

shaped band of hairs on inside not extending onto base of

anterior lip. Stamens : posticous filaments ± 0.7 mm long,

posticous anthers ± 0.8 mm long; anticous filaments ±

1.2 mm long, anticous anthers ± 1.2 mm long; all fila-

ments distally slightly bearded with clavate hairs;

anthers included. Ovary 2-3 x 0.8-1. 5 mm, glabrous;

style 10-15 mm long; stigma ± 1 mm long, appearing

simple, though minutely bifid, included; nectary not dis-

tinct, apparently comprising sterile basal '/4 portion of

ovary, no nectar secretion observed. Capsules 4-7 x 2-3

mm, slightly shorter, as long as. or just protruding from

persistent calyx lobes, glabrous (Figure 2B). Seeds ±

elliptic to unevenly shaped (Figure 2C). ± 0.5 mm long, grey-

142

Bothalia 33,2 (2003)

FIGURE 1. — Distribution of Jamesbrittenia bergae P.Lemmer in

Limpopo [Northern Province],

brown; testa with polygonal cells clearly visible, arranged

in ± longitudinal rows, outer periclinal walls collapsed

inwards forming a shallow pit, radial walls smooth

(Group la sensu Hilliard 1994: 85).

The glandular pubescence and the medium-textured

corolla lobes are reminiscent of Jamesbrittenia Candida

Hilliard, the habit and general vegetative morphology of

J. accrescens (Hiem) Hilliard and the flower shape with

their arrangement towards tips of stems of J. grandiflora

(Galpin) Hilliard and J. macrantha (Codd) Hilliard. In J.

bergae, however, the flowering branches, although ter-

minal, do not form racemes with leaves degenerating

into bracts. The flower colour is a brilliant scarlet, not

white or mauve as in J. grandiflora and J. macrantha,

nor varying from green through yellow, orange-brown,

maroon to dark brown as in J. accrescens. In the most

recent comprehensive key to the species of Jamesbrit-

tenia (Hilliard 1994), J. bergae keys out as J. accrescens.

The new species is easily distinguished from the latter by

its different floral colour, medium-textured corolla limb

(the limb is thick-textured in the case of J. accrescens ),

shorter corolla tube (that of J. accrescens being 18-25

mm long and thicker), corolla lobes with the apices usu-

ally retuse (not ± rounded as in J. accrescens ) and the margin

FIGURE 2. — A, 1/s corolla tube showing position of stamens, x 3; B,

ripe fruit capsule, x 5; C, seed, x 20.

without a thin-textured, pale, marginal zone (that of J.

accrescens having a thin, cream to greenish yellow mar-

gin). Furthermore, the ovary and capsule of J. bergae are

glabrous, whereas those of J. accrescens are minutely

glandular-puberulous. The capsules of J. bergae are

smaller (4-7 x 2-3 mm) than those of J. accrescens,

which are 7-10 x 3. 5^1. 5 mm. J. bergae, in addition, has

a different habitat and geographical distribution than J.

accrescens, which is known to grow in rocky places in

scrub, often on margins of patches of forest, from the

Zoutpansberg eastwards to the eastern highlands of

Mpumalanga.

Distribution and ecology : although often locally abun-

dant, J. bergae appears to be a rare species, known only

from the type locality at present. The plants grow in full

sun in crevices on ferricrete outcrops with a southern

aspect, in the southern foothills of the 'Berg van Winde’

near Thabazimbi, Limpopo [Northern Province] (Figure

1). The altitude varies from 1 056 m to 1 106 m. The area

receives an average of 550 mm of summer rain per

annum. The vegetation type is Mixed Bushveld (Low &

Rebelo 1996).

Herbs and grasses associated with J. bergae include

Aristida congesta subsp. barbicollis, Elionurus muticus,

Evolvulus alsinoides, Merremia tridentata subsp. angus-

tifolia var. angustifolia and Tephrosia longipes subsp.

longipes var. longipes. Trees and shrubs growing in the

vicinity of the new species are Combretum molle, Di-

plorhynchus condylocarpon, Englerophytum magalis-

montanum, Eaurea saligna , Ozoroa paniculosa var. pan-

iculosa, Spirostachys africana, Strychnos madagascari-

ensis, and Ziziphus nmcronata subsp. mucronata. The

species referred to were recorded in summer after an

average rainy season.

Additional material examined

LIMPOPO. — 2427 (Thabazimbi): Farm Brakvallei, southern foot-

hills of Berg van Winde, (-CB), S.E. Strauss 803, 809, 811, 816, 819

(PRE), S.E. Strauss 812 (E).

ACKNOWLEDGEMENTS

I sincerely thank Dr O.M. Hilliard for her initial guid-

ance and her encouragement. Prof. A.E. van Wyk for his

unstinted co-operation and support, Mr P.W. van

Schalkwyk for permission to take herbarium specimens,

and his hospitality during field work, Mrs S.E. Strauss

and Mr A.S. Berga for collecting herbarium specimens,

Mr D.A. McCallum of the C.E. Moss Herbarium and Mr

T. Trinder-Smith of the Bolus Herbarium for assistance

with the genus Jamesbrittenia, Dr H.F. Glen for translat-

ing the diagnosis into Latin and Ms G. Condy for the

painting and text figure.

REFERENCES

HILLIARD, O.M. 1994. The Manuleae. A tribe of Scrophulariaceae.

Edinburgh University Press, Edinburgh.

LOW, A.B. & REBELO, A.G. 1996. Vegetation of South Africa,

Lesotho and Swaziland. Department of Environmental Affairs

and Tourism, Pretoria.

Bothalia 33,2 (2003)

143

FIGURE 3. — Jamesbrittenia bergae. Artist: G. Condy. Specimen cultivated from seed collected in the wild.

Digitized by the Internet Archive

in 2016

https://archive.org/details/bothaliavolume3333unse_0

Bothalia 33,2: 145-147 (2003)

A new C yrtanthus species (Amaryllidaceae: Cyrtantheae) endemic to

the Albany Centre, Eastern Cape, South Africa

D.A. SNIJMAN*

Keywords: Amaryllidaceae, Eastern Cape, Cyrtanthus L.f., new species, South Africa, transitional vegetation type

ABSTRACT

Cyrtanthus macmasteri Snijman is a rare new species from the Albany Centre of endemism, Eastern Cape, South Africa.

Most closely related to C. galpinii Baker, and autumn-flowering species with a single or rarely-flowered inflorescence from

the northern regions of southern Africa, C. macmasteri is distinguished by a 3 to 6-flowered inflorescence. It grows on steep

banks of the Great Kei River and its tributaries and flowers in summer.

INTRODUCTION

Cyrtanthus L.f. is an exclusively sub-Saharan African

genus in the family Amaryllidaceae, found only in the

southern and eastern parts of the continent and frequent-

ly restricted to habitats with distinctive soil types, eco-

tones or isolated gorges. Dyer (1939) and later Reid &

Dyer (1984) reviewed the 50 species known in southern

Africa, whereas Nordal (1979) studied the two species

and four subspecies from East Africa.

The Eastern Cape is floristically rich, phytogeographi-

cally complex and poorly collected (Gibbs Russell &

Robinson 1981; Gibbs Russell et al. 1984). In 1994 Mr

J.C. McMaster. a keen naturalist who repeatedly visited

remote localities to record Eastern Cape plants, sent

pressed and living plants of this new species collected

from near the confluence of the Bolo and Kei Rivers

(Figure 1), to the Compton Herbarium. Mrs M.A. Holmes

had first recorded matching material ( Holmes NBG70846 )

from the same area in 1963. Although previously identi-

fied by Miss W.F. Barker as C. galpinii Baker, the acqui-

sition of Cameron McMaster's specimens enabled this

rarely seen species to be described and illustrated.

Cyrtanthus macmasteri Snijman , sp. nov., quoad

colorem, magnitudinem et formam florum C. galpinii

Baker similis, sed floribus multis (3-6) differt. Figura 2.

TYPE. — Eastern Cape, 3227 (Stutterheim): Farm

Keibolo, above Bolo River Gorge near its confluence

with the Kei River, (-BC), 6 February 1994, McMaster

s.n. ( NBG192230 , holo.; PRE).

Deciduous, bulbous herb. 200-250 mm tall when

flowering. Bulb solitary, hypogeal, depressed globose,

25-35 x 30-35 mm diam.; outer tunics papery and brown,

extended into a slender neck up to 90 mm long. Leaves 1

or 2 each season, most often dying back before flowering

and emerging anew after flowering, strap-shaped,

150-320 x 6-7 mm. narrowing up to 2 mm at base and

apex, glabrous, somewhat glaucous, reddened at base;

adaxial surface channelled; abaxial surface distinctly

keeled; margins and midrib minutely papillate towards

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

7735 Claremont, Cape Town.

MS. received: 2003-05-12.

base. Inflorescence 3-6-flowered; scape erect, up to 190

mm long, 4—5 mm diam., tapering distally, glaucous

green, flushed with pink near base, solid at base but hol-

low above; spathe valves 2, equitant, lanceolate, 50-57 x

8-9 mm, herbaceous, reddish turning pale brown, inner

withering and reflexing before outer; bracteoles filiform,

± 5 mm long; pedicels erect, ranging from 15-55 x

1. 5-2.0 mm in each inflorescence, pale green. Flowers

secund, spreading horizontally or slightly drooping,

opening ± synchronously; perianth funnel-shaped, 37-42

mm long, scarlet, inner surface of tube whitish pink with

6 scarlet streaks leading downwards from tepal sinuses,

not scented, producing nectar; tube 25-30 mm long,

lower 10 mm narrow, 2-4 mm diam., curved at ± 90°,

upper 15-20 mm flaring open to 10-13 mm at throat;

tepals ovate, 10-14 x 6-8 mm, outspread at anthesis,

outer and inner almost equal, 5-7-veined, firm-textured,

without a thickened midrib; apex obtuse, mucronulate;

mucro white. Stamens biseriate, regular, arising ± three

quarters up tube, outer attached 3—4 mm below inner;

anthers dorsifixed, oblong, up to 3 mm long, yellow.

Ovary ellipsoidal, 5-6 x 2-3 mm, green; ovules axile, ±

25 per locule; style pressed against lower tepals, up to ±

9 mm long, exserted beyond stamens, white to scarlet dis-

tally; stigma 3-branched, branches slender, truncate, ± 2.0

x 0.5 mm, papillate on inner surface. Capsule narrowly

FIGURE 1. — Known distribution of Cyrtanthus macmasteri in the

Great Kei River Basin, Eastern Cape, South Africa.

146

Bothalia 33,2 (2003)

elliptical, ± 30 x 8 mm, 3-valved, valves reflexing when

dry. Seeds flattened, up to 10x4 mm, blackish.

Phenology, the main flowering time is February but

sometimes late January and early March if conditions are

favourable. Seed ripens at the end of March and by mid-

April it has mostly dispersed. The leaves emerge several

months prior to flowering and mostly die back before the

flowers appeal-. Under cultivation at Kirstenbosch Botani-

cal Garden, a new leaf is often produced within a month

after flowering.

Diagnostic features', it is possible to group the species

of Cyrtanthus by their flowers. Several species have a

perianth in which the tube is narrow at the base and then

flares widely to the throat. Reid & Dyer (1984) grouped

together 13 species defined by this floral form. Of these,

C. elatus (Jacq.) Traub, C. guthrieae L.Bolus and C. mon-

tanus R.A.Dyer have regular flowers without markings.

The other 10 species have distinctive lines running into

the throat from the tepal sinuses, the style and often the

stamens flex downwards against the lower tepals, and the

stigma has three long branches. These include five red-

flowered species allied to Cyrtanthus sanguineus (Lindl.)

Walp. and three species with lemon to pale pink flowers

related to C. loddigesianus (Herb.) R.A.Dyer.

Based on its flared tube, paler throat with darker streak-

ing, and deflexed style with a three-branched stigma, C.

macmasteri falls clearly into the group comprising C.

thomcroftii C.H. Wright, C. galpinii, C. eucallus R.A.Dyer,

C. sanguineus , and C. flammosus Snijman & Van Jaarsv.

The inclusion here more recently of C. wellandii Snijman

in which the stamens and style are included low down in

the tube is discussed in Snijman & Batten (2003).

Of the smaller-flowered species belonging to the C.

sanguineus group, C. macmasteri is most similar to C.

galpinii, a species from Zimbabwe, Mozambique, Mpu-

malanga, Swaziland and northern KwaZulu-Natal. Both

have narrow (less than 7 mm wide), hysteranthous leaves

that taper to a narrower base (1-2 mm wide), and flowers

that flare widely to a pale perianth throat (at least 10 mm

across) with contrasting streaks leading down into the

tube. In C. galpinii the flowers are variable, although

mostly red they may also be pink or orange. In southern

Africa the perianth is 50-75 mm long, whereas the spec-

imens from Zimbabwe that are usually called C. galpinii

have shorter perianths, 30-60 mm long (Nordal 1979). In

plants from Zimbabwe C. galpinii may rarely have two

flowers per inflorescence, however, elsewhere the inflo-

rescence is consistently single-flowered and appears from

March to July. In C. macmasteri the length of the perianth

(37-42 mm) overlaps the smaller forms of C. galpinii, but

the species differs in having a 3-6-flowered inflorescence

which is produced mainly in February.

It is possible to confuse poorly prepared specimens of

C. macmasteri with those of C. collinus Ker Gawk, an

unrelated species of the Cape Region that extends along

the Cape Fold Mountains from the Riviersonderend

Range, Western Cape to the Zuurberg, Eastern Cape.

This is due to the perianth tube in C. collinus sometimes

being slightly flared towards the mouth. However, the

flowers are uniformly red and without nectar guides, the

style and trifid stigma arch against the upper tepals and

the leaves are V-shaped in cross section, lacking the dis-

tinct keel found in C. macmasteri.

Distribution and habitat'. Cyrtanthus macmasteri is

known only along the steep east- and south-facing banks of

the Great Kei River and its tributaries, the Bolo and

Nqancule Rivers (Figure 1). It grows singly, is never com-

mon and is confined to altitudes of 640 to 1 100 m, where

Valley Bushveld gives way to Dohne Sourveld (Acocks

1975), mostly amongst sandstone rocks of the Beaufort

Group and less frequently in soils weathered from dolerite

intrusions, with less than 500 mm annual rainfall. Woody

shrubs and small trees scattered throughout the habitat

include Acacia karroo, Carissa bispinosa, Cussonia spicata,

Diospyros whyteana, Encephalartos princeps, and various

species of Rhus. Other species of Cyrtanthus , which are sym-

patric with C. macmasteri, are C. macowanii Baker

(. McMaster NBG182475) and C. obliquus (L.f.) Aiton

( Snijman 1639, NBC) but their peak flowering times do not

overlap. Following the IUCN Red List criteria the species is

categorized here as LR (lc): lower risk and of least concern.

Endemism', many authors (Nordenstam 1969; Cowling

& Hilton-Taylor 1997; Van Wyk & Smith 2001) have

reported high levels of endemism in Eastern Cape, particu-

larly in the Albany Centre. Following Van Wyk & Smith

(2001), C. macmasteri is a noteworthy example of an

Albany Centre endemic that has affinities with the subtropi-

cal eastern regions of sub-Saharan Africa.

ACKNOWLEDGEMENTS

Mr J.C. and Mrs R. McMaster are thanked for their invalu-

able help with collecting data and specimens of Eastern Cape

Cyrtanthus. Mr G. Duncan is gratefully acknowledged for

data from the bulb collection at Kirstenbosch. Mrs A. Batten

generously made available her painting of C. macmasteri for

publication. Tire colour illustration was kindly sponsored by

Starke-Ayres Nursery.

Other specimens examined

EASTERN CAPE. — 3227 (Stutterheim): steep eastern aspect near

highest point of Farm The Deeps, (-BC), 5 March 1998; McMaster s.n.

(PRE), 24 January 2000, McMaster NBG192237: Farm Keibolo above

Bolo River Gorge near its confluence with the Kei River, (-BC), 29

January 1995, McMaster s.n (K); slopes down to Nqancule and Kei

Rivers, (-BD), Holmes NBG70846.

REFERENCES

ACOCKS, J.P.H. 1975. Veld types of South Africa. Memoirs of the

Botanical Survey of South Africa No. 40: 1-128.

COWLING, R.M. & HILTON-TAYLOR, C. 1997. Phytogeography,

flora and endemism. In R.M. Cowling, D.M. Richardson &

S.M. Pierce. Vegetation of southern Africa. Cambridge

University Press, Cambridge.

DYER, R.A. 1939. Areview of the genus Cyrtanthus. Herbertia 6: 65-103.

GIBBS RUSSELL, G.E. & ROBINSON, E.R. 1981. Phytogeography

and speciation in the vegetation of the eastern Cape. Bothalia

13: 467-472.

GIBBS RUSSELL, G.E., RETIEF, E. & SMOOK, L. 1984. Intensity of

plant collecting in southern Africa. Bothalia 15: 131-138.

NORDAL, I. 1979. Revision of the genus Cyrtanthus (Amaryllidaceae)

in East Africa. Norwegian Journal of Botany 26: 183-192.

NORDENSTAM, B. 1969. Phytogeograpiiy of the genus Euryops

(Compositae) a contribution to the phytogeography of southern

Africa. Opera Botanica 23: 7-77.

REID, C. & DYER, R.A. 1984. A review of the southern African

species of Cyrtanthus. American Plant Life Society, La Jolla.

SNIJMAN, D A. & BATTEN, A. 2003. Cyrtanthus wellandii.

Flowering Plants of Africa 58: 50-55, t. 2189.

VAN WYK, A.E. & SMITH, G.F. 2001. Regions offloristic endemism

in southern Africa: a review with emphasis on succulents.

Umdaus Press, Hatfield, Pretoria.

Bothalia 33,2 (2003)

147

FIGURE 2. — Cyrtanthus macmasteri, x 1. A, bulb; B, mature leaves; C, inflorescences; D, 1/s flower; E, capsule; F, seed. Painted from McMaster

NBG192330 collected at Farm Keibolo (NBG). Artist: Auriol Batten.

Bothalia 33.2: 149-154(2003)

Understanding Erica xwillmorei , a nineteenth century English garden

hybrid

E. CHARLES NELSON* and E.G.H. OLIVER**

Keywords: Erica L., garden history, nomenclature

ABSTRACT

The application of the binomial Erica willmorei Knowles & Westc. is discussed and the name is typified by an illustration.

It is demonstrated that the name was altered to E. wilmoreana by Bentham, misapplied and misspelled soon after publication

and that various plants known by this binomial and its numerous variants do not represent the original hybrid.

INTRODUCTION

A brief history of hybrid Cape heaths

In Britain during the late eighteenth and early nine-

teenth centuries there was immense interest, at least

among wealthy garden owners, in forming collections of,

and in cultivating, species of Erica from southern Africa,

the so-called Cape heaths. Collectors, especially Francis

Masson (1741-1805) (Bradlow 1994) and James Niven

(1776-1827) (Nelson & Rourke 1993). were engaged to

go to the Cape of Good Hope and to search for new

species, to gather seeds and to send these to their spon-

sors and patrons in Europe. These collectors were re-

markably successful, and so were the gardeners and nursery-

men who received the wild-collected seeds. Seedlings of

many hitherto unknown species were raised and substan-

tial numbers were grown on to flowering stage. These

plants provided the subjects for a series of splendid illus-

trated books, of which probably the most notable was

Henry Charles Andrews’s Coloured engraving of heaths,

a part-work issued between 1794 and about 1830

(Cleevely & Oliver 2002).

Cape heaths were among the most fashionable plants

of this period. Wealthy enthusiasts indulged their passion

for them to the utmost, even building special glasshous-

es to accommodate and protect the plants, and amassing

collections of hundreds of different species and varieties.

In such circumstances, the nurserymen, on whom the

overwhelming majority of the Cape heath fanciers de-

pended for novelties, had to strive to ensure that new

species were continually available. While wild-collected

seed was the principal means of achieving this, the cost

of obtaining the seeds of more and more new species was

undoubtedly enormous in both financial terms and in

terms of the seed collectors' time. There came a point

* Cultivar Registrar. The Heather Society, Tippitiwitchet Cottage, Hall

Road. Outwell. Wisbech, Cambridgeshire, PEI 4 8PE, United Kingdom.

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

7735 Claremont, Cape Town.

f We have deliberately omitted the authorities for many of the binomi-

als used herein for two reasons: firstly, because it is impossible often to

establish the true identity of the plants concerned, and secondly,

because some of the names were first published in nurserymen's lists

and the correct attribution of the specific or varietal epithets has yet to

be established.

MS. received: 2002-09-10.

when employing a collector and basing him at the Cape

of Good Hope became entirely uneconomical.

By the early 1 800s it is certain that at least two hundred

different Cape heaths were in cultivation in Europe — the

list of heathers (including European species) offered in

1 802 by Lee and Kennedy of Hammersmith included 228

species and varieties. Less than a decade later Smith

(1809) remarked that while it was ‘difficult to guess at the

number of the real species of Erica ... our gardeners reckon

about 300, many of which are merely varieties Even if

the plants did not represent that number of distinct species,

they still constituted a remarkably diverse assembly of

plants, and there can be little doubt that the nurserymen

who grew Cape heaths made special efforts to propagate

the best ones. Many could be increased only by cuttings,

but the fact that the shrubs bloomed did mean that seeds

could potentially be harvested without the necessity of

sending collectors to the Cape.

Many Cape Erica species have exacting pollination

requirements — pollinating agents in the wild range from

nectar-feeding birds to insects and the wind (Rebelo et

al. 1985). Specialized animal pollinators are inevitably

absent when plants are cultivated in regions far-distant

from their natural habitats, and thus, in cultivation, polli-

nation may never be effected and viable seed will not

form. On the other hand, considering that in the early

nineteenth century in Europe, an individual Cape heath

collection could contain more than a hundred species,

and that these would have been contained in a single

glasshouse, the potential even for accidental cross-fertiliza-

tion by ‘alien’ pollinators was enormous. However, the

absence of ‘natural’ pollinators can be overcome by

hand-pollination, and it is evident that European garden-

ers were soon adept at pollinating — and cross-pollinat-

ing— the Cape heaths in their care.

No precise record has been traced recording the first

instance of garden-harvested seed, nor of the first

seedlings from such seed, but there is indirect evidence

that around 1790 at least one English nurserymen had

succeeded in obtaining and germinating home-produced

seed from Cape heaths. Moreover, he had also artificially

cross-pollinated two species and produced a hybrid. The

nurserymen was William Rollisson (his surname is fre-

quently misspelled Rollinson), and he used two Cape

heaths named Erica grandiflora 1 and E. vestita as the

150

Bothalia 33,2 (2003)

FIGURE 1. — The illustration x 1, by

R. Mills, published by

Knowles & Westcott (1838),

here designated the iconotype

of Erica xwillmorei. By cour-

tesy of Ian A. Nex, Archivist

and Librarian, The Birming-

ham Botanic Gardens, and

reproduced by permission of

The Director, The Birmingham

Botanic Gardens.

Bothalia 33,2 (2003)

151

parents of the hybrid which was released under the names

E. vestita var. fiilgida or E. fulgida (Rollisson 1843).

Rollisson kept this a secret, and it is clear from subse-

quent comments by his contemporaries, for example

Henry Andrews, that Rollisson passed off his hybrid

plants as new species raised from seeds gathered in the

wild at the Cape of Good Hope.

That others also carried out artificial pollination and

cross-pollination is highly likely, yet the first explicit

reference to it dates from 1817 when the Dean of Man-

chester. the Rev. William Herbert, who is best known for

his work with Amaryllidaceae, casually remarked that he

had raised some hybrids of Cape heaths: ‘and the new

heathers I have already obtained, are most distinct and

remarkable, the individuals of each new species [v/c]

being perfectly uniform’ (Herbert 1818).

Various comments published by Andrews (1794 to

± 1830) reveal probable hybrids. Of Erica rupestris var.

rubra he wrote: ‘This Erica was raised from British seed

gathered from the E. rupestris , in the Autumn of 1807, by

Mr. Knight, nurseryman in the King’s Road. Chelsea, and

is the only one, out of a number sown, that vegetated.’

Using the rather obtuse phrase "seminal variety’, Andrews

(1794 to ± 1830) signalled some more: E. coventrya was

‘one of those fine seminal varieties raised from Cape seed

at the Hammersmith Nursery’; E. refulgens was ‘consid-

ered as a seminal variety of the E. versicolor ’ ; E. stellifera

‘supposed to have been raised from seed of E. ventricosa ,

or praegnans ... is also another seminal variety'.

The original Erica xwillmorei

Erica willmorei was one of the plants featured in

Floral Cabinet and Magazine of Exotic Botany, pub-

lished in Birmingham, a short-lived periodical edited by

George Beauchamp Knowles (± 1800-1860), who was

Professor of Botany at the Birmingham School of

Medicine, and Frederic Westcott (d. 1861). Their aim was

admirable, to provide ‘accurate, and at the same time,

highly finished representations of such plants as are

remarkable for their beauty, their rarity, or their peculiar-

ity of structure’ (see Desmond 1977). The periodical last-

ed just three years, and among the plant portraits was one

of a new hybrid heather which Knowles & Westcott

(1838) named ‘ Erica willmorei, Mr Willmore’s Erica’:

"This hybrid variety of Erica is in the collection of John

Willmore, Esq. of Oldford, and was raised by that gen-

tleman’s gardener, Mr. John Williams, whose skill and

perseverance have been that means of introducing an

immense number of beautiful hybrids, particularly

Calceolarias, many of which are infinitely more beauti-

ful than any of the original species’.

The coloured plate (Figure 1) was accompanied by a

Latin description and an English translation, so the bino-

mial E. xwillmorei Knowles & Westc. is validly pub-

lished. and while no contemporary herbarium specimens

matching the descriptions and illustration have yet been

traced, the name can be typified by the published plate.

E. Xwillmorei Knowles & Westc., Floral Cabinet

and Magazine of Exotic Botany 2: 115, t. 73 (1838).

Iconotype: Floral Cabinet and Magazine of Exotic Botany

2: t. 73 (1838) selected here.

E. xwilmoreana [sic] Benth. in DC.: 661 (1839).

‘Leaves in threes, linear, channelled, bracteas close to

the calyx; sepals ovate-lanceolate, acute; flowers axil-

lary; corolla tubular-ventricose, semi-pellucid; anthers

included; stigma somewhat exserted. Flowers rather

transparent, in shape somewhat intermediate between

tubular and ventricose, of a bright but rather pale red, the

mouth 4-parted, divisions rounded, of a beautiful green

with a distinct white margin.’ Figures 1; 4A.

Knowles & Westcott (1838) did not provide any

information about the parentage of E. xwillmorei, except

to comment that it was ‘interesting as affording a proof

of the possibility of obtaining hybrid varieties between

Heaths with tubular flowers, and those with globose

flowers; or, to use the language of gardeners, between the

pill and the tube, a fact which we believe has been very

generally denied’.

We suggest that E. xwillmorei was a hybrid derived

perhaps from E. blenna Salisb., a Cape heath that was

certainly in cultivation during the early nineteenth centu-

ry principally under the synonym E. vernix Andrews, and

another undetermined species with a long, tubular corol-

la. John Willmore, after whom this hybrid was named,

lived at Oldford, a suburb of Birmingham where he had

a fine garden with a ‘rich collection’ of exotic plants.

Erica wilmoreana and other orthographic variants

It is necessary at this point to note that Bentham

(1839), citing Knowles & Westcott (1838) as the author-

ity, altered the name to E. wilmoreana, not only mis-

spelling it — a difficulty that recurs far too frequently

when this name is employed — but also, unnecessarily,

amending the termination. Thus E. wilmoreana (and all

variant spelling of it) is a synonym of E. xwillmorei (and

all its variants) and is not available for use for any plant

other than the one illustrated by Knowles & Westcott

(1838). Concerning these orthographic variants, it is

apposite to note Dauthenay’s (1900) remarks on the

name: ‘Nous avons designe l’espece de Bruyere qui fair

le sujet de cet article sous le nom d 'Erica Wilmorei. On

l’appelle aussi E. Wilmoreana, mais quelques personnes,

evidemment mal renseignees, en ont fait Y Erica

Vilmoriniana. Ces deux mots ont evidemment une cer-

taine ressemblance dans leur consonnance. Pour eviter

cette confusion, il suffirait, dans ce cas particulier, de

respecter la loi de priorite’.

The orthographic variants and typographic errors,

recorded (to September 2002) in The Heather Society’s

database of heather names for E. xwillmorei, are as

follows (only the earliest records are noted, arranged

alphabetically):

E. villmoreana : Garten Magazin 51: 203 (1898).

E. villmoriniana: Flora capensis 4,1: 315 (1905).

E. vilmoreana: Journal of the Linnean Society 24: 180

(1887).

152

Bothalia 33,2 (2003)

FIGURE 2. — This speci-

men from Joseph

Dickinson’s her-

barium (LIV) is

labelled 'Erica

willmorei hybrid

1837’. It is not

the same plant as

that illustrated

by Knowles &

Westcott (1838).

It is exactly 104

mm long from

base to tip. By

courtesy Dr L.

Wolstenholme,

and reproduced

by permission of

The Board and

Trustee of the Na-

tional Museums

& Galleries on

Merseyside, Li-

verpool.

E. vilmoriniana: Revue Horticole 1891: 235 (1891).

E. wilmoreana : DC., Prodromus 7: 661 (1839).

E. willmoreana: The Gardeners ’ Chronicle 22 October:

711 (1842).

E. willmoriana : Revue Horticole 1891: 235 (1891).

E. wilmorea : Deutsches Magazin fiir Garten 36: 69

(1883).

E. wilmorei: Catalogue, James Veitch & Sons, 70 (1873,

1874).

E. wilmoreii: Erica stock list, Ericaflora, Monbulk,

Victoria, Australia, not dated [± 2000].

E. wilmoriana : The Gardeners ’ Chronicle 27 August:

569 (1842).

Other Erica “ willmorei' ’ and Erica “ wilmoreana ”

To complicate matters, the first author recently became

aware of the existence in the herbarium of Liverpool

Museum (LIV) of a specimen labelled, in the handwrit-

ing of Joseph Dickinson, ‘Erica Willmorei Hybrid 1 837’

(Eigure 2). This particular specimen was most probably

taken from a plant growing in Liverpool Botanic Garden —

Dickinson was Secretary of the Liveipool Botanic Garden

Trust, and the specimen derives from his herbarium of culti-

vated plants (J. Edmondson pers. comm. 2002). Beyond

doubt, the specimen does not represent the plant illustrated

by Knowles & Westcott (which will be referred to hence-

forth as K&W73 [i.e. Knowles & Westcott, Floral Cabinet

and Magazine of Exotic Botany 2: 115, t. 73 (1838)].

The Dickinson specimen represents an Erica with hir-

sute flower buds (not glabrous as in K&W73 ), ± 15 mm

long, club-shaped with distinctly swollen tips (not elon-

gate-ovoid, swollen towards the base, as in K&W73).

The leaves, ± 5 mm long, are ciliate (not glabrous as in

K&W73). The specimen has not been examined, but a

xerox print, enlarged twice, on which these details can be

very clearly seen, was studied by us.

The first and obvious conclusion is that by 1837

Wilson, John Willmore's gardener, had succeeded in pro-

ducing no fewer than two hybrids, probably of entirely dif-

ferent parentage, and that these had been distributed to

other gardeners either under the name Erica “ willmorei ”

or, more probably, without a name in which case the recip-

ients subsequently labelled these for their own conve-

nience as Erica “ willmorei ” (this is not an uncommon

occurrence, even today, in gardens and nurseries).

While Knowles & Westcott (1838) chose, for whatev-

er reason, to illustrate a seedling with elongate-ovoid

flowers and glabrous foliage, which perhaps soon be-

came extinct in cultivation, it seems that the plant repre-

sented by the specimen in Joseph Dickinson’s herbarium

was a seedling (probably from a group of seedlings) that

was destined to continue to flourish in various gardens.

That it was successfully propagated and distributed can-

not be doubted — there are several different pieces of evi-

dence indicating this.

In 1842 there was sufficient doubt about the exact

identity of at least one plant labelled Erica “willmorei 7

“ willmoreana ” for Regel ( 1842) to list ‘ Erica wilmoreana

hortulanis Dresden’ and comment as follows: ‘E. Wil-

moreana der Dresdener Garten ist eine selbst als Bastard

nur schwierig von E. Linnaeoides zu unterscheidende

Pflanze und einzig durch noch buschigeren Wuchs und in

der Witte etwas angeschwollene Blumenrohre ver-

schieden.’

Dickinson’s specimen exhibits superficial resem-

blance to an illustration (Figure 3) labelled ‘Wilmore’s

heath (Erica wilmoreana)' published by Step (1897). The

flower buds in the illustration are similar in shape to

those of Dickinson’s specimen, and while they are not

depicted as being hirsute, there is again ample evidence

that plants labelled Erica “ wilmoreana ” were rather vari-

able (see Figure 4).

The earliest description that has been traced of Erica

“ wilmoreana ” which is not directly derived from

Knowles & Westcott (1838) was published by Paquet

(1844): ‘Belle plante d'un aspect blanchatre, a feuilles

velues, quaternees; fleurs en tube, bilabiees, roses a la

partie inferieure, blanches au sommet.' This albeit brief

characterization matches later descriptions and also

Step’s illustration (1897) (Figure 3), although the leaves

depicted therein are not hairy (‘a feuilles velues’) as

described by Paquet (1844). However, Nicholson

(1885) noted that the linear leaves of Erica “ wilmore-

ana” were ‘covered in short white hairs, as also are the

branches.’ Carriere (1892), describing Erica “ wilmore-

ana” and three of its cultivars, noted that in Erica

“ wilmoreana glauca": ‘La villosite sur la corolle est

beaucoup plus pronouncee sur cette plante que sur

toutes les autres de ce meme groupe, chez la plupart

desquelles la corolle est glabre, tandis que chez cette

variete glauca la corolle est parfois presque hispide, ce

qui suffirait pour la differencier.' Otherwise, most de-

Bothalia 33,2 (2003)

FIGURE 3. — ‘Wilmore’s heath ( Erica wilmoreana)' from Step (1897:

t. 164). Main shoot, x 0.68. Note the shape of the flowers, with

the swollen tips, and compare with those on the specimen in

Figure 2. Reproduced by courtesy of the Lindley Library, Royal

Florticultural Society, London.

scriptions of this heath are perfunctory, concentrating

on flower colour.

Erica “ wilmoreana ” was listed in English nursery-

men’s catalogues in the 1870s and 1880s — for example,

James Veitch & Sons (1873-1874); William Rollisson &

153

Sons (1877); B.S. Williams (1881) — hut what Cape heath

they were actually growing and selling is impossible to

determine. Illustrations of so-called Erica “ willmorei ”

appeared in The Garden (1889) (see Figure 5), Revue

Horticole (1892, 1900) and, as already noted, in Step’s

Favourite flowers for gardens and greenhouse (189: 3: t.

164) (see Figure 3). In The Gardeners' Chronicle on 15

February 1896, Hudson (1896) described and illustrated

Erica “ wilmoreana The present is a most fitting season

for drawing attention to one of the finest of all the early

spring-flowering Heaths. It is one of the earliest hybrids

raised in this country, being brought into notice about

sixty years ago. Of its parentage I have no record, but as

regards the value of the offspring there can be no ques-

tion, whether for the greenhouse or conservatory. Under

good cultivation it is a long-lived variety, being a most

vigorous grower ... The colour of the flowers is deep

pink, tipped white, and they last in good condition sever-

al weeks.' The accompanying illustration (Figure 5)

showed a floriferous plant, nothing whatever like that

illustrated by Knowles & Westcott (1838) [the right-hand

specimen was shown in the plate published seven years

earlier in The Garden (1889)].

CONCLUSION

This paper touches on an almost intractable subject,

the correct application of the numerous binomials given

to cultivated plants of Cape heaths by European, but

especially English, nurserymen, gardeners and horticul-

tural authors during the first half of the nineteenth centu-

ry when erica-mania was in its most efflorescent phase.

These names are recorded in a database compiled by The

Heather Society, acting as ICRA (International Cultivar

Registration Authority) for Erica, which will form the

source for the second volume of the International regis-

ter of heather names [the first volume (Nelson & Small

2000) listed only the names of hardy cultivars and

European species]. In many, if not most cases, the names

must remain designated as incertae sedis.

The International Code of Botanical Nomenclature

establishes rules for the naming of plants including pri-

mary hybrids of garden origin. Even though its parentage

may not be recorded, the application of any hybrid bino-

mial is precisely determined by reference to protologue

FIGURE 4. — Details from the illustrations: A, published by Knowles & Westcott (1838: see Ligure 1), x 2.2; B, Step (1897: see Figure 3), x 2.7;

and C, of the specimen in LIV (Figure 2), x 2.7. The flowers in A are red with green lobes prominently margined with white; note the

emergent anthers. The flowers of B are red with white lobes; the anthers are included.

154

Bothalia 33,2 (2003)

FIGURE 5. — 'Erica willmoreana x’. One of several illustrations evi-

dently derived from a photograph; first published in The

Garden on 2 February 1889 (only the right-hand shoot was

shown), this version appeared in The Gardeners' Chronicle on

15 February 1896, x 0,56.

(the original published description) and the associated

materials which may include herbarium specimens and

illustrations. In this case no herbarium material has been

traced but an excellent illustration which accompanied

the protologue, and now designated as the iconotype,

serves to establish the exact application of the binomial

Erica xwillmorei. The fact that an herbarium specimen

bearing the name ‘Erica Willmorei' and dated 1837 is

extant is not relevant, because it indubitably is an entire-

ly different plant and is not connected in any way to the

protologue. It is clear that this binomial was misapplied

soon after its publication in 1838, and it was also subject

to misspelling.

It remains to be established what the correct identities,

and thus the correct names, are of cultivated and natural-

ized plants presently called Erica willmorei (or E. will-

moreana, etc.). These include several cultivars, as well as

plants reported as naturalized in several parts of

Australia. Until specimens can be obtained and studied,

the identity of these plants cannot be resolved.

ACKNOWLEDGEMENTS

We are grateful to Dr John Edmondson (Liverpool

Museum) for bringing Dickinson’s specimen to our atten-

tion; to the librarians of The Linnean Society of London,

The Natural History Museum (Botany Library) and the

Lindley Library (Royal Horticultural Society), London, for

access to reference works; and to David Small (President,

The Heather Society). Lor assistance with the illustrations

published in this paper, we are grateful to Dr L.

Wolstenholme (Curator of Botany, Liverpool Museum)

and Ian Nex (Archivist and Librarian, Birmingham Botanic

Garden).

REFERENCES

ANDREWS, H.C. ‘1802-1809’ [recte 1794 to ± 1830], Coloured en-

graving of heaths. London. 4 volumes.

BRADLOW, F.R. (ed.). 1994. Francis Masson's account of three journeys

at the Cape of Good Hope 1772-1775. Cape Town.

CARRIERE, E.-A. 1892. Erica wilmoreana et varieties. Revue Horti-

cole 1892: 201-203.

CLEEVELY, R.J. & OLIVER, E.G.H. 2002. A preliminary note on the

publication dates of H.C. Andrews’ Coloured engravings of

heaths ( 1794—1830). Archives of Natural History 29: 245-264.

DAUTHENAY, H. 1900. L’ Erica wilmorei et ses varieties. Revue Horti-

cole 1900: 484.

DE CANDOLLE, A. P. 1839. Ericaceae. Prodromus 7: 661.

DESMOND, R. 1977. Victorian gardening magazines. Garden History

5,3: 47-66.

HERBERT, W. 1818. Instructions on the treatment of the Amaryllis

longifolia ... with some observations on the production of hybrid

plants. Transactions of the Horticultural Society of London 3:

187-196.

HUDSON, J. 1896. Erica wilmoreana. The Gardeners' Chronicle, ser.

3, 19 (15 February): 200, fig. 27.

KNOWLES, G.B. & WESTCOTT, F. 1838. Erica willmorei. Floral Cabi-

net and Magazine of Exotic Botany 2: 115, 1 16, t. 73.

NELSON, E.C. & ROURKE, J.P. 1993. James Niven (1776-1827), a

Scottish botanical collector at the Cape of Good Hope: his lior-

tus siccus at the National Botanic Gardens, Glasnevin, Dublin

(DBN), and the Royal Botanic Gardens, Kew (K). Kew Bulletin

48: 663-682.

NELSON, E.C. & SMALL, D.J. (eds). 2000. International register of

heather names. Volume 1. Hardy cultivars and European species.

4 parts. Creeting St Mary.

NICHOLSON, G. 1885. Erica wilmoreana. The illustrated dictionary

of gardening vol. 1 (A-E): 526. London.

PAQUET, V. 1 844. Traite de la culture des plantes de terre de bruyere.

Paris.

REBELO, A.G., SIEGFRIED, W.R. & OLIVER, E.G.H. 1985. Polli-

nation syndromes of Erica species in the south-western Cape.

South African Journal of Botany 51: 270-280.

REGEL, E. 1842. Erica wilmoreana, in Die Kultur und Aufzahlung der

... Eriken. Verhandlungen des Vereins zur Beforderung des

Gartenbaues in den Koniglich Preufiischen Staaten 16: 244.

ROLLISSON, G. 1 843. Hybrid plants. The Gardeners' Chronicle 8 July: 461 .

SMITH, J.E. 1809. Erica. In A. Rees, The cyclopeedia; or, universal

dictionary of arts, sciences and literature 13: unpaginated [Erica].

London.

STEP, E. 1897. Wilmore’s heath ( Erica wilmoreana). Favourite flowers

of garden and greenhouse vol. 3: t. 164. London.

Bothalia 33,2: 155-171 (2003)

Notes on African plants

VARIOUS AUTHORS

EUPHORBIACEAE

EXCOECARIA MADAGASCARIENSIS ; A FIRST RECORD FOR THE FLORA OF SOUTHERN AFRICA REGION

While collecting specimens for the compilation of the

Tree atlas of Swaziland , an unknown plant was collected

by R Loffler & L. Loffler in the Lubombo Mts near the

Mozambique border. A literature search led to the speci-

men’s tentative identity as Excoecaria madagascariensis

(Baill.) MUll.Arg (1866). Since the Swaziland locality

was so far distant from its nearest locality in Chirinda

Forest, Zimbabwe, the site was re-visited in November

2002. Despite intensive searching of the area, only a sin-

gle specimen of Excoecaria was found, although not the

same as the original plant, which was not re-located.

However, it's identity was undisputed, the new glossy

red leaves which gives it the Zimbabwean common name

of ‘red-ears’ being conspicuous. The plant was in both

flower and fruit (Figure 1A, B).

Described from Madagascar, E. madagascariensis

also occurs in Somalia (Thulin 1993), the coastal forests

of Kenya and isolated inland forests in Tanzania

(Radcliffe-Smith 1987). A disjunct locality is represent-

ed by its occurrence as a fairly common understorey

species in Chirinda Forest in southeastern Zimbabwe.

The new locality extends its distribution by almost 700

km and represents a further considerable disjunction for

the species, as well as a new record for the Flora of

southern Africa region.

The habitat in which the Excoecaria occurs in the

Lubombo Mountains, is in dry, evergreen forest situated on

the floor of a valley near the Mtibhlati River at an altitude

of 240 m a.s.l. The canopy is dominated by Atalaya alata

(Sapindaceae), Balanites maughamii (Balanitaceae),

Chionanthus foveolatus subsp .foveolatus (Oleaceae), Ficus

polita , F. petersii (Moraceae), Homalium dentatum (Fla-

courtiaceae), Margaritaria discoidea subsp. fagifolia,

Spirostachys africana (Euphorbiaceae), Strychnos usam-

barensis , S. gerrardii (Strychnaceae) and Wrightia natal-

ensis (Apocynaceae). Understorey small trees and shrubs

include Diospyros natalensis subsp. nummularia (Eben-

aceae), Erythroxylum emarginatum (Erythroxylaceae),

Hyperacanthus amoenus (Rubiaceae), Salacia leptoclada

(Celastraceae), Teclea gerrardii (Rutaceae), Tinnea barba-

ta (Lamiaceae) and Uvaria lucida (Annonaceae). A herb

layer is almost absent.

The immediate area in which the E. madagascarien-

sis grows is severely threatened by the uncontrolled

spread of alien invader plants, particularly Melia azeder-

ach and Chromolaena odorata , both of which form pure

stands along the nearby flood-damaged river. Despite

being in relatively undisturbed climax forest, the entire

area around the single Excoecaria plant was dotted with

small Melia seedlings. Further down the river the river-

ine vegetation and adjacent forest is being cleared for

cultivation, a process which may well reach the Excoe-

caria site. In addition, certain trees (notably Wrightia

natalensis) are being felled either for medicinal plant

material or construction purposes. In view of the appar-

ent extreme rarity of this plant, Excoecaria madagas-

cariensis must be regarded as critically threatened in

Swaziland.

Excoecaria madagascariensis {Baill.) Miill.Arg.

in DC., Prodromus systematis naturalis regni vegetabilis

15,2: 1219 (1866); Radcl.-Sm.: 383, t. 72 (1987); Thulin:

306, t. 176 E-G (1993); Radcl.-Sm.: 316 (1996); M.

Coates Palgrave: 518 (2002). Stillingia madagascarien-

sis Baill.: 522 (1858). Spirostachys madagascariensis

Baill.: t. 8/19, 21 (1858). Sapium madagascariensis

(Baill.) Prain: 1010(1913), non Pax (1890); Brenan: 226

(1949). Type: Madagascar, Nosy Be (Nossi Be), Pen’ille'

475 (P, holo., G. K).

Excoecaria sylvestris S. Moore in Rendle et al.: 204 (1911).

Syntypes: Zimbabwe, Chipinge Dist., Chirinda Forest, 31 Jan. 1906,

Swynnerton 72 (BM, K, SRGH) & Oct 1908, Swynnerton 72a (BM).

SWAZILAND. — 2632 (Bela Vista) Lubombo Mts, Siteki Dist.,

Mtibhlati/Mtibalati River, 26°33T3"S, 32°06'22"E, 240 m, 22 Nov.

2002, Burrows & Loffler 7893 (Buffelskloof Herb., PRE. SDNH).

REFERENCES

BAILLON, H.E. 1858. Etude generate du groupe des Euphorbiacees.

Paris.

FIGURE 1. — Excoecaria madagas-

cariensis, Burrows & Loffler

7893: A, fruiting branch; B,

inflorescence.

156

Bothalia 33,2 (2003)

BRENAN, J.RM. 1949. Check-lists of the forest trees and shrubs of the

British Empire 5. Tanganyika Territory. Imperial Forestry

Institute, Oxford.

COATES PALGRAVE, M. 2002. Keith Coates Palgrave Trees of south-

ern Africa , edn 3. Struik, Cape Town.

DE CANDOLLE, A.P. 1866. Prodromus systematis naturalis regni

vegetabilis, vol. 15. Treuttel & Wiirtz, Paris.

MULLER ARGOVIENSIS, J. 1866. Euphorbiaceae (in part). In A.P. de

Candolle, Prodromus 15: 1219. Treuttel & Wiirtz, Paris.

PAX, F.A. 1890. Euphorbiaceae. In A. Engler & K. Prantl, Die natiir-

lichen Pflanzenfamilien 3,5: 1-1 19. Engelmann, Leipzig.

PR AIN, D. 1913. Euphorbiaceae. In W.T. Thiselton-Dyer, Flora of

tropical Africa 6, 1 . Reeve, London.

RADCLIFFE-SMITH, A. 1987. Euphorbiaceae (part 1 ). In R.M. Polhill,

Flora of tropical East Africa. Balkema, Rotterdam, Boston.

RADCLIFFE-SMITH, A. 1996. Euphorbiaceae. In G.V. Pope, Flora

zambesiaca 9,4. Royal Botanic Gardens, Kew.

RENDLE, A.B., BAKER, E.G., MOORE, S. & GEPP, A. 1911. A con-

tribution to our knowledge of the flora of Gazaland; being an

account of the collections made by C.F.M. Swynnerton. Journal

of the Linnean Society, Botany 40: 1-245.

THULIN, M. 1993. Excoecaria. Flora of Somalia 1: 305-307. Royal

Botanic Gardens, Kew.

J.E. BURROWS*, S.M. BURROWS*, P. LOFFLER+ & L. LOFFLER+

* Buffelskloof Herbarium, P.O. Box 710, 1120 Lydenburg, South Africa.

+ P.O. Box 764, Mbabane HI 00, Swaziland.

MS. received: 2003-01-28.

LYCOPERDACEAE— GASTEROMYCETES

CALVATIA SECT. MACROCALVATIA REDEFINED AND A NEW COMBINATION IN THE GENUS CALVAT1A

INTRODUCTION

Whilst accepting Kreisel’s (1992) incorporation of the

genus Langermannia Rostk. into Calvatia Fr., we, like

others (e.g. Calonge & Martin 1990; Demoulin 1993;

Lange 1993; Calonge 1998), are not convinced of the

generic status of his segregate genus Hcmdkea (Kreisel

1 989). Kreisel ( 1 989) referred to Handkea those members

previously placed in Calvatia, but characterized by ‘asep-

tate, slit-like, pitted capillitium’, and included species

with sterile bases (compact or cellular) and without.

For the same reasons as expressed in Lange (1993), and

from conclusions based on our own comparative morpho-

logical and anatomical studies of the Lycoperdaceae of

southern Africa (to be reported elsewhere), we agree with

Lange (1993) in treating Handkea as a section of Calvatia.

To that purpose Lange (1993) relegated the genus

Handkea Kreisel to sectional rank under the name

Calvatia sect. Handkea (Kreisel) M. Lange. Lange’s sec-

tional name was not validly published, however, since the

requirements of ICBN Art. 33.3 (Greuter et al. 2000) were

not met. Moreover, Lange apparently overlooked the fact

that, long before the establishment of the genus Handkea,

Kreisel (1962) had established Calvatia sect. Macro-

calvatia Kreisel, to which he assigned Calvatia excipidi-

formis (Scop.: Pers.) Perdeck |= Handkea excipulifotmis

(Scop.; Pers.) Kreisel] and Calvatia utriformis (Bull.:

Pers.) Jaap [= Handkea utriformis (Bull.: Pers.) Kreisel],

the latter being the type species of the genus Handkea.

Calvatia sect. Macrocalvatia is thus also pertinently cited

as an earlier synonym of Handkea by Kreisel (1989).

Therefore, if the latter two species, (and a number of other

species of Handkea) are to be accommodated in Calvatia,

there already exists a section for them, namely Calvatia

sect. Macrocalvatia.

A problem arises, however. The original diagnosis for

Calvatia sect. Macrocalvatia, as defined by Kreisel

(1962: 163), is unambiguous: ‘ Subgleba distincte cel-

losa. Capillitium non septatumJ This diagnosis clearly

provides for species with fruit bodies characterized by

cellular sterile bases only and excludes those with com-

pact bases as well as those without sterile bases. Calvatia

sect. Macrocalvatia in the sense of Kreisel can therefore

neither accommodate all of the species assigned to

Handkea by Kreisel (1989) nor all of those assigned to

Calvatia sect. Handkea by Lange (1993). Instead of

establishing yet another new section to accommodate the

excluded species, however, we here emend the circum-

scription of Calvatia sect. Macrocalvatia to accommo-

date also those Handkea species currently excluded by

the original diagnosis.

Emended sectional description

Calvatia sect. Macrocalvatia Kreisel in Feddes

Repertorium 64: 163 (1962), emend. J.C.Coetzee, Eicker

& A.E.van Wyk. Type species: Calvatia excipuliformis

(Scop.: Pers.) Perdeck.

Handkea Kreisel: 282 ( 1989); Calvatia sect. Handkea

(Kreisel) M. Lange: 143 (1993), nom. inval.

Fruit bodies with sterile base or not. Capillitial septa

extremely rare to essentially absent, easily missed; capil-

litial threads fragmenting at septa or more commonly by

irregular rupturing of walls; capillitial walls fragile, with

small perforations and conspicuous slit-like fissures,

often between perforations.

New combination in Calvatia

The inclusion of Handkea into Calvatia necessitates

the following new combination (description based on dry

herbarium material; colour codes and colour terms fol-

low Kornerup & Wanscher 1981):

Calvatia capensis (Lloyd) J.C.Coetzee, Eicker &

A.E.van Wvk, comb. nov. Type: South Africa, Stellenbosch,

A.V. Duthie 403 (Lloyd Myc. Coll. 7567 in BPI 706162,

holo.!; Herb. v.d. Byl in PREM 31472, iso.!).

Lanopila capensis Lloyd in Mycological Writings 7: 1177 (1923);

Verwoerd: 25 ( 1925); Bottomley: 579 ( 1948).

Handkea capensis (Lloyd) Kreisel & G. Moreno: 84 (1996).

Illustrations: Lloyd: 1. 230, fig. 2352 (1923); Kreisel & G. Moreno:

86, figs 1-7 (1996).

Bothalia 33,2 (2003)

157

Basidiocarp epigeous, with prominent rooting base, ±

globose (sensu Verwoerd 1925; Bottomley 1948), holotype

appearing to have been ± 30 mm wide and 25-30 mm high,

dehiscing by irregular fragmentation of peridium. Peridium

very thin, 0.09-0.2 mm, rigid but brittle and extremely

fragile, differentiated into an exo- and endoperidium (two

layers not discernible from type material, however).

Exoperidium dark brown with reddish tinge (7F6) to paler

and concolorous with endoperidium, fugacious, remaining

as tiny, weft-like mycelial patches on some specimens.

Endoperidium consisting of a thin, amorphous crust over-

laying a slightly thicker hyphal layer, surface colour vari-

ous shades of pale to darker brown [± 5C5 (topaz) and 5D5

(clay) to 6E6 (cocoa brown/leather brown/tan) to almost

7D5], with an evanescent metallic gloss, disappearing with

time (persistent only in folds on holotype), leaving surface

dull brown, colour of inside surface ± 5D6 (honey yel-

low/oak brown) to 5E6 (mustard brown), outer surface of

holotype ornamented with numerous, tiny, off-white, stel-

late ridges. Hyphal layer of endoperidium composed of

cyanophilic, branched hyphae mostly up to 5 pm diam., but

inflated in places, especially at branches and apices, mod-

erately thick-walled (mostly 0.75-1 pm), not tapering but

ending in rounded to often inflated tips, true septa infre-

quent but not uncommon, often perforated with slit-like

pits but much less conspicuously so compared to capilli-

tium. Gleba cottony to powdery, very fragile, brown (5D5

to 6E6) consisting of spores and capillitium. Capillitium

golden brown in clear lactophenol, strongly cyanophilic in

lactophenol with aniline blue, inamyloid in Melzer’s solu-

tion, dichotomously branched, commonly up to 7 (rarely

up to 11) pm diam., gradually tapering to thin-walled,

rounded tips, as little as 1 pm diam., often undulating

towards apices, very fragile, breaking up into fragments of

varying length, septa not observed; capillitial wall smooth,

moderately thickened, mostly varying between 0.5 and

1.25 pm, L/H averaging 0.69 (n = 18) for hyphae 5-7 pm

diam., with numerous, very conspicuous, small to large fis-

sures or slit-like pits. Paracapillitium absent Basidiospores

golden brown in clear lactophenol, cyanophilic reaction

variable, inamyloid in Melzer’s solution, globose, with

short hyaline apiculus mostly less than 1 pm long, but up

to ± 1.3 pm not uncommon, uniguttulate, radial symmetric,

isopolar, 3-5 pm diam., generally appearing to have a

diameter slightly less than much of capillitium; spore wall

± 0.5 pm thick, glabrous under light microscope but dis-

tinctly verruculose under SEM, verrucae not more than 0.2

pm high. Subgleba present, small (up to 10 mm high), yel-

lowish brown in holotype (5D5) to brown in isotype (6E6),

compact, composed of yellow-brown, cyanophilic,

branched, much contorted and bent hyphae, commonly up

to 7.5 pm diam., tapering like capillitial hyphae, moderate-

ly thick-walled (mostly between 1.0-1. 5 pm), apparently

aseptate, false septa observed but rare; diaphragm absent,

boundary with gleba poorly defined.

Distribution: Western Cape, South Africa.

Habitat: soil in a temperate climatic zone with mild,

wet winters and hot, dry summers (Mediterranean cli-

mate). Fynbos Biome.

Discussion: the material from which Lloyd (1923)

first described this fungus consists of one half of a longi-

tudinally bisected specimen sent to him by Miss A.V.

Duthie from Stellenbosch, South Africa. In his original

description Lloyd made no mention of the subgleba, a

feature first described from the other half (the isotype) by

Verwoerd (1925). In a footnote to his often overlooked

work (in Afrikaans) on South African Gasteromycetes,

Verwoerd (1925) had the following to say regarding this

puffball (our translation): 'According to Lloyd ... it does

not have a sterile base. The half in my possession, how-

ever, clearly shows one. Lloyd described it from the

other half.' The half described by Lloyd does have a ster-

ile base, however, and it is difficult to understand how he

could have overlooked this structure. Contrary to the

statement by Kreisel & Moreno (1996), Bottomley

(1948) also acknowledged the presence of a sterile base.

The holotype is in a very poor state, consisting only of

the sterile base with very little gleba still attached to it;

the peridium has disintegrated almost completely into

tiny fragments.

Although Verwoerd (1925) did not recognize the

numerous, very conspicuous slit-like perforations in the

capillitial walls of Lanopila capensis for what they real-

ly were, he did notice the resultant appearance of the

capillitium, which he described as '... with a marbled sur-

face’, emphasizing the diagnostic value of this character.

Bottomley (1948) described the capillitium as having 'a

watered appearance’.

Numerous long, thin, angular, needle-shaped crystals,

not visible with the naked eye but very conspicuous under

the SEM, occur on the peridial surface of the holotype.

These crystals were not observed on the isotype, howev-

er, and are assumed to be an artefact of unknown origin.

Although Ponce de Leon (1981) excluded this fungus

from Lanopila Fr., he did not designate it to another

genus. On a slip dated 1991, inserted with the type speci-

men, he assigned it to Calvatia Fr., however, albeit with

a question mark. Calonge, also on a herbarium slip dated

1992, placed it in Langermannia. Based on Bottomley ’s

description of the gleba being ‘septate but fragmenting at

the septa’, Kreisel (1992, 1994) reduced Lanopila capen-

sis to synonymy under Calvatia flava (Massee) Kreisel.

After having examined the holotype, however, he recog-

nized it as a distinct species and placed it in the genus

Handkea Kreisel on the basis of its slit-like capillitial

perforations (Kreisel & Moreno 1996). We prefer to

retain it in Calvatia, however, assigning it to sect.

Macrocalvatia emend.

Prior to this study Calvatia capensis was known from

the type collection only, but a re-examination of the puff-

balls in the E.L. Stephens collection, recently transferred

from BOL to PREM, as well as some specimens from the

Lloyd collection in BPI, brought to light at least five

more collections of this fungus.

Specimens examined

WESTERN CAPE. — 3318 (Cape Town): Rietvlei, 10 June 1951,

(-CD or DC), herb. Stephens 925 (PREM); Rietvlei, (-CD or DC),

herb. Stephens 1931 (PREM); Devil’s Peak near wattles, 7 July 1962,

(-CD), Chapman s.n. sub herb. Stephens 4301 (PREM); Stellenbosch,

Papegaaiberg, 19 June 1921, (-DD), A.V. Duthie 304 sub herb. Lloyd

7567 ( BPI706162 , holo., PREM31472, iso.). 3418 (Simonstown): Smits-

winkel Bay, by roadside, 28 April 1937, (-AD), R.S. Adamson s.n. sub

158

Bothalia 33,2 (2003)

herb. Stephens 452 (PREM). Locality unknown: South Africa, A. V.

Duthie s.n. sub herb. Lloyd 51765 ( BPI709920 ).

The following specimens, all with slit-like capillitial

pores and semblances of sterile bases are very similar to

C. capensis, but the material is too scanty to allow for

definite identification:

WESTERN CAPE — 3318 (Cape Town): Glen, in grass, 12 May

1954, (-CD), Chapman 424 sub herb. Stephens 1387 (PREM); between

Klipheuwel and Bellville, 17 July 1955, (-DA to DC), herb. Stephens

1500 (PREM). Locality unknown: herb. Stephens 2032 (PREM).

ACKNOWLEDGEMENTS

The curators of BPI and PREM are thanked for the

loan of specimens in their care.

REFERENCES

BOTTOMLEY, A.M. 1948. Gasteromycetes of South Africa. Bothalia

4: 473-810.

CALONGE, F.D. 1998. Flora mycologica lberica 3. Gasteromycetes 1.

Lycoperdales, Nidulariales, Phallales, Sclerodermatales,

Tulostomatales. Cramer, Stuttgart.

CALONGE, F.D. & MARTIN, M.P 1990. Notes on the taxonomical

delimitation in the genera Calvatia, Gastropila , and Langer-

mannia (Gasteromycetes). Boletin Sociedad Micologica de

Madrid 14: 181-190.

DEMOULIN, V. 1993. Calvatia pachyderma (Peck) Morg. and Gastro-

pila fragilis (Lev.) Homrich & Wright, two possible names for

the same fungus. Mycotaxon 46: 77-84.

GREUTER, W„ MCNEILL, J„ BARRIE, F.R.. BURDET, H.M., DE-

MOULIN, V„ FILGUEIRAS, T.S., NICOLSON, D.H., SILVA.

PC., SKOG. J.E., TREHANE, P, TURLAND, N.J. & HAWKS-

WORTH. D.L. 2000. International Code of Botanical Nomen-

clature (Saint Louis Code). Regnum vegetabile 138. Koeltz

Scientific Books, Konigstein.

KORNERUP. A. & WANSCHER. J.H. 1981. Methuen handbook of colour,

edn 3, 1st reprint. Eyre Methuen, London.

KREISEL, H. 1962. Die Lycoperdaceae der Deutschen Demokrati-

schen Republik. Feddes Repertorium 64: 89-201.

KREISEL, H. 1989. Studies in the Calvatia complex (Basidiomycetes).

Nova Hedwigia 48: 281-296.

KREISEL, H. 1992. An emendation and preliminary survey of the genus

Calvatia (Gasteromycetidae). Persoonia 14: 431^-39.

KREISEL, H. 1994. Studies in the Calvatia complex (Basidiomycetes)

2. Feddes Repertorium 105: 369-376.

KREISEL, H. & MORENO, G. 1996. The genus Handkea Kreisel (Basi-

diomycetes, Lycoperdaceae) in the southern hemisphere. Feddes

Repertorium 107: 83-87.

LANGE, M. 1993. Classifications in the Calvatia group. Blvttia 51:

141-144.

LLOYD, C.G. 1923. The genus Lanopila. Mycologica I Writings 7: 1177.

PONCE DE LEON, P. 1981. Langertnannia bicolor (Lev.) Demoulin &

Dring. Phytologia 48: 373-383.

VERWOERD, L. 1925. Suid-Afrikaanse Lycoperdaceae en Nidulari-

aceae. Annale van die Universiteit van Stellenbosch. Reeks A,

wis- en natuurkunde 3: 1 — 4-5 .

J.C. COETZEE* and A.E. VAN WYK**

* Department of Horticulture and Food Technology, Peninsula

Technikon. P.O. Box 1906. 7535 Bellville.

** H.G.W.J. Schweickerdt Herbarium, Department of Botany, Uni-

versity of Pretoria, 0001 Pretoria.

MS. received: 2002-12-05.

ASTERACEAE

A NEW COMBINATION IN THE GENUS MACLED1UM (MUTISEAE)

Klatt (1866) described a new variety of Dicoma zey-

heri , var. thyrsiflora , for a taxon collected in Barberton

(Figure 2). Thellung (1921) subsequently raised this vari-

ety to the species level. Dicoma thyrsiflora. In the same

year Moore (1921) described the same taxon as Dicoma

kirkii var. microcephala. Pope (1991) considered this

taxon to be an example of variation in Dicoma zeyheri

subsp. zeyheri. While revising the genus Dicoma , the first

author studied the variation in Dicoma zeyheri , and var.

thyrsiflora was found to be distinct and morphologically

different from the typical subspecies. Ortiz (2001) rein-

stated the genus Macledium Cass, and Dicoma zeyheri

was transferred to this genus. He, like Pope (1991), also

recognized only two subspecies: M. zeyheri subsp. argy-

rophyllum and M. zeyheri subsp. zeyheri and included the

above discussed taxon under the typical subspecies. It is

here proposed that the name of this taxon be reinstated

with the correct nomenclatural combination as follows:

Macledium zeyheri (Sond.) S. Ortiz subsp. thyrsi-

florum (Klatt) N.C.Netnou, comb. nov.

Dicomu zeyheri Sond. var. thyrsiflora Klatt in Bulletin de l’Herbier

Boissier 4: 844 (1866). Dicoma thyrsiflora (Klatt) Thell. in Thell. &

Schinz: 25 (1921). Type: South Africa, Transvaal [Mpumalanga].

Kaapriver Valley, Barberton, Galpin 91 1 (PRE!, holo.; K, BOL!, iso.).

Dicoma kirkii Harv. var. microcephala S. Moore: 231 (1921). Type:

South Africa, Transvaal [Mpumalanga], Barberton, Thorncroft 1074

(BM, holo.; PRE!, iso.).

According to K. Balkwill (pers. comm.), Macledium

zeyheri subsp. thyrsiflorum is a serpentine endemic.

The subspecies of Macledium zeyheri are keyed out as

follows:

la Involucral bracts extending 60-280 mm down the capitulum

stalk; leaves located basally or absent

M. zeyheri subsp. argyrophyllum

1 b Involucral bracts confined to the head or extending less than

60 mm down the capitulum stalk; plants leafy:

2a Heads very large, more than 30 mm wide, solitary or up to

three on a peduncle; involucral bracts silvery tinged

purple, with narrow whitish margins; leaves up to 35

mm wide M. zeyheri subsp. zeyheri

2b Heads relatively small, less than 20 mm wide, often

numerous in a panicle; involucral bracts light green

with broad, whitish margins; leaves relatively narrow,

up to 20 mm wide M. zeyheri subsp. thyrsiflorum

ACKNOWLEDGEMENTS

This note is part of an M.Sc. thesis submitted to Rand

Afrikaans University. Financial assistance from the NBI,

Bothalia 33,2 (2003)

159

TYPE

HERBARIUM E E. GALPIN. SOUTH f,? RICA.

N..( Tt If.

u(ftt t,.„ 't<jO Ujj alt

FIGURE 2. — Isotype of Macledium

zeyheri subsp. thyrsiflorum ,

Galpin 911 (BOL).

NRF and RAU is acknowledged. Dr Santiago Ortiz and

Dr H.F. Glen are thanked for their valuable comments.

REFERENCES

KLATT, F.W. 1866. Dicoma zeyheri var. thyrsiflora. In H. Schinz,

Beitrage zur kenntnis der Afrtkanischen Flora (Neue Folge).

Bulletin de VHerbier Boissier 4: 844.

MOORE, S. LE M. 1921. Alabastra Diversa — Part 34. Journal of

Botany 59: 231.

ORTIZ, S. 2001. The reinstatement of the genus Macledium Cass.

(Asteraceae, Mutiseae): morphological and phylogenetic argu-

ments. Taxon 50: 733-744.

POPE, G.V. 1991. Notes on Dicoma Cass. (Compositae). Kew Bulletin

46: 699-709.

THELLUNG, A. 1921. Dicoma thyrsiflora. In A. Thellung & H.

Schinz, Vierteljahisschrift der Naturforscenden Gesellschaft in

Zurich 66: 51.

N.C. NETNOU* and B.-E. VAN WYK**

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

** Department of Botany, Rand Afrikaans University, P.O. Box 524,

2006 Auckland Park, Johannesburg.

MS. received: 2002-11-15.

160

Bothalia 33,2 (2003)

LYCOPERDACEAE— GASTEROMYCETES

AUTHOR CITATION AND PUBLICATION DATE OF THE NAME CALVATIA CRANIIFORMIS

In the course of our taxonomic studies on the genus

Calvaria Fr. emend. Morgan in southern Africa, we occa-

sionally come across interesting nomenclatural inconsis-

tencies. An important and noteworthy case addressed in

this contribution concerns the origin of the name Calvaria

craniiformis, the type of the genus Calvaria Fr., a con-

served generic name. Originally described as Bovista

cranifonnis by Von Schweinitz ( 1 832), this taxon has sub-

sequently been treated as a member of Calvaria.

Throughout the literature, Fries (1849: 442), where

the genus Calvaria was first established, is consistently

cited also as the place of valid publication of the combi-

nation Calvatia craniiformis. In terms of ICBN Article

33.1 (Greuter et al. 2000: 56), however, that is incorrect,

since, when Fries ( 1 849) established the genus Calvaria ,

he merely remarked: ' Hujus loci Bovista cranif. Schw.\

and nowhere in the original publication did he actually

definitely associate the specific epithet craniiformis with

the genus name Calvaria or its abbreviation as required

by Art. 33.1. Two similar cases are discussed in Art. 33.1,

Ex. 2, which make it very clear that the name Calvaria

craniiformis was not validly published in Fries (1849).

As far as we could ascertain, the combination Calva-

tia craniiformis was first validly published by De Toni

(1888: 106) in Saccardo’s Sylloge fungorum 7. Although

De Toni also ascribed this name to Fries (1849), it should

be cited either as Calvaria craniiformis (Schwein.) Fr. ex

De Toni or merely as Calvatia craniiformis (Schwein.)

De Toni, and not as Calvatia craniiformis (Schwein.) Fr.

as has been the exclusive practice in the past.

ACKNOWLEDGEMENTS

The hospitality extended to the senior author by Dr

D.N. Pegler and staff of the Mycological Herbarium at

Kew during a brief study visit to that institution is

acknowledged with appreciation. Funding granted by the

Peninsula Technikon research committee is also

acknowledged with gratitude.

REFERENCES

DE TONI, J.B. 1888. Nidulariaceae, Lycoperdaceae et Hymenogastra-

ceae. In P.A. Saccardo, Sylloge fungorum omnium hucusque

cognitorum 7,1: 28-180. Sumptibus auctoris typis seminarii,

Patavii.

FRIES, E.M. 1849. Summa vegetabilium Scandinaviae. Sectio Poste-

rior. Typographia Academica, Uppsala.

GREUTER, W„ MCNEILL, J„ BARRIE, F.R., BURDET, H.M.,

DEMOULIN, V., FILGUEIRAS, T.S., NICOLSON, D.H.,

SILVA, PC., SKOG, J.E., TREHANE, P„ TURLAND, N.J. &

HAWKSWORTH, D.L. 2000. International Code of Botanical

Nomenclature (Saint Louis Code). Regnum vegetabile 138.

Koeltz Scientific Books, Konigstein.

VON SCHWEINITZ, L.D. 1832. Synopsis fungorum in America bore-

ali media degentium. Transactions of the American Philo-

sophical Society held at Philadelphia for promoting useful

knowledge n.s. 4: 141-316 (plus plate XIX & legend).

J.C. COETZEE* and A.E. VAN WYK**

* Department of Horticulture and Food Technology, Peninsula Tech-

nikon, P.O. Box 1906, 7535 Bellville.

** H.G.W.J. Schweickerdt Herbarium, Department of Botany, University

of Pretoria, 0001 Pretoria.

MS. received: 2002-12-05.

MESEMBRYANTHEMACEAE

BRIANHUNTLEYA, A NEW GENUS ENDEMIC TO THE WORCESTER-ROBERTSON KAROO, SOUTH AFRICA

INTRODUCTION

A fundamental principle emerging from Aristotle’s

biological work, in about 335 BC, and upon which the

earliest classification systems were built, is translated as

'the species is defined by the genus and the difference ’

(Greuter 2002). This Aristotelian logic is at the core of

the Linnaean classification system and can be applied to

any hierarchical system. The concept becomes more

challenging in the case of genera containing single

species, and, unlike early classification systems, most

modern ones do contain high numbers of monotypic gen-

era (Williams 1964). This has also been shown for

Mesembryanthemaceae (Chesselet et al. 1995), and cur-

rently, 27 genera out of 124, including the one described

here, are monotypic. It is one thing to lament monotypes,

quite another to find a logical alternate home for, e.g.,

Didymaotus N.E.Br. When a species is sufficiently dif-

ferent from all others, it is necessary to create a new

genus to accommodate the ‘difference’.

A new monotypic genus, Brianhuntleya Chesselet,

S.A.Hammer & I. Oliver is here described. It comprises a

single species, Brianhuntleya intrusa (Kensit) Chesselet,

S.A.Hammer & I. Oliver. Formerly placed in Ruschia

Schwantes, it is removed from the genus on the basis of

a unique suite of characters. The morphology, distribution,

ecology and phylogenetic affinities of Brianhuntleya are

discussed.

Our new genus is named in honour of Prof. Brian

John Huntley, Chief Executive Officer of the National

Botanical Institute, South Africa, since 1990. We feel

that it is appropriate to name this genus Brianhuntleya in

appreciation of his considerable contribution to the

advancement of biodiversity research and conservation

in southern Africa.

In recent years a number of new genera have been

described in Mesembryanthemaceae, of which several

pay tribute to botanists, notably Ihlenfeldtia H.E.K.Hart-

Bothalia 33,2 (2003)

161

mann (Hartmann 1992); Hartmanthus S. A. Hammer

(Hammer 1995a) and Hammeria Burgoyne (Burgoyne et

al. 1998). Curiously, all three of these are bitypic.

A live plant in fruit ( Chesselet & Oliver 485 ), collected

near Worcester, on the road to Robertson, in December

2002, was brought to the Compton Herbarium at

Kirstenbosch for identification. The singular combina-

tion of characteristics, particularly of the pedicel, calyx

and fruit, precluded a suitable generic placement in the

current system for mesembs. However, it was necessary

to search for a species name, as Louisa Bolus had named

most mesembs, and often more than once, during the

years of intensive collecting and describing of new

species in the first half of the 20th Century, and indeed

the plant was finally identified as Ruschia intrusa

(Kensit) L. Bolus. It is significant that we had trouble

finding a 'Bolus name’: it didn't occur to us that anyone

would ever have placed this plant in Ruschial

The following combination of features distinguishes

B. intrusa from all other mesembs: ‘intruse’ calyx tube;

unique, finely striate wax cover of the leaves; echinate,

highly papillate seed; sigmoid, succulent pedicel that

withers, allowing fruit to break off from the mother plant

and disperse seed from the well-separated fruit; flat bowl-

shaped capsule base enabling rain-splash dispersal in

detached fruit (usually found in some multilocular fruit).

Brianhuntleya Chesselet , S. A. Hammer & I. Oliver,

gen. nov. (Ruschieae Schwantes; Mesembryanthemaceae

Fenzl). Type species: B. intrusa (Kensit) Chesselet,

S.A.Hammer & I. Oliver.

Fruticulus perennis caespitosus repens, radice palari

centrali radicibusque vadosis e nodis basalibus enatis;

rami prostrati lignosi reliquuis dessicatis parum priorum

foliorum tectis. Folia trigona connata, basin versus rubi-

cunda, carinata griseo-viridia, tegmine cereo crasso stri-

ato, pro parte in lamellas foliis crescendibus delabente,

unum vel duo pares foliorum ad extremitates ramulorum

brevium. Flos solitarius in pedicello longo crasso sig-

moideo camoso, fructu ubi maturo exarescenti desiccan-

ti disrumpentique. Pedicellus ebracteatus, sed evolutio

pedicelli florisque concurret cum productione paris

foliorum quod primo bracteas simulare potest. Calyx

truncatus, tubum formans, sepalis 5 subaequilongis, 3

membranaceis. Petala magenteo-purpurea, uniseriata,

staminodia filamentosa nulla; staminum classes staturae

tres, stamina basin versus papillata, apicibus roseis, conum

centralem circum stigmata formantes, polline pallide

flavo; styli subulati, longitudine mediocri. Nectarium

cristatum annulare (holonectarium lophomorphum).

Fructus capsula 5-locularis hygrochastica, basi crateri-

formi, supra elevata; valvae apertae horizontales, corpo-

ra claudentia magna, loculorum exitos obstantia, propter

texturam spongiosam albida, membranae tegentes cen-

trum fructus versus elevatae, liminibus claudentibus in

superficie inferiora distali; dispersio seminum ope ejec-

tionis per membranes tegentes carinae dilatantes diver-

gentes, a laminis dilatantibus distinctae, brunneae, lacer-

atae, in subulam desinentes; alae valvarum nullae.

Semina rubiginosa, ± 1 mm longa, propter papillas longas

echinata.

Perennial, tufted, creeping dwarf shrub with a central

taproot and shallow roots arising from nodes at base of

plant; branches prostrate, woody, covered with dried

remains of previous leaf pairs. Leaves trigonous, fused,

reddish at base, keeled, grey-green with thick, striate,

wax cover that flakes off partially as leaves expand, one

or two leaf pairs at ends of short shoots. Flowers solitary,

borne on long, thick, sigmoid succulent pedicel that

shrivels, dries out and breaks off when fruit is ripe.

Pedicel without bracts, but development of pedicel and

flower coincides with production of a leaf pair that in

early stages may resemble bracts. Calyx truncate, form-

ing a calyx tube, 5 sepals more or less of equal length, 3

membranous. Petals magenta-purple, in a single whorl;

filamentous staminodes absent; three size classes of sta-

mens, with papillate bases and pink tips, forming a cen-

tral cone around stigmas; pollen pale yellow; styles subu-

late, of medium length. Nectary crested, annular (lopho-

morphic holonectary). Fruit 5-locular, hygrochastic cap-

sule, base bowl-shaped and top raised: valves opening to

horizontal position, closing bodies large, blocking exits

of locules, whitish from spongy tissue, covering mem-

branes raised towards centre of fruit, with closing ledges

on distal undersurface; ejection dispersal through cover-

ing membranes; expanding keels diverging, distinct from

expanding sheets, brown, lacerate, ending in an awn;

valve wings absent. Seeds reddish, ± 1 mm long, echinate

from long papillae.

The genus includes a single species:

Brianhuntleya intrusa ( Kensit ) Chesselet, S.A.Ham-

mer & I. Oliver, comb. nov.

Mesembryanthemum intrusion Kensit in Bolus & Kensit in Trans-

actions of the Royal Society of South Africa 1: 151 (1909). Ruschia

intrusa (Kensit) L. Bolus: 220 (1950). Lectotype: Cape, hills near

Robertson, July 1901, Marloth 4592 (BOL!).

Emended species description

Tufted plant, 70-100 mm high: branches decumbent,

woody, old leaves persistent on plant. Leaves trigonous,

50-60 x 7-8 mm, grey-green, reddish at bases, waxy,

keel indistinct; arising from between 2 pairs of young

green leaves in axil of older pair. Flowers up to 35 mm

diam., spreading; pedicel 30-50 mm long, succulent.

Sepals 5, of subequal length, 3 membranous, up to 3 mm

long; petals up to 17 x 1.5 mm, 1 -seriate, pale rose-purple;

nectary green, annular and crested; top of ovary raised,

convex; stigmas 5, subulate, acuminate, 2 mm long; sta-

mens in 3 rows, 2-4 mm long, pinkish at tips; anthers

and pollen pallid. Fruit 5-locular, hygrochastic capsule,

7. 5-9.0 mm diam., 4 mm deep, valves raised, 4 mm high;

seed 1 mm long. Flowers open for a few hours in the

afternoon. Flowering time : late May-June in cultivation

and July in the wild. Figure 3.

Etymology, the specific epithet intrusum, from the

Latin intrusus , is used in the botanical context to

describe a form that appears pushed or thrust inwards. In

this context Bolus used the term to describe the ‘very

peculiar truncate somewhat intruse calyx tube’ character-

istic of this species (Bolus & Kensit 1909).

162

Bothalia 33,2 (2003)

cel; I, lop view; J, open fruit with large whitish closing bodies, covering membranes and expanding keels. K, echmate seed. Scale bars. A J, 5 ,

K, 1 mm.

Bothalia 33,2 (2003)

163

FIGURE 4. — Known geographical distribution of Brianliuntleya intrusa.

Distribution and ecology, the genus Brianhuntleya is

endemic to the Worcester-Robertson Karoo, part of the

Succulent Karoo Biome, from which 11 genera and 37

species of mesembs are known (Chesselet et al. in prep.).

The monotypic genus Stayneria L. Bolus is the only other

mesemb genus endemic to the Worcester-Robertson

Karoo, where it is confined to Renosterveld in the

Breede River Valley, from Worcester to McGregor.

B. intrusa is known from populations at Worcester,

Robertson, Montagu, McGregor and Bonnievale, all in

the Breede River Valley, where it is restricted to eroded

Malmesbury Shale sometimes interspersed with small

chunks of surface quartzite (Figure 4). Plants grow on

low hills at altitudes ranging from 200-250 m, in full

sun, with mixed succulent vegetation that includes

Ruschia caroli, Adromischus filicaulis, Poellnitzia rubri-

flora and Conophytum ficiforme. Near Bonnievale it

grows with Acrodon purpureo stylus.

In horticulture B. intrusa is very resilient, not sensi-

tive to over-watering, and it produces its splendid flow-

ers at a time when most vygies are without flowers. It

becomes quite luxuriant if well watered, forming a grey-

green ground cover, reminiscent of some species of

Carpobrotus N.E.Br. It is so undeservedly obscure in

horticulture that a recent plant list noted that, until

recently, no one had ever purchased it.

With only a few known populations, B. intrusa is

already in a conservation programme at the Karoo Desert

NBG. Live plants were collected in a salvage operation

for ex situ and in situ conservation from a site of planned

road works at the Gorees Hoogte Pass near Robertson.

Once road works are complete, the area will be restored.

Systematic affinities: when Bolus (nee Kensit) describ-

ed Mesembryanthemum intrusum (Bolus & Kensit 1909),

she suggested that the new species is affiliated to M.

divergens Kensit, now Antegibbaeum fissoides (Haw.)

Schwantes ex C.Weber from near Matjiesfontein, and to

M. brevipes Schltr., now known as Argyroderma fissum

(Haw.) L. Bolus, from the Knersvlakte. This latter associa-

tion seems highly unlikely. However, both Klak et al.

(2003) and Hartmann (2001a) follow Bolus’s suggestion

that B. intrusa may be closely related to the monotypic

Antegibbaeum Schwantes ex C.Weber. Echinate seeds,

which are characteristic of B. intrusa , feature in a num-

ber of genera in Mesembryanthemaceae including

Acrodon N.E.Br., Antegibbaeum , Braunsia Schwantes

and Namaquanthus L. Bolus.

It may be deduced from a recent molecular study of

relationships in the Lampranthus Group (Klak et al.

2003), that B. intrusa would resolve as a clade with

Hammeria , Antegibbaeum , Smicrostigma N.E.Br., Zeukto-

phyllum N.E.Br., Vlokia S. A. Hammer and 1 Braunsia ’

vanrensburgii (L. Bolus) L. Bolus. Diagnostic features of

the fruit, however, conflict significantly with the pro-

posed grouping of the above genera by Klak et al. (2003).

One cannot rule out the possibility that B. intrusa may

belong to the mysterious ‘Calamophyllum’ , a genus erect-

ed by Schwantes (1927) based on cultivated plants.

Species included in Calamophyllum were originally

described by Haworth in the 1790s under Mesembry-

anthemum and have not been definitely identified since.

Problems of correct identification in this genus are not

surprising, as descriptions are conflicting and distribu-

tions unknown (Jacobsen 1960; Herre 1971; Smith et al.

1998; Hartmann 2001b). Some features that may support

this possibility include the grey-green leaves of ± cylin-

drical shape, the long pedicel of C. teretifolium (Haw.)

Schwantes, the leaves of C. teretiusculum (Haw.)

Schwantes that may be impunctate (Hartmann 2001b), and

the flowers seen in Mesembs of the World (Smith et al.

1998), as well as the illustration of the ovary with raised

top and five stigmas shown in R. Darroll’s illustration in

Herre’s (1971) Genera of the Mesembryanthemaceae .

Brianhuntleya grows sympatrically with Acrodon

purpureo stylus (L. Bolus) Burgoyne near Bonnievale,

and it is remarkable how easily these plants can be con-

fused with each other, both having similar grey-green

leaves with reddish bases, creeping growth forms and

detachable fruit. Nonetheless, A. purpureo stylus is easily

distinguished by its unwaxed leaves, Acrodon- type flow-

ers (striate pink petals, stamens collected in a cone,

plumose stigmas and pale pollen), and shiny orange, well-

exposed stems reminiscent of species of Jordaaniella

H.E.K. Hartmann. When Dehn (1992) revised the Ruschii-

nae, he annotated specimens of Ruschia intrusa as

belonging to the genus Acrodon although this was later

rejected. However, B. intrusa has pale pollen and echi-

nate seed in common with species of Acrodon.

The transfer of Ruschia purpureostyla (L.Bolus) Bruyns

to Acrodon (Burgoyne 1998) raised much controversy

(Klak 2000; Hartmann 2001b). The lophomorphic

holonectary of A. purpureo stylus, shown to be a signifi-

cant taxonomic character by Chesselet et al. (2002), pre-

cludes its inclusion in Cerochlamys N.E.Br, a genus with

a lophomorphic meronectary, as suggested by Hartmann

(2001b). The fruit of A. purpureostylus is similar to that

of B. intrusa as it also has a withering pedicel and

detaches from the mother plant, unlike the rather solid

fruit with thick persistent pedicels of other species of

Acrodon N.E.Br. The unusual fruit of A. purpureostylus

either casts some doubt on its current generic placement

or highlights strong selection for detachable fruit in their

164

Bothalia 33,2 (2003)

mutual environment. The base is rounded and not flat as

in B. intrusa and the covering membranes are not as

raised or hardened, closing bodies are larger and spongi-

er, the valves open to the upright position (as in other

Acrodon and Ruschia species) and do not fold back to the

horizontal position as in B. intrusa, thus excluding the

possibility of these two species being congeneric.

A link to the Leipoldtia Group of Hartmann (1991) is

suggested by Brianhuntleya' s unique fruit structure and

overall resemblance to members of the genera Cephalo-

phyllum N.E.Br. and Cheiridopsis N.E.Br. The floral

resemblance may be more than superficial; it is remark-

able how closely the flowers of B. intrusa resemble those

of a typical ‘showy’ Cephalophyllum and how little they

look like those of most Ruschia species (those being

much smaller, and often bunched). In cultivation, B.

intrusa crosses with Cephalophyllum subulatoides

(Haw.) N.E.Br., a Little Karoo species, providing further

support for its affinity to members of the Leipoldtia

Group. The flat, bowl-shaped base is only known from

genera such as Cheiridopsis and Cephalophyllum, how-

ever, these genera have multilocular fruits, whereas

Brianhuntleya has a five-locular fruit. It is indeed a sin-

gular entity, and its placement has intrigued mesemb spe-

cialists for many years (Hammer 1995b; Hartmann

2001a; Klak et at. 2003).

ACKNOWLEDGEMENTS

We thank Overhex Wineries and Haw & Inglis (Pty)

Ltd for taking an active interest in our indigenous flora

and Dr Hugh Glen for the Latin translation.

Other specimens examined

WESTERN CAPE. — 3319 (Worcester): Langvlei, (-DC), Bruyns

9057 (BOL); Robertson, between McGregor and Bonnievale, (-DD),

Glen 624 (BOL). 3320 (Montagu): Bonnievale, (-CC), R.H. Compton

NBG 11 38/24 (BOL); ./. Lewis NBG 1975/33 (BOL).

REFERENCES

BOLUS, H.M.L. 1950. Notes on Mesembryanthemum and allied genera

3: 220.

BOLUS, H.M.L. & KENSIT, L. 1909. Contributions to the African Flora.

Transactions of the Royal Society of South Africa 1: 147-163.

BURGOYNE. P.M. 1998. Finding a place in the sun: where does

Ruschia purpureostyla belong? Aloe 35: 60, 61.

BURGOYNE, P.M.. SMITH, G.F. & CHESSELET, P. 1998. Hammeria,

a new genus of Aizoaceae from South Africa. Cactus and

Succulent Journal 70: 203-208.

CHESSELET, R, MOSSMER, M. & SMITH, G.F. 1995. Research priori-

ties in the succulent plant family Mesembryanthemaceae Fenzl.

South African Journal of Science 91: 197-209.

CHESSELET, P, SMITH, G.F. & VAN WYK, A.E. 2002. A new tribal

classification of Mesembryanthemaceae: evidence from floral

nectaries. Taxon 51: 295-308.

CHESSELET, P, VAN WYK, A.E., GRIFFIN, NJ. & SMITH, G.F., in

press. Patterns of floristic diversity in Mesembryanthemaceae.

Aloe.

DEHN, M. 1992. Untersuchungen zum Verwandtschaftskreis der

Ruschiinae (Mesembryanthemaceae Fenzl). Mitteilungen aus

dem Institut fiir Allgemeine Botanik, Hamburg 24: 91-198.

GREUTER, W. 2002. The ancient Greek roots of biological sciences.

Flora Mediterranea 12: 5-10.

HAMMER, S.A. 1995a. Hartmanthus, a new genus in Aizoaceae.

Haseltonia 3: 77-82.

HAMMER, S.A. 1995b. A few observations on Ruschia schwantii.

Piante Grasse 15: 51-58.

HARTMANN, H.E.K. 1991. Mesembryanthema. Contributions from

the Bolus Herbarium 13: 75-157.

HARTMANN, H.E.K. 1992. Ihlenfeldtia, a new genus in Mesembry-

anthema (Aizoaceae). Botanische Jahrbucher 114: 29-50.

HARTMANN, H.E.K. 2001a. Illustrated handbook of succulent plants.

Aizoaceae A-E. Springer.

HARTMANN, H.E.K. 2001b. Illustrated handbook of succulent plants.

Aizoaceae F-Z. Springer.

HERRE. H. 1971. The genera of the Mesembryanthemaceae. Tafel-

berg, Cape Town.

JACOBSEN, H. 1960. A handbook of succulent plants, vol. III. Mesem-

bryanthemums (Ficoidaceae): 913-1441. Blanford, Poole.

KLAK, C. 2000. Taxonomic studies in the Aizoaceae from South

Africa: three new species and some new combinations.

Bothalia 30: 35 — 42.

KLAK, C„ HEDDERSON, T.A. & LINDER, H.P. 2003. A molecular

systematic study of the Lampranthus Group (Aizoaceae) based

on the chloroplast TrnL-trnF and nuclear ITS and 5S NTS

sequence data. Systematic Botany 28: 70-85.

SCHWANTES, G. 1927. Zur Systematik der Mesembrianthemen.

Zeitschrift fiir Sukkulentenkunde 3: 15, 28.

SMITH, G.F.. CHESSELET, R, VAN JAARSVELD, E.J.. HARTMANN,

H„ HAMMER, S„ VAN WYK, B-E„ BURGOYNE, P„ KLAK,

C. & KURZWEIL, H. 1998. Mesembs of the World. Briza

Publications, Pretoria.

WILLIAMS, C.B. 1964. Patterns in the balance of nature and related

problems in quantitative ecology. Academic Press, London.

P. CHESSELET*, S. HAMMER** and I. OLIVER***

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

7735 Claremont, Cape Town.

** Sphaeroid Institute, 845 Mason Road, Vista, California 92084, USA.

*** Karoo Desert NBG, P.O. Box 152, 6850 Worcester, South Africa.

MS. received: 2003-07-07.

FABACEAE

THE CORRECT NAME FOR ACACIA MONTANA

The author published a new species. Acacia montana

within the Acacia karroo complex in Coates Palgrave

(2002). Bentham ( 1842) used this specific epithet for an

Acacia species in Australia that was transferred to

Racosperma Martius by Pedley (1987). The Australian

species is now known as R. montana which is the base

name for A. montana. Therefore A. montana is a

homonym and the correct name is:

Acacia theronii P.P.Swartz, nom. nov.

A. montana P.P.Swartz in M. Coates Palgrave, Keith Coates

Palgrave Trees of southern Africa, edn 3: 19, 289 (2002) non A. mon-

tana Benth: 360 ( 1842).

TYPE. — KwaZulu-Natal, 283 1 (Nkandla): Hlabisa Dist.,

Feb. 1976, (-BB), Swartz 178 (PRE, holo.; PRU).

Acacia theronii is named after Prof. G.K. Theron, pre-

viously from the Botany Department of the University of

Pretoria, who did many years of research on the vegeta-

tion of the Loskopdam Nature Reserve, where this

Bothalia 33,2 (2003)

165

FIGURE 5. — Acacia theronii P.P.Swartz, Codd 2009, Acocks 13077, ±

18 km SE of Hlabisa, KwaZulu-Natal.

species is found. The author investigated the Acacia karroo

complex in detail for many years and realised that this

new taxon is well defined (Swartz 1982). It is a tall, sleder

tree, branching high above the ground, with a relatively

pale trunk (Figure 5). It is well adapted to grow in dry

and very hot areas and is always associated with shale

rock formations. It has the ability to revive well after fire.

The wood is dense, hard, compact and relatively heavy.

Chemical analysis of the wood confirms that it is a sepa-

rate taxon from A. karroo and A. natalitia (Malan &

Swartz 1995). The leaves of A. theronii have a robust,

mat appearance, as the leaflets (pinnules) are relatively

large and densely packed with a thick wax layer as seen

under the electron-microscope (Swartz 1982). Anatomy

of young stems show that the periderm actively divides

at an early stage, and new cells are constantly being

formed, pushing the older ones off, and resulting in a rela-

tively thick and powdery bark (Swartz 1982). Chemical

analysis of the fragrances of the flowers, the seed and

leaves done by Brain (1986, 1987), also show evidence

that this is a well-defined new species.

Acacia theronii was found growing on the hills of

Hlabisa in northern KwaZulu-Natal, southeast of

Vryheid, as well as on the hills around Groblersdal in

Mpumalanga in the Loskopdam Nature Reserve. It

occurs in a transitional area in Umfolozi, Hluhluwe and

the Lebombo Mountains and on the bushy hills around

Pongola, Magudu, Mkuze and Nongoma (Ross 1975).

ACKNOWLEDGEMENTS

The advice of Profs P.J. Robbertse, P.D.F. Kok and

A.E. van Wyk. Ms Meg Coates Palgrave and Ms Marie

Jordaan is greatly appreciated.

REFERENCES

BENTHAM, G. 1842. Notes on Mimoseae, with a synopsis of species.

Hooker’s London Journal of Botany 1: 360.

BRAIN, P. 1986. Leaf peroxidase types in Acacia karroo. Geographical

distribution and influence of the environment. South African

Journal of Botany 52: 48-52.

BRAIN, P. 1987. Immunology and phylogeny: a preliminary study of

Acacia. South African Journal of Science 83: 422^-27.

MALAN, E. & SWARTZ, P. 1995. A comparative study of the pheno-

lic products in the heartwood of Acacia karoo from two differ-

ent localities. Phytochemistry 39: 791-794.

PEDLEY, L. 1987. Racospenna Martius (Leguminosae: Mimosoideae)

in Queensland: a checklist. Austrobaileya 2: 344—357.

ROSS, J.H. 1975. Fabaceae. Mimosoideae. Flora of southern Africa,

vol. 16, part 1. Department of Agricultural Technical Services,

Pretoria.

SWARTZ, P. 1982. ’n Numeries-taksonomiese evaluering van die tak-

son Acacia karroo Hayne in suidelike Afrika. M.Sc. thesis.

University of Pretoria.

SWARTZ, P.P. 2002. Acacia montana. In M. Coates Palgrave, Keith

Coates Palgrave Trees of southern Africa, edn 3. Struik, Cape

Town.

P.P. SWARTZ*

* Formerly National Botanical Institute, Pretoria. Present address: 362

Brook Street, 0181 Menlo Park, Pretoria.

MS. received: 2003-07-01.

CLEVEACEAE -MARCH ANTIALES

SAUTERIA NYIKAENSIS, A NEW LIVERWORT SPECIES FROM MALAWI

INTRODUCTION

In the Cleveaceae the dorsal air pores of the thalli are

simple and the radial walls of the surrounding cells are

often thickened. Three genera have traditionally been

grouped together in this family, namely Athalamia Falc.,

Peltolepis Lindb. and Sauteria Nees. Peltolepis and

Sauteria have not been reported from Africa, but

Athalamia (formerly Clevea) has long been known from

this continent, with two species that occur here, namely A.

spathysii (Lindenb.) Nees and A. pulcherrima (Steph.)

Hatt. (Vanden Berghen 1965). A third species, Clevea

(. Athalamia ) crassa Trabut, from the Atlas Mountains

(Magreb), is considered to be a nom. inval. (Grolle 1976).

Sauteria is a small genus of ± five (Bischler 1998) or six

species (Schuster 1992), although some authorities recognize

only three species worldwide (Gradstein et al. 2001). The

Bothalia 33,2 (2003)

. „ , ,o7A a_f thnllns- A dorsal face with young female receptacle, stolon cut off apically; B, ventral

nsjz ;h;rr & o. a, -s - ** —

furrow; P-R, paleae. Drawings by M. Steyn.

Bothalia 33,2 (2003)

167

genus is widespread, but is restricted to high mountain envi-

ronments, often above 3 000 m, although it also occurs at

lower elevations on islands (Gradstein etal. 2001). In Europe,

Russia, Siberia, Tibet, the Himalayas and northern Japan, the

distribution of its members is arctic or montane. Its western

range includes areas in Iceland, east and west Greenland, as

well as east and west Canada (Muller 1951-1958). Hitherto,

the only known records in the southern hemisphere are those

from the Andes of Pern, northern Argentina and Chile, as well

as the Galapagos Islands, where it was found at an altitude of

about 1 200 m (Gradstein et al. 2001 ).

During April 2000, an unusual thallose liverwort was col-

lected on the Nyika Plateau at an altitude of 2 343 m, on soil,

in a cavity under a rock overhang. Unfortunately, neither of

the two female receptacles present is mature, and antheridia

are absent. Nevertheless, in order to draw attention to this

plant, it is described here and has been referred to the genus

Sauteria for the following reasons: 1, the thalli are light green

and fragile, the assimilation tissue is spongy and there is no

trace of pigmentation; 2, the cells surrounding the simple air

pores are strongly thickened; 3, the air chambers are empty,

lacking both filaments and papillae, medianly in 2 or 3 layers

and, visible beneath the epidermal cells in the wings, are the

parallel outlines of what appears to be a single layer, oblique-

ly orientated toward the thallus margins; 4, the ventral scales

do not extend to the thallus margins; they are hyaline, with a

single, tapering appendage, and are arranged in ill-defined

rows, mostly confined to the prominent midrib; oil bodies are

very rare and apical slime papillae are absent; 5. the female

receptacle originates from a deep notch at the apex of the thal-

lus; 6, in cross section the stalk of the receptacle has a single

rhizoid furrow, whereas Atlialamia species have none and

Peltolepis species have two.

According to Schuster (1992), ‘ Sauteria is separated

from the other two genera of the Cleveaceae by one

absolute feature (solitary rhizoid furrow of carpocephalum

stalk)'. He also mentions ‘distinct, scattered oil cells' in the

ventral scales, in some (but not all) cases, adding that iso-

lated ventral scales in Greenland Sauteria alpina often lack

oil cells. Shimizu & Hattori (1954) describe the oil cells in

S. alpina as ‘scattered in the ventral scales and the dorsal

epidermis of thallus, rare’. In their description of S. alpina

var. japonica (later elevated to S. yatsuensis ), they note

that, ‘oil-cells scattered in ventral tissue of thallus and

bractlets of female receptacle (and also in ventral scales of

thallus?), very rare", their question mark clearly indicating

uncertainty. In a later description of S. alpina, Hattori &

Shimizu ( 1955) remark that ‘oil-cells rare, scattered in ven-

tral tissue, ventral scales and bractlets’. They do not, how-

ever, illustrate oil cells in their ‘Text-fig. XXI', although

figs F-H of the ventral scales show groups of 5-7 cells sur-

rounding a much smaller central cell, which does not con-

tain an oil body. This is also seen in Figure 6K-M of the

Nyika plant. Oil bodies in Sauteria have been observed to

be long-persisting; those in the scales of 5. alpina , leg.

S.O.Lindberg & E.Rettig (held at PRE), are still present

120 years after collection.

Sauteria nyikaensis Perold , sp. nov.

Thalli magnitudine media vel sat magna, apice semel

dichotome ramificantes, interdum irregulariter; laete

virides, sine pigmento, fragiles spongiosique. Cavemulae

aeriae, circumscriptione dare visibile, in medio thalli

parallele, apicem versus dispositae, sed marginem thalli

versus oblique dispositae. Pori dorsales non elevati, sim-

plices, ab 1 vel 2 annulis concentricis cellularum cir-

cumscripti, pro parte vel pro parte maxima incrassation-

ibus conspicuis tecti. Squamae ventrales hyalinae appen-

diculo acuminato non semper manifesto, in seriebus in-

certis supra costam dispositae. Costa saepe producta

stolonem magnum geotropicum formans. Antheridia non

visa. Receptaculum gynoeciale immaturum, in incisura

apicali setae brevi tereti, sulco uno rhizoidali insidens.

Guttae olei omnino absentes, semel tantum in squama

ventrali visae.

TYPE. — Malawi, 1033: Nyika National Park, Jalawe

viewpoint, (-BD), on soil, in a cavity under rock over-

hang, at altitude 2 343 m, 3 April 2000, Koekemoer 1874

(PRE, holo.) with Lunularia cruciata (L.) Dumort. ex

Lindb. and Plagiochasma eximium (Schiffn.) Steph.

Thalli prostrate, in crowded patches, obovate, apical-

ly notched or incised, on either side with rounded lobes

(Figure 6A, B), continuing sometimes as smaller lobes

along slightly decurved (Figure 7A) attenuate margins;

medium-sized to fairly large, up to 12 mm long and 5-9

mm wide distally, narrowing gradually or abruptly up to

± 4 mm wide proximally, branching dichotomously

once, but in young plants often irregularly or rather dif-

fusely; light green, without any pigmentation, margins

colourless; fragile and spongy, with clearly visible out-

lines of elongated, empty air chambers medianly running

parallel toward apex, but soon becoming obliquely orien-

tated toward thallus margins (Figure 6C), each one

apparently opening by a simple air pore; along dorsal

midline, slightly concave and not grooved, laterally mar-

gins acute, flanks sloping obliquely, ventral face median-

ly keeled with a prominent midrib, rounded distally but

flattening proximally, covered with rhizoids and ill-

defined rows of hyaline scales; midrib rarely branched at

its apex, ensuing laminae irregularly shaped, most com-

monly continuing growth distally and occasionally proxi-

mally as well, forming very large tuberous, geotropic

stolons (Figures 6AB; 7D), up to 850 pm diam., filled

with starch grains. Dried plants with flanks sometimes

flat, incurved or ascending.

Dorsal epidermal cells rarely chlorophyllose, unis-

tratose, thin-walled, without trigones, 4- to 6-sided,

shorter than wide, generally 30^-5 x 60.0-72.5 pm, in

cross section 32.5M-0.0 pm thick; margins unistratose,

with 2 juxtaposed cell rows (Figure 6D), mostly rectan-

gular, others 5-sided, outermost cells 22.5-50.0 x 15.0-

32.5 pm, some with thickened walls between adjoining

cells; second row of cells 4- or 5-sided, 22.5-45.0 x

27.5^15.0 pm, walls not thickened; air pores (Figures

6F-H; 7B; 8 A) one per air chamber, not raised, simple,

oval or rounded, 10-15 x 10-20 pm, with or without

faint inner ring of small cells and then bounded by 1,

occasionally 2 concentric rings of cells, variable in num-

ber, and covered partly to sometimes almost entirely by

conspicuous thickenings, 15-25 x 12.5-20.0 pm, also

obscuring several of the radial cell walls, width of air

pore together with surrounding cells 62.5-100.0 pm; row

of dorsal epidermal cells adjoining thickened cells often

168

Bothalia 33,2 (2003)

FIGURE 7. — Sauteria nyikaensis , Koekemoer 1874. SEM micrographs. A, margin of dorsal face of thallus decurved over lateral part of ventral

face, arrows indicating positions of some dorsal air pores with thickenings; B, thickened cells around contracted air pore in concentric

rings, partly shown; C, irregular rows of ventral scales mostly over midrib, between rhizoids; D, massive ventral stolon with scales and

rhizoids, partly overlying apical part of ventral face of thallus; E, young female receptacle from above; F, young female receptacle from

side. A, x 58; B, x 430; C, x 18.5; D, x 8.5; E, x 18; F, x 21.

somewhat smaller than average, partly arranged in a con-

centric ring, occasionally the thickenings extending

slightly onto a few of them as well.

Assimilation tissue, as seen below and through dorsal

epidermis, with parallel outlines of empty air chambers

200-350 pm apart, running obliquely across wings to mar-

gins of thallus, partitioned in wings by slanting, unistratose

cross walls at intervals of up to 600 pm between them, at

thallus margins air chambers somewhat smaller; in cross

section (Figure 6E), thallus over midrib 700-1 150 pm thick,

± upper half with polyhedral air chambers in 2 or 3 layers,

65-175 x 150-220 pm, with lower ones smaller, unistratose

walls consisting of chlorophyllose cells, spherical or ovoid,

37.5-50.0 x 25.0-42.5 pm; storage tissue occupying ± */6 of

width of thallus medianly and ± 1 2 rows of cells in lower

half of thickness of thallus, decreasing laterally, soon disap-

pearing and flanks bounded beneath by ventral epidermis

only, cells crowded together, angular, 27.5^10.0 x 50-65

pm, no sclerotic cells, oil bodies or mucilage cavities pre-

sent; rhizoids densely covering midrib, fewer beneath

wings, some smooth (Figure 6J), 27.0-47.5 wide, others

pegged (Figure 61), 15-25 pm wide. Ventral scales (Figure

6K-M), hyaline, in 2-4 poorly defined rows over midrib

and extending onto adjacent ventral face of thallus (Figure

1C), also on continuation(s) of midrib as geotropic stolon(s)

(Figures 6B; 7D); inconspicuous, asymmetrically triangu-

lar, one side obliquely rounded, margins entire, tapering

gradually and not constricted where joined with acuminate,

apically pointed and not sharply differentiated appendage,

725-1025 pm long (including appendage), width across

base 375-725 pm, cells 4-6-sided, 45-75 x 30-45 pm, in

each scale 1-3 groups of cells surrounding 1 much smaller,

central cell, not containing an oil body; oil bodies very rare

(Figure 8A, B), 22.5-25.0 x 20.0-27.5 pm, light brown and

finely granular.

Monoicous? Antheridia unknown. Gynoecial recepta-

cle terminal, raised on short stalk, (Figures 6N; 7E, F) at

crotch of apical incision up to 2.5 mm long, separating 2

thallus lobes, immature, rounded above, ± 1675 pm

wide, with 8 lobes below, air pores not seen, but may

develop later, as air chambers visible in cross section of

receptacle, a single archegonium also seen; stalk terete,

with one rhizoid furrow (Figure 60), 625 pm long at this

stage of development, 875 pm wide, without assimila-

tory strip, naked below, but with paleae at apex; paleae

elongated and narrow (Figures 6P-R; 9B) 375-800 x

75-120 pm, inner cells 25-40 x 12.5-25.0 pm, with 1 to

3 papillae apically and sometimes 1 at margin, thicker-

walled at tip. Chromosome number for the genus

Sauteria : n = 36 (Muller 1951-1958; Hattori & Shimizu

1955, count by Dr S. Tatuno for Sauchia japonica, later

transferred to Sauteria yatsuensis\ Bischler 1998). The

Nyika material was no longer living when examined, and

a chromosome count could not be done.

DISCUSSION

In the absence of antheridia, the single archegonium

seen in the above specimen, would not have been ferti-

lized. The development of antheridia may have been

delayed for some reason, or else they had already disap-

peared. Bischler (1998) states that in families of the

Marchantiales with archegoniophores other than the

Marchantiaceae, the stalk elongates after fertilization. In

the Nyika plant, however, the stalk is still very short,

almost sessile, and fertilization had not taken place.

Bothalia 33,2 (2003)

169

U. . ■' / "

, //ffi :

- , :: /•

A /»' . '

v / r / .r ‘

. V :>'VX P-**.

-v

-■vV

■/. J .r"

' 0-1 ■ " ' - < -V

>•' ."/ %v v

A

: Vv\{i ir

FIGURE 8. — Sauteria nyikaensis, Koekemoer 1874. LM micrographs.

A, part of ventral scale with 2 oil bodies; B. much enlarged

micrograph of one of the oil bodies. A, x 198; B. x 790.

Because of the delicate structure of the thalli, the

plants are thought to be drought intolerant and to peren-

nate during the dry season by means of the tuberous

geotropic stolons.

At the site in northern Malawi (Figure 10), where the

plant was collected near the Jalawi View Point, it grew

on calcareous soil containing slivers of mica, in a small,

cave-like cavity at the base of a large rock. Not much

direct sunlight could have reached it there, but species of

Sauteria, except for S. cliilensis, lack secondary pigmen-

tation even when growing in open, strongly illuminated

sites (Schuster 1992). Shimizu & Hattori (1954) do not

regard the presence of thickened radial walls around the

dorsal air pores as being of generic value, since they had

observed pores with both thickened and thin radial walls

on the same plant. All the air pores in the Nyika speci-

men had thickenings, not just on the radial walls but part-

ly or entirely covering the cells surrounding them, some-

what like those in Athalamia pulcherrima, as illustrated

by Vanden Berghen (1965). When stained with periodic

acid-Schiff’s (PAS) reaction (Jensen 1962), the thicken-

ings became intensely pink, much more so than the rest

of the tissues.

> \

A \

V

B

FIGURE 9. — Sauteria nyikaensis, Koekemoer 1874. LM micrographs.

A, dorsal air pore with thickenings on surrounding cells; B,

palea. A, x 500; B, x 100.

same family, the Cleveaceae. but adds that, ‘other char-

acters argue against such a classification". She does not

elaborate further.

Sauteria nyikaensis is separated from the other

species in the genus by collectively considering the fol-

lowing characters listed in Table 1 .

The genus Sauteria was first described by Nees (1838)

and named for the Austrian physician, Anton E. Sauter,

1800-1881, who also collected and studied liverworts.

If more material of Sauteria nyikaensis with ripe

sporophytes is collected, the above description will,

undoubtedly, have to be emended.

ACKNOWLEDGEMENTS

Bischler (1998) states that the genera Athalamia,

Sauteria and Peltolepis are traditionally grouped in the

I wish to sincerely thank the curator of PRE, Dr M.

Koekemoer, for collecting this specimen while on a South-

Bothalia 33.2 (2003)

171

FIGURE 10. — Locality of Sauteria nyikaensis in Malawi.

em African Botanical Diversity Network (SABONET)

Expedition to the region; also Dr David Long for advice.

The referees, the artist, Ms M. Steyn, and the typist, Ms D.

Maree, are warmly thanked for their valued contributions,

and so is Dr H.F. Glen for the Latin diagnosis.

REFERENCES

BISCHLER, H. 1998. Systematics and evolution of the genera of the

Marchantiales. Bryophytorum Bibliotheca 51: 1-201.

EDWARDS, D. & LEISTNER. O.A. 1971. A degree reference system for

citing biological records in southern Africa. Mitteilungen der

Botanischen Staatssammlung Miinchen 10: 501-509.

GAO. C., ZHANG. G.C. & CAO, T. 1981. Taxa nova bryophytorum tibeti-

canum. Acta Botanica Yunnanica 3: 389-399.

GRADSTEIN, S.R.. CHURCHILL, S.P. & SALAZAR-ALLEN, N. 2001.

Guide to the bryophytes of tropical America. Memoirs of the

New York Botanical Garden 86: 1-577.

GROLLE. R. 1976. Verzeichnis der Lebermoose Europas und benach-

barter Gebiete. Feddes Repertorium 87: 171-279.

HASSEL DE MENENDEZ. G.G. 1963. Estudio de las Anthocerotales

y Marchantiales de la Argentina. Opera Lilloana 7: 1-279.

HATTORI. S. & SHIMIZU. D. 1955. Marchantiales of Japan IV.

Journal of the Hattori Botanical Laboratory 14: 91-107.

JENSEN, W.A. 1962. Botanical histochemistry: 1 — 408. Freeman. San

Francisco & London.

KASHYAR S.R. 1929. Liverworts of the western Himalayas and the

Panjab Plains. 1: 1-129. Lahore.

MULLER, K. (Mull.Frib.) 1951-1958. Die Lebermoose Europas. Dr

L. Rabenhorst's Kryptogamen-Flora 6, edn. 3: 368-382.

NEES AB ESENBECK. C.G. 1838. Naturgeschichte der europaischen

Lebermoose 4: 1-540.

SCHUSTER. R.M. 1992. The Hepaticae and Anthocerotae of North

America 6: 1-937. Field Museum of Natural History. Chicago.

SHIMIZU, D. & HATTORI. S. 1954. Marchantiales of Japan. III.

Journal of the Hattori Botanical Laboratory 12: 53-75.

VANDEN BERGHEN, C. 1965. Hepatiques recoltees par le Dr J.-J.

Symoens dans la region peri-Tanganyikaises. Bulletin de la

Societe Royale de Botanique de Belgique 98: 129-174.

S.M. PEROLD*

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

MS. received: 2003-07-01.

Bothalia 33,2: 173-193 (2003)

Inflorescences of Cliff ortia L. (Rosaceae) and related vegetative branch-

ing patterns

A.C. FELLINGHAM* and H.P. LINDER**

Keywords: branching, Cliffortia L., dichogamy, dioecy, herkogamy, inflorescence, monoecy, morphological plasticity, sex change

ABSTRACT

The inflorescence construction of eight species, representative of the types found in the 119 species of the rosaceous

genus Cliffortia L. is described, based on stereo microscopic examination of fresh and dried specimens, combined with exten-

sive field observations. In its simplest form the inflorescence is a reduced short shoot, bearing a lateral ebracteate flower and

a potentially viable apical bud. Variations in the basic structure can be in the number of flowers, the mix of the sexes of the

flowers and the number and type of short shoots as primary, secondary and tertiary axes. A high incidence of structural plas-

ticity of the inflorescence occurs. This can be either throughout the development of the inflorescence or only at the onset of

the vegetative stage. These changes occur in the short shoot(s) constituting the axes of the inflorescence, causing either an

increase in the length of the intemodes, apical proliferation of the axes or a combination of these two effects. A specific com-

bination of changes is linked to a specific inflorescence type. The vegetative elements of the inflorescence thus modified, are

retained as an integral part of the vegetative branching system, with extensive influence on the branching pattern. This can

also result in the predominance of one sex over the other over time, so that an individual, initially of the one sex, can become

one of the opposite sex by the end of the season. Erroneous interpretation of a single point in the process of sex change as if

it is a permanent state of sexuality, led to the prevalent acceptance of dioecy as the norm for the genus. Monoecy with

dichogamy (or herkogamy at inflorescence level) was observed in this genus, as in many other wind-pollinated taxa.

CONTENTS

Abstract 173

Introduction 173

Materials and methods 174

Results 175

Cliffortia

1 . niveniodes Fellingham 175

2. crenata L.f 176

3. ruscifolia L 178

4. heterophylla Weim 179

5. odorata L.f 182

6. arborea Marloth 182

7. dichotoma Fellingham 184

8. conifera E.G.H.Oliv. & Fellingham 185

Discussion 186

Basic inflorescence construction 187

Wind pollination 191

Impact on vegetative branching 191

Specimens examined 191

Acknowledgements 192

References 192

INTRODUCTION

Cliffortia L. includes 119 species of woody evergreen

plants. The genus was last revised by Weimarck in 1934,

and most of the publications since then have dealt with

the description of occasional new species (Weimarck

1940, 1946, 1953, 1959; Oliver & Fellingham 1991,

1994; Fellingham 1993a, b, 1994, 1995), or developed

the taxonomy to incorporate new species (Weimarck

1946, 1948). Koutnik (1987) listed the genus as being the

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

7735 Claremont, Cape Town. Present address: 23 Moreson Avenue,

Valmary Park, 7550 Durbanville, South Africa.

** Institute for Systematic Botany, University of Zurich. Zollikestrasse

107. CH 8008 Zurich, Switzerland.

MS. received: 2002-05-08.

largest genus of wind-pollinated plants in the Cape flora,

but added the comment that many of these species might

be apomictic. This has not yet been corroborated. Oliver

& Fellingham (1994), in a detailed discussion of the

inflorescence construction and phenology of section

Arborea in Cliffortia hinted at the biological and mor-

phological complexity found in the genus.

Cliffortia is a typical member of the temperate Cape

Flora, and with ±114 species in the southwestern tip of

South Africa (Fellingham 2000). It is one of the 13 gen-

era in the remarkably rich Cape Floral Region (CFR)

with more than 100 species (Goldblatt & Manning,

2000). Although the genus is found from sea level to the

highest mountain summits in the CFR, further north it is

restricted to high altitude areas. At least eight species are

found in the Drakensberg of Lesotho and KwaZulu-Natal

(Jacot Guillarmod 1971; Hilliard & Burtt 1987). Only

three species occur in Zimbabwe and Malawi (Mendes

1978), one of which reaches north to the East African

highlands (Graham 1960).

Cliffortia is usually included in section Sanguisor-

beae , on account of the reduced carpel numbers, pre-

dominantly uni-ovulate ovaries and missing petals (Mel-

chior 1964; Takhtajan 1997). The entire section includ-

ing Cliffortia is wind pollinated and has, therefore, uni-

sexual and obscure flowers with stigmas or stamens as

their most prominent parts, and the flowers are often var-

iously aggregated into dense spikes or heads. Eriksson et

al. (1998) found the elements of the section Sangui-

sorbae to have strong links with each other and not sim-

ply an artifact of convergent evolution. A phylogenetic

analysis of this section, based on ITS sequence data

(Helfgott et al. 2000), indicated that some species of a

paraphyletic genus, Sanguisorba (a northern hemisphere

genus), are the closest relatives of Cliffortia. These in

turn are related to Acaena and Polylepis, both, like

Cliffortia , southern genera. The other African wind-pol-

174

Bothalia 33,2 (2003)

linated genera, Hagenia and Leucosidea, are sisters to

each other, but are rather more distantly related to

Cliffortia. Furthermore, they differ morphologically by

the presence of petals. Morphological data are thus con-

sistent with the classifications proposed by Melchior

(1964) and Takfrtajan (1997).

Cliffortia is remarkably uniform in flower structure,

but diverse in leaf morphology. Aspalathus (Fabaceae)

and Anthospermum (Rubiaceae) exhibit morphological

convergence in vegetative parts with Cliffortia. Nineteen

species of Aspalathus are indistinguishable in the sterile

state from as many species of Cliffortia, except for the

absence of leaf sheaths and stipules (Dahlgren 1971).

Several species of Anthospermum, sharing habitats with

Cliffortia species, are indistinguishable from these spe-

cies, but for their opposite leaves. The similarity extends

beyond the presence of leaf sheaths and stipules to include

flower morphology, both having small unisexual flowers,

prominent calyx lobes and remarkably similar fruits.

To date, the inflorescences of Cliffortia have received

scant attention. Weimarck's (1934) rather superficial

descriptions of the inflorescences indicate that he misun-

derstood the basic construction of the inflorescences (not

difficult, if working with limited material, especially if

most of it is herbarium material, which has to be treated

with great care). This lack of understanding of the inflores-

cence construction affected the interpretation of the pat-

terns of sexuality in Cliffortia. Often only overtly male

or overtly female plants are found in any population at

any specific time, leading to suggestions that at least

some of the species may be apomictic (Koutnik 1987),

and the general impression that most species are dioe-

cious. Weimarck (1934) noted that ‘the distribution of

male and female flowers has been shown to be very

irregular so that, as far as is known, not less than 47

species out of 78 have been found at least capable of

being monoecious’. Nevertheless, he still held the notion

of dioecy as the rule for the genus.

These findings established the existence of a wide

range of inflorescence types in Cliffortia, varying from

solitary flowers to fascicled flowers and highly con-

densed cones containing large numbers of flowers. In

this paper we wish to address several questions: (a) is

there a common basic construction discemable in the dif-

ferent inflorescences of Cliffortia ; (b) can these common

units be used to compare different types of inflorescence

construction; (c) how can this account for the apparent

dioecy in the genus; and (d) how does the inflorescence

type influence the branching pattern?

Terminology

The terminology we use is largely derived from

Weberling (1983, 1989). Inflorescences refer to ‘the

shoot system which serves for the formation of flowers

and which is modified accordingly’ (Troll in Weberling

1989). On a single plant or branch system, separate inflores-

cences are separated by sets of vegetative branches.

Short shoots [brachyblasts in Weimarck (1934)] are

defined here by their short internodes. A short shoot is

not, however, always a permanent element of the plant

morphology but can disappear through abortion, or trans-

formation into a long shoot. Neither is it always a patent-

ly visible structure but can be completely and perma-

nently hidden in the axil of its subtending leaf. The

matrix for the flower-bearing short shoots is the long

shoot, which is a vegetative shoot with long intemodes.

Ordinary vegetative leaves on the long shoots subtend

these fertile short shoots. The main involvement of the

short shoot is with flower bearing, with the flowers either

bracteate or ebracteate but never subtended by ordinary

vegetative leaves. Vegetative short shoots do occur, but

only rarely and then they bear ordinary vegetative leaves,

as long shoots do. A short shoot is thus a permanent or

temporary, patent or obscure shoot with short intemodes,

has the main function of flower bearing and occurs in the

axil of a vegetative leaf on a long shoot, or rarely on a

vegetative short shoot. Such short shoots are then equiv-

alent to ‘florescences’. A group of florescences aggregat-

ed on a vegetative short or long shoot would constitute a

‘synflorescence’. Where several florescences are adja-

cent, they form double racemes or a diplobotryum.

The inhibition zone lies directly below the inflores-

cence where the formation of florescences is inhibited,

and which is therefore sterile. Axillary buds in the inhi-

bition zone need to have at least the potential to make

florescences, but are prevented from doing so by the

actual inflorescence. By contrast the proliferation zone is

above the synflorescence, and occurs where the flores-

cence apex grows out and returns to vegetative growth

(Weberling 1989).

Vegetative leaves are usually borne on long shoots,

and never subtend flowers. Since floral bracts are recog-

nized as being different from vegetative leaves, the cor-

rect identification of these is important. Floral bracts are

defined as subtending flowers, but the loss of true bracts

may lead to the first leaf below a flower being vegetative,

thus making the application of the definition potentially

misleading. It therefore seems more practical (though not

typologically correct) to recognize bracts as being mor-

phologically different from vegetative leaves. This iden-

tification is important, as florescences are defined as

being subtended by vegetative leaves (rarely modified),

and as containing floral bracts.

In spite of the fact that Weberling’s (1989) new termi-

nology had been used to describe the inflorescence in C.

conifera (Oliver & Fellingham 1994), we have been

reluctant to use it in this paper. His definition of a raceme

or botrys (botryum?) as having ‘clearly developed inter-

nodes’ and ‘stalked flowers’, seems to preclude its appli-

cation to the short shoot inflorescence typical in the

genus Cliffortia. It would appear that Weberling’s earlier

(1983) definition of long shoots and short shoots in terms

of their respective function is more appropriate to the

subject of this paper.

MATERIALS AND METHODS

Observations on inflorescence structure were based

on both fresh and dried specimens, and where possible,

collections of fresh material were made repeatedly from

the same populations, over a season or more. Fertile

branches were dissected, examined and sketched with

the aid of a stereo microscope fitted with a camera luci-

Bothalia 33,2 (2003)

175

da. The dried material was softened, by boiling and soak-

ing in diluted dishwashing solution before dissection.

The majority of drawings were done using the camera

lucida, but larger, fresh specimens were drawn free-hand.

Species were selected to reflect the diversity of inflores-

cence structure in the genus. The most specialized are the

three cone-bearing species, C. conifera, C. dichotoma

and C. arborea. A highly condensed inflorescence,

though with an amazing plasticity, is found in C. odora-

ta, which, furthermore, appears to have no short shoots.

In C. heterophylla the inflorescence is clearly demarcat-

ed and initially condensed but extremely plastic and

impermanent. The multiflowered form of the short shoot

inflorescence, with bracteate flowers, is represented by

C. ruscifolia. The apparent lack of short shoots (and

flowers) in C. crenata , makes it an interesting and impor-

tant subject. C. nivenioides is an exceptional and there-

fore very interesting species for two reasons. It has a pre-

ponderance of short shoots, the majority of which are

vegetative and the rarer fertile ones appear to bear single

flowers directly in the axils of ordinary vegetative leaves.

RESULTS

1 . Cliff ortia nivenioides Fellingham

This species is known from the type locality only: a

very discrete locality on an open sunny marsh at an alti-

tude of almost 2 000 m, north of Blesberg in the Swart-

berg Mountains. The general appearance of this small

shrub is that of a compact bunch of long branches emerg-

ing from a central point. These long branches are them-

selves sparsely branched to totally unbranched. The

leaves are arranged imbricately on short shoots to form

flat, slightly elongated fans. The combined effect is that

of a species of the genus Nivenia Vent. (Iridaceae)

(Figure 1A), hence the derivation of the specific epithet.

Flowers occur on the apical short shoots only, or

rarely also on one or two lateral short shoots just below

the apex. Each flower appears to be subtended by an

ordinary vegetative leaf. These subtending leaves are

bilaterally flattened and closely arranged on the short

shoots, with the leaf sheaths imbricate, completely

obscuring the intemodes of the short shoot (Figure 1A,

C). As these subtending leaves are not modified in any

way, they are not bracts. The male flowers occur togeth-

er with the female flowers on the same short shoots, but

in the lower, and therefore older leaf axils. The young

female flowers are borne singly and totally hidden in the

axils of the leaves, and fertile plants can easily be mis-

taken as sterile. Being more mature, and having larger

calyx lobes and protruding stamens, the male flowers are

more readily visible than the totally hidden, immature

female flowers (Figure 1 A, C). This difference in develop-

mental stages between male and female flowers excludes

the possibility of self-fertilization.

The fertile short shoot in the axil of a vegetative leaf,

on a long shoot, appears to be the inflorescence (Figures

1A; 2A: il). The presence of vegetative leaves on the

short shoot, however, indicates that it is a vegetative

shoot. This is confirmed by the lengthening, in ascending

order, of the hitherto very short internodes of the short

shoot with imbricate leaves, changing it into a long shoot

with distant leaves at the time of fruiting. This change

clearly reveals the true nature of this short shoot as equiva-

lent to the vegetative long shoots supporting the inflores-

cences encountered in other species (Figures IB; 2B).

At the end of the flowering season the male flowers and

the fruits are shed and the vegetative stage (Figure 2C) is

entered. This is marked by two events. The apical bud of

the newly formed vegetative long shoot forms a new apical

short shoot (na) and new lateral short shoots (nl) develop in

the axils of the leaves which subtended the past season’s

flowers. Thus longitudinal growth occurred on the main

FIGURE 1 —Cliffortia nivenioides,

Viviers & Vlok 470 (BM, K,

MO, NBG, P, PRE). A. long

shoot bearing short shoots

with female flowers in axils

of leaves; B, fruit-bearing

long shoot, developed out of

short shoot; C, short shoot

with lower node elongated

and lower leaf axil bearing

male flower, a, female

flower; b, fruit; c, male

flower; d, visible intemodes.

All copied from Inge

Oliver’s drawings. Scale bar:

1 mm.

176

Bothalia 33,2 (2003)

FIGURE 2. — Cliffortia nivenioides : longitudinal diagrams of inflorescence and branch development, covering two fertile seasons spanning a

vegetative season. A, young fertile short shoot season with young single-flowered inflorescences; B, same as A, matured: a long shoot with

mature flowers; C, same as B, vegetative stage: new short shoots developed out of remaining inflorescence; D, same as C, second season:

the apical and one lateral short shoot bearing flowers; E, same as D, matured: apical and one lateral long shoot (lateral long shoot to become

vegetative branch). O, subtending leaf on main branch; 1-6, subtending leaves of first season’s inflorescence; 1' — 6', subtending leaves of

second season’s inflorescences; da, dormant apical buds; f, female flower; fl, flowering long shoot; fs, fertile short shoot; il, fertile short

shoot in A becoming long shoot in B; i2, fertile short shoot in D becoming long shoot in E; m, male flower; na, new apical short shoot; nl,

new lateral short shoot; ss, sterile short shoot; yf, young flower. Broken lines: expanded axes of short shoots.

stem and a new branch is formed. The short shoots thus

formed, appear to replace the single flowers of the recent

flowering season, but are in fact, proliferation shoots of

minute, single-flowered inflorescences each with an

ebracteate flower. Most of these newly formed, short

shoots remain vegetative and do not undergo any further

vegetative growth. Their apical buds have become dor-

mant. Only a small number near the apex of the plant

develop further, viz. the new apical short shoot (na) and

the uppermost new lateral short shoot (nl) and become the

new flowering short shoots of the second flowering season

(Figure 2D: i2). Their imbricate leaves become the sub-

tending leaves for the new single-flowered inflorescences

(Figure 2C-E). As in the previous fertile stage (Figure 2A:

fs (il), B: fl), these fertile short shoot(s) change from short

shoots with imbricate leaves, into long shoots with distant

leaves as they mature (Figure 2D: i2, E: i2).

The number and positions of the long shoots thus

formed, are determined by the number and positions of the

short shoots involved in flower bearing. With only the api-

cal short shoot and one lateral involved, the result will be

one apical long shoot and one side branch (lateral long

shoot) (Figure 2D, E). As before, the next stage of develop-

ment is the shedding of the flowers leaving the new long

shoots with their now no longer imbricate leaves (1-61),

ready to subtend a new generation of short shoots, some of

which are destined to become future branches.

Branching occurs only where lateral long shoots

develop. The sparse vegetative branching pattern in this

species, is thus directly linked to the structure of the fer-

tile shoot (synflorescence) of the previous fertile season,

limiting the development of long shoots. Where only the

new apical short shoot (na) develops into the new fertile

shoot (i2), no side branch will develop, so that solely

longitudinal growth takes place. It is only when one (or

more) of the lateral short shoots (nl) are also flower bear-

ing, that one (or more) lateral long shoots develop and

branching occurs (Figure 2D, E).

2. Cliffortia crenata L.f.

This species occurs on mountain slopes and plateaux

at altitudes ranging from 1 300 m in the northern, drier

areas of the CFR, to 300 m in the Montagu area. The

plants appear to be totally without short shoots, being

virgate, lanky shrubs up to 2 m tall, very sparsely branch-

ed in the upper parts only. At no time are any flowers

visible. The two large leaflets are attached to the stem at

an acute angle, completely enveloping it, so that, in order

to examine the leaf axils for flowers, at least one of the

leaflets has to be folded back or removed.

An immature (primordial) inflorescence (Figures 3G;

4 A) consists of a pair of immature flowers, one slightly

above the other, attached to a swollen basal structure,

with an apical bud between the two flowers. The flowers

are subtended by small but unmistakably leafy elements.

The size of the immature flowers relative to that of the

inflorescence axis, facilitates the interpretation of the

structures. It is quite clear that the flowers are borne on

the swollen basal structure and not adjacent to it. In terms

Bothalia 33,2 (2003)

177

FIGURE 3. — Cliffortia crenata, Oliver 10387 (NBG). A, portion of

main branch with condensed inflorescence: B, proliferating

inflorescence (branchlet from axil of a vegetative leaf), bearing

one male flower and three fruits; C-E, subtending leaves, in

ascending order, on branchlet B; F, apical growth point from

axil of fourth leaf on branchlet B; G, primordial inflorescence

(short shoot from axil of vegetative leaf near apex of a main

branch), showing a basal contracted axis bearing two lateral

flowers and a central apical growth point. Scale bars: 1 mm.

of the elements of a simple inflorescence consisting of a

short shoot, the swollen basal structure is the inflores-

cence axis bearing two lateral, alternate flowers with the

apical portion of the axis continuing beyond. In an inflores-

cence, condensed to such a degree as this, the subtending

leafy elements can be expected to be bracts rather than

vegetative leaves.

A primordial inflorescence can develop into one of

two types of mature inflorescences: a condensed type or

a proliferating type. The condensed type contains a sin-

gle flower or a cluster of mature flowers, on a much con-

densed inflorescence axis without any apical prolifera-

tion. In a typical instance, three flowers, each laterally in

the axil of a bract, are found in a cluster: the two lower

female and upper male. These three mature flowers

obscure the inflorescence axis and its ancillary vegetative

parts viz. leaves or bracts. The two female flowers are

much more advanced in development than the male. The

one female is a fully developed fruit without calyx lobes,

whereas the other is a younger fruit with the calyx still in

place. The male flower is still in the bud stage and there-

fore, much younger than the female flowers (Figures 3A;

4B). In this type of inflorescence there is evidence of the

presence of an apical bud but no proliferation takes

place. The diminutive size of the subtending leaves sug-

gests that these would not develop into vegetative leaves

but rather into bracts, subtending flowers of both sexes.

The proliferating type of mature inflorescence pre-

sents not only clusters of fruit and flowers in a more open

arrangement than those in the condensed type, but has a

viable apical bud which proliferates into a young vegeta-

tive shoot, which eventually develops into a branch. The

structure that originates in the leaf axil is the shoot and

the flowers are borne laterally on the basal part of the

shoot and not directly in the axil (Figures 3B; 4C). The

lower two flowers are female and close to each other,

with their small imbricate subtending modified leaves

completely covering the very short intemodes. The next

slightly more distant flower is female and followed by

younger male flower(s). The internodes separating the

flowers become increasingly longer upward, but remain

much shorter than those higher up on the vegetative part

of the shoot (Figure 3B ). The leaves subtending the basal

female elements (fruits and/or flowers), show progres-

sive development from the base upwards. The lowest are

merely scale-like structures, consisting mainly of a nar-

row sheath without any stipules and with only the vestige

of a blade attached directly to the sheath, without articu-

lation (Figure 3C). Above these are leaves in progressive

stages of development from very small and bract-like

with just the indication of stipules and a single articulat-

ing leaf blade (Figure 3D), to almost normal, small veg-

etative leaves with two leaflets (Figure 3E). These small

leaves subtend the male flower in Figure 3B. Above the

male flowers, the shoot becomes a normal long shoot

with vegetative leaves. In occasional leaf axils near the

apices of these vegetative shoots, small buds which

could develop into secondary primordial inflorescences,

are found (Figures 3F; 4C: pi).

The proliferating inflorescence consists of a basal

flower-bearing short shoot (ssh) in which the apical bud

proliferates into a distal long shoot (lsh) which will

lengthen to extend beyond the subtending leaf on the

main stem to form the new branch (Figure 4C). This type

of inflorescence occurs less frequently than the contract-

ed type and usually near the apices of the branches. These

then, are the only inflorescences that give rise to vegeta-

tive branches, as the formation of branches depends

entirely on the occurrence of inflorescences with subse-

quent distal long shoot development. The sparseness of

the branching pattern, resulting in the virgate growth form

of this species, can thus be directly related to the small

number of inflorescences that undergo the full range of

changes. The positions of the branches are determined by

the position of the proliferating inflorescences which is

usually near the apices of existing branches.

A

A

A*

i

4

ssh

vl

B

FIGURE 4. — Cliffortia crenata : lon-

gitudinal diagrams of inflo-

rescences. A, primordial,

with immature flowers; B,

mature contracted, with

mature male and female flow-

ers; C, proliferating, with

mature male and female

flowers, apical proliferating

branch/long shoot (lsh) with

secondary primordial inflo-

rescence (pi), bt, bract; f,

female flower; m, male

flower; ssh, short shoot; vl,

vegetative leaf. Broken lines:

expanded axes of short shoots.

178

Bothalia 33,2 (2003)

3. Cliffortia ruscifolia L.

This is a widespread and probably the most familiar

species of Cliffortia, occurring in well-drained habitats

over a wide range of altitudes. It occurs in a great variety

of vegetative forms from low, sparse, sprawling plants

with yellowish green leaves, to erect bushy plants up to

1.5 m tall and with grey-green leaves, from the same

locality. The size and shape of the vegetative leaves are

also very variable but always unifoliolate and pungent.

The inflorescence is generally bracteose and multi-

flowered, rarely single-flowered with the flower appar-

ently subtended by an ordinary vegetative leaf. The latter

type could be mistaken for the immature form of the

FIGURE 5. — Cliffortia ruscifolia. A,

well-developed inflorescence

exhibiting male and female

flowers subtended by triden-

tate bracts, overtopped by

proliferating shoot with lin-

ear vegetative leaves; B, C,

vegetative leaves from just

below inflorescence in A,

each successively shorter and

broader than the one below;

D-F, bracts subtending flow-

ers in inflorescence A; G-K.

leaves of proliferation zone

in A; L, growth point of pro-

liferating shoot; M. inflores-

cence with male and female

flowers subtended by triden-

tate bracts and vegetative

leaves below inflorescence,

shed to form inhibition zone

(i); N, rare single-flowered

inflorescence (one fruit) sub-

tended by a shortened vege-

tative leaf (no bract); i, inhi-

bition zone; j, proliferation

zone. A-L, Oliver 10567

(NBG); M, Oliver 10569

(NBG); N, Oliver 10574

(NBG). Scale bars: 1 mm.

inflorescence if it was not for the fact that the female ele-

ment in it was a mature fruit. In essence, the inflores-

cence is a short shoot, occurring in the axil of an ordinary

vegetative leaf on a long shoot. Male and female flowers

are borne singly in the axils of bracts, in a single inflores-

cence, usually with the female flowers below the males

(Figure 5 A, M).

Two distinct zones can be distinguished in the fertile

short shoot. The lower zone consists of ± five vegetative

leaves, which graduate in form and size from long and

narrow (lowest) to much shorter and wider in the high-

est. All of these leaves are typical vegetative leaves, plain

with slightly hairy edges and nerves and sharp apical

spinelets. With the leaves diminishing in length upwards.

Bothalia 33,2 (2003)

179

FIGURE 6. — Cliffortia ruscifolia : longitudinal diagrams of inflorescence development and related branches. A, young inflorescence in axil of veg-

etative leaf on long shoot; B. mature inflorescence with proliferation of apical bud; C, vegetative stage: new branch consisting of remains

of the short shoot (rs) plus new long shoot (Is); D. new inflorescences in axils of vegetative leaves on long shoot, constituting potential

new branches developing as illustrated in B and C. az, apical proliferation zone; br, bract: f, female flower; fp, fertile part of short shoot;

ini, inflorescence of first season; in2, inflorescences of second season; iz, inhibition zone; Is, long shoot; m, male flower; ns, new short

shoot; riz, remains of inhibition zone; rs, remains of short shoot; vl, vegetative leaf; 1-5, vegetative leaves on long shoot becoming sub-

tending leaves for inflorescences of second season. Broken lines: expanded axes of short shoots.

this zone seems to fit the description for Troll's ‘field of

inhibition' (Weberling 1989). For such a small area of

inhibition as is encountered here, the term inhibition

zone seems appropriate. As seen in Figure 5M, the leaves

of this area are shed to reveal the fertile part of the

inflorescence, a zone of imbricate bracts, each subtend-

ing a flower. The bracts are much shorter and more hairy

than the leaves of the inhibition zone, and dentate to

trilobate. In some instances the apical bracts are imma-

ture. All multi-flowered inflorescences contain male and

female flowers in equal proportions (Figures 5A-F, M; 6A).

The apical buds of the mature inflorescence prolifer-

ate to form comas of young vegetative leaves, some with

minute buds in their axils, on a developing long shoot.

The appearance of this vegetative long shoot marks the

onset of the new vegetative stage. This phenomenon can

be defined in terms of Weberling’s (1989) concept of

‘late proliferation’, except that he sets the prerequisite of

a frondose inflorescence for the use of this term. It does

otherwise fit the requirement of marking the return of the

‘inflorescence apex to the vegetative condition’. In spite

of the bracteose character of the inflorescence, we pro-

pose the use of the term proliferation zone for the early

stage of development of the long shoot (Figures 5A, G-

K: 6B: az, 6C: Is).

With the appearance of the proliferation zone the veg-

etative stage is entered. At the same time, or in some

instances, before this event, the leaves of the inhibition

zone are shed, followed by the shedding of the flowers

and their subtending bracts (Figures 5N; 6B, C). The

short shoots are thus left bare with the short intemodes

(which do not elongate) permanently visible. The prolif-

eration zones are thus the entire source of vegetative

growth. These then develop into long shoots with proxi-

mal zones of bare short intemodes, as the only indica-

tions of the sites of the old inflorescences. Though these

zones are a permanent feature of the basal (proximal)

portions of all branches thus derived, they do not con-

tribute any further to their development (Figure 6B-D).

As the fertile stage is characterized by the development

of short shoots, the vegetative stage is characterized by

the development of long shoots proliferating out of the

short shoots (Figure 6B. C). Potentially every short shoot

could develop into a vegetative branch. Furthermore, all

the leaves on a long shoot could become subtending

leaves for the next fertile season's short shoots (Figure

6D). However, usually only a limited number of leaves

develop short shoots in their axils. Of these short shoots

only a limited number give rise to long shoots.

4. Cliffortia heterophylla Weim.

This species is locally common in sunny situations

near streams in the Betty's Bay area. In the vegetative

stage, the plants have the general appearance of saplings

up to 3 m tall, with willow-like leaves and some sec-

ondary thickening of the main stems. In spring a single

conspicuous cylindrical inflorescence develops on the

apex of the main stem (Figure 7 A). Plants under stress,

as those from which the apical inflorescences have been

removed or those having their main stems constricted by

twiners, may develop multiple lateral inflorescences.

Inflorescences are initially female, then progress

through a bisexual phase to the male stage. Only minimal

traces of the fertile stage, in the form of amplexicaul

180

Bothalia 33,2 (2003)

FIGURE 7. — CUffortia heterophyl-

la, Fellingham 1640 (NBG).

A, female inflorescence on

apex of main stem; B, female

flower in situ on node (bracts

removed); a, bracteoles; b, ca-

lyx lobes; h, intemode. Scale

bar: 1 mm.

sheaths around the main axis of the plant, are left in the

vegetative stage, after the shedding of the fugacious

bracts (Figure 8: rb). With the secondary thickening of

the main stem, these sheaths also disappear as they are

stretched and broken.

In the bisexual phase, the inflorescence demonstrates

all the different developmental stages in the form of rec-

ognizable zones, which are dealt with here in the chrono-

logical order of development, viz. female zone, male

zone and the two vegetative zones: the inhibition zone

and the proliferation zone. The longitudinal diagram

(Figure 8) depicts all of these zones as one entity. It is,

however, rare to find specimens showing all the stages:

usually the proliferation of the apical bud takes place

only after the shedding of all sexual elements.

The cylindrical, apical inflorescence is in essence a

short shoot up to 240 mm long, the uppermost internodes

about 1 mm long, gradually increasing in length to about

7 mm basally. All internodes of the short shoot are totally

obscured by the lanceolate-acuminate bracts, which are

up to 60 x 15 mm. In the apical part of the inflorescences,

the female flowers occur singly on very short asymmetri-

cal structures (Figure 7B). In the older (lower) part of the

inflorescence, up to six female flowers are borne on lat-

eral short shoots, totally hidden by the imbricate bracts.

(Figure 8: Is). The flowers are lateral on these short

shoots, and subtended by leafy elements that are much

smaller than, and different in shape to the normal vegeta-

tive leaves, and therefore, bracts (bt). The short asymmet-

rical structures bearing the single flowers higher up on the

same inflorescence, can thus be interpreted as primordial

short shoots. Such a short shoot has a fully viable apical

bud while only the lowest node is mature enough to sus-

tain a fully developed flower, without the bract being in

evidence yet. Already at this stage, the inflorescence can

be described as a diplobotryum or double raceme, with a

central short shoot bearing lateral short shoots in the axils

of its bracts; the lateral short shoots bearing (female)

flowers in the axils of their bracts.

Two events mark the onset of the development of the

male zone. The hitherto short internodes of the apical

Bothalia 33,2 (2003)

181

FIGURE 8. — Cliffortici heterophylla :

longitudinal diagram of in-

florescence. as, apical short

shoot; az, apical zone; br,

branch; bt, bract; da, dormant

apical bud; dl, developing

main axis long shoot; f,

female flower; fz, female

zone; ib, imbricate bract; iz,

inhibition zone; 11, lateral

long shoot; Is, lateral short

shoot; m, male single-flow-

ered inflorescence; mz, male

zone; rb, remains of imbri-

cate bract; rls, remains of

lateral short shoot; va, viable

apical bud; vbr, vegetative

branch; vz, vegetative zone.

Broken lines: expanded axes

of short shoots.

short shoot of the main stem start lengthening, the lower

ones first and then the others in sequence upwards. The

result is the separation of the previously imbricate bracts

to reveal the intemodes, thus changing the short shoot

into a long shoot (Figure 8: mz). At the same time, the

apical buds on all lateral short shoots (Is) give rise to

long shoots (11), in sequence from below upwards. These

lateral long shoots lengthen and male flowers (m) devel-

op on their distal parts, whereas the female flowers (f) on

the proximal part, (short shoot), mature into fruits before

dropping off, starting with the lowest, then involving

those higher up, again in ascending order (Figure 8: mz).

The formation of the vegetative zone is initiated by

the shedding of the male flowers as well as any fruits that

might still be present. This leaves the remains of the lat-

eral short shoot (rls) in the form of a region of short

intemodes, interspersed with ridged nodes, on the proxi-

mal part of the branch (br). This is accompanied by the

loss of the bracts (ib) on the main stem, leaving their

remains (b) in the form of the leaf sheaths which are

eventually destroyed by secondary thickening of the

main stem. Thus the vegetative zone (vz) comes into

being, and develops upwards as the fruit and flowers

mature and are shed. With the shedding of the last flower,

the vegetative zone spreads to envelop the whole of the

inflorescence, reducing it to a framework of newly formed

vegetative branches on an apical extention of the original

main stem. With this, the vegetative stage has been

entered (Figure 8: vz).

The vegetative stage is completed by the development

of the apical proliferation zone (az). This is achieved by

the proliferation of the apical bud of the main inflores-

cence axil, which is just a highly condensed portion of

the main stem, giving rise to an apical long shoot with

vegetative leaves. In the axils of the vegetative leaves of

this zone, purely vegetative branches (vbr) develop.

These consist of long shoots only, without any remains

of lateral short shoots on them as in the case of the

branches originating in the inflorescence and are shorter

than the side branches originating in the inflorescence (Figure

182

Bothalia 33,2 (2003)

8: az). The apical proliferation zone thus gives rise to the

only purely vegetative growth of the plant. Proliferation

of the apical bud can take place early in some instances,

producing a coma above the apical cylindrical female

zone (Figure 8: az, fz). Normally this occurs only after

the vegetative zone is fully developed and the inflores-

cence has disappeared completely.

The vegetative stage ends when the apical bud pro-

duces the new apical cylinder of imbricate bracts, mark-

ing the onset of the following fertile stage. This places

the old apical proliferation zone directly below the new

inflorescence. The youngest (apical) branches of the pro-

liferation zone, will still be relatively short when the new

inflorescence starts developing. This can be interpreted

as the old apical proliferation zone becoming the inhibi-

tion zone of the new inflorescence (Figure 8: iz).

5. Cliff ortia odorata L.f.

Dense mats of this species occur on the banks of

rivers and dams in the southwestern, southern and south-

eastern coastal districts, usually at low altitudes but also

up to 1 000 m on Table Mountain. The upright shrub is

up to 2 m tall with the main stems slightly zig-zagging at

the nodes, where the branches originate. This irregulari-

ty develops as a result of the displacement of the stem by

the enlarging highly condensed inflorescences and the

subsequent development of the branches in the axils of

the amplexicaul leaves at these nodes.

The young inflorescence first appears as a single

flower in the axil of a young leaf near the tip of a main

stem or branch. This single flower is soon joined on the

broad flower base, which is the primordial inflorescence

axis, by small clusters of much younger flowers with

their bracts and bracteoles developing between the first

flower and its bracteoles, thus causing the first flower to

appear ebracteate (Figure 9A). These inflorescences are

subtended by vegetative leaves, and develop into clusters

of flowers of both sexes. Each cluster is composed of

several subclusters, each subtended by a bract-like, high-

ly modified and extended membranous leaf sheath with-

out a leaf blade. The individual flowers are interspersed

with membranous bracts in varying states of develop-

ment, with an occasional ebracteate flower. The subclus-

ters are implanted onto a flattened disc, the condensed

inflorescence axis, by means of short stalks. Removal of

the subclusters reveals the discreet implantation sites left

on the disc. In the centre of an occasional inflorescence

in this stage of development, a small raised area in the

shape of a pyramid occurs. This pyramid is the apical

bud of the condensed inflorescence axis, starting to pro-

liferate (Figures 9E; 10A).

In a significant proportion of inflorescences, further

changes follow, culminating in production of a central

branch in each inflorescence. In some cases, the central

pyramidal section of the initially much-condensed inflores-

cence axis, develops into a very short vegetative shoot

with one or two of the subclusters attached to it, thus sep-

arating these subclusters somewhat from the rest of the

inflorescence. At the same lime, vegetative leaf blades

develop on the sheaths subtending the subclusters, thus

marking the nodes in the condensed inflorescence axis

more clearly (Figures 9C; 10B). This interspersing of

partial inflorescences with vegetative shoots, results in

the formation of a branch complete with small lateral

inflorescences. (Figures 9B; 10C).

More often though, the inflorescence remains tightly

compact when the apical bud proliferates above the sub-

clusters, generating a new branch. The amplexicaul

sheath of the vegetative leaf on the main stem, subtending

this compact inflorescence with its central branch,

becomes much extended (Figure 9D). In large mature

inflorescences with the central branch well developed, the

implantation sites of the subclusters are so numerous as to

be closely packed around the base of the branch, in a dis-

tinctly spiraling pattern (Figure 9F). This compact inflo-

rescence with the apical proliferation, giving rise to a

branch, represents the basic structure of the typical inflores-

cence. It exhibits a clear distinction between the produc-

tion of short shoots, or as in this species, a system of short

shoots, associated with the fertile stage and the long shoot

formation of the vegetative stage (Figure 10A).

With two highly modified short shoots viz. the primary

inflorescence axis as a flattened disc and the secondary axis

as the stalk to a subcluster of flowers, the inflorescence can

be defined as a double raceme or a diplobotryum. The

arrangement of these subclusters (diplobotrya) within the

total inflorescence is, however, not a fixed state. As seen

above, in the instance of vegetative shoots and leaves

developing in the inflorescence, a certain degree of plastic-

ity exists, allowing the initially compact inflorescence to

become more open (Figures 9C; 10B).

6. Cliffortia arborea Marloth

For a considerable time this species was the only

known ‘tree’ in the genus. Marloth (1905) noted the height

as occasionally up to 1 0 m. There was considerable vari-

ability in growth form and sexuality in the studied popu-

lation in the Nuweveld Mountains above Beaufort West.

On terraces against rock ledges about 100 m away from

the sheer cliffs, the growth form was that of upright trees

up to 4 m tall, with main trunks up to 150 mm in diame-

ter and with reddish brown, flaking bark, and mostly

female cones. On the edge of the cliffs, the population

was smaller and the growth form more stunted and com-

pact, and the plants were almost totally covered in male

flowers from just below the occasional apical female

cones to the lowest branches brushing the ground.

Female cones appear as apical thickenings on the ends

of main branches. On older plants, the female inflores-

cence cones can be clearly seen to occur in series, from

the older ones below to the one of the current season

above (Figure 11 A). Occasionally cones are also found

on side branches originating from older cones lower

down on main branches (Figure 12A).

The female inflorescence forms an oblong cone-like

polytelic synflorescence (Figure 12A, B: SN) consisting

of an aggregated and spirally arranged condensed lateral

double racemes (homothetic diplobotrya). In the young

(current season’s) cone, each diplobotryum culminates

in a coma of vegetative leaves forming a star-shaped

rosette. Together, the rosettes cover the cone in a shroud

Bothalia 33,2 (2003)

183

FIGURE 9. — Cliffortia odorata, Fellingharn 1660 (NBG). A, young, apparently single-flowered, inflorescence with central female flower and two

lateral clusters of immature flowers in axils of its bracteoles; B, main stem with three well-developed vegetative leaves, each subtending a dis-

crete inflorescence containing flowers of both sexes; C. proximal side of inflorescence with two pairs of subclusters separated by short vege-

tative shoot; D. main stem with compact inflorescence of three clusters (flowers and bracts removed, leaving subtending leaf sheaths) envelop-

ing base of side branch originating within uppermost leaf sheath; E, semischematic drawing of young condensed inflorescence axis in leaf axil

on main stem with three implantation sites of flower clusters and central raised pyramid marking onset of branch proliferation; F, mature form

of E with well-developed central branch surrounded by numerous spirally arranged implantation sites of flower clusters. Scale bars: 1 mm.

of neatly arranged stars, from which the common name

for this species viz. sterboom (= star tree) is derived.

On the main axis of the cone, trifoliolate primary cone

leaves with broadened curved sheaths with or without

pungent stipules, are spirally arranged (Figures 1 1 B ;

12B; pci). Each primary cone leaf subtends and sur-

rounds a diplobotryum (DB). A diplobotryum consists of

a number of co-florescences (CoF) on a secondary axis,

which is basally much condensed but apically prolifer-

ates into a vegetative shoot, bearing the star-like coma of

vegetative leaves. The co-florescences (botrya) are

arranged on the basal condensed part of the axis and sur-

rounded by involucre-like groups of unifoliolate sec-

ondary cone leaves (scl), which form the firm part of the

cone underneath the shroud of comas (Figures 1 1 C; 12B:

184

Bothalia 33,2 (2003)

FIGURE 10. — Cliffortia odorata. longitudinal diagrams of inflores-

cences in different stages of development. A, compact mature

inflorescence with apical branch; B, open inflorescence sepa-

rated into two parts by central vegetative short shoot; C, later

developmental stage with three lateral inflorescences, br,

branch; cvs, central vegetative short shoot; li, lateral inflores-

cence; Is, leaf sheath subtending subcluster; ms, main stem; sc,

subcluster; si, subtending leaf. Broken lines: considerably

expanded axes of short shoots.

scl). The botrya (co-florescences) are highly condensed

with the axes not more than a slightly raised flattened

area bearing a number of tiny flowers in a capitulum-like

arrangement. The flowers are almost completely hidden

by the secondary cone leaves with not much more than

the strap-shaped styles and stigmas showing.

Mature cones proliferate apically into a vegetative

continuation of the main axis, which will give rise to the

following season’s cone. Furthermore, the star-like

comas of vegetative leaves marking the apices of the

diplobotrya, proliferate into vegetative side branches.

These can bear female cones, male flowers or eventually

thicken and continue the vegetative development of the

plant (Figures 1 1 A; 12A: VB). Old cones are retained on

main stems for a number of years but eventually disinte-

grate. This happens in stages. Firstly, after three or four

years, abscission of the secondary cone leaves takes

place at the top of the sheaths, leaving the woody bases

on the main branches for several years (Figure 11 A).

Eventually the clusters of secondary leaf bases get sepa-

rated from each other, as the main branches bearing them

undergo longitudinal growth and secondary thickening.

By that time, the leaf bases will have been reduced to a

few scattered scales on the main stem and will be hardly

recognizable as the remains of the original cone.

Male inflorescences are simple clusters of a few

ebracteate flowers in the axils of vegetative leaves on lat-

eral branches. These are borne below the female cones

but on the same main branches (e.g. Acocks 18625,

Fellingham 1625, Marloth 9730). Occasionally male

flowers also occur on the thin side branches originating

from older female cones ( Fellingham 1625), but always

lower down on the plant than the female inflorescences

of the same season. While distinctly monoecious, this

species never has the male and female flowers occurring

in the same inflorescence. Where male flowers do occur

on the thin vegetative side branches of the occasional

older female cones, as seen in Fellingham 1625 , they are

more than a year younger than these female cones and

borne on vegetative shoots, and not on the female cones

themselves.

7. Cliffortia dichotoma Fellingham

This new species of Cliffortia section Arboreae, from

near Nieuwoudtville, Namaqualand, is closely related to C.

arborea. It occurs on rocky ledges, on slight northern and

southern slopes on the Oorlogskloof escarpment, south of

the Oorlogskloof Nature Reserve. The whole population is

old and moribund, with part of every plant dead and dry.

No seedlings have been found for at least 25 years.

The plants appear more tree-like than any other

species in the genus and are up to 5 m tall. The main

trunks are buttressed and therefore irregular in shape, up

to 500 x 300 mm in cross section, and bare for the lower

0.5 m. Branching is dichotomous from below a cone,

with the new season’s cones at the tips of the new

branches (Figure 14A). This branching pattern results in

a tree with a spreading canopy.

In spring female cones are initiated as swellings cov-

ered by imbricate, hard, spiky, tridentate leaves with

broad, amplexicaul sheaths, at the tips of some of the

branches (Figure 13 A). By early summer these are fully

developed young female cones with numerous circles of

long, maroon, strap-shaped stigmas marking rosettes of

conical, hairy and pointed modified leaves, the centres of

which soon become raised and punctuated by ‘stars’ of

about 5 vegetative leaves. The two lateral branches

directly below the cone, start to thicken and curve

upward in preparation for the production of the follow-

ing season’s pair of apical cones (Figure 13C). By the

following spring the old female inflorescence cones are

shrouded in numerous upwardly curving vegetative

branchlets, having originated from the ‘stars’ of the pre-

vious spring, plus one apical, straight and slightly more

robust, branchlet. The two branches just below the cone

are close together and clearly more robust than the ones

lower down (Figure 13B).

The female inflorescence cones consist of a number

of lateral condensed double racemes (homothetic

diplobotrya) (Figure 14B: DB), each consisting of a few

sessile capitulum-like co-florescences (botrya) (CoF).

These are aggregated and spirally arranged on a con-

densed main axis to form an oblong, cone-like, polytelic

synflorescence on the end of a main branch, originating

as one of a pair from below a cone of the previous sea-

son (Figures 13C; 14A-D). Each diplobotryum is sub-

tended by a primary cone leaf (pci) with a grossly

extended sheath segregating the individual diplobotrya

from each other, and a tridentate, often stipulate, woody

blade just emerging between the involucres of secondary

cone leaves (Figures 13E; 14B-D). The secondary cone

leaves have thickened woody sheaths without stipules

and swollen trifoliolate blades, with the leaflets of each

Bothalia 33,2 (2003)

185

FIGURE 11. — Cliffortia arborea. A,

apical portion of main stem

with three female inflores-

cences (cones) in situ, oldest

one lowest, weathered to

basal portions of primary and

secondary cone leaves and

the remains of two lateral

branches; B, primary cone

leaf, astipulate type, adaxial

view; C, secondary cone leaf,

lateral-adaxial view; D, leaf

from region of stem directly

below female cone, abaxial

view; E, vegetative leaf,

adaxial view. A, Fellingham

1624 ; B-E, Oliver 10054.

Scale bars: 1 mm.

leaf arranged in triangles (Figures 13D; 14B, C, D: scl).

Both the primary and secondary cone leaves differ from

the ordinary vegetative leaves, which are small, unifoli-

olate and ligulate in shape (Figure 13G). Each secondary

cone leaf subtends a capitulum-like botrys, bearing up to

12 flowers on minute pedicels, subtended by greatly

modified bracts (Figure 13F). The flowers, each with

three or four calyx lobes, are totally hidden amongst the

secondary cone leaves with only their ligulate styles pro-

truding (Figure 13C). The fully developed fruits are

irregularly angular and closely packed in the concavity in

the adaxial side of the sheath of the secondary cone leaf,

which subtends the capitulum-like flower-bearing struc-

ture, the co-florescence (Figure 13D). In the mature

young cone in which proliferation of the apical bud has

already given rise to a conspicuous vegetative continua-

tion of the main stem, spirally arranged circles of long

maroon stigmas — each circle with a tridentate, woody,

primary leaf below it — demarcate the double racemes

(diplobotrya) (Figure 13C).

In the slightly more mature cone with withered stig-

mas, the centres of these circles are raised as the second

order axes proliferate, and each circle becomes crowned

with a star-like arrangement of vegetative leaves. The

appearance of the ‘stars' that are to develop into the

shroud of side shoots, seems to coincide with fertiliza-

tion, as at that point the styles begin to dry and shrivel.

As the second order axes elongate in the more mature

cone, the ‘stars' are replaced by a shroud of vegetative

side shoots (Figure 13B). These side shoots as well as the

apical proliferation of the main stem, are shed in the

older cones. It seems that the shrouds of shoots are in

place during the development of the fruits, which is the

rest of the current season and the following one.

The male inflorescences are in the axils of ordinary

vegetative leaves on branchlets lower down on the main

branches bearing the female cones, and consist of solitary

male flowers or small groups of very shortly pedicelate

male flowers with much reduced bracteoles (Figure 13G).

8. Cliffortia conifera E.G.H.Oliv. & Fellingham

The only known population of this species is on an

east-facing cliff edge, on the Anysberg. The small group

of tree-like shrubs up to 4 m tall and with the main trunks

up to 150 mm in diameter, resemble conifers in their

upright growth form, as well as in their cone-like inflores-

cences, each terminally on a short lateral branchlet.

186

Bothalia 33,2 (2003)

These cone-bearing branchlets occur in groups of up to

10, in subterminal zones on main branches; rarely on

secondary branches. Older branches have several zones

of cones interspersed with vegetative regions consisting

of well-developed, leaf-bearing lateral branches.

The obovoid or occasionally spheroid female cone con-

sists of many condensed double racemes (homothetic

diplobotrya) aggregated on a shortened main axis, which is

the condensed end of a lateral branchlet. The cone is devoid

of proliferating shoots, either from the ends of the double

racemes or the apex of the cone (Figures 15 A; 16A, B). The

cone can be interpreted as a polytelic synflorescence com-

posed of numerous (up to 50), condensed, sessile racemes

(botrya), the co- florescences, grouped together in highly

condensed, double racemes (diplobotrya). The co-flores-

cences resemble capitula with up to 16 flowers all arising at

the same level from the truncated end of the very short, 3rd

order, florescence axis. The flowers are interspersed with

long erect hairs from which the variously modified and

reduced bracts just emerge. Except for the protruding stig-

mas and occasionally the small calyx lobes, the flowers (and

eventualy the fruit) are concealed by the bulbous secondary

cone leaves on the 2nd order cone axes, which are involu-

crally arranged and form the matrix of the cone. The sec-

FIGURE 12. — Cliffortia arborea :

longitudinal diagrams of in-

florescences. A, three female

cones on main branch and

two on lateral branches; B,

part of cone showing three

dibotrya with vegetative

branches; C, D, single bot-

ryum or co-florescence (solid

square), expanded to show

individual flowers (circles),

hr, bract of single flower;

CoF, co-florescence or bot-

ryum; DB, diplobotryum;

pci, primary cone leaf; scl,

secondary cone leaf; SN, syn-

florescense; VB, vegetative

secondary branch. Broken

lines: considerably expanded

axes of short shoots.

ondary cone leaves are unifoliolate, lack both stipules and

sheaths but have a demarcation between the upper bulbous

part with velvety indumentum, and the basal flattened

glabrous part (Figures 15C, D; 16B: scl). The primary cone

leaves originate on the main cone axis, subtend the lateral

second order diplobotrya and are also concealed by the sec-

ondary cone leaves. The primary cone leaves differ from the

secondary cone leaves in having extended leaf sheaths with

stipules and, being uni- or trifoliolate (Figures 15E; 16B:

pci). Both the primary and secondary cone leaves differ

from the ordinary vegetative leaves which are trifoliolate,

the leaflets flat with rolled edges and plain to tridentate and

the sheath amplexicaul and stipulate (Figure 15A, B).

The male inflorescence is a much simpler structure,

consisting of a cluster of 3 or 4, occasionally 5 flowers,

on a highly reduced short shoot, in the axil of a subapi-

cal vegetative leaf, on lower lateral branches, well below

zones of female cones (Figure 15B).

DISCUSSION

Although only eight species of the 119 in Cliffortia

were included in this study, they are representative of the

Bothalia 33,2 (2003)

187

FIGURE 13. — Cliffortia dichotoma. A, apical swelling on main stem; B, mature inflorescence cone shrouded in vegetative lateral branchlets and

apical continuation of main stem; C, young female inflorescence cone with numerous circles of strap-shaped stigmas; D, secondary trifo-

liolate cone leaf with exstipulate wooden sheath, subtending capitulum-like botrys bearing three remaining fruits; E, primary cone leaf with

extended curved sheath; F, capitulum-like botrys, flowers and surrounding hairs removed exposing pedicels, each with subtending modi-

fied bract; G, vegetative branch with male flowers. A. B. D, F, Fellingham 1684 (BOL, NBG); C, E, G, Fellingham 1689 (BOL, NBG).

Scale bars: 1 mm.

types of inflorescences in the genus, covering the range

from the single-flowered inflorescence in C. nivenioides,

to the highly condensed and discrete female cone in C.

conifera. There is no hypothesis of the phylogenetic rela-

tionships in Cliffortia , so our sampling could not be based

on the phylogenetic patterns in the genus. Nonetheless, it

seems likely that the sampling captured most of the vari-

ation in the inflorescence morphology in the genus.

Basic inflorescence construction

The basic inflorescence unit in Cliffortia is a short

shoot with one or more intemodes, bearing one or sever-

al lateral, unisexual flowers, which may be bracteate or

ebracteate. This is therefore an indefinite polytelic flo-

rescence (Weberling 1989). The short shoot itself is

borne in the axil of a vegetative leaf. In its most common

form, the shoot is single-flowered, and this misled

Weimarck (1934) into interpreting them as ‘solitary

flowers in the axils of ordinary vegetative leaves’. There

are a number of ways in which this can be modified to

generate the diversity of inflorescences observed in

Cliffortia: these are detailed below.

Given that all inflorescences are basically racemose

short shoots, there are a number of parameters that vary

among species in the genus, and that are to some extent

responsible for the variability in the inflorescence con-

structions:

1, number of flowers per inflorescence. These vary

from one (probably the most common situation in the

genus), to several as in C. ruscifolia , to numerous as in

C. heterophylla, C. odorata and the three coniferous

species of the section Arboreae. In some cases, initially

only one or a few flowers are formed (usually female).

When these flowers have matured, growth resumes in the

florescence, to form a further set of usually male flowers.

In C. heterophylla , though, the female florescence con-

tinues growth as a long shoot, which then bears single-

flowered lateral male florescences;

188

Bothalia 33,2 (2003)

FIGURE 14. — Cliffortia dichotoma :

longitudinal diagrams of in-

florescences. A, three female

cones, two on dichotomous

branches, central one older,

having lost its apical and lat-

eral branches; B, part of cone

showing three dibotrya with

vegetative branches; C, D,

single botryum or co-flores-

cence (solid square), expand-

ed to show individual flowers

(open circles), br, bract of

single Bower CoF, co-flores-

cence or botryum; DB, diplo-

botryum; pci, primary cone

leaf; scl, secondary cone leaf;

SN, synflorescence; VB,

vegetative secondary branch-

es. Broken lines; consider-

ably expanded axes of short

shoots.

2, the short shoot internodes are generally about equal

in length. This is so in most primary short shoots, sec-

ondary short shoots and even in systems containing ter-

tiary short shoots as well. In C. oclorata , however, the

axis of the primary short shoot is very condensed, form-

ing a flattened disc borne in the axil of a vegetative leaf.

This disc-like short shoot retains its viable apical bud and

bears secondary short shoots in the form of stalked discs

as the ultimate flower-bearing short shoots. Disc-like

ultimate flower-bearing short shoots also occur in C.

arborea, C. conifera and C. dichotoma, but in these

species as the tertiary short shoots;

3, aggregation of short shoots can be variously devel-

oped. In some cases, the florescences are borne solitary,

although generally in distinct zones on the plant.

However, they may be aggregated into clusters (in C.

oclorata), elongated synflorescences (in C. heterophylla),

or aggregated into dense cone-like structures (in C.

arborea, C. dichotoma and C. conifera). Some of these

synflorescences can be remarkably complex, and diverse

in structure, while exhibiting great plasticity;

4, short shoots transforming into long shoots. After the

flowers on the short shoot have matured, the internodes

elongate, thus transforming the short shoot into a long

shoot, as seen in C. nivenioides and C. heterophylla;

5, viability of the apical bud of short shoot. In C.

conifera all apical buds of short shoots have lost their

viability. In the other species, continued growth from at

least some of these apical buds variously affects the syn-

florescence, and often the whole plant architecture.

Firstly, continuing the growth of the florescence general-

ly leads to the bearing of a later generation of male flow-

ers after the initial generation of female flowers. This

was observed in a number of species, such as C. crenata,

C. heterophylla and C. oclorata. This extends the period

of time over which the florescence unit remains produc-

tive, and allows for the evolution of herkogamy at an

inflorescence (rather than flower) level. Secondly, this

continued growth may lead to the formation of new veg-

etative shoots, thus true proliferation, where the repro-

ductive apical bud transforms to a sterile (vegetative)

bud. This is common in the genus, and is discussed in

more detail below.

The approach of searching for a basic pattern in the

inflorescence, and then establishing how the inflores-

cence can be modified to produce a remarkable diversity

of apparently different types, can be very productive.

Weber (1995) used this approach to show that in

Phyllagathis superficial description of inflorescences

can be quite misleading, and that this can be revealed by

searching for the basic pattern (in this case a monotelic

homocladic thyrse). Here we attempt to account for the

variation in inflorescences in Cliffortia by finding

‘switches’, simple ‘on or off’ devices. Modifications in

the behaviour of the apical buds of the inflorescences.

Bothalia 33,2 (2003)

189

FIGURE 15, — Cliffortia conifera,

Oliver 10055 (NBG). A,

branch bearing female cones;

B. branchlet bearing male

flowers. C, D, secondary

cone leaf: C, adaxial view; D,

lateral view. E, primary cone

leaf, adaxial view. A, B,

copied from Inge Oliver’s

drawings. Scale bar: 1 mm.

and the timing of intemode elongation of the synflores-

cence and florescence axes, should be adequate. Such

modifications have also been used to explain the varia-

tion in the inflorescences of the legume tribes Ingeae and

Acacieae (Grimes 1999). These ‘switches’ could then be

used as characters for a cladistic analysis. This could lead

to a much better set of inflorescence homologies, than

using descriptive terms directly, since using the switches

allows for more intermediate conditions.

The possible characters that could be used are listed

below. We have not attempted to score these for the

species, since our sample of species is too small to be

able to interpret the result: 1, florescence number of

flowers: one/several; 2, florescence flower arrangement:

protogynous/mixed; 3, florescence growth: continuous

(all flowers formed more or less at the same time)/ inter-

rupted (leading to sex changes); 4, florescence short

shoot: permanent/later elongating into long shoots; 5,

florescence apical bud: terminating/proliferating; 6, syn-

florescence apical bud: terminating/proliferating; 7,

florescences: solitary/aggregated/forming cones; 8, male

and female florescences: in the same synflorescence/in

separate synflorescences.

Sexuality

The flowers are always unisexual, and there are sev-

eral ways in which the male and female flowers can be

combined to form a monoecious species. The following

three patterns were observed:

Firstly, the male and female flowers may be found in

separate synflorescences, and would therefore be spatial-

ly separated. This is exemplified by the cone-bearing

species, C. arborea, C. dichotoma and C. conifera , in

which the female florescences are many-flowered and

are aggregated into cone-like synflorescences which are

completely female. Male florescences are borne in sepa-

rate, very different looking synflorescences with very

few flowers. This situation is comparable to that found in

most conifers.

Secondly, single-flowered florescences where the

male and female florescences are combined into a com-

mon synflorescence. The only species with this pattern

investigated is C. nivenioides. Male florescences mature

first, followed by a time lag before the females mature.

190

Bothalia 33,2 (2003)

FIGURE 16. — Cliffortia conifera :

longitudinal diagrams of in-

florescences. A, main branch

with two cone-bearing zones;

B. part of female cone show-

ing three dibotrya; C, D, sin-

gle botryum or co-flores-

cence, (solid square) expand-

ed to show individual flowers

(circles), br, bract of single

flower; CoF1, co-florescence

or botryum; DB. diplobot-

ryum; FB, flowering sec-

ondary branches; pci. prima-

ry cone leaf; scl, secondary

cone leaf; SN, synflores-

cence; VB, vegetative secon-

dary branches. Broken lines:

considerably expanded axes

of short shoots.

Thirdly, male and female flowers are found in the

same florescence (raceme), the flowers are separated

usually with the female flowers borne below the male

flowers (thus on the older parts). In general, the females

then flower before the males, so that there is no overlap

in the flowering time between male and female flowers.

Thus the inflorescences are dichogamous. This has been

referred to as sex changes or sequential hermaphrodism

of the individual (Freeman et al. 1980), and is particular-

ly well illustrated by C. heterophylla. The individual

always starts out as overtly female, then becomes bisex-

ual, and eventually is overtly male. Single observations

would suggest that the species is dioecious, and bisexual

individuals could be taken to be anomalous. This sequen-

tial hermaphrodism is only possible because of the mor-

phological plasticity of the inflorescences, and in partic-

ular because the apical buds retain viability. Sex chang-

ing (or sequential hermaphroditism) as ‘a phenomenon

of at least some dioecious species’, has more recently

been based on findings in 66 species in 25 families, doc-

umented from as early as 1910 (Freeman et al. 1980).

‘Apparent dioecy’ in any one season, masking the true

labile nature of the sex of the plant which can change

with changes in the physiological state, which in turn

depends on external factors, was described in Arisaema

triphyllum (Araceae) by Bierzychudek (1982). She also

found that all Arisaema species change sex, though only

some change from male to female without any hermaph-

rodites. Similar findings of sex changes were reported in

Myrica gale (Myricaceae) (Proctor et al. 1996).

Two of the species investigated here have bisexual

florescences, but show no segregation of the sexes either

in time or space. In C. ruscifolia the female flowers were

in the fruit stage when the males still had good anthers

[Oliver 10387 (NBG)]. The same phenomenon is seen in

C. odorata [Fellingham 1664 (BOL), 1678 ( BOL )•] .

It appears as if all species of Cliffortia may be

monoecious. Previously it had been suggested that the

Bothalia 33,2 (2003)

191

standard condition in Clijfortia is that the plants are dioe-

cious, and Weimarck (1934) suggested that ± 60% of the

species in the genus are ‘capable of being monoecious’.

This could be a misinterpretation based on single sam-

ples of species exhibiting sex changes, such as C. het-

erophylla. Oliver & Fellingham (1991) described C.

burgersii as dioecious, but this might reflect yet another

example of the possible erroneous interpretations of mor-

phologically different stages of a monoecious plant, as

morphologically different plants of a dioecious species.

It seems essential to follow a plant through its flowering

sequence to uncover sequential female and male phases.

Wind pollination

Many of the specializations in the inflorescences of

Clijfortia can be interpreted as adaptations to wind pollina-

tion. The genus as a whole shows typical wind-pollinated

flowers: a reduced, dull perianth, and unisexual flowers

(Faegri & Van der Pijl 1979; Proctor et al. 1996; Linder

1998). Since most species and genera in the Sanguisorbae

are wind pollinated, the anemophilous syndrome was

mostly inherited by Clijfortia, and indeed many of the typ-

ical wind-pollination features of Clijfortia are also found in

the related genera Acaena, Sanguisorba , and others. Wind

pollination is generally seen as a common alternative to

biotic pollination in species-poor or cool habitats where the

survival of biotic pollinators would be difficult (Whitehead

1969, 1983). Yet, wind pollination has frequently evolved

in the species-rich and structurally dense Fynbos of the

Cape Floral Kingdom, and is found in the Ericaceae,

Proteaceae, Asteraceae ( Tarchonanthus ), Rubiaceae

( Anthochomis ) and Rosaceae ( Clijfortia ) (Koutnik 1987;

Le Maitre & Midgley 1992). Accepted as a primitive con-

dition in the gymnosperms, it had been confirmed in the

Cape genera Podocatpus and Widdringtonia (Koutnik

1987). Within Clijfortia , most inflorescences appear to

separate male and female flowers either in time (through

sequential hermaphroditism) or spatially (in the species

with cones, where female and male synflorescences are

separated). However, the coniferous species of Clijfortia

have a rare reversal of the location on the plant of the male

and female elements, with the male below the female. This

rare condition is shared with only one evergreen gym-

nosperm, viz. Pinus sylvestris (Proctor et al. 1996).

Impact on vegetative branching

In Clijfortia the florescences generally proliferate,

that is. the apical bud transforms to a vegetative bud, and

growth continues into the next season as a vegetative

shoot. This can be a major source of vegetative branch-

ing to the plant, and as such have a major impact on the

plant architecture. The degree of involvement of the var-

ious orders of short shoots, determines the abundance

and positions of long shoots produced.

A sparsely branched, lanky growth form is produced

where flower bearing is limited to a few apical vegetative

short shoots. These are the only short shoots that eventually

develop into long shoots (vegetative branches) by the

lengthening of their intemodes (e.g. C. nivenioides). By

contrast, the rich and random branching pattern in C. rusci-

folia originates from its numerous inflorescences. Most of

these proliferate into vegetative long shoots. A sparse and

virgate branching pattern results from the small percentage

of inflorescences, all near the apex of the main stem, that

proliferate. This is seen in C. crenata. The monopodial

branching in C. odorata is the result of proliferation of the

primary short shoots of numerous axillary inflorescences on

main stems, which themselves grow vegetatively only. The

monopodial branching of C. heterophylla, however, is the

result of a combination of two factors. The fast-growing

main stem gains from the lengthening of the primary short

shoot of the apical inflorescence as well as the eventual api-

cal proliferation. The limited linear increase in the lateral

branches, originating from the proliferation of the second

order short shoots, enhances this effect. Similarly, in the

young plant of C. arborea , the monopodial branching is regu-

lated by the proliferation of the first order short shoot, which

is the main stem of the plant. As the plant ages and side

branches become involved in flower bearing, the branching

pattern is obscured. A peculiar involvement of the position

of the inflorescence determining the branching pattern is

found in C. dichotoma. With the abortion of all initial pro-

liferation shoots, two vegetative branches, close to each

other and just below the terminal inflorescence, take over

the function of continuing linear growth. These develop into

a pair of thick main branches, one on each side of the cen-

tral apical inflorescence. This results in the dichotomous

branching pattern, unique in the genus. It is only in C.

conifera , the species that exhibits no morphological plastic-

ity of the inflorescence, that there is no effect of the inflores-

cence on the branching pattern, other than that a zone of

branches are dedicated exclusively to the bearing of the

female cones. In this aspect, it approaches the condition in

the gymnosperms.

In inflorescences consisting of a system of short shoots

of primary and secondary or primary, secondary and ter-

tiary short shoots, it is always only the ultimate short

shoots that are dedicated to flower bearing, to the extent to

which the viability of their apical buds is lost. The apical

buds of the lower order short shoots, forming the matrix of

the inflorescence, however, retain their viability. These are

the short shoots that can undergo one of two changes, or a

combination of these two, at the onset of the vegetative

stage or, in some instances, during the development of the

inflorescence. The changes to the short shoot are the

lengthening of the intemodes rendering it a long shoot, or

the proliferation of the apical bud into an apical long

shoot, or a combination of the two processes. With the

lengthening of the intemodes, it becomes a long shoot

without leaving a trace of the original short shoot. With the

addition of apical proliferation to the lengthened shoot, the

ensuing long shoot would gain greatly in longitudinal

growth. It is only in the case of apical proliferation with-

out any lengthening of the internodes, that the original

short shoot can still be discerned. It is this inherent mor-

phological plasticity of the matrix of the inflorescence that

renders it capable of contributing to the vegetative growth

both in substance and pattern.

SPECIMENS EXAMINED

Fresh material

Notes

1. Fresh specimens of Clijfortia heterophylla (4): Karin Behr sub

Fellingham 1638 ; Jane Forrester sub Fellingham 1640 (4) NBG, BOL,

192

Bothalia 33,2 (2003)

were examined in the early Spring of 1994. Six field trips were under-

taken between the 3-01-1995 and the 22-06-1995 to the Harold Porter

Nature Reserve in Betty’s Bay, to study this species in situ. Sketches

are from Forrester sub Fellingham 1640 (4) NBG, BOL.

2. Field work was done on the two populations of C. arborea (6) on the

Nieuweveldberg escarpment, at Beaufort West and fresh material taken

for study were Fellingham 1624 female (6) MO, NBG, PRE; 1625

male & female (6) BM, NBG, PRE.

3. Field work on C. dichotoma (7) was done in September 1995,

November 1995, November 1996, and October 1997, on the Farm

Papkuilsfontein in the Lokenberg area of Namaqualand. Specimens

were collected on all of these occasions. The fresh specimens studied

and sketched were Fellingham 1684 (7) BOL, K, NBG, PRE and

Fellingham 1689 (7) BOL. K, NBG, collected on 27-09-1995 and lb-

11 - 1 995 respectively.

4. Fresh material of the type specimen of C. conifera (8), Oliver 10055

male & female (8) BOL, K, NBG, PRE, was examined and sketched.

The type locality, the eastern end of the Anysberg in the Ladismith

District, was visited and more fresh material, Fellingham 1531 female

(8) NBG, collected for examination in the herbarium.

Other fresh material

Fellingham 1647-1652, 1654-1658, 1663, 1668, 1673, 1675, 1677(4)

NBG; 1660, 1664, 1678 (5) NBG; 1662, 1674, 1676 (4) BOL.

Fellingham & Vlok 1588(1) NBG.

Oliver 10387 (2) NBG; 10567, 10569, 10574 (3) NBG.

Viviers & Vlok 470 ( 1) NBG.

Herbarium specimens

Acocks 18621 male & female (6) BOL; male & female (6) K; male (6)

NBG; 19280(2) NBG.

Barker 5593 (3) BOL; 8263 (2) NBG. Barnes s.n. (3) BOL. Bean &

Trinder-Smith 2733 (2) BOL. Bolus 2754 (2) NBG, 7600 (2) BOL,

NBG; 10603, 12674 (3) BOL; 19854, 19855 (5) BOL. Bond 1754 (1)

NBG, PRE. W.Bond 1511 (2) BOL. Boucher 209 (4) NBG, PRE; 502

(3) NBG, PRE; 4193, 5019 (2) NBG. Barman 895 (3) BOL.

Compton 3039, 9492 (3) BOL; 3040, 3459, 3920 (2) BOL. 5685, 7412,

8662, 10830, 11831, 18425, 18523, 22216, 22247, 22843 (2) NBG;

14374, 51589 (5) BOL. Cooper 1451 (5) BOL.

De Villiers s.n. immature female (6) NBG. De Vos 624, 1213 (4) NBG.

Dickin 83 (5) BOL. Durand 263 (3) NBG, PRE.

Ebersohn 15/68 (4) NBG. Ecklon & Zeyher 1720 (3) BOL. Esterhuysen

3447, 5139, 5263a, 10346, 13959, 25911, 27596 (2) BOL; 5241a (2) NBG;

s.n., 3030, 3443, 9540, 29097 (3) BOL; 13677 (4) BOL; 29003 (4) BOL,

NBG; 1831, 15311, 19430, 23890, 29064, 30049, 33509, 33759 (5) BOL.

Fellingham 1533 (2) NBG; 1702 male & female (7) BOL, NBG. male

(7) PRE, K; 1705, 1709 (7) BOL, NBG, PRE; 1706, 1707, 1708, 1710,

1711, 1712 (1) BOL. Fourcade 24 (5) BOL; 3089 (3) BOL.

Gillen 1908 (2) BOL, NBG; 4244 (3) BOL. Goldblatt 2162 (2) NBG.

Grobler 20181 (4) NBG.

Haynes 362 (3) NBG, PRE. Hennecart 54 (3) BOL. Hugo 911 (2)

NBG. Hutchinson 1123 (2) BOL.

Kerfoot 5397 (3) NBG, PRE. Kruger 801 (4) NBG; 1204, 1346 (2)

NBG, PRE.

Leighton 12 (3) BOL; 13 (4) BOL; 2977 (5) BOL. Levyns 482, 2918,

3016, 4061, 4091, 4776, 7938, 8056, 9141, 11634 (3) BOL; 2051, 2460,

6167, 6444, 6511, 8010, 8975, 9210, 11194 (2) BOL; 6466 (5) BOL;

7789, 7795, 8095 (4) BOL; Lewis 6065 (2) NBG. Loubser 825 (5) BOL.

Maguire 155 (3) BOL. Manson 130 (3) NBG, PRE. Marloth 1977 (2)

NBG, 9089 (2) BOL; 3907 female (6) BOL, male (6) K; 9730 male &

female (6) NBG; 9770 female (6) NBG, PRE; 14106 (4) NBG; Mauve,

Van Wyk & Pare 28 (2) NBG. McDonald 1688, 1732 (2) NBG.

McOwan STEU13375 (2) NBG. Michel! BOL16091 (5) BOL. Moffett

& Steensma 4067 male (6) NBG.

Net & Boucher 73 (5) BOL, K, NBG, PRE.

Oliver 4252, s.n. (2) NBG; 9730 female (8) NBG, PRE, S; 10054

female (6) BOL. K, MO, NBG, PRE, S.

Parker 3663 (5) BOL; 3717 (3) BOL. Paterson 2015 (5) BOL. Pillans

s.n. (2) BOL; 7358, 9292 (3) BOL. Pretorius 396 (7) BOL, K, MO,

NBG, PRE.

Rehmann 2716 (2) BOL. Rodin 3091 (3) BOL.

Salter 6335 (2) BOL. Schoncken 269 (2) NBG. Shearing 893 female

(6) PRE. Simpson 93 (2) NBG. Sims 2514 (5) BOL. Stephens 7125 (3)

BOL. Stokoe 405 (4) BOL, PRE, SAM; 7260 (3) BOL; 7264, 9048,

17238 (4) BOL; SAM61486, SAM59988 (4) NBG. SAM; s.n. (5) BOL.

Taylor 4728, 8023, 9517 (2) NBG. Thode 4818 (2) NBG. Thompson

176 (3) NBG, PRE; 2275 ( 1 ) NBG, PRE. Tyson 766 (2) BOL, NBG.

Van der Merwe 847 (4) NBG. Van Niekerk 763 (2) BOL; 787 (3) BOL.

Van Wilgen 162 (3) NBG, PRE. Van Wvk536 (2) NBG; 1072 female (8)

NBG. Yvette Van Wyk 626 (7) NBG. Van Zyl 3363 (2) NBG. Vlok 1326

(1) NBG, PRE. Von Willard s.n. (7) NBG.

Walters 1741 (2) NBG. Whitman s.n. (3) BOL. Wollev Dod 5 (3) BOL.

Wurts 1366 (2) BOL, NBG.

ACKNOWLEDGEMENTS

We would like to thank the staff of the Harold Porter

Nature Reserve, Betty’s Bay, and in particular Ms J.A.

Forrester and Ms C.M. Behr, for allowing free access to

study material. For assistance with fieldwork, our thanks

go to the Conservation Officers Mr J.H.J. Vlok in the

Swartberg area, and Mr J. Afrika and Mr W.A. Pretorius

in the Oorlogskloof area. The friendly reception by Mr

and Mrs W.N.D. van Wyk on their Farm Papkuilsfontein,

near Nieuwoudtville, is much appreciated. Our thanks to

the staff of the Compton Herbarium (NBG), particularly

Dr J.B.P. Beyers and Mr C.N. Cupido for making study

material available to us and Dr E.G.H. Oliver for encour-

agement and helpful suggestions from the time that my

(ACF) interest in the genus Clijfortia was first kindled

and more latterly, editing and providing the Latin diag-

nostic. I appreciate the helpful suggestions regarding my

(ACF) drawings, received from Mrs I.M. Oliver. Also to

the curators of the Bolus Herbarium of the University of

Cape Town (BOL) and the National Herbarium, Pretoria

(PRE), our thanks for the loan of specimens.

REFERENCES

BIERZYCHUDEK, P. 1982. The demography of Jack-in-the-pulpit, a

forest perennial that changes sex. Ecological monographs 52:

335-351.

DAHLGREN, R. 1971. Multiple similarity of leaf between two genera

of Cape plants, Cliffortia L. (Rosaceae) and Aspalathus L.

(Fabaceae). Botaniska Notiser 124: 292-304.

ERIKSSON, T.. DONOGHUE, M.J. & HIBBS, M.S. 1998. Phylo-

genetic analysis of Potentilla using DNA sequences of nuclear

ribosomal internal transcribed spacers (ITS), and implications

for the classification of Rosoideae (Rosaceae). Plant Syste-

matics and Evolution 211: 155-179.

FAEGRI, K. & VAN DER PUL. L. 1979. The principles of pollination

biology. Pergamon Press, Oxford.

FELLINGHAM, A.C. 1993a. Cliffortia fasciculata, a superfluous name

for C. amplexistipula. Bothalia 23: 67, 68.

FELLINGHAM, A.C. 1993b. Observations on Cliffortia micrantha.

Bothalia 23: 65, 66.

FELLINGHAM, A.C. 1994. Cliffortia longifolia, a ‘good’ species or

should it be a variety under C. strobilifera. Bothalia 24; 31-35.

FELLINGHAM, A.C. 1995. A new species of Cliffortia from the

Swartberg. Bothalia 25: 104-107.

FELLINGHAmTa.C. 2000. Cliffortia. In P. Goldblatt & J.C. Manning,

Cape plants. A conspectus of the Cape flora of South Africa.

Strelitzia 9: 608-616. National Botanical Institute, Cape Town

and Missouri Botanical Garden.

FELLINGHAM, A.C. 2003. A new serotinous species of Cliffortia

(Rosaceae) from the Northern Cape. Bothalia 33: 4 1 — 47.

FREEMAN, D.C., HARPER. K. T. & CHARNOV, E. L. 1980. Sex

change in plants: old and new observations and new hypothe-

ses. Oecologia 47: 222-232.

GOLDBLATT, P. & MANNING, J.C. 2000. Cape plants. A conspectus

of the Cape flora of South Africa. Strelitzia 9. National Botani-

cal Institute, Cape Town and Missouri Botanical Garden.

Bothalia 33,2 (2003)

193

GRAHAM. R.A. 1960. Rosaceae. In C.E. Hubbard & E. Milne-

Redhead, Flora of tropical East Africa: 1-61 . Crown Agents for

Oversea Governments and Administrations. London.

GRIMES, J. 1999. Inflorescence morphology, heterochrony, and phy-

logeny in the Mimosoid tribes Ingeae and Acacieae (Legumi-

nosae: Mimosoideae). The Botanical Review 65: 317-347.

HELFGOTT. D M., FRANCISCO-ORTEGAM, J„ SANTOS-GUER-

RA. A.. JANSEN. R.K. & SIMPSON, B.B. 2000. Biogeo-

graphy and breeding system evolution of the woody Bencomia

alliance (Rosaceae) in Macaronesia based on ITS sequence

data. Systematic Botany 25: 82-97.

HILLIARD, O.M. & BURTT. B.L. 1987. The botany of the southern

Natal Drakensberg. National Botanic Gardens, Cape Town.

JACOT GUILLARMOD, A. 1971. Flora of Lesotho. Cramer, Lehre.

KOUTNIK, D. 1987. Wind pollination in the Cape Flora. In A.G.

Rebelo, A preliminary synthesis of pollination biology in the

Cape flora: 126-133. South African National Scientific

Programmes Report, Pretoria.

LE MAITRE. D.C. & MIDGELEY, J.J. 1992. Plant reproductive ecol-

ogy. In R.M. Cowling. The ecology offynbos: 135-174. Oxford

University Press, Cape Town.

LINDER. H.P 1998. Morphology and the evolution of wind pollina-

tion. In S.T. Owens & P.J. Rudall, Reproductive Biology:

123-135. Royal Botanic Gardens. Kew.

MARLOTH. R. 1905. Eine neue interessante Cliffortia vom

Roggeveld. Botanische Jahrbucher 39: 318, 319.

MELCHIOR, H. 1964. Syllabus der Pflanzenfamilien. Bomtraeger,

Berlin.

MENDES, E.J. 1978. Rosaceae. In E. Launert, Flora zambesiaca:

7-33. Flora Zambesiaca Managing Committee. London.

OLIVER. E.G.H. & FELLINGHAM, A.C. 1991. A new species of

Cliffortia from the southwestern Cape. Bothalia 21: 60-62.

OLIVER, E.G.H. & FELLINGHAM, A C. 1994. A new serotinous

species of Cliffortia (Rosaceae) from the southwestern Cape

with notes on Cliffortia arborea. Bothalia 24: 153-162.

PROCTOR. M„ YEO, P. & LACK, A. 1996. The natural history of pol-

lination. Harper Collins, London.

TAKHTAJAN, A. 1997. Diversity and classification of flowering plants.

Columbia University Press, New York.

WEBER. A. 1995. Inflorescence morphology of the Malayan species of

Phvllagathis (Melastomaceae). Feddes Repertorium 106: 445^161.

WEBERLING, F. 1983. Fundamental features of modem inflorescence

morphology. Bothalia 14: 917-922.

WEBERLING, F. 1989. Morphology of flowers and inflorescences.

Cambridge University Press, Cambridge.

WEIMARCK, A.H. 1934. Monograph of the genus Cliffortia. Lund.

WEIMARCK, A.H. 1940. Some notes on the genus Cliffortia.

Botaniska Notiser 51: 399-402.

WEIMARCK, A.H. 1946. Further notes on the genus Cliffortia.

Botaniska Notiser 80: 407—420.

WEIMARCK, A.H. 1948. The genus Cliffortia, a taxonomic survey.

Botaniska Notiser 1948: 167-203.

WEIMARCK, A.H. 1953. Two new Cliffortia species from the Cape

Province. Botaniska Notiser 102: 78-80.

WEIMARCK, A.H. 1959. Four new Cliffortia species. Botaniska Notiser

112: 73-79.

WHITEHEAD, D.R. 1969. Wind pollination in the angiosperms: evo-

lutionary and environmental considerations. Evolution 23:

28-35.

WHITEHEAD, D.R. 1983. Wind pollination: some ecological and evo-

lutionary perspectives. In L. Real, Pollination Biology: 97-107.

Academic Press, Ontario.

>

Bothalia 33,2: 195-198 (2003)

More accurate publication dates for H.C. Andrews’ The Heathery ,

particularly volumes 5 and 6

R.J. CLEEVELY*, E.C. NELSON** and E.G.H. OLIVER***

Keywords: dates of publication. Erica L.

ABSTRACT

Using three types of evidence — dated watennarks; dates of flowering of Erica L. species in cultivation in England; dates

on which Andrews prepared the original drawings — it is concluded that the six volumes of Henry Charles Andrews’ The

Heathery were published as follows: volume l,not earlier than June 1805; volume 2, not earlier than 1806; volume 3, 1806;

volume 4, 1807; volume 5, not before 1816; volume 6, late in 1828. Consequently, most, if not all, of Andrews’ new names

for Erica species were first published in Coloured engravings of heaths.

INTRODUCTION

Henry Charles Andrews published The Heathery as a

convenient, working edition of his lavish, large-format

part-work Coloured engravings of heaths. When com-

pleted, The Heathery comprised six volumes, each con-

taining 50 hand-coloured plates with accompanying

brief text, and an alphabetical index numbering the

plates. The introductory pages were essentially similar

to those in Coloured engravings but the text was revised

according to experience. Inevitably, the ‘Introduction'

(in vol. 1) was quite different and solely an explanation

of the new work; it contained a statement that The

Heathery’ would be published in parts (as all Andrews’

other publications had been) and that ‘when one volume

is completed, every necessary requisite for binding will

be given’. However, it is our contention that this was

not the case, and that the individual volumes of The

Heathery were published as separate, complete vol-

umes. There is no evidence known to us in the form of

wrappers or partial sets of unbound parts that The

Heathery were issued, like Coloured engravings of

heaths , in sequential fascicles for later collation and

binding.

As with Coloured engravings of heaths , which has

been discussed in detail by Cleevely & Oliver (2002), the

dates of issue of The Heathery , especially the last two

volumes, are problematic. The title pages of the six vol-

umes are dated as follows; 1. 1804; 2, 1804; 3, 1806; 4,

1807; 5, 1809; 6, 1804 (for comment on the title page of

the last volume, see below). Hitherto, taxonomists (e.g.

Dulfer 1965) have accepted those dates as the correct

publication dates, except for that of vol. 6; for the sixth

volume, 1812 has been the date generally cited in botan-

ical monographs following Pritzel (1872). Kerkham

(1988) gave 1804—1809 as the date range for publication

of the six volumes of The Heathery but also remarked

* High Croft, Gunswell Lane, South Molton, Devon, EX36 4DH, United

Kingdom.

** Cultivar Registrar, The Heather Society, Tippitiwitchet Cottage, Hall

Road, Outwell, Wisbech, Cambridgeshire, PE 14 8PE, United Kingdom.

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

7735 Claremont, Cape Town.

MS. received: 2002-10-11.

that ‘Vol. 6 must have been published much later as there

are 1826 watennarks in it.’ Stafleu & Mennega (1992)

accounted for only the first four volumes stating the

dates from the title pages.

At present, we have been unable to discover how The

Heathery was advertised, sold or distributed. We have

not been able to trace any published references to it in

contemporary periodicals.

In this paper we present evidence supporting Kerk-

ham ’s (1988) remarks and contradicting the title page

dates for vols 1, 2, 5 and 6. We conclude that those dates

are far too early, and in particular that instead of publi-

cation of vols 5 and 6 of The Heathery preceding publi-

cation of Coloured engravings of heaths , those two vol-

umes, like the previous four, were published after the rel-

evant fascicles of Coloured engravings of heaths.

We have examined the sets of The Heathery’ held in

the following libraries and a private collection; data

extracted from these copies are held by the authors on a

database: 1, Lindley Library, Royal Horticultural Society,

London (complete); 2, The Linnean Society of London

(partial set, vols 1-3); 3, Royal Botanic Gardens, Kew

(partial set, vols 1-3, parts of vols 4 and 5); 4, Bolus

Herbarium, University of Cape Town (cited as BOL by

Kerkham (1988)) (complete); 5, Compton Herbarium,

National Botanical Institute, Kirstenbosch (formerly in

Cape Government Herbarium) (complete); 6, Molteno

Library, National Botanical Institute, Kirstenbosch (par-

tial set, vols 1-3); 7, South African National Library (for-

merly called South African Library), Cape Town (cited

as SAL by Kerkham (1988)) (complete); 8, Walter J.

Middelmann, Rondebosch, Cape Town (complete).

It is surprising that so few complete copies seem to

exist, and the occurrence of incomplete sets of the plates,

especially from the later volumes, might indicate that

these could be obtained separately, or even in fascicles,

although, as noted, we do not consider this to be the case.

Alternatively, it might be that the volumes really were

heavily used as ‘working’ tools — Andrews described The

Heathery as a ‘green-house companion’ — and suffered

accordingly.

196

Bothalia 33,2 (2003)

THE HEATHERY: PLATES

The plates published in The Heathery are directly

related to the plates in Coloured engraving of heaths.

Comparison of the individual illustrations indicates that

The Heathery plates, which, as noted below, are invari-

ably dated after the companion ones in Coloured engrav-

ings of heaths , show only a small portion of the plant as

portrayed in the main work. Many, but not all, The

Heathery illustrations are reversed and in some instances

alterations are evident. The enlarged dissections, arrayed

along the bottom of each plate in The Heathery are also

either mirror images or redrawn or new. The redrawn

dissections sometimes differ in details- broader append-

ages, filaments omitted, or the ovary now included in the

small gynoecium/androecium dissection — and in most of

these cases the whole drawing is redone with the anthers

in different positions and with a different aspect. Re-

versed images corroborate evidence of the engraved pub-

lication dates that Andrews prepared new plates for The

Heathery , often, but not always, using the printed plates

from Coloured engraving of heaths as the templates. Had

he used his original drawings as templates for the

engravings published in both books, the images would

all have had the same orientation.

Regarding engraved dates on botanical illustrations

published in Great Britain during the eighteenth and

nineteenth centuries, under an Act of Parliament (8

George II.c.13) passed in 1734, authors of an engraved

work were granted copyright protection ‘to commence

from the day of the first publishing thereof, which shall

be truly engraved with the name of the proprietor on each

plate, and printed on every such print or prints’ (see

Henrey 1975: II, p. 664; Steam 1940). Thus the legally

required engraved date should be the exact date of publi-

cation. However, by the early nineteenth century,

although this act and several subsequent ones were still

in force, authors, including Andrews, began to abandon

the practise of engraving dates on their illustrations,

probably because the necessarily lengthy process of pro-

duction of these hand-coloured illustrations meant that

engraved dates were not the true dates of publication.

THE HEATHERY: VOLUMES 1-4

It is generally accepted that the first four volumes of

The Heathery are not of nomenclatural significance.

While the dates on the title pages are not always consis-

tent with the contents, amending the dates of publication

of these volumes will not cause shifts in the priority of

names.

Volume 1 has a title page dated 1804, but according to

the dates engraved on the 50 plates, almost two-thirds

(29) of the plates were not prepared until 1805. The lat-

est engraved date is June 1805 (Erica walker ia\ t. 50). It

should be noted that with a single exception, all the

engraved dates in this volume post-date those on the cor-

responding plates in Coloured engravings of heaths. The

one exception is the plate depicting Erica pubescens

minima which is dated May 1 805 in The Heathery (t. 39)

but 1 June 1805 in Coloured engravings of heaths (vol.

2: t. 124).

The watermarks in copies of this volume are confus-

ing. While the dates 1804 and 1805 predominate in the

copies we have examined, it is clear from paper dated

1808, 1810, 1812 and 1822 that both plates and text

pages were reprinted on more than one occasion, proba-

bly as customers requested copies or back numbers.

We conclude that contrary to the date on the title page,

vol. 1 was issued not earlier than June 1805.

Volume 2 has a title page dated 1804, but according to

the engraved dates this could not have been issued before

1806. Forty-two plates are dated 1805, six are dated

1 806. Without exception, all the engraved dates coincide

with or post-date those on the corresponding plates in

Coloured engravings of heaths. Watermark dates range

between 1803 and 1805, although one text page with

1 808 has been noted.

We conclude that contrary to the date on the title page,

vol. 2 was issued not earlier than 1806.

Volume 3 has a title page dated 1806 and 36 plates are

also so dated. Two plates are dated 1805 but none is

dated 1807. As with vol. 2, all the engraved dates coin-

cide with or post-date those on the corresponding plates

in Coloured engravings of heaths. We have not detected

any anachronistic watermarks in this volume; the dates

that can be determined are 1803, 1804 and 1805.

It is probable that this volume was completed in 1 806,

and as it would be odd if it was published before vol. 2,

the two could have been issued together.

Volume 4 has a title page dated 1807. Five plates are

dated 1806 and 26 are dated 1807. One 1809 watermark

date has been detected but could indicate a reprinting of

that plate (Erica canaliculata minor: t. 1 57). As with vols

2 and 3 all the engraved dates coincide with or post-date

those on the corresponding plates in Coloured engrav-

ings of heaths.

These suggest that vol. 4 was completed and perhaps

published in 1 807.

In a catalogue issued in 1813, Andrews stated that he

had ‘finished his various botanical works’ and sought ‘to

remind ... his Patrons’ that they ‘may now complete their

sets’ — Coloured engravings of heaths was stated to com-

prise three volumes and The Heathery just four volumes

(Andrews 1813: [i]). Judging by Andrews’ phraseology,

he deemed these works complete. Thus it seems that pub-

lication of volume 5 of The Heathery and volume 4 of

Coloured engravings of heaths had not been commenced

in 1813.

THE HEATHERY: VOLUMES 5 AND 6

The evidence for much later dates than those stated

for vols 5 (‘1809’) and 6 (‘1804’) is three-fold; first,

dated watermarks [as noted by Kerkham (1988)]; sec-

ond, the dates when the portrayed heathers flowered in

England; third, the dates on which the original drawings

were stated to have been prepared by Andrews.

Bothalia 33,2 (2003)

197

As far as we can ascertain, the sixth volume did not have

a separate, new title page. For some unknown reason,

Andrews re-used the title page of volume 1 but had it mod-

ified by hand, the Roman numeral ‘V' being neatly inserted

before the original numeral ‘I’. Thereby the volume number

was amended from one to six, whereas the date ‘1804'

remained unaltered. He followed the same procedure with

vol. 4 of Coloured engravings of heaths, re-using the title

page of the first volume and inserting by hand the Roman

numeral ‘V’ after the T (see Cleevely & Oliver 2002: 249).

Watermarks

When using watermarks to date publications, several

points have to be borne in mind. Despite the require-

ments of the Paper Act of 1794 that enabled paper manu-

facturers to qualify for exemption from the payment of

Excise Duty on any paper produced for export, they were

not meticulous in changing the date used in their moulds.

The date 1794, in particular, occurs in many publications

printed between that year and 1801. Secondly, in a study

of dated papers, Heywood (1950) concluded that the

average interval between the making of a paper and its

actual use was a little under three years.

Although virtually all the watermarks found in The

Heathery are of Whatman papers, these were made at two

different mills and by two separate businesses. The water-

mark ‘J. WHATMAN’ was retained by William Balston

when he sold the premises of Turkey Mill to the

Hollingworth Brothers in 1806, on the dissolution of the

partnership he had had with them since 1794. The name of

the mill was added to the Whatman name in order that it

could continue to be used by the Hollingworths from 1 807.

Balston briefly used ‘WRB’ upon opening his new

Springfield Mill (one of the first to rely on steam power)

in 1807. After 1814, he adopted various other designa-

tions with the Whatman mark which reflected the man-

agement of the new business. These included ‘J.

Whatman & W. Balston’ and later ‘J. Whatman & W.

Balston & Co.’, followed, after 1822 when his son joined

him, by ‘W. Balston & Son’ (Balston 1954).

The occurrence of the dated watermarks ‘W. Bjalston

& Co. 1815’ and ‘Balston 1815’ in copies of The

Heathery, apart from the date itself, support the thesis

that the publication date of vol. 5 cannot be earlier than

1816. It has been claimed by Balston (1954: 124) that

‘there is only one known instance of a Balston water-

mark after 1816’, although other examples in that work

and our current examination refute this.

As already noted, Kerkham (1988) remarked that

‘vol. 6 must have been published much later as there are

1 826 watermarks in it.' However, it must be kept in mind

that we have detected watermark dates as late as 1 822 in

vol. 1, so we urge treating the watermark evidence with

circumspection.

The following is a summary of the watermark evidence:

Volume 5: in the examined copies, we have detected six

plates with dated watermarks for 181 2, eight text pages and

two plates with 1815, and one text page and two plates

dated 1816; Volume 6: in the examined copies, we have

detected the following dates: 1824 (four: one plate; three

text pages); 1825 (five plates); 1826 (ten; two plates; eight

text pages); 1828 (one plate).

Dates of raising and flowering

All the plants portrayed in The Heatheiy and in

Coloured engraving of heaths were cultivated in English

gardens, and Andrews proclaimed on the title page of

Coloured engraving of heaths (1802: vol. 1) that ‘the

drawings [were] taken from living plants only.’ Given

this, the dates of raising and of flowering are clearly of

importance in determining dates of publication because,

put simply, a plant could not be included in any illustrat-

ed work until it had been raised in an English garden or

nursery and grown until it bloomed. This is especially

significant when it is recognized that many of the

heathers included in vols 5 and 6 of The Heatheiy were

horticultural variants, sometimes artificial hybrids, pro-

duced in England. Some plants were even unique, known

to Andrews only from a single individual.

There is however one difficulty with Andrews’ dates,

as demonstrated by this example. Regarding E. beau-

montia, Andrews stated that ‘This fine new species of

Erica was raised from Cape seed last Autumn, 1827, and

flowered the ensuing summer for the first time at the

Nursery of Messrs Rollisson, Lower Tooting ...’. To raise

seedlings of any heather and get them to bloom within

nine months, which is the approximate interval between

‘late Autumn 1827’ and the ‘ensuing summer’, is impos-

sible. Even with modem glasshouse facilities the mini-

mum time from seedlings appearing to flowering is ±

three years. Thus Andrews' dates of raising are suspect,

although he may only have reported the information that

Messrs Rollisson gave him. As for flowering dates, there

is no apparent reason to doubt their accuracy because

only plants that had bloomed are illustrated.

The dates that Andrews gave in vol. 4 of Coloured

engraving of heaths for plants which he also featured in

vol. 5 of The Heatheiy and which post-date its stated

year of publication, 1 809, are summarized in Table 1 . For

vol. 6, only those which post-date 1819 are summarized.

TABLE 1 . — Dates of Erica plants featured in vols 5 and 6 of The Heathery

198

Bothalia 33,2 (2003)

Drawing dates

Given the relationship between the plates in Coloured

engravings of heaths and The Heathery when Andrews

gave a date for the illustration in the former (Table 2), it

can be argued that this date must also apply to The

Heathery. Andrews dated the following illustrations (see

Cleevely & Oliver 2002).

TABLE 2. — Dates of Erica illustrations in Coloured engravings of heaths

and The Heathery

Thus there are illustrations in vol. 5 that were not pre-

pared until 1816, and the final illustration in vol. 6 was

not completed until the summer of 1828.

CONCLUSION

As long as it is accepted that The Heathery was pub-

lished as a sequence of separate, intact volumes, the

watermarks in vol. 5 indicate a publication date not earli-

er than 1815 and the raising and blooming dates corrobo-

rate that year, whereas the dates on which the plates were

drawn, imply that this volume cannot have been issued

before 1816 when Andrews drew E. splendens (t. 240).

Regarding vol. 6, the watermarks signal that it cannot

have appeared earlier than 1 826, whereas the raising and

blooming dates point to publication late in 1828, a date

that is confirmed by the fact that Andrews only drew E.

undulata (t. 300) in the summer of 1828.

We conclude that the dates of publication of H.C.

Andrews’ The Heathery are: Volume 1: not before June

1 805; Volume 2: not before 1 806; Volume 3: 1 806; Volume

4: 1807; Volume 5: not before 1816; Volume 6: late 1828.

As a consequence, it can be concluded that most, if

not all, of H.C. Andrews’ new names for Erica species

were first published in Coloured engravings of heaths.

ACKNOWLEDGEMENTS

We are grateful to Tony Swann and Walter Middel-

mann for their co-operation, and to the staff of the several

libraries noted for access to copies of The Heathery.

REFERENCES

ANDREWS, H.C. 1794-± 1828. Coloured engravings of heaths. The

author, London. 4 volumes.

ANDREWS, H.C. ‘ 1 804’-‘ 1 809’. The Heathery; or a monograph of the

genus Erica. The author, London. 6 volumes.

ANDREWS, H.C. 1813. A catalogue (raisonne) of the original and

select pictures: now exposed to the public, at the London gallery,

22. Piccadilly, late Bullock's Museum. The author, London.

BALSTON, T. 1954. William Balston, papermaker (1759-1849). Methuen,

London.

CLEEVELY, R.J. & OLIVER, E.G.H. 2002. A preliminary note on the

publication dates of H.C. Andrews’ Coloured engravings of

heaths ( 1 794-1830). Archives of natural history 29: 245-264.

DULFER, H. 1965. Revision der siidafrikanischen Arte der Gattung

Erica L. 2 Teil. Annalen des Natiirhistorischen Museums in

Wien 68: 25-177.

HENREY, B. 1975. British botanical and horticultural literature

before 1800. II. The eighteenth century history. Oxford Uni-

versity Press, London.

HEYWOOD, E. 1951. Watermarks, mainly of the seventeenth and eigh-

teenth centuries. Hilversum.

KERKHAM, A.S. 1988. Southern African botanical literature 1600-

1988. South African Library, Cape Town.

PRITZEL, G.A. 1872. Thesaurus literaturae botanicae, edn 2. Brockhaus,

Leipzig.

STAFLEU, F.A. & MENNEGA, E.A. 1992. Taxonomic literature.

Supplement 1: A-Ba. Koeltz Scientific Books, Konigstein.

STEARN, W.T. 1940. Bibliographical notes: CX Schneevoogt and

Schwegman’s 'leones Plantarum Rariorum’. Journal of Botany

78: 67 (note 7).

Bothalia 33.2: 199-206 (2003)

Embryology and systematic relationships of Kiggelaria (Flacourtiaceae)

E.M.A. STEYN*t, A.E. VAN WYK** and G.F. SMITH*

Keywords: Achariaceae, embryology, Flacourtiaceae, Kiggelaria L., ovule structure, seed coat

ABSTRACT

Kiggelaria L. is endemic to Africa and the only representative of tribe Pangieae (Flacourtiaceae) on the continent.

Molecular genetics, phytochemistry and host-parasite relationships have indicated a relationship between this woody,

pantropical tribe of Flacourtiaceae and a small, trigeneric family of herbaceous, southern African endemics, the

Achariaceae. In the present study, ovule and seed structure in Kiggelaria were investigated and compared with relevant data

recently reported for Achariaceae. Support for an alliance with Achariaceae were found in the presence of anatropous,

bitegmic, sessile ovules with zigzag micropyles, deep-lying embryo sacs covered by an epistase in the ovule and seed,

endotestal-exotegmic seed coat layers, suspensorless embryos and sarcotestal seed with a prominent, plate-like hypostase.

Unlike Achariaceae, Kiggelaria seeds do not have chalazal seed lids, pronounced raphal ridges, a fringe layer, or stomata

in the outer epidermis of the sarcotesta. Structural dissimilarities in seeds of ornithochorous Kiggelaria and myrmeco-

chorous Achariaceae were regarded as adaptations linked to different strategies for seed germination and dispersal.

INTRODUCTION

Kiggelaria L. is a widely distributed, wholly African

genus of dioecious trees and shrubs. It occurs in all the

provinces of the Republic of South Africa and in

Lesotho, Swaziland, Namibia, south tropical Africa and

tropical East Africa as far north as Mount Kilimanjaro

(Killick 1976a). In these areas Kiggelaria represents an

important floristic element of forest in the archipelago-

like Afromontane Region of Endemism (White 1978).

This African phytocorion is of special biogeographical

and evolutionary significance because of its putative

ancient status. Although the genus shows considerable

variability in juvenile and mature vegetative morpholo-

gy, only one polymorphic species is currently recog-

nized, namely Kiggelaria africana L. (Killick 1976b).

In most modem suprageneric classification systems of

Flacourtiaceae, Kiggelaria is placed within the tribe

Pangieae. This tribe of about 10 genera has a disjunct,

pantropical distribution (Wendt 1988) with most genera

reported from Malaysia (Chase et al. 2002). Chiangio-

dendron T. Wendt occurs in southeastern Mexico (Wendt

1988). Baileyoxylon C.T. White is found in Australia and

Kiggelaria is the only representative from Africa. Al-

ternatively, Kiggelaria and other genera of the Pangieae

have been treated as a separate family, Kiggelariaceae

(Savolainen et al. 2000; Soltis et al. 2000). Within

Flacourtiaceae, the Pangieae form a more or less homo-

geneous group that, in terms of generic content, has

remained consistent in taxonomic treatments (Hutchin-

son 1967; Lemke 1988; Takhtajan 1997). On account of

comparative macromorphology, wood anatomy, palynol-

ogy and the distribution of selected classes of chemical

constituents, Lemke ( 1988) regarded the cyanogenic, rel-

atively unspecialized flacourtiaceous tribes Berberidop-

sideae, Erythrospermeae, Pangieae and Oncobeae as

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

** H.G.W.J. Schweickerdt Herbarium, Department of Botany,

University of Pretoria, 0002 Pretoria,

t Corresponding author, e-mail: elsie@nbipre.nbi.ac.za

MS. received: 2002-12-11.

closely related. He suggested that phylogenetic relation-

ships among them and other cyanogenic families of

Violales should be examined more carefully.

A possible phylogenetic link between one such fami-

ly, namely Achariaceae, a family of three monogeneric

genera from southern Africa, and the tribe Pangieae, par-

ticularly Kiggelaria , was first suggested by the breeding

behaviour of a butterfly, Acraea horta (Nymphalidae:

Heliconiinae: Acraeini) in the botanical garden of the

University of Pretoria. This was reported by Dahlgren &

Van Wyk (1988; see also Steyn et al. 2002b). Based on

evidence from molecular biology. Chase et al. (1996)

also suggested a linkage between the herbaceous

Achariaceae and the woody tribe Pangieae. More recent

phylogenetic molecular data reported by e.g. Soltis et al.

(2000) and Savolainen et al. (2000) have provided addi-

tional support for a close association between Acharia

Thunb., Guthriea Bolus (Achariaceae) and Kiggelaria.

The third genus, Ceratiosicyos Nees, was not included in

these investigations. Lately, Chase et al. (2002) proposed

splitting Flacourtiaceae sensu Lemke (1988) and emend-

ing the circumscription of the families Achariaceae

Harms and Salicaceae Mirb. Achariaceae sensu Chase et

al. (2002) thus include Acharieae Benth. & Hook.f.,

Pangieae ( sensu auct.), Erythrospermeae DC. and a

newly described tribe, Lindackerieae Zmarzty (in Chase

et al. 2002: 172, 173), all cyanogenic. The non-cyano-

genic tribes of Flacourtiaceae sensu Lemke [Flacourtieae

DC., Samydeae (Vent.) Dumort., Homalieae (R.Br.)

Dumort., Scolopieae Warb., Prockieae Endl., Abatieae

Hook.f. and Bembicieae Warb.] have been placed with

Saliceae (Salix L. and Populus L.) and Scyphostegieae

(Hutch.) Zmarzty in a much enlarged Salicaceae (Chase

et al. 2002).

Apart from data provided by molecular phylogenet-

ics, phytochemistry and host-parasite relationships, sup-

port for an alliance between Pangieae sensu auct. and

Achariaceae Harms has rarely been offered. Steyn et al.

(2002a, b) pointed out that the development and structure

of mechanical seed coat layers in Achariaceae closely

match those of Kiggelaria and Hydnocarpus Gaertn.

200

Bothalia 33,2 (2003)

(Pangieae) as described by Van Heel (1979). This resem-

blance suggests that a detailed comparison of ovule and

seed characters might reveal additional similarities

between Kiggelario and Achariaceae. However, ernbry-

ological data on Kiggelaria obtainable from literature

were scant and ambiguous; we therefore re-investigated

ovule and seed structure in Kiggelaria africana. The

results of our study are given in the present report.

(adnate adaxial basal scales, according to Chase et al.

2002). The position of the placental bundles in relation to

the petals suggests that they represent the marginal bun-

dles of the congenitally fused carpels. During the matu-

ration of ovules, the space in the locule becomes restrict-

ed. The ovules are pressed against each other and the

ovary wall, and pushed out of alignment with the

micropyles pointing in all possible directions.

MATERIAL AND METHODS

Open flowers and developing fruit were collected

from female trees in a dioecious, natural population of

Kiggelaria africana growing in the National Botanical

Garden, Pretoria, South Africa. To facilitate penetration

of chemicals into ovules and seeds, the thick ovary wall

was partly removed and developing seeds were removed

from the locule. All material was fixed and stored in a 0.1

M cacodylate buffered solution containing 4% formalde-

hyde and 2.5% gluteraldehyde. Following the methods

of Feder & O'Brien (1968), material was dehydrated in

an alcohol series and impregnated with glycol methacry-

late (GMA). All material was imbedded in GMA, sec-

tioned transversely or longitudinally at 2-3 pm and sub-

sequently stained with the periodic acid/Schiff reaction

and toluidine blue O (pH 4.4) by using the protocols of

O'Brien & McCully (1981).

RESULTS

Placentation and orientation of ovules

The unilocular, thick-walled and densely pubescent

ovaries in female flowers (Figure 1) contain many ovules

borne singly or in pairs on four or five parietal placentas.

The sessile ovules are supplied by vascular strands con-

nected to the large vascular bundles lying opposite the

petals with their conspicuous, adaxial nectary glands

Structure of mature ovule

Ovules are anatropous, bitegmic, crassinucellate and

ovoid in shape (± 530 x 400 pm) when the flowers open.

Sagittal sections show that there is no funicle and the

raphe is pronounced (Figure 2A). The vascular bundle of

the raphe branches as it enters the overgrown chalaza,

but the ramifications do not enter the integuments. Both

integuments are multi-layered. The outer is thicker than

the inner, especially at its distal rim where it overtops the

inner integument to take part in the formation of the

slightly zigzag micropyle canal (Figure 2A, B). At anthe-

sis, the outer and inner epidermal layers of the outer

integument are separated by about four layers of thin-

walled, isodiametric mesophyll cells containing many,

small starch grains and occasionally showing periclinal

as well as anticlinal divisions. Periclinal divisions also

occur in the inner epidermal cells of this integument,

whereas the cells of the outer epidermis and the epider-

mis of the chalaza mostly divide anticlinally to form a

layer of radially flattened cells (Figure 2 A, B). The inner

integument is about five layers thick and its cells are

smaller than those of the outer integument, except for the

outer epidermal cells of the inner integument which are

elongate in the direction of the longitudinal axis of the

ovule. A distinct cuticle layer separates the integuments

from each other.

The ovoid, relatively small nucellus with enclosed

embryo sac, lies in the centre of the ovule, occupying

FIGURE 1. — Female flower of Kigge-

laria in t/s (sepals removed),

a, petal with adnate adaxial

nectary gland; b, ovary with

parietal placentae in ante-

petalous position; c, ovule.

Note arrows indicating large

vascular bundles that supply

ovules. Scale bar: 500 pm.

Bothalia 33.2 (2003)

201

about one-third of its length (Figure 2A, B). The micro- tissue by dividing once, periclinally. Below the nucellus

pylar nucellus consists of about eight cell layers and the epidermis, the parietal nucellus stains darkly on account

nucellus epidermis contributes to the formation of this of the thicker walls and large numbers of starch grains in

FIGURE 2. — Ovule and developing seed of Kiggelaria. A, structure of mature ovule in sagittal section. Note thick nucellus cap (white bracket)

and filiform apparatus of synergids (white arrows); B. mature ovule in median 1/s showing zigzag micropyle; C, micropylar part of young

seed during resting stage of zygote; D, suspensorless embryo in micropylar part of developing seed. Curved arrows in B & C indicate

zigzag micropyle; d, raphe; e, embryo sac; f, chalaza; g, antipodal cells; h, nucellus epidermis; i, inner integument; j, inner epidermis of

testa; k, outer epidermis of tegmen; 1, epistase; m, young embryo; n, outer epidermis of inner integument; o, outer integument; s, zygote;

v, vascular bundle. Scale bars: 100 pm.

V«Vv-

202

Bothalia 33,2 (2003)

FIGURE 3. — Seed coat development in

Kiggelaria. A, zygote stage of embryo;

B, when endosperm becomes cellular; C,

dispersed seed, j, inner epidermis of

testa; k, outer epidermis of legmen; n,

nucellus epidermis; p, outer epidermis of

testa; q, succulent mesophyll of sar-

cotesta; r, mesophyll of tegmen; t, inner

epidermis of tegmen; u, darkly stained

layer of sarcotesta. Scale bars: 100 pm.

the cells. The inner layers of the lateral nucellus and the

column-like nucellus below the embryo sac (postament,

according to Shamrov 1998) show a similar staining

reaction.

The pear-shaped embryo sac of the mature ovule thus

lies deeply imbedded in nucellar tissue. The narrow cha-

lazal end contains three small antipodal cells (Figure

2B). The nucleus of the central cell lies in the centre.

Bothalia 33,2 (2003)

203

adjacent to the wall, whereas the egg apparatus occupies

the wider, micropylar region. The position of the two

synergids with darkly stained, well-formed filiform

apparatus (Figure 2A), clearly indicates the distal limits

of the embryo sac. Reports that the tip of the embryo sac

of Kiggelaria breaks through the nucellus tissue and

enters the micropyle (Johri etal. 1992) are not substantiat-

ed by the present study.

Seed and seed coat development

Fertilized ovules enlarge rapidly and endosperm for-

mation precedes embryo development. During the free-

nuclear stage of the endosperm, the zygote remains in a

resting phase directly below the micropyle (Figure 2C)

and is separated from the endostome by the thick- walled

remnants of the nucellus cap that form an epistase. This

tissue plugs the micropyle (Bouman 1984) and it persists

in later stages of embryo development (Figure 2D).

The embryogeny of Kiggelaria was not studied in detail;

pro-embryos were not found in our material and the embryo

could not be typified. When the seed reaches its final size of

about 6 mm, the free-nuclear endospenn becomes cellular.

The embryo then is in the early heart-shaped stage and has

no suspensor (Figure 2D). By the time the capsule splits

open to release the ripe, bright orange-red seeds, the embryo

lies in the centre of the endospenn. has thin, spatulate

cotyledons and is of medium size, i.e. it extends about

halfway up into the endosperm.

Contribution of the outer integument (testa) to the seed coat

The outer integument is strongly multiplicative and

all cell layers contribute to the formation of the mature

seed coat. The outer epidermis divides periclinally once

or twice (Figure 3 A) and the resulting epidermal tissue

eventually acquires unevenly thickened cell walls to

form a collenchymatous pellicle (Figure 3B. C). This

protective layer also covers the chalaza and raphal region

of the seed and contains no stomata. The cells of the meso-

phyll below the epidermis of the integument, raphe and

chalaza divide in various planes to develop into the

multi-layered, thin-walled, succulent tissue of the sar-

cotesta (Figure 3A-C). In the ripe seed, the contents of

the innermost cells of the sarcotesta reacts strongly with

PAS and toluidine blue. These dark-staining cells form a

layer that separates the sarcotesta from the mechanical

tissue (Figure 3B, C). It is in this position that a fringe

layer occurs in the three genera of the Achariaceae

(Steyn et al. 2002a, b), but such a layer is absent in Kig-

gelaria.

The outer integument contributes to the protective,

mechanical tissue of the seed coat. This tissue has a dual

derivation and consists of two sub-layers. The outer layer

originates from the inner epidermis of the outer integu-

ment; the inner layer from the outer epidermis of the

inner integument (see further on). The cells of the inner

epidermis start dividing periclinally just after fertiliza-

tion to form radial rows of cells (Figures 2C; 3A). The

rows later become disrupted, because the outer cells

undergo further divisions in various planes, the inner

cells increase in size and the developing exotegmic fibres

start intruding into the endotestal layers (Figures 2D; 3B,

C). At seed dispersal stage, the large cells of the endotesta

have developed into thick-walled, isodiametric sclereids

(Figure 3C). The outer, smaller cells of the endotesta

remain relatively thin-walled and separate the sclereids

from the layer of dark-staining sarcotestal cells (Figure

3C). At the chalazal end of the seed, the layers of rela-

tively thin-walled, endotestal cells continue into the cha-

laza, surround the vascular tissue and separate the latter

from a thin, plate-like layer of lignified cells that repre-

sents a hypostase (not illustrated).

Contribution of the inner integument (tegmen) to the seed

coat

The inner sub-layer of the mechanical tissue is

exotegmic and originates from the outer epidermis of the

inner integument. The cells divide periclinally and cells

are formed centripetally (Figures 2C; 3A). The deriva-

tives rapidly stretch in a direction parallel to the longitu-

dinal axis of the seed (Figures 2C, D; 3B, C). At this

stage the cuticle between the two developing sub-layers

is still distinct. When the endosperm becomes cellular,

the tips of the elongated exotegmic cells start intruding

into the adjacent sub-layer of endotestal tissue (Figures

2D; 3B) and the cuticle becomes disrupted. At seed dis-

persal stage the endotegmic sub-layer has matured into

thick-walled, longitudinally orientated fibres (Figure

3C). This sub-layer does not continue into the chalaza.

The mesophyll and inner epidermis of the inner

integument do not play a significant role in the structure

of the mature seed coat; the layers disintegrate when the

endosperm tissue matures (Figure 3C). The cuticle be-

tween the nucellus and inner epidermis of the inner

integument, so prominent in seeds of Guthriea (Steyn et

al. 2001), is thin. A pigment layer that, according to Van

Heel (1979), differentiates on the inside of the sclereid

layer in Kiggelaria (also Hydnocarpus) and persists

when the mesophyll disintegrates later on, was not seen

during this investigation.

DISCUSSION

For most of the 80-95 genera traditionally placed in

Flacourtiaceae (e.g. Hutchinson 1967; Lemke 1988) and

recently split in two groups to expand the families

Achariaceae and Salicaceae (Chase et al. 2002), embryo-

logical characters are completely unknown. The meagre,

scattered data used to compile accounts of Flacourtiaceae

in compendia dealing with comparative embryology (e.g.

Davis 1966; Johri et al. 1992) provide no support for the

‘sweeping taxonomic changes’ proposed by Chase et al.

(2002). However, two embryological studies do suggest

that such changes might have merit. Firstly, a recent study

on members of Flacourtiaceae sensu lato with a multi-

whorled androecium (Bernhard & Endress 1999), showed

that the stamens are initiated centrifugally in Flacourtieae

and Scolopieae (non-cyanogenic tribes) as reported for

Populus in Salicaceae Mirb. (Kaul 1995). In Erythrosper-

meae, Oncobeae (except Oncoba) and Pangieae (cyano-

genic tribes) initiation is centripetal or simultaneous.

Achariaceae, however, are characterized by a single whorl

of 3-5 stamens (Dahlgren & Van Wyk 1988; Bernhard

Ovule position and number Parietal, multi-ovular. Parietal, multiovular in Guthriea and Ceratiosicyos, tetra-ovular in Acharia.

Ovule type Anatropous, bitegmic, crassinucellate. Anatropous, bitegmic, crassinucellate.

204

Bothalia 33,2 (2003)

stomata present ( Acharia , Guthriea).

Mesophyll of sarcotesta Succulent outer layers, fringe layer absent in inner layer. Succulent outer layers, inner layer forms fringe layer.

Mechanical layers in seed Endotestal sclereids + exotegmic longitudinal fibres. Endotestal sclereids + exotegmic longitudinal fibres.

Chalazal seed lid Absent. Absent in Ceratiosicyos, present in Acharia, Guthriea.

Dispersal mechanism Autochory + ornithochory. Autochory ( Ceratiosicyos ), autochory + myrmecochory (Acharia, Guthriea).

Bothalia 33,2 (2003)

205

1999). Secondly, an earlier work on seed coat structure by

Corner (1976) indicated that the family should be divided

in two groups, namely a Flacourtia group ( Casearia

Jacq., Flacourtia L'Her., Idesia Maxim., Oncoba Forssk.)

and a Hydnocarpus group ( Flydnocarpus , Kiggelaria ,

Pangium Reinw., Scaphocalyx Ridl.).

Comer (1976) mistakenly interpreted the seeds of all

investigated members of his Hydnocarpus group as

pachychalazal with no contribution of the integuments

towards the formation of mechanical seed coat layers.

Van Heel’s (1974, 1979) studies showed that the seed

coat of only Pangium is pachychalazal; Erythrospermum

Lam., Hydnocarpus and Kiggelaria have seed coats with

endotestal-exotegmic mechanical layers, but the integu-

mental derivation and dual origin of the mechanical lay-

ers are masked by the early disappearance of the cuticle

boundary. In Caloncoba Gilg and Camptostylus Gilg

(Lindackerieae) the cuticle separating the inner and outer

integument also vanishes entirely during seed formation

(Van Heel 1977). Results obtained during the present

study confirm Van Heel’s (1979) observations on

Kiggelaria. Seed coat structure in Pangieae, Erythro-

spermeae and Lindackerieae therefore corresponds to

Achariaceae Harms as described by Steyn et al. (2002a,

b), providing embryological support for emending the

circumscription of Achariaceae as proposed by Chase et

al. (2002).

A second embryological character of Kiggelaria that

needed clarification is the shape of the embryo sac.

Davis (1966) reported that the embryo sac of Kiggelaria

does not become much elongated, while Johri et al.

(1992) found the embryo sac in Kiggelaria similar to that

of Azara Ruiz & Pav. (= Arechavaletaia Speg.) with an

embryo sac breaking through the nucellus and reaching

the exostome of the micropyle. An ‘extra-nucellar

embryo sac’ (Johri et al. 1992: 549) also occurs in

Flacourtia (Johri et al. 1992) and approaches the state

described for Salicaceae Mirb. where the tip of the

embryo sac comes into contact with the integument

(Chamberlain 1897; Maheshwari & Roy 1951; Johri et

al. 1992). This type of elongated embryo sac may well be

characteristic for Salicaceae sensu Chase et al. (2002).

Our results on Kiggelaria show that the tip of the embryo

sac, clearly indicated by the filiform apparatus of the

synergids, remains covered by the nucellus cap (Figure

2C, D) as reported for Achariaceae (Steyn et al. 2001,

2002a, b). The nucellus cap later forms an epistase, sep-

arating the zygote and embryo from the micropyle. The

presence of an epistase is another clear indication that the

tip of the embryo sac stays inside the nucellus. An epistase

is only known in a few angiosperm families (Bouman

1984) and was recently reported for Achariaceae (Steyn

et al. 2001, 2002a).

A detailed comparison of ovule and seed structure in

Kiggelaria and Achariaceae (Table 1) shows additional

noteworthy similarities, such as the contribution of the

nucellus epidermis to nucellus cap formation, sessile

ovules with zigzag micropyles (not Ceratiosicyos), sus-

pensorless embryos and sarcotestal seeds. The latter two

characters may be important markers for Achariaceae

sensu Chase et al. (2002); the embryo of at least Idesia

(Flacourtieae) has a long suspensor (Johri et al. 1992),

while sarcotestal seeds were also reported for Caloncoba

Gilg and Camptostylus Gilg (Lindackerieae) by Van Heel

(1977). In the recently circumscribed Salicaceae the

seeds are not sarcotestal, but are often arillate (Chase et

al. 2002).

With regard to ovule characters, Kiggelaria seems

closer to Acharia and Guthriea than to Ceratiosicyos

(Table 1 : Nos 3, 5, 6 & 7). Some structural dissimilarities

between Kiggelaria and Achariaceae Harms (see Nos 6,

16 & 17) can possibly be linked to specific adaptations

for seed dispersal (see Steyn et al. 2002a for details);

Kiggelaria seeds are bird-dispersed (Palmer & Pitman

1972) whereas Acharia and Guthriea , but not Ceratio-

sicyos, are adapted to dispersal by ants. It is noteworthy

that seeds of Kiggelaria and Ceratiosicyos , both having

relatively large embryos, lack the seed lid found in seeds

of Acharia and Guthriea. The absence of such a lid in

Kiggelaria supports our hypothesis (Steyn et al. 2002a,

b) that this peculiar device is a specific adaptation to seed

germination for the small, slow-maturing embryos in

seeds of Acharia and Guthriea.

ACKNOWLEDGEMENT

We are indebted to the National Botanical Institute,

Pretoria for providing the infrastructure to execute this

study.

REFERENCES

BERNHARD, A 1999. Floral structure and development of Ceratiosicyos

laevis (Achariaceae) and its systematic position. Botanical

Journal of the Linnean Society 131: 103-113.

BERNHARD, A. & ENDRESS, RK. 1999. Androecial development

and systematics in Flacourtiaceae s.l. Plant Systematics and

Evolution 215: 141-155.

BOUMAN, F. 1984. The ovule. In B.M. Johri, Embryology of angio-

sperms. Springer, Berlin.

CHAMBERLAIN. C.J. 1897. Contribution to the life history of Salix.

Botanical Gazette 23: 147-179.

CHASE, M.W., LLEDO, M.D., CRESPO, M.B. & SWENSEN, S.M.

1996. Abstract: ‘When in doubt, put it in the Flacourtiaceae’:

molecular systematics of Flacourtiaceae. American Journal of

Botany 83 (6, Suppl.): 146.

CHASE, M.W., ZMARZTY, S., LLEDO, M.D., WURDACK, K.J.,

SWENSEN, S.M. & FAY, M.F. 2002. When in doubt, put it in

Flacourtiaceae : a molecular phylogenetic analysis based on

plastid rbcL DNA sequences. Kew Bulletin 57: 141-181.

CORNER, E.J.H. 1976. The seeds of dicotyledons. Cambridge University

Press, Cambridge.

DAHLGREN. R. & VAN WYK, A.E. 1988. Structures and relationships

of families endemic to or centred in southern Africa. Monographs

in Systematic Botany 25: 1-94.

DAVIS, G.L. 1966. Systematic embryology of the angiosperms. Wiley,

London.

FEDER, N. & O’BRIEN. T.P. 1968. Plant microtechnique: some prin-

ciples and new methods. American Journal of Botany 55:

123-142.

HUTCHINSON, R. 1967. The genera of flowering plants. Clarendon

Press, Oxford.

JOHRI, B.M., AMBEGAOKAR, K.B. & SRIVASTAVA, PS. 1992.

Comparative embryology of angiosperms, vol. 1. Springer,

London.

KAUL, R.B. 1995. Reproductive structure and organogenesis in a cot-

tonwood, Populus deltoides (Salicaceae). International Journal

of Plant Science 156: 172-180.

KILLICK, D.J.B. 1976a. Kiggelaria africana. The Flowering Plants of

Africa 43: t. 1702.

KILLICK, D.J.B. 1976b. Flacourtiaceae. 4. Kiggelaria. In J.H. Ross, Flora

of southern Africa 22: 60-63. Botanical Research Institute,

Pretoria.

206

Bothalia 33,2 (2003)

LEMKE, D.E. 1988. A synopsis of Flacourtiaceae. Aliso 12: 29—43.

MAHESHWARI, P. & ROY, S.K. 1951. The embryo sac of Salix.

Phytomorphology 1: 70-72.

O'BRIEN^ T.P. & MCCULLY, M.E. 198 1 . The study of plant structure :

principles and selected methods. Termacarpi, Melbourne.

PALMER, E. & PITMAN, N. 1972. Trees of southern Africa. Balkema,

Pretoria.

SAVOLAINEN, V., FAY, M.F., ALBACH, D.C.. BACKLUND, A.. VAN

DER BANK, M„ CAMERON, K.M., JOHNSON, S.A., LLEDO.

M.D., PINTAUD. J.-C., POWELL, M„ SHEAHAN, M.C.,

SOLTIS, D.E., SOLTIS, P.S., WESTON. P, WHITTEN, W.M.,

WURDACK, K.J. & CHASE, M.W. 2000. Phylogeny of the eudi-

cots: a nearly complete familial analysis based on rbch gene

sequences. Kew Bulletin 55: 257-309.

SHAMROV, 1. 1. 1998. Ovule classification in flowering plants — new

approaches and concepts. Botanische Jahrbiicher 120: 377-407.

SOLTIS. D.E., SOLTIS PS., CHASE, M.W., MORT, M.E., ALBACH,

D.C., ZANIS, M„ SAVOLAINEN, V., HAHN, W.H., HOOT,

S.B., FAY, M.F., AXTELL, M„ SWENSEN, S.M., PRINCE,

L.M.. KRESS, W.J., NIXON, K.C. & FERRIS, J.S. 2000.

Angiosperm phylogeny inferred from a combined data set of

1 8S rDNA, rbch, and atpB sequences. Botanical Journal of the

Linnean Society 133: 381 — 46 1 .

STEYN, E.M.A., VAN WYK, A.E. & SMITH, G.F. 2001 . A study of the

ovule and young seed of Guthriea capensis (Achariaceae). South

African Journal of Botany 67: 206-213.

STEYN, E.M.A., VAN WYK, A^E. & SMITH, G.F. 2002a. Ovule, seed

and seedling characters in Acharia (Achariaceae) with evidence

of myrmecochory in the family. South African Journal of

Botany 68: 143-156.

STEYN, E.M.A., VAN WYK, A.E. & SMITH, G.F. 2002b. A study of

ovule-to-seed development in Ceratiosicyos (Achariaceae) and

the systematic position of the genus. Bothalia 32: 201-210.

TAKHTAJAN, A. 1997. Diversity and classification of flowering

plants. Columbia University Press, New York.

VAN HEEL, W.A. 1974. Flowers and fruits in Flacourtiaceae. II. The

seeds of Pangium edule Reinw. Blumea 22: 15-19.

VAN HEEL, W.A. 1977. Flowers and fruits in Flacourtiaceae. III.

Some Oncobeae. Blumea 23: 349-369.

VAN HEEL, W.A. 1979. Flowers and fruits in Flacourtiaceae. IV.

Hydnocarpus spp., Kiggelaria africana L., Casearia spp., Berberi-

dopsis corallina Hook.f. Blumea 25: 513-529.

WENDT, T. 1988. Chiangiodendron (Flacourtiaceae: Pangieae), a new

genus from southeastern Mexico representing a new tribe for

the New World flora. Systematic Botany 13: 435^141.

WHITE, F. 1978. The Afromontane Region. In M.J.A. Werger,

Biogeography and ecology of southern Africa. Junk, The

Hague.

Bothalia 33,2: 207-228 (2003)

Vegetation of the rock habitats of the Sekhukhuneland Centre of Plant

Endemism, South Africa

S.J. SIEBERT*1, A.E. VAN WYK*, G.J. BREDENKAMP* and F. SIEBERT*

Keywords: biodiversity, Braun-Blanquet, conservation, endemism, phytosociology, Sekhukhuneland. syntaxonomy, ultramafic rock habitats

ABSTRACT

A hierarchical classification, description, and ecological and floristic interpretations are presented on the vegetation

types of the ultramafic rock habitats of the Sekhukhuneland Centre of Plant Endemism. Releves were compiled in 100 strati-

fied random plots. A TWINSPAN classification, refined by Braun-Blanquet procedures, revealed 17 plant communities,

which are classified into 13 associations belonging to four proposed alliances. Many new syntaxa are ecologically inter-

preted and described. For each syntaxon, the species richness, endemism and conservation status was determined. Much of

the plant community distribution can be ascribed to specific habitat preference. The floristic and habitat information, pro-

posed classification, general description and vegetation key are provided to aid future identification of conservation areas,

land use planning and research. An ordination (DECORANA) based on floristic data confirmed potential relationships that

could exist between the plant communities and associated habitats and environmental gradients.

INTRODUCTION

Although several phytosociological studies have

been conducted on vegetation types of the northeast-

ern Drakensberg Escarpment and adjacent areas of

Limpopo [Northern Province] and Mpumalanga (Deall

1985; Bloem 1988; Matthews 1991; Burgoyne 1995),

the vegetation of the Sekhukhuneland Centre of Plant

Endemism (SCPE) (Van Wyk & Smith 2001; Siebert

& Van Wyk 2001) has never been studied in detail.

Recently much work has been done to document the

plant diversity and vegetation of this poorly known

micro-regional Centre of Plant Endemism (Siebert et

al. 2002a-e; Siebert et al. 2003). The vegetation of

rock habitats is the last major vegetation type of

Sekhukhuneland to be described. Its extent of occur-

rence is ± 4 000 km: and is characterized by consider-

able diversity in geology (Kent 1980) and physiogra-

phy (Land Type Survey Staff 1987). In South Africa

rock outcrop communities have received very little

attention from botanists and environmentalists in the

past (Bredenkamp & Deutschlander 1995), probably

due to their low agricultural potential. However,

ultramafic rock habitats are floristically noteworthy

and have high conservation significance, because

many endemics with distributions correlated with this

geological substrate occur here (Madulid & Agoo

1995; Meirelles etal. 1999; Reddy et al. 2001; Siebert

et al. 2001).

Various vegetation types have been identified for the

rock habitats of the northeastern Drakensberg Escarp-

ment (Matthews et al. 1991; Matthews et al. 1992) and

western Rustenburg Layered Suite (Coetzee 1975; Van

der Meulen 1979), areas that have a strong floristic rela-

tionship with the SCPE (Siebert 1998). The areas where

the rocky outcrops of the SCPE occur were mapped as

* Department of Botany, University of Pretoria, 0002 Pretoria.

f Present address: Department of Botany, University of Zululand, 3886

Kwadlangezwa.

MS. received: 2002-02-27.

three major Veld Types by Acocks (1988), namely

Mixed Bushveld (18), Sourish Mixed Bushveld (19) and

North-Eastern Sandy Highveld (57). A more generalized

classification of the same region’s vegetation is given by

Low & Rebelo (1996), who recognize one broad

Vegetation Type for the study area, namely Mixed Bush-

veld (18).

The vegetation of the SCPE can be broadly described

as mountain bushveld that forms a mosaic with moist

grassland in the south and semi-arid bushveld in the

north. In this region of undulating hills and mountains, a

predominant and characteristic feature is the scattered

rock habitats. The vegetation of rock habitats described

here only includes those plant communities identified as

the Hippobromus pauciflorus-Rhoicissus tridentata

Rock Outcrop Vegetation by Siebert et al. (2002a). This

vegetation forms a mosaic distribution with the other

major vegetation types of the SCPE. Forty-one of the

Sekhukhuneland endemics/near-endemics and threat-

ened taxa are known to occur in rock habitats (Siebert

1998).

The present paper forms part of a comprehensive veg-

etation and floristic survey of the SCPE (Siebert 2001).

It is envisaged that the identification, classification and

description of the various vegetation units will contribute

to the knowledge of the plant diversity and biological

intricacies of the region. This paper provides ecological

and floristic data of the region’s rocky habitat communi-

ties on ultramafic outcrops and associated habitats, by

characterizing and interpreting the vegetation units.

Classification of the vegetation is basic to the formula-

tion of management policies to co-ordinate and imple-

ment proper land use in bioregional planning activities.

An assessment of the plant diversity, endemism and Red

Data List taxa of the plant communities of the study area

is supplied to aid future conservation actions and ecosys-

tem management strategies. Such information is needed

to build an adequate database of natural features and

other land uses related to sustainable development

(Bedward et al. 1992).

208

Bothalia 33,2 (2003)

STUDY AREA

The study area is located in the SCPE and is defined

in Siebert et al. (2002a) as an area more or less restrict-

ed to the Rustenburg Layered Suite, one of the strati-

graphic units of the Bushveld Complex (Figure 1). The

Rustenburg Layered Suite is known for its concentric

belts of norite and pyroxenite (Visser et al. 1989). In

addition, large quantities of heavy metals such as iron,

chromium and platinum are present in the different zones

of the suite (Schurmann et al. 1998; Viljoen &

Schurmann 1998).

The study area lies in the summer rainfall region and

the mean annual rainfall is 578 mm (South African

Weather Bureau 1998), of which nearly half (48%) is

received between December and February (summer); a

mean of 283 mm for these three months (Erasmus

1985). However, the rainfall pattern is strongly influ-

enced by the local topography and varies from as little

as 400 mm in the Steelpoort and Olifants River valleys

in the north, to an estimated 700 mm on the Leolo

Mountains and the Dwarsrivier Mountains in the south

FIGURE 1. — Extent of occurrence

of rock habitats of Sekhukhu-

neland Centre of Plant Ende-

mism in Limpopo [Northern

Province] and Mpumalanga,

South Africa.

(Siebert 1998). Perhaps the most outstanding climatic

feature of the drier central and northern parts of the

SCPE is that it lies in the rainshadow of the northeastern

Drakensberg Escarpment.

Annual temperatures of the study area range from

-4.5°C to 38°C, with a daily mean of 18.5°C (South Afri-

can Weather Bureau 1998). The northern and western

parts of the study area are on average warmer than the

southern and eastern parts (Siebert 1998). The northern

parts of the region exhibit mean daily temperatures of

28.3°C maximum and 7.2°C minimum. Temperatures

vary at different localities within the area, also correlat-

ing strongly with physiographic features, being higher in

the low-lying valleys and lower on the high-lying plateau

(Buckle 1996).

METHODS

A first approximation of a vegetation classification,

based on the total Holistic data set of 4 1 5 stratified ran-

dom sample plots (Siebert et al. 2002a) was obtained by

the application of Two-Way Indicator Species Analysis

Bothalia 33,2 (2003)

209

(TWINSPAN) (Hill 1979a). This first step of an objec-

tive multivariate classification identified six major vege-

tation types for the Sekhukhuneland Centre (Siebert et

al. 2002a). These results were used to subdivide the data

set into phytosociological tables. One of these tables rep-

resented floristic and habitat data of rock habitats (100

releves) and was again subjected to TWINSPAN. The

resultant classification was further refined by Braun-

Blanquet procedures in the MEGATAB computer pro-

gramme (Hennekens 1996b).

For sampling purposes, stratification was based on

terrain type, aspect and vegetation structure. To stan-

dardize the plot size and to counter the bias of different

scale (Jonsson & Moen 1998), plot size was fixed at

400 m2. Within each sample plot, all species were

recorded and a cover-abundance value was assigned to

each species according to the Braun-Blanquet scale

(Mueller-Dombois & Ellenberg 1974). Plant species

names conform to those of Retief & Herman (1997),

and practice followed in the H.G.W.J. Schweickerdt

Herbarium (PRU), University of Pretoria. Terminology

to describe vegetation structure follows Edwards

(1983). Environmental data recorded for each sample

plot include terrain type (Land Type Survey Staff 1987,

1988, 1989), aspect, slope, geology (Visser et al. 1989),

soil type (Mac Vicar et al. 1991) and rockiness of soil

surface. Longitude and latitude readings were also

recorded for each sample plot using a Global

Positioning System (GPS). All releve data are stored in

the TURBO VEG database (Hennekens 1996a), man-

aged by the Department of Botany, University of

Pretoria (Mucina et al. 2000). Syntaxa are named in

accordance with the Code of Phytosociological Nomen-

clature (Weber et al. 2000).

The ordination algorithm Detrended Correspondence

Analysis (DECORANA) (Hill 1979b) (Figure 2) was

applied to highlight potential gradients in the vegetation,

and the relationship between these plant communities

and the physical environment.

To facilitate the identification of areas of high conser-

vation potential, the alpha diversities of the different

plant communities were calculated. The alpha diversity

(plant species richness) is defined as the number of

species per unit area within a homogeneous community

or the total number of species per community (Whittaker

1977). A 400 nr sample plot was taken as the unit area.

The geographical distribution of all the taxa was ver-

ified at the National Herbarium (PRE), Pretoria, to iden-

tify any taxa endemic/near-endemic to the region

(Siebert 1998). All taxa were also checked against the

Red Data List of southern African plants (Hilton-Taylor

1996) to determine their conservation status.

RESULTS

Classification of vegetation

The floristic data analysis resulted in the identifica-

tion of 17 plant communities of the Hippobromus pauci-

florus-Rhoicissus tridentata Rock Outcrop Vegetation

(Table 1). These were subsequently hierarchically classi-

fied into 13 associations. Four major syntaxa are recog-

nized on the grounds of the physical environment and are

proposed as potential alliances, with all 13 associations

and eight subassociations classified under them. No

macro-climatic or geological variation plays a role in

FIGURE 2. — Relative positions of all releves along first and second axis of ordination of rock habitat vegetation in Sekhukhuneland Centre of

Plant Endemism. Numbers refer to plant communities in Table 1 .

TABLE 1 . — Phytosociological table of vegetation of rock habitats in Sekhukhuneland Centre of Plant Endemism

210

Bothalia 33,2 (2003)

-s—

as os

+ 02 02 02 . + . . . .

06 + 06 + . . + 06 06 06

+ + 06 06 06 . . . .

— . < + 06 06 06

o' + o>-

+ 02 02

g ".s

“III 1 §

1 1 1 1

•3 .Is is •

,.5-S » ' s-5 e 5

•! 5 s 5 S S 2 E S

b | 5 3'^ £ 3. | e 5?'

u §

3 3

f .5

| S s |

a .s .

§■!§,■

1 1 3 O 'g. § 3

II

as

* 1 "g. s "I £

■S S .2 fc 3

a | g s

b ■£ s §

,3 £■•§ ,s a 3 'g .a

^ jS|||ss

: bs o-f-? i s

: o O ijj •?: cAi(oo^^ujcoi:^k)'o5:*T:k,

TABLE 1. — Phytosociological table of vegetation of rock habitats in Sekhukhuneland Centre of Plant Endemism (cont.)

Bothalia 33,2 (2003)

211

— .0606 .06

■ 06 c6 + c6 + 06 c6

•2 to ^ a >>■

— + 06

+

06 06

— 06 06 06 06 06

“ Q iS £ ii $ £ c«iijQfNG6^(jO^ wO 2 to ^

s-g §■

TABLE 1. — Phytosociological table of vegetation of rock habitats in Sekhukhuneland Centre of Plant Endemism (cont.)

212

Bothalia 33,2 (2003)

. + 06 . o6 + 06 + 06 '

06 . 06 C< . + + + Oi

+ + o6 Q6 06 . . . .

06 06 06 . 06 06 O'

06 . . . . 06 06

06 + 06

% %

§ 3 | * It

Q ^ S a. -3 5 S'

I * o = 2 S «>

© .2 -2 ^ T; =:

Z *5 -2 -s £ *5o

■g 1 g J ■§ -s

s- a. g = g 0

?• l.i S 1 1

I &§■£•§ &1

2 ^ ,

C5 cl. -t K. cz)Coo,(3tocCc£coU. !/) Q ^ S 6 G 6 '

LI ».p4| »

S 1 ! E :! S. 3 3 | 3 1 1 ‘S o' I ‘I || 3 I

a js ,g

S t< |

f^.'C

-a a | ^ J §- £?

•3 "I ! 'g § I I =

^"to'coli;21a<oP.

TABLE 1 . — Phytosociological table of vegetation of rock habitats in Sekhukhuneland Centre of Plant Endemism (cont.)

Bothalia 33,2 (2003)

213

. QZ . QZ QZ QZ l

■ QZ qz

. QZ QZ

. QZ QZ

TABLE 1. — Phytosociological table of vegetation of rock habitats in Sekhukhuneland Centre of Plant Endemism (cont.)

214

Bothalia 33,2 (2003)

■ DC — DC DC DC DC •

■ — oc

— . DC DC

DC DC

< DC +

— + DC

“2 ' . DC <

DC + DC

+ .DC . —

"S . DC DC DC

DC < DC + —

tits

,-g -g* s I

= 5 -S J

■ I II HI

s g Si i •§>

I a | 3 a g,-S

•S3 .C 2 O u .2

e .g § f. s u

i A" Ci. w, C C ^

,aj ^3

‘c o

£ 8sj

1 a "s.

,3| i'£,a"§

1-fe i sa«,

I | S t £ '.5 E

j g g. d 3 5 c J

tS’l. 3 '

. £, a §"5 £ §•

SjS-sxlii^st ’§ 3 "f ~ ^ .1' If

TABLE 1. — Phytosociological table of vegetation of rock habitats in Sekhukhuneland Centre of Plant Endemism (cont.)

Bothalia 33,2 (2003)

215

‘ QZ + QZ

< QZ . Oi ■

— CC QZ +

OS . CC

OL +

QC C6

cc + az

. + . & &

. . os cm +

: az + as os

. C£ CC . + o' 0!S

a: + + a*

- QZ

+ 06 cc o£

+ + . oz oc

~ ^2 C

! ^ | .§* 5 '

^ 6'-

i ^ -

1> H. |

gu 5 .g. '

Q ^ O O (zi 6' Q ^ ^ Q Q

216

Bothalia 33,2 (2003)

local differentiation of the plant communities. Plant com-

munities relate to soil type, rockiness and terrain type,

with aspect and slope also of importance. The four major

communities (potential alliances) are distinctive and eas-

ily distinguishable in the field. This might be attributed

to the uniformity of the environmental factors for each of

the major communities, causing a distinct distribution

pattern of habitats and associated vegetation.

The hierarchical classification of the vegetation is

reinforced by the correlation between habitat and plant

communities (Figure 2). The distribution of Sekhukhu-

neland Centre endemic/near-endemic and Red Data List

taxa among various plant communities is listed in Table

2. A summary of selected community attributes is sup-

plied in Table 3. Plant communities of the Hippobromus

pauciflorus-Rhoicissus tridentata Rock Outcrop Vege-

tation recognized in the SCPE are classified as follows:

Four types of rock habitats are recognized for the

region, namely rocky outcrops (I), rocky ridges (II),

rocky flats (III) and rocky refugia (IV). Rocky outcrops

are defined as a mass of exposed stacked boulders on

hills. Rocky ridges are defined as exposed reefs and scat-

tered groups of rocks and boulders on the sides of moun-

tains/hills. Rocky flats are defined as rock beds exposed

at ground level. Rocky refugia are defined as sheltered

rocky areas of scattered boulders around caves, in kloofs

and below cliffs.

I. Setario lindenbergianae-Crotion gratissiini Alliance

of Rocky Outcrops

1. Vepro reflexae—Mimusopetum zeyheri Association

2. Setario lindenbergianae-Crotonetum gratissimi Asso-

ciation

II. Themedo triandrae-Combretion mollis Alliance of

Rocky Ridges

3. Grewio monticolae-Elephantorrhizetum praetermis-

sae Association

4. Setario lindenbergianae-Combretetum mollis Associa-

tion

4. 1 . Setario lindenbergianae-Combretetum mollis cathe-

tosttm edulis Subassociation

4.2. Setario lindenbergianae-Combretetum mollis dios-

pyretosum nitensis Subassociation

4.3. Setario lindenbergianae-Combretetum mollis aloe-

tum aculeatae Subassociation

5. Brachiario serratae—Viticetum wilmsii Association

6. Cymbopogono validi-Brachylaenetum rotundatae Asso-

ciation

7. A loo pretoriensis-Xerophytetum retinervis Association

8. Tephrosio purpureae-Rhoicissetum tridentatae Association

9. Sporobolo fimbriati-Rhamnetum prinoidis Association

10. Eragrostio lehmannianae-Hippobrometum pauciflori

Association

1 0.4. Eragrostio lehmannianae-Hippobrometum pauciflori

enteropogono macrostachyos Subassociation

III. Ursinio nanae-Myrothamnion flabellifoli Alliance

of Rocky Flats

1 1 . Ursinio nanae-Myrothamnetum flabellifoli Association

11.1. Ursinio nanae-Myrothamnetum flabellifoli euphorbie-

tosum cooperi Subassociation

1 1 .2. Ursinio nanae-Myrothamnetum flabellifoli xerophyto-

sum villosae Subassociation

IV. Combreto erythrophylli-Celtion africanae Alliance

of Rocky Refugia

12. Acacio ataxacanthae-Celtidetum africanae Association

12.1. Acacio ataxacanthae-Celtidetum africanae clausene-

tosum anisatae Subassociation

12.2. Acacio ataxacanthae-Celtidetum africanae acacie-

tosum galpinii Subassociation

13. Andrachno ovalis-Allopliylletum transvaalensis Asso-

ciation

Plant community descriptions

The Hippobromus pauciflorus-Rhoicissus tridentata

Rock Outcrop Vegetation is largely restricted to the slopes

and plateaus of undulating ultramafic hills. Surface rocks

are predominant and abundant in these habitats, with

rock percentage varying from 25% on the rocky flats to

more than 50% in the rocky refugia. The vegetation can

structurally be classified into forest/woodland (rocky

refugia), woodland/thicket (rocky outcrops and ridges)

and herbland (rocky flats) (Edwards 1983). The rock

habitats of Sekhukhuneland, like those in other parts of

the world (Madulid & Agoo 1995), constitute an impor-

tant feature which occurs in islands differing significant-

ly from surrounding areas.

I. Setario lindenbergianae-Crotion gratissimi all. nova

hoc loco

Nomenclatural type: Setario lindenbergianae-Crotonetum

gratissimi (holotypus), Association 2 described in this

paper. This proposed alliance is floristically related to

the Croton gratissimus-Setaria lindenbergiana Wood-

land of Van der Meulen (1979).

Species group D (Table 1).

Environmental data : the vegetation of this alliance of

rocky outcrops is a thicket or woodland. It is found on all

aspects of gently sloped (1-5°) rock intrusions on mid-

slopes, scarps and occasionally in valleys (Table 3). Soil

forms are shallow and rocky. The soil surface is covered

by 60-90% of rock with a large average diameter of

2.5-8 m (Table 3).

Diagnostic taxa : the trees Ficus abutilifolia, Homa-

lium dentation, Pouzolzia mixta and Vepris reflexa and

the herbaceous shrubby climbers Asparagus buchananii,

A. intricatus and Rhoicissus sekhukhuniensis.

Dominant/prominent taxa : trees are Croton gratis-

simus and Maytenus undata , the dominant grass is Pani-

cum deustum and frequently occurring herbaceous taxa

include the forb Commelina africana , the fern Pellaea

calomelanos and the succulent Sarcostemma viminale.

Notes on floristic diversity, floristic links with the rest

of the data set are visible in species groups K, Z and AG

(Table 1). These few and weak links support the propos-

al of this major group as an alliance. The mean number

of species encountered per sample plot in this group is

29, with a total number of 107 plant taxa (13 releves)

(Table 3). There are 16 plant taxa of conservation value,

ten are SCPE endemics and six are SCPE near-endemics.

One of the endemics. Euphorbia sekukuniensis, is a Red

Data List taxon assessed as Rare (Table 2). Eight ende-

mics of conservation value are restricted to this commu-

nity of the SCPE.

TABLE 2. — Endemic/near-endemic and Red Data List plant taxa of rock habitats of Sekhukhuneland Centre of Plant Endemism

Bothalia 33,2 (2003)

217

o

o

X

Z

hS

04

oi

£

o

c

Blocks represent taxa restricted to a specific syntaxon.

218

Bothalia 33,2 (2003)

3 &

^ o

^ Q

< -X-

Bothalia 33,2 (2003)

219

1. Vepro reflexae-Mimusopetum zeyheri ass. nova hoc loco

Nomenclatural type: releve 190 (holotypus).

Species group A (Table 1).

Related to the Ilex mitis—Pittosporum viridiflorum Forest

(Coetzee 1975).

Environmental data : the vegetation is a sparse, tall

thicket of rock outcrops on the banks of seasonal streams

and rivers. It is mostly found along watercourses that

flow from the mountains towards the valley of the Steel-

poort River. The habitat has gentle, sloped sides (1-3°)

(Table 3). The dominant soil type is the Bonheim Form,

a melanic A-horizon underlain with a pedocutanic B.

Mean rock size varies from 5.5 to 7 m in diameter and

covers 70-80% of the soil surface.

Diagnostic taxa : the woody species Heteropyxis natal-

ensis , Mimusops zeyheri and Pittosporwn viridiflorum and

the climbers Abrus laevigatas , Rhoicissus tomentosa and

Secamone fHifonnis: there are no diagnostic grasses.

Dominant/prominent taxa: dominant woody species

include Ficus abutilifolia and Vepris reflexa. Cymbopogon

validus and Panicum deustum are common grass species;

prominent forbs are Orthosiphon labiatus and Ruellia patula.

Notes onfloristic diversity: this community is rare in the

SCPE, and exhibits a slight floristic affinity with other rock

habitats of the SCPE in species groups K and Z (Table 1).

The mean number of species per releve is 31, and the total

number of species recorded for the association is 46 (three

releves) (Table 3). Five plant taxa of conservation value

occur (Table 2), namely one SCPE endemic, Rhoicissus

sekhukhuniensis , and four SCPE near-endemics.

2. Setario lindenbergianae-Crotonetum gratissimi ass.

nova hoc loco

Nomenclatural type: releve 195 (holotypus).

Species group B (Table 1 ).

Related to the Croton gratissimus-Setaria lindenbergiana

Woodland (Van der Meulen 1979).

Environmental data: the vegetation is a sparse, short

thicket on rocky outcrops in the Steelpoort River Valley.

The plant community is associated with exposed norite

or pyroxenite outcrops on level to gently sloped (0-1°)

midslopes and scarps of hills. Surface rock cover per-

centage is 60-90%, with the stacked exposed boulders

reaching diameters between 2.5 and 8 m (Table 3). The

dominant soil type is the Mispah Form, indicating very

shallow soils over rock.

Diagnostic taxa: trees/shrubs include Commiphora mar-

lothii , Euphorbia sekukuniensis , Nuxia congesta , Premna

mooiensis and Steganotaenia araliacea ; woody forbs are

Abutilon pycnodon, Ruttya ovata and Turraea obtusifolia,

and the succulents Cyphostemma sulcatum and Tetradenia

brevispicata: Stylochaeton sp. nov. (Siebert 1332) is an

undescribed taxon and a neo-endemic of the SCPE.

Dominant/prominent taxa: Anclropogon schirensis , Cym-

bopogon excavatus , Eragrostis nindensis and Panicum

deustum are the dominant grasses; other prominent plants

are the woody species Barleria rotundifolia, Combretum

molle, Croton gratissimus and Maytenus undata.

Notes onfloristic diversity: this plant community ex-

hibits a slight floristic link with the rocky ridges of the

Centre in species groups K, T and Z (Table 1 ). The mean

number of species encountered per sample plot is 27,

with the total number for this association being 91 (10

releves) (Table 3). A high number of fourteen taxa with

conservation status are recorded for the association

(Table 2). Ten are SCPE endemics, which is the highest

number recorded for any plant community of rock habi-

tats in the SCPE. Four are SCPE near-endemics. Three

Red Data List taxa are recorded. Of all the rock habitats,

this association has the highest number of plant taxa with

conservation value restricted to it (seven).

II. Themedo triandrae-Combretion mollis all. nova hoc loco

Nomenclatural type: Setario lindenbergianae-Combre-

tetum mollis (holotypus), Association 4 described in

this paper. This proposed alliance is floristically relat-

ed to the Combretum molle-Diheteropogon amplectens

Woodland Order of Van der Meulen ( 1 979).

Species group T (Table 1).

Environmental data: in the SCPE this alliance of

rocky ridges is characterized as open to closed moist

bush clumps, with predominantly shallow, black and red

clay soil forms. This vegetation occurs on midslopes and

scarps of undulating ultramafic hills, on varying slopes

of 1-15° on all aspects. Rocks cover 45-80% of the soil

surface and vary in diameter from 1 to 5 m (Table 3).

Diagnostic taxa: the trees Acacia caffra , Olinia emar-

ginata and Scolopia zeyheri , the shrubs Elephantorrhiza

praetermissa and Pavetta zeyheri , the forbs Ruellia patu-

la and R. stenophylla: the grass Cymbopogon excavatus.

Dominant/prominent taxa: other prominent species of

the alliance include the trees Combretum molle, Cussonia

transvaalensis , Euclea crispa and Hippobromus pauci-

florus , and the ground layer is dominated by the grasses

Themeda triandra and Setaria sphacelata.

Notes on floristic diversity: this major group is domi-

nant and floristic relationships exist with all the other

plant communities of rock habitats, indicating that it

forms the core of this vegetation type in the SCPE (Table

1). Associations 4 to 9 represent bush clumps in the

Brachiario serratae-Melhanietum randii Rocky

Grassland of the Roossenekal Subcentre (Siebert et al.

2002d) and Associations 3, 5 and 10 represent bush

clumps in the Loudetio simplicis-Eucleion linearis and

Setario sphacelatae-Acacietum caffrae of the Steelpoort

Subcentre (Siebert et al. 2002b). The mean number of

species encountered per sample plot is ± 35, and the total

number of plant species is 150 taxa (62 releves) (Table

3). Thirty-one taxa of conservation value are part of the

proposed alliance, and 13 are restricted to it (Table 2).

Thirteen SCPE endemics and 15 SCPE near-endemics

were recorded. Nine taxa are listed on the Red Data List.

3. Grewio monticolae-Elephantorrhizetum praetermissae

ass. nova hoc loco

Nomenclatural type: releve 130 (holotypus).

Species group E (Table 1).

Environmental data: this association represents bush

clumps on rocky ridges on warm north and northeast

aspects of norite and pyroxenite hills in the Steelpoort

Subcentre. It occurs on midslopes and scarps on red loam

220

Bothalia 33,2 (2003)

soils of the Glenrosa and Mispah Forms. It covers gentle

to moderate sloped areas (3-7°). Rock cover on the sur-

face is 70-90%, with rocks reaching 2^4.5 m in diameter

(Table 3).

Diagnostic taxa : herbaceous taxa include forbs such

as the herbs Aspilia mossambicensis and Orthosiphon

fruticosus and the succulent Kleinia stapeliiformis’,

grasses are Aristida rhiniochloa and Sporobolus stapfi-

anus: woody species are Englerophytum magalismon-

tanum and Grewia monticola.

Dominant/prominent taxa : important dominant taxa are

shrubs, namely Elephantorrhiza praetermissa , Hippobromus

pauciflorus, Pavetta zeyheri and Xerophyta retinervis (form);

grasses such as Aristida transvaalensis , Panicum deustum

and Themeda triandra are dominant.

Notes on floristic diversity, this association is not

strongly linked with the other associations of its group

and is probably more related to the Englerophytum mag-

alismontanum-Acacia cajfra Mountain Bushveld

(Winterbach et al. 2000). The mean number of species

encountered per sample plot is 30, with a total number of

89 plant taxa (nine releves) (Table 3). Four SCPE en-

demics, seven SCPE near-endemics and one Red Data

List taxon are present in this association (Table 2). None

of the 1 1 taxa of conservation value are restricted to it.

4. Setario lindenbergianae-Combretetum mollis ass. nova

hoc loco

Nomenclatural type: releve 41 (holotypus).

Species group J (Table 1 ).

Related to the Setaria lindenbergiana—Combretum molle

Woodland Community (Bredenkamp et al. 1994).

Environmental data : this association represents bush

clumps on moderate to steep-sloped (1-15°) midslopes

and scarps of norite (sometimes pyroxenite) hills. It

occurs on deeper red and black clay soils of the Mayo

and Milkwood Forms, which are interspersed with shal-

low gravel soils of the Glenrosa Form. Rock cover is

45-80% of the soil surface with sizes between 0.5 and 7

m in diameter (Table 3).

Diagnostic taxa: herbaceous taxa include forbs such

as the herbs Commelina benghalensis and Orthosiphon

labiatus and the succulent Tetradenia riparia ; the grass

is Eragrostis chloromelas ; woody species are Cussonia

paniculate i, Dombeya rotundifolia , Ficus ingens , Grewia

occidentalis and Seemannaralia gerrarclii.

Dominant/prominent taxa : important dominant taxa

are trees, namely Allophyllus africanus, Apodytes dimi-

eliata, Croton gratissimus , Combretum mode and Cussonia

transvaalensis ; grasses such as Cymbopogon excavatus ,

Setaria lindenbergiana, S. sphacelata and Themeela trian-

elra are dominant.

Notes on floristic diversity: the association has a

species combination typical for rocky outcrops in species

group K, shared with Associations 1, 2 and 3 (Table 1).

In species group P it shares a species combination typi-

cal of rocky ridges with Association 5 to 8 (Table 1). This

community is also unique in that it shares forest species

with Associations 12 and 13 in species group AF (Table

I ). The mean number of species encountered per sample

plot is ± 44, the highest mean number recorded per releve

for any of the major vegetation units of rock habitats in

the SCPE. A total number of 1 22 plant taxa were record-

ed (22 releves) (Table 3) of which 20 have conservation

status (the highest number for any association of SCPE

rocky habitats). Nine SCPE endemics, eight near-

endemics (the highest number for rocky habitats) and six

Red Data List taxa (the highest number for rocky habi-

tats) are recorded for this association (Table 2). A high

number of six taxa is restricted to it.

4.1. Setario lindenbergianae-Combretetum mollis cathe-

tosum edulis subass. nova hoc loco

Nomenclatural type: releve 41 (holotypus).

Species group F (Table 1 ).

Environmental data: this sub-association represents

dry, warm bush clumps on northern aspects of hills. It

occurs on deeper red and black clay soils of the Mayo

and Milkwood Forms and lies on moderately sloped

(5-15°) midslopes and scarps. Rock cover on the surface

is 45-50%, with rock sizes between 0.5 and 1 m in diame-

ter (Table 3).

Diagnostic taxa: trees include Acacia robusta , Catha

edulis , Euphorbia ingens , Ficus craterostoma , Schrebera

alata and Sclerocarya birrea: herbaceous taxa include the

forbs Commelina e recta. Helichrysum intricatum, Hy-

poestes aristata and Sanseviera hyeicinthoides: the grasses

are Aristida canescens and Eragrostis heteromera.

Dominant/prominent taxa: conspicuous dominant taxa

are trees such as Acacia ataxacantha , Hippobromus pauci-

florus and Ziziphus mucronata, and grasses such as Panicum

deustum, Setaria sphacelata and Themeda triandra.

Notes on floristic diversity: the subassociation has a

grassland-savanna species combination in species group

H which it shares with Subassociation 4.2 (Table 1). The

mean number of species encountered per sample plot is

46, together with Subassociation 4.2, the highest number

recorded per releve for any of the vegetation units of

rock habitats. A total number of 105 plant taxa were

recorded (eight releves) (Table 3). Five SCPE endemics,

of which two are Red Data List taxa namely Ele-

phantorrhiza praetermissa and Zantedeschia pentlandii,

and five near-endemics are found in this subassociation

(Table 2).

4.2. Setario lindenbergianae-Combretetum mollis diospy-

retosum nitensis subass. nova hoc loco

Nomenclatural type: releve 81 (holotypus).

Species group G (Table 1).

Environmental data: see description of Subassocia-

tion 4.1. This subassociation represents bush clumps of

rocky ridges, which is restricted to grassland on moist,

cool southern aspects (Table 3).

Diagnostic taxa: predominantly woody ones, namely

Buddleja auriculata, B. salviifolia, Diospyros lycioides

subsp. nitens, Jasminum quinatum, Rhus rigida and Triaspis

glaucophylla ; the only diagnostic forb is Pupalia lappacea.

Dominant/prominent taxa: conspicuous woody species

are Apodytes dimidiata, Combretum mode, Hippobromus

pauciflorus and Rhoicissus tridentata: grasses include

Heteropogon contortus, Panicum deustum, Setaria sphace-

lata and Themeda triandra.

Bothalia 33,2 (2003)

221

Notes on floristic diversity. Species group H (Table 1)

shows a strong floristic connection with Subassociation 4. 1

due to the similar geographical distribution, just on different

aspects of the same hills. The mean number of species

encountered per sample plot is 46, the highest mean number

recorded per releve in the data set. The total number of plant

species for this subassociation is 1 1 0 (seven releves), the rich-

est diversity of species recorded for any rock habitat in the

SCPE (Table 3). Of the 10 taxa of conservation value in this

subassociation, four are SCPE endemics and five are SCPE

near-endemics. Of these, three are Red Data List taxa (Table

2). Three taxa are restricted to the subassociation, namely

Berkheya insignis (endemic form), Eucomis montana (Rare

in the Red Data List) and Gnidia cajfra (endemic form).

4.3. Setcirio lindenbergianae-Combretetum mollis oloetum

aculeatae subass. nova hoc loco

Nomenclatural type: releve 20 (holotypus).

Species group I (Table 1).

Environmental data: a subassociation dominated by a

dense herbaceous cover on level ridges situated on mid-

slopes and scarps of grass-covered norite hills. It occurs

on red clay soils of the Mispah Form, with the soil sur-

face covered by 70-80% rock of a relatively large size of

4 to 7 m in diameter (Table 3).

Diagnostic taxa : the fern Cheilanthes hirta : the succu-

lents Aloe aculeata , Kalanchoe rotundifolia , Plectranthus

xerophilus and Sansevieria aethiopica: the forbs, Gloriosa

superba and Tripteris auriculata , dominate the communi-

ty; two undescribed Cyphostemma species; Sporobolus

ioclados and Trachypogon spicatus are the grasses.

Dominant/prominent taxa: woody species are the trees

Barleria rotundifolia, Catha transvaalensis, Croton gratis-

simus, Kirkia wilmsii and the small shrub Chrysanthemoides

monilifera: conspicuous grasses include Andropogon

schirensis and Eragrostis nindensis.

Notes on floristic diversity: a noteworthy floristic

relationship exists with Association 2 in species group C

(Table 1), which can be ascribed to occurrence on the

scarps of hills. The mean number of species encountered

per sample plot is 39 and the total number for the sub-

association is 103 (seven releves) (Table 3). Six SCPE

endemics, six near-endemics and five Red Data List taxa

are recorded (Table 2). Of these 14 taxa of conservation

value, three near-endemics are restricted to it, namely

Aloe reitzii var. reitzii (Indeterminate in Red Data List),

Chlorophytum cyperaceum and Plectranthus xerophilus.

5. Brachiario serratae-Viticetum wilmsii ass. nova hoc loco

Nomenclatural type: releve 71 (holotypus).

Species group L (Table 1).

Environmental data: this association is a dry bush

clump of any aspect, situated on exposed iron-rich norite

and magnetite ridges, on midslopes and scarps of hills. It

occurs on red and black clay soils of the Mispah Form

(ortic A-horizon) and Milkwood Form (melanic A-hori-

zon) underlain by hard rock. The soil surface is covered by

50-60% rock, of an average size of 2.5-3 m in diameter

(Table 3). Slope of the habitat is usually moderate (3-7°).

Diagnostic taxa: dominant species are forbs such as

Rhynchosia spectabilis, Ruellia cordatci and Pearsonia aris-

tata ; the succulent Aloe verecunda: the sedge Bulbostylis

burchellii and the grasses Aristida junciformis, Brachiario

serrcita and Tristachya rehmannii.

Dominant/prominent taxa: woody species are Apodytes

dimidiata , Catha transvaalensis , Olea capensis subsp. ener-

vis and Vitex obovata subsp. wilmsii', conspicuous taxa

include Aloe castanea and Sphedamnocarpus pruriens.

Notes on floristic diversity: the community has a marked

floristic grassland affinity in species group X with Asso-

ciation 1 1 (Table 1 ). The mean number of species encoun-

tered per sample plot is 40, with the total number for this

association being relatively high at 109 species (seven

releves) (Table 3). Six SCPE endemics and six SCPE near-

endemics, of which three are Red Data List taxa, were

recorded (Table 2).

6. Cymbopogono validi-Brachylaenetum rotundatae ass.

nova hoc loco

Nomenclatural type: releve 51 (holotypus).

Species group M (Table 1).

Environmental data: this association represents bush

clumps in grassland on all aspects of hills with norite and

ferrogabbro rocky ridges. It usually occurs at higher alti-

tudes than the other associations, and is found on mid-

slopes and scarps on red clay soils of the Mayo and

Mispah Forms. It lies on gently sloped areas (1-5°).

Rock cover on the surface is 60-80%, with rocks reach-

ing an average size of 3-5 m in diameter (Table 2).

Diagnostic taxa: herbaceous taxa include the forbs

Pachvcarpus transvaalensis, Pearsonia sessilifolia, Rhyn-

chosia hirta, Senecio oxyriifolius and Solatium supinunr,

woody species include the tree Brachylaena rotundata, the

shrub Grewia villosa and the bushy Eelicia filifolia.

Dominant/prominent taxa: important dominant taxa are

trees such as Catha transvaalensis, Olea capensis subsp.

enervis and Ziziphus mucronata, and grasses such as Cym-

bopogon excavatus, C. validus, Heteropogon contortus and

Themeda triandra.

Notes on floristic diversity: no noteworthy floristic

links are evident with associations outside the major

group (alliance). The mean number of species encoun-

tered per sample plot is 34, with a high total number of

109 plant taxa (seven releves) (Table 3). Five SCPE

endemics and five near-endemics, of which three are Red

Data List taxa, are found in this association (Table 2).

7. Aloo pretoriensis-Xerophytetum retinervis ass. nova hoc

loco

Nomenclatural type: releve 136 (holotypus).

Species group N (Table 1).

Related to the Zantedeschio pentlandi-Aloetum castaneae

(Siebert et al. 2002d).

Environmental data: this association is an open,

sparse bush clump of norite and pyroxenite ridges, on

midslopes and scarps of hills. It occurs on black and red

clay soils of the Glenrosa and Mispah Forms on cool

south and southeast aspects. The soil surface is covered

by 60-80% rock of an average size of 2. 5-3. 5 m in diam-

eter (Table 3). Slope of the habitat is usually 1-5°.

Diagnostic taxa: dominated by forbs such as Con-

volvulus sagittatus, Dalechampia galpinii, Gnidia varia-

222

Bothalia 33,2 (2003)

bilis, Jatropha latifolia and Justicia protracta: the diag-

nostic shrub is Gymnosporia buxifolia ; the succulent Aloe

pretoriensis .

Dominant/prominent taxa: woody species include the

shrubs Hippobromus pauciflorus, Pavetta sp. nov. and Vitex

obovata subsp. wilmsiv, important conspicuous grasses

include Aristida transvaalensis , Cymbopogon excavatus and

Themeda triandra.

Notes onfloristic diversity, the association has sever-

al floristic relationships with the rest of the data set. The

mean number of species encountered per sample plot is

33, with 89 species the total number for this association

(six releves) (Table 3). Of the 12 taxa of conservation

value, no taxa are restricted to it. Six SCPE endemics and

five SCPE near-endemics were recorded, of which four

are Red Data List taxa (Table 2).

8. Tephrosio purpureae-Rhoicissetum tridentatae ass. nova

hoc loco

Nomenclatural type: releve 320 (holotypus).

Species group O (Table 1).

Environmental data: this vegetation type is a typical

ecotone between Brachiario serratae-Melhanietum randii

Rocky Grassland and Themedo triandrae-Combretion

mollis of Rocky Ridges in the southern region of the

SCPE. It is open shrubland in moist grassland on mid-

slopes and scarps of undulating norite or pyroxenite hills.

The habitat has a level slope of 1-3°, restricted to south

and west aspects. Soils are typical red and black clays of

the Mayo and Milkwood Forms. Rock size is 1-3.5 nr in

diameter and cover 50-70% of the soil surface (Table 3).

Diagnostic taxa : the grassland forb species Dioscorea syl-

vatica, Helichrysum albilanatum , Rhynchosia minima ,

Tephrosia purpurea and Zomia linearis: the grasses Digitaria

argyrograpta, Eragrostis curvula and Hyparrhenia filipendu-

la\ the trees Canthium mundianum , Rhus sekhukhuniensis

and Rhus discolor (suffrutex).

Domincmt/prominent taxa : the trees/shrubs Euclea

crispa , Hippobromus pauciflorus, Olea capensis subsp.

enervis, Rhoicissus tridentata and R. sp. nov.; the grass

Cymbopogon excavatus.

Notes on floristic diversity: this association’s floristic

relationships are typical for the alliance (Table 1 ). The

mean number of species encountered per sample plot is

34 and the total number of plant species for this associ-

ation is 75 (three releves) (Table 3). A high number of

taxa of conservation value are found in this association

( 1 3), and of these seven are SCPE endemics and six are

SCPE near-endemics. Four of these are Red Data List

taxa (Table 2). Two taxa with conservation status are

restricted to the association, namely the near-endemic

Helichrysum albilanatum and the Rare (R) endemic

Rhus sekhukhuniensis.

9. Sporobolo fiimbriati-Rhamnetum prinoidis ass. nova hoc

loco

Nomenclatural type: releve 4 (holotypus).

Species group Q (Table 1 ).

Environmental data: this is a distinct association of

moist riverbank thicket on raised banks of mountain

streams in the valleys between undulating grass-covered

norite and pyroxenite hills. It lies on a gentle slope of

1-3°. Soils are characteristically a moist ‘humus-rich’

sandy loam on a rocky substrate. Approximately 60-70%

of the soil surface is covered by rocks, with a size of

3. 5-5. 5 m in diameter (Table 3).

Diagnostic taxa: tree species are prominent, namely

Cassinopsis ilicifolia, Leucosidea sericea and Rhamnus

prinoides: forbs are Freesia laxa, Kalanclioe paniculata

and Thunbergia atriplicifolia : a sedge Scleria dieterlenii

and the grass Sporobolus fimbriatus.

Dominant/prominent taxa: other important trees are

Chionanthus foveolatus and Olinia emarginata ; the gras-

ses are Aristida transvaalensis , Cymbopogon validus,

Eragrostis racemosa and Panicum deustum.

Notes onfloristic diversity: the association follows the

Holistic affinities of the alliance, but is characterized by

the absence of the taxa in species group P and Y due to

the presence of moist, lowland soils (Table 1). The mean

number of species encountered per sample plot in this

association is 34, with the total number of 64 plant

species (five releves) (Table 3). Four taxa of interest

occur in this association, namely two SCPE endemics

and two near-endemics (Table 2).

10. Eragrostio lehmannianae-Hippobrometum pauciflori

Association (Siebert et al. 2002b)

10.1. Eragrostio lehmannianae-Hippobrometum pauciflori

rhoetosum batoplivllae Subassociation (Siebert et al.

2002b)

10.2. Eragrostio lehmannianae-Hippobrometum pauciflori

sorgetosum bicoloris Subassociation (Siebert et al.

2002b)

10.3. Eragrostio lehmannianae-Hippobrometum pauciflori

eucleetosum crispae Subassociation (Siebert et al.

2002b)

10.4. Eragrostio lehmannianae-Hippobrometum pauciflori

enteropogono macrostachyos subass. nova hoc loco

Nomenclatural type: releve 334 (holotypus).

Species group S (Table 1 ).

Environmental data: this vegetation type represents

transformed savanna on cool, south and east aspects of

norite hills. It covers moderate midslopes and scarps

(3-9°). The community is restricted to sandy loam soils

in the northern parts of the SCPE. Approximately

2CM-0% of the soil surface is covered by rocks, with a

diameter of >500 mm (Table 2).

Diagnostic taxa: the shrubby Psiadia punctulata ; the

grasses Cynodon dactylon, Enteropogon macrostachys,

Eragrostis lehmanniana and Panicum coloration.

Dominant/prominent taxa: prominent trees are Acacia

caffra , Euclea crispa, Hippobromus pauciflorus and Rhoi-

cissus tridentata.

Notes on floristic diversity: no clear-cut floristic links

exist with other plant communities of rocky ridges and it

is characterized by the absence of the taxa of rocky habi-

tats listed in species group Y (Table 1). Intense harvesting

of wood and overgrazing of the veld is indicated by the

low frequency of taxa in species groups T, Z and AG

Bothalia 33,2 (2003)

223

(Table 1). The mean number of species encountered per

sample plot is 20 and the total number is 35 (three releves)

(Table 3). These numbers are the lowest recorded for the

rock habitats of the SCPE. None of the five plant taxa with

conservation value are restricted to it and comprise three

SCPE endemics and two near-endemics, of which one is a

locally common Red Data List taxon (Table 2).

III. Ursinio nanae-Myrothamnion flabellifoli all. nova

hoc loco

Nomenclatural type: Ursinio nanae-Myrothamnetum flabelli-

foli (holotypus). Association 11 described in this paper.

This proposed alliance is floristically related to the

Myrothamnus flabellifolius-Ursinia nana Community

of Smit et al. (1997).

Species group W (Table 1).

Environmental data : alliance of rocky flats on foot-

slopes. midslopes and scarps of predominantly ultramaf-

ic hills and to a lesser extent also hills of the Transvaal

Sequence. The habitat occurs on all aspects and is gently

to moderately sloped (1-9°). Approximately 60-90% of

the soil surface is covered by flat rocks exposed at

ground level with a relatively large diameter of >10 m

(Table 3). Soils are sandy and ‘humus-rich' in hollows

and fissures in the bedrock.

Diagnostic taxa : see description of Association 1 1 .

Dominant/prominent taxa: see description of Associa-

tion 11.

Notes on floristic diversity: a strong floristic affinity

exists with the alliance of rocky ridges, which is con-

firmed by species groups Y and Z (Table 1). The mean

number of species encountered per sample plot is ± 29, and

the total number of plant species is 83 taxa (14 releves)

(Table 3). This alliance has 12 plant taxa of conservation

value, of which three are SCPE endemics and seven near-

endemics. Six Red Data List taxa occur. Together with

Association 4, this is the highest number of Red Data List

taxa recorded for rock habitats in the SCPE. None of these

taxa are restricted to the proposed alliance.

11. Ursinio nanae-Myrothamnetum flabellifoli ass. nova hoc

loco

Nomenclatural type: releve 35 (holotypus).

Species group W (Table 1).

Related to the Myrothamnus flabellifolius-Ursinia nana

Community (Smit et al. 1997).

Environmental data: the vegetation structure is shrub-

by and grassy and widespread throughout hillsides of the

study area and variations of this association are probably

widespread along the north-eastern escarpment. It occurs

on all aspects of footslopes. midslopes and scarps. It lies

on gentle to moderate slopes (1-9°) and is found pre-

dominantly on moist, ‘humus-rich' sandy soils. Approxi-

mately 60-90% of the soil surface is covered by rocks,

with a mean size of >10 m in diameter (Table 3).

Diagnostic taxa: dominated by forbs including Cras-

sula swaziensis, Craterostigma wilmsii , Oldenlandia herba-

cea and Pearsonia cajanifolia; grasses are Aristida ad-

scensionis , Eragrostis capensis , E. pseudosclerantha and

Melinis repens.

Dominant/prominent taxa: the fern Pellaea calome-

lanos\ the shrubby Xerophyta retinerx’is ; the succulents

Aloe castanea and Crassula sarcocaulis ; the grasses

Aristida transvaalensis and Eragrostis racemosa.

Notes on floristic diversity: the same as the alliance.

Association is common throughout the Bankenveld.

11.1. Ursinio nanae-Myrothamnetum flabellifoli euphor-

bietosum cooperi subass. nova hoc loco

Nomenclatural type: releve 35 (holotypus).

Species group U (Table 1).

Environmental data: widespread throughout the

Roossenekal Subcentre on north, south and west aspects

of footslopes, midslopes and scarps of undulating norite

hills. It lies on gentle slopes (1-3°) and is found pre-

dominantly on sandy soils. Approximately 70-80% of

the soil surface is covered by rocks, with a mean diame-

ter of >10 m (Table 3).

Diagnostic taxa: herbaceous species are the fern

Cheilanthes involuta , the geophyte Stylochaeton natal-

ensis, and the forbs Dioscorea dregeana and

Orthosiphon amabilis; trees/shrubs include Vangueria

infausta and a short-stemmed form of the succulent

Euphorbia cooperi ; grasses are Aristida scabrivalvis and

Microchloa caffra.

Dominant/prominent taxa: the succulent Crassula

swaziensis and woody Myrothamnus flabellifolius are domi-

nant forbs; prominent shrubs are Euclea crispa , Mundulea

sericea and Rhoicissus tridentata ; frequently occurring

grasses are Aristida transvaalensis , Eragrostis pseudoscler-

antha, E. racemosa and Heteropogon contortus.

Notes on floristic diversity: a strong floristic relation-

ship exists with Subassociation 11.2 in species group W

and Association 5 species group X (Table 1). The mean

number of species encountered per sample plot is 30 and

the total number of plant species is 66 taxa (six releves)

(Table 3). None of the ten plant taxa of conservation

value, namely two SCPE endemics, six near-endemics

and five Red Data List taxa, are restricted to the sub-

association (Table 2).

11.2. Ursinio nanae-Myrothamnetum flabellifoli xerophy-

tosum villosae subass. nova hoc loco

Nomenclatural type: releve 97 (holotypus).

Species group V (Table 1 ).

Environmental data: this community is widespread

throughout the SCPE. It occurs on rocky flats with

‘humus-rich’, sandy soils. It covers moderately sloped

footslopes, midslopes and scarps of 3-9° on all aspects

of undulating norite and pyroxenite hills. Approximately

60-90% of the soil surface is covered by large rocks,

with a mean size of >10 m in diameter (Table 3).

Diagnostic taxa: herbs are most diagnostic and include

the fem Cheilanthes eckloniana and fern ally Selaginella

dregei , the forbs Kedrostis foetidissima, Thesium burkei and

Xerophyta villosa: the succulents Crassula alba , Euphorbia

schinzii , Kalanchoe luciae and Kleinia long flora: Rhus

wilmsii is a woody suffrutex.

Dominant/prominent taxa: taxa of importance are the

grasses Aristida transvaalensis and Melinis nenhglumis ;

the succulents Aloe castanea and Crassula sarcocaulis

are conspicuous members.

224

Bothalia 33,2 (2003)

Notes on floristic diversity, floristic relationships are

similar' to that of Subassociation 11.1 (Table 1). The

mean number of species encountered per sample plot is

27 and the total number of plant species is 75 taxa (eight

releves) (Table 3). There are nine taxa of conservation

value in the subassociation, namely three SCPE ende-

mics, five near-endemics and four Red Data List taxa

(Table 2).

IV. Combreto erythrophylli-Celtion africanae (Siebert et

al. 2002c)

Species group AF (Table 1).

Environmental data : this well-documented alliance

(Coetzee 1975; Van der Meulen 1979; Matthews et al.

1992) represents forests and dense woodlands in rocky

refugia of the SCPE. It is a rare vegetation type and can

be found on southern aspects of valleys, and mountain

footslopes, midslopes and crests. The habitat is charac-

terized by large norite boulders of minimum 2 m high

and the mean rock diameter ± 0.5-2. 5 m and covers ±

10-70% of the soil surface. It is characterized by gentle

to moderate slopes ( 1—7°). Soil types are a red or black

clay base on unconsolidated material and include the

Mayo (lithocutanic B-horizon) and the Oakleaf (neocu-

tanic B-horizon) Forms.

Diagnostic taxa : the trees Calodendrum capense and

Celtis africana, the shrub Diospyros whyteana and the

succulent Aloe arborescens.

Dominant/prominent taxa: prominent taxa include the

woody species Acacia ataxacantha , Allophylhis africanus ,

Ehretia rigida , Halleria lucida, Hippobromus pauciflorus

and Ziziphus mucronata: Panicum deustum is the domi-

nant grass.

Notes on floristic diversity: a strong floristic affinity

exists with Association 4 in species group AF (Table 1),

which indicates its relationship with the tall bush clumps of

rocky ridges. The mean number of species encountered per

sample plot is ± 36 and the total number of plant species is

79 taxa (1 1 releves) (Table 3). There are five taxa of con-

servation value associated with this major group, namely

two SCPE endemics and three SCPE near-endemics (Table

2). Of these taxa, an undescribed Gymnosporia, is restrict-

ed to the alliance.

12. Acacio ataxacanthae-Celtidetum africanae (Matthews et

al. 1992)

12.1. Acacio ataxacanthae-Celtidetum africanae clauseneto-

sum anisatae subass. nova hoc loco

Nomenclatural type: releve 67 (holotypus).

Species group AA (Table 1 ).

Environmental data: in the Roossenekal Subcentre this

subassociation represents wooded rocky refugia, mostly

associated with boulders around caves, in kloofs and below

cliffs, or stone walls of old kraals. It is a vegetation unit on

red clay soils of the Mayo and Oakleaf Forms. These units

occur on footslopes and midslopes of undulating norite

hills. The gentle slopes vary from 3-7° and east-south-west

aspects are predominant. Rock cover percentage varies

from 10 to 40% and rock diameter is 1—1.5 m (Table 3).

Diagnostic taxa: trees are most dominant and include

Clausena anisata, Clerodendrum glabrum, C. myricoides,

Ficus thonningi , Obetia tenax and the succulent Aloe mar-

lothii: forbs are Abutilon austro-africanum, Cyathula cylin-

drica, Hermannia floribunda and Scadoxus puniceus;

grasses include Brachiaria brizantha, Digitaria san-

guinalis, Setaria verticillata and Urochloa mossambicensis.

Dominant/prominent taxa: important dominant taxa

include the woody species Acacia ataxacantha , Allophylhis

transvaalensis , Celtis africana and Diospyros whyteana , the

forb Pavonia burchellii, and the grasses Panicum deustum

and P. maximum.

Notes on floristic diversity: strong floristic links exist

with Subassociation 12.2 in species group AC and a spe-

cific Afromontane link with Association 13 in species

group AE (Table 1 ). Species shared with other rock habi-

tats are few. The mean number of species encountered

per sample plot is a high 41 (Table 3). The total number

of plant species for this subassociation is 68 (six releves).

Two taxa of conservation value occurs in this subassoci-

ation, namely an undescribed endemic Cyphostemma

species (Siebert 1383) and the near-endemic Euphorbia

lydenburgensis.

12.2. Acacio ataxacanthae-Celtidetum africanae acacieto-

sum galpinii subass. nova hoc loco

Nomenclatural type: releve 182 (holotypus).

Species group AB (Table 1).

Environmental data: this vegetation type is slightly

transformed woodlands of rocky banks along the larger

rivers in the valleys. The habitat lies between norite out-

crops on black and red clay soils derived from alluvium.

It lies on gentle slopes of 3-5°. Soils are predominantly

the Mayo and Oakleaf Forms. Approximately 20-70% of

the soil surface is covered by rocks, with an average

diameter of 0.5-2. 5 m (Table 3).

Diagnostic taxa: the woody species Acacia galpinii ,

Combretum erythrophyllum. Ficus sur, Flueggea virosa ,

Melia azedarach (naturalized alien) and Spirostachys

africana ; forbs are Achyranthes aspera, Barleria obtusa,

and the climbers Cardiospermum corindum and Secamone

acutifolia.

Dominant/prominent taxa: Celtis africana and Schotia

brachypetala are prominent trees; important dominant

grasses include Panicum deustum and P. maximum.

Notes on floristic diversity: the subassociation shows

the same floristic links as Subassociation 12.1 (Table 1).

The mean number of species encountered per sample

plot is 33 and the total number is 64 (three releves)

(Table 3). It has two taxa of conservation value, namely

a locally common near-endemic and an undescribed en-

demic Gymnosporia (Siebert 458) (Table 2).

13. Andrachno ovalis-Allophylletum transvaalensis ass. nova

hoc loco

Nomenclatural type: releve 406 (holotypus).

Species group AD (Table 1).

Environmental data: this association comprises two

relict Afromontane Forest patches on the summit of the

Leolo Mountains ( I 800 m a.s.l.). It is associated with norite

substrates and boulders of 2-6 m high. The habitat has a

southern aspect and a gentle slope of 1-3°. Approximately

20-40% of the soil surface is covered by rocks with a mean

Bothalia 33.2 (2003)

225

diameter of 500-750 mm (Table 3), which can be up to 3 m

high. Soil is black clay of the Oakleaf Form.

Diagnostic taxa : dominated by woody species, name-

ly Andrachne ovalis , Gymnosporia sp. nov. ( Van Wxk &

Siebert 13351), Ilex mitis, Kiggelaria africana. Primus

africana and Senna Occident alis: forbs include Polygala

virgata , Senecio tamoides, Solanum aculeastrum and

Urtica lobulata.

Dominant/prominent taxa: conspicuous taxa are the

tree Halleria lucida, the climber Clematis brachiata and

the grass Panicum deiistum.

Notes on floristic diversity: strong floristic affinities

exist with Association 12 in species groups AE and AF

(Table 1). It also shares woodland species with Asso-

ciation 4 in species group AF (Table 1 ). However, this is

not a true vegetation type of rock habitats, but due to the

subsequent undersampling and existence of only two forest

patches (2 releves). The community is described here for

lack of better placement. The mean number of species

encountered per sample plot is 35 and the total number

of plant species is 65 taxa (two releves) (Table 3). This

association has one taxon with conservation status,

namely the undescribed, near-endemic Gymnosporia sp.

nov. {Van Wyk & Siebert 13351) (Table 2). Another note-

worthy taxon is Nemesia zimbabwensis with its disjunct

distribution shared with the Eastern Highlands of

Zimbabwe. Both these taxa are restricted to the associa-

tion. Its relict status gives the community special conser-

vation significance as a rare plant community (perhaps

the rarest in the SCPE).

DISCUSSION

Ordination and environmental factors

The naturally sparsely vegetated appearance of rocky

habitats can be ascribed to the relatively high surface

cover of rock, comprising many taxa typical for this

habitat in the northern provinces of South Africa. When

compared with other habitats of the SCPE (Siebert

2001), its environmental factors are relatively homoge-

neous. A combination of factors such as terrain type

(slope), soil texture (clay/sand content) and rockiness

(rock size and rock cover), affect the species composition

of these plant communities. The ordination supported a

gradient which is mainly the consequence of rockiness.

The scatter diagram displays the distribution of

releves along the first and second ordination axes (Figure

2). The vegetation units are represented as groups, their

distribution on the scatter diagram corresponding with

certain physical environmental conditions. The rocki-

ness, slope and soil texture determine a definite gradient

that is depicted by both the first (eigen value = 0.669)

and second axis (eigen value = 0.456). Rockiness, slope

and soil texture determine the moisture retention and

drainage of the habitat. The gradient on the x-axis

expresses rock cover as a percentage of the soil surface,

with the left extreme of the scatter diagram representing

rocky flats with its continuous layers of rock at the soil

surface and the right depicting the large boulders with

large areas of open soil between them, which are typical

for rocky refugia. On the y-axis, the gradient indicates

higher moisture availability over the long term at the top

of the diagram, because clayey soils on moderate slopes

with large areas covered with rock remain moist over a

longer period. Steep slopes with sandy soils and low rock

cover dry out quickly and are at the bottom of the dia-

gram. The x-axis also exhibits a gradient with deep soils

at the right and shallow soils at the left.

All these gradients correlate closely with each other

and have a strong influence on the vegetation structure

and species composition. The three most dominant and

conspicuous taxa of each growth form (trees/shrubs/suf-

frutices. forbs/sedges and grasses) are given for each of

the four major vegetation types (alliances) depicted in

the scatter diagram (Table 4).

A vegetation key is presented to aid with the identifi-

cation of the various plant communities (Table 5). The

definitions are broad indications of typical groups and

should be seen as a guideline. A diagnostic characteristic

of the vegetation or habitat is given, followed by a most

diagnostic and a most visual species of the plant com-

munity. The first species is restricted to the specific com-

TABLE 4. — Nine most dominant and conspicuous plant taxa of each of major vegetation types depicted in DECORANA scatter diagram

la Tall, moist woodland (Celtis africana & Panicum deustum)

b Short woodland, herbland and grassland ( Aloe castanea & Euclea crispa)

226

Bothalia 33,2 (2003)

"S 1

X)

£ o

s s

PS -g

4s, 4s,

'a, a,

s s

c a a a

X o

^ a,

oijj .a,

c a;

a

o g

s s

ON On

A V

<u <u

•fe. 1

£ *c

f*S

Co ^

00 H CN — '

&0

| §

•a -a s S

§ §

<D <L»

2 2

•§ ^

I £

<D c/5

•r C

£ ‘c3

>? c

1 >2

> ^

«

8 a -S 1

* (§ § g

“ “ c "o -o

D- D- (L) ^ > r-<

O £ Q- „2 1> 3

co co O U Q co

cd X c3 x

&■§

§ a

s,

S IS £ C

p o » s

.a s

3? a

•a '-a

8 -2

»o -§

s p

tq

Nt Co

°0 S

R *5.

a

^ o

u h n «

S,

5 IS

S <$s

•S- s

§. Six

S E'l

*S 5 E

<u ?3 Sc

' I #

>< a

< o

§ u

Si •s

a o

3 .O 3 JO

-^0 -id -5 •£>

U U

0 O o 22

01 ei m Z

£ £

V V

■S T -S -S

a a .S .2

G‘ oq J J

CN I-H

NO 1/3 <t ^

G &

a 4s,

O c

v C

Q- <V

3 •£

g S

'C -a a

a a ^ 2

2 1? P as

« S £.§

o « -S 1

_o o Q ^

‘S "S

>, tj

f i

> Oi < CO 00 CO

cd X cd X 03 X

E E

a cu o o

5 5

Bothalia 33,2 (2003)

227

munity only, and the second is dominant, but also occurs

in other communities. Where one species is given, no

species was restricted to the group only.

Diversity and conservation

Much of the vegetation distribution of the SCPE is

attributed to the vegetation dynamics and historic evolu-

tion of the floristics of the region, which is influenced by

three factors, climate, topography and soils (Siebert

1998). It has been shown that the vegetation of rocky

outcrops is specifically adapted to the chemical environ-

ment (Morrey et al. 1989; Bredenkamp & Deutschlander

1995; Tyler 1996). When considering the large number

of associations described in this paper, it is speculated

that the soil chemistry of the rock habitats in the SCPE is

partly responsible for the unique communities and plant

diversity that occur there (Siebert 2001).

Rock habitats are characterized by high spatial het-

erogeneity due to the range of differing aspects and

slopes (topography), all of which result in differing soil,

light and hydrological conditions. Variation in aspect and

soil drainage has proven to be an important predictor of

plant diversity in the SCPE (Siebert 2001). This is attrib-

uted to landscapes with spatially heterogeneous abiotic

conditions, which provide a diversity of potential niches

for plants. Plant species richness and diversity is signifi-

cantly higher in sites with high geomorphological het-

erogeneity (Burnett et al. 1998). However, rock habitats

are usually isolated, which means that should they be

disturbed, the species richness will be influenced nega-

tively due to the long distances between similar plant

communities (Bruun 2000).

Vegetation units or rock habitats in the SCPE require

special protection against disturbances and should be

considered for conservation purposes in the light of the

rapidly developing mining industry of the region.

Already, some plant communities of rock habitats in the

SCPE (especially the rocky refugia) are not as diverse as

they were in the past, due to extensive harvesting of fire-

wood and building material (Crookes et al. 2000). These

Afromontane forests of rocky refugia are of continental

conservation importance (White 1981). In addition, cer-

tain SCPE plant endemics of rock habitats are restricted

to specific communities, and once removed, will proba-

bly not return. If it is considered that ‘disclimax’ in dry

woodlands (<1 000 mm/annum) created through past land

use activities, will not develop into stands similar to the

previous state (Roth 1999), and given the present state of

fragmentation and continued disturbance taking place

within SCPE ecosystems, the rock habitats of the SCPE

require immediate attention. All the plant communities

of rock habitats presented in this paper are of conserva-

tion value, as they provide important islands to sustain

biodiversity.

CONCLUSIONS

The classification obtained by TWINSPAN and

refined by Braun-Blanquet procedures resulted in 17 veg-

etation units (plant communities) that can be related to

environmental factors. These vegetation units should be

considered as ecologically interpretable plant communi-

ties for the area concerned. The classification of these

vegetation units as associations is supported by the results

of the ordination that pointed towards meaningful rela-

tionships between the vegetation and habitat gradients.

It is hoped that the classification and description of

the different vegetation units will make a significant con-

tribution towards the understanding of the plant commu-

nities of rock habitats of the SCPE and in southern Africa

as a whole. The information supplied in this paper can be

meaningfully applied in the management and conserva-

tion of the respective areas. Proper and sound future

assessment of the region’s vegetation should include

aspects such as species richness, rarity and habitat pref-

erence. Such studies will further contribute and benefit

decisions on land use management and conservation, and

might hold solutions for the rehabilitation of areas dis-

turbed by mining activities in this region.

ACKNOWLEDGEMENTS

Our thanks to Ms Martie Dednam, H.G.W.J. Schwei-

ckerdt Herbarium (PRU), University of Pretoria, for the

processing of plant specimens, and the Curator and staff

of the National Herbarium (PRE) in Pretoria, for assis-

tance with plant identification. The National Research

Foundation. University of Pretoria, Edward Mellon

Foundation and Department of Environmental Affairs

and Tourism financially supported this project. The

South African Weather Bureau provided the rainfall and

temperature data.

REFERENCES

ACOCKS, J.P.H. 1988. Veld Types of South Africa, edn 3. Memoirs of

the Botanical Survey of South Africa No. 57.

BEDWARD, M„ PRESSEY, R.L. & KEITH. D A. 1992. A new approach

for selecting fully representative reserve networks: addressing

efficiency, reserve design and land suitability with an iterative

analysis. Biological Conser\>ation 62: 115-125.

BLOEM, C. 1988. n Plantsosiologiese studie van die Verlorenvaleinatuur-

reservaat, Transvaal. M.Sc. thesis. University of Pretoria, Pretoria.

BREDENKAMP. G.J., BEZUTDENHOUT. H„ JOUBERT, H. & NAUDE,

C. 1994. The vegetation of the Boskop Dam Nature Reserve,

Potchefstroom. Koedoe 37: 19-33.

BREDENKAMP, G.J. & DEUTSCHLANDER, M.S. 1995. New azonal

syntaxa from the hills and riverbanks of the Manyeleti Game

Reserve, Northern Transvaal Province, South Africa. Koedoe

38: 41-58.

BRUUN, H.H. 2000. Deficit in community species richness as explained

by area and isolation of sites. Diversity and Distributions 6: 129-

OS.

BUCKLE. C. 1996. Weather and climate in Africa. Addison Wesley Long-

man. Harlow.

BURGOYNE, P.M. 1995. Phytosociology of the north-eastern Trans-

vaal high mountain grasslands. M.Sc. thesis, University of Pretoria,

Pretoria.

BURNETT, M R., AUGUST. P.V., BROWN, J.H. & KILLINGBECK,

K.T. 1998. The influence of geomorphological heterogeneity on

biodiversity. I. A patch-scale perspective. Consen’ation Biology

12: 363-370.

COETZEE, B .J. 1975. A phytosociological classification of the Rusten-

burg Nature Reserve. Bothalia 4: 561-580.

CROOKES. D„ IRELAND. C„ DE WIT, M„ GEACH, B„ KGAME.

W. & NETSH1LUVH1. T. 2000. Benefits and costs of natural

resource use in Sekhukhuneland, Northern Province: case study

evidence from Managaneng and Makua. Unpublished Report

No. ENV-P-I-2000-0 1 3 . CSIR. Pretoria.

DEALL. G.B. 1985. A plant-ecological study of the Eastern Transvaal

Escarpment in the Sabie area. M.Sc. thesis, University of Pretoria,

Pretoria.

228

Bothalia 33,2 (2003)

EDWARDS, D. 1983. Abroad-scale structural classification of vegeta-

tion for practical puiposes. Bothalia 14: 705-712.

ERASMUS, J.F. 1985. Rainfall deciles for the Transvaal Region.

Report GB/A/87/18: 147-216. Soil and Irrigation Research

Institute, Pretoria.

HENNEKENS, S. 1996a. TURBOVEG: software package for input , pro-

cessing, and presentation of phytosociological data. User’s

guide. EBN-DLO, Lancaster University.

HENNEKENS, S. 1996b. MEGATAB: a visual editor for phytosocio-

logical tables. User’s guide. Giesen & Geurts, Ulft.

HILL, M.O. 1979a. TWINSPAN: a FORTRAN program for arranging

multivariate data in an ordered two way table by classification of

individuals and attributes. Cornell University, Ithaca, New York.

HILL, M.O. 1979b. DECORANA: a FORTRAN program for detrended

correspondence analysis and reciprocal averaging. Cornell

University, Ithaca, New York.

HILTON-TAYLOR, C. 1996. Red Data List of southern African plants.

Strelitzia 4. National Botanical Institute, Pretoria.

JONSSON, B.G. & MOEN, J. 1998. Patterns in species associations in

plant communities: the importance of scale. Journal of Vegeta-

tion Science 9: 327-332.

KENT. L.E. (ed.). 1980. Stratigraphy of South Africa: lithostratigraphy

of the Republic of South Africa, South West Africa/Namibia and

the Republics of Transkei, Bophuthatswana, Venda. Government

Printer. Pretoria.

LAND TYPE SURVEY STAFF. 1987. Land types of the maps 2526

Rustenburg, 2528 Pretoria. Memoirs on the Agricultural Natu-

ral Resources of South Africa No. 8. Department of Agriculture

and Water Supply, Pretoria.

LAND TYPE SURVEY STAFF. 1988. Land types of the maps 2426 Tha-

bazimbi, 2428 Nylstroom. Memoirs on the Agricultural Natural

Resources of South Africa No. 10. Department of Agriculture and

Water Supply, Pretoria.

LAND TYPE SURVEY STAFF. 1989. Land types of the map 2530

Barberton. Memoirs on the Agricultural Natural Resources of

South Africa No. 13. Department of Agriculture and Water

Supply, Pretoria.

LOW, A.B. & REBELO, A.G. (eds). 1996. Vegetation of South Africa,

Lesotho and Swaziland. Department of Environmental Affairs

& Tourism, Pretoria.

MACVICAR, C.N., BENNIE, A.T.P. & DE VILLIERS, J.M. 1991 . Soil

classification: a taxonomic system for South Africa. Department

of Agricultural Development, Pretoria.

MADULID, D A. & AGOO. E.M.G. 1995. Assessment of the vegeta-

tion of Philippine mountains on ultramafic rocks. In T. Jaffre,

R .D. Reeves & T. Becquer, The ecology of ultramafic and met-

alliferous areas : 113-118. Centre de Noumea, Orstrom.

MATTHEWS, W.S. 1991. Phytosociology of the North-eastern Moun-

tain Sourx’eld. M.Sc. thesis. University of Pretoria, Pretoria.

MATTHEWS, W.S., BREDENKAMP, G.J. & VAN ROOYEN, N. 1991.

The grassland associated vegetation of the Black Reef Quartzite

and associated large rocky outcrops in the North-eastern Moun-

tain Sourveld of the Transvaal Escarpment, South Africa. South

African Journal of Botany 57: 143-150.

MATTHEWS, W.S., BREDENKAMP, G.J. & VAN ROOYEN, N. 1992.

The vegetation of the dry dolomitic regions of the north-eastern

Mountain Sourveld of the Transvaal escarpment. South Africa.

Pliytocoenologia 20: 467^188.

MEIRELLES, S.T., PIVELLO, V.R. & JOLY, C.A. 1999. The vegetation

of granite outcrops in Rio de Janeiro, Brazil, and the need for

its protection. Environmental Conservation 26: 10-20.

MORREY, D.R., BALKWILL, K. & BALKWILL, M-J. 1989. Studies

on serpentine flora: preliminary analyses of soils and vegetation

associated with serpentine rock formations in the southeastern

Transvaal. South African Journal of Botany 55: 171-177.

MUCINA, L„ BREDENKAMP. G.J., HOARE, D.B. & MCDONALD,

D.J. 2000. A national vegetation database for South Africa.

South African Journal of Science 96: 497, 498.

MUELLER-DOMBOIS, D. & ELLENBERG, H. 1974. Aims and methods

of vegetation ecology. Wiley, New York.

REDDY R.A., BALKWILL, K. & MCLELLAN, T. 2001. Is there a

unique serpentine flora on the Wilwatersrand? South African

Journal of Science 97: 485—494.

RETIEF. E. & HERMAN, P.P.J. 1997. Plants of the northern Provinces

of South Africa: keys and diagnostic characters. Strelitzia 6.

National Botanical Institute, Pretoria.

ROTH, L.C. 1999. Anthropogenic change in subtropical dry forest dur-

ing a century of settlement in Jaiqui Picado, Santiago Province,

Dominican Republic. Journal of Biogeography 26: 739-759.

SCHURMANN, L.W.. GRABE, P-J. & STEENKAMP, C.J. 1998.

Chromium. In M.G.C. Wilson & C.R. Anhaeusser, The mineral

resources of South Africa'. 90-105. Handbook No. 16, Council

for Geoscience. CTP Book Printers, Cape Town.

SIEBERT, S.J. 1998. Ultramafic substrates and floristic patterns in

Sekhukhuneland, South Africa. M.Sc. thesis. University of Pretoria.

SIEBERT, S.J. 2001. Vegetation on the ultramafic soils of the Sekhu-

khuneland Centre of Endemism. Ph.D. thesis. University of

Pretoria.

SIEBERT. S.J. & VAN WYK, A.E. 2001. Sekhukhuneland: floristic

wealth versus platinum and chromium riches. Veld & Flora 87:

168-173.

SIEBERT, S.J., VAN WYK, A.E. & BREDENKAMP, G.J. 2001.

Endemism in the flora of ultramafic areas of Sekhukhuneland.

South Africa. South African Journal of Science 97: 529-532.

SIEBERT, S.J., VAN WYK, A.E. & BREDENKAMP, G.J. 2002a. The

physical environment and major vegetation units of the

Sekhukhuneland Centre of Plant Endemism. South African

Journal of Botany 68: 127-142.

SIEBERT. S.J., VAN WYK, A.E. & BREDENKAMP, G.J. 2002b. Vege-

tation ecology of Sekhukhuneland, South Africa: Combretum

hereroense-Grewia vemicosa Open Mountain Bushveld. South

African Journal of Botany 68: 475^196.

SIEBERT, S.J., VAN WYK. A.E. & BREDENKAMP, G.J 2002c.

Vegetation ecology of Sekhukhuneland, South Africa: Kirkia

wilmsii—Terminalia prunioides Closed Mountain Bushveld.

South African Journal of Botany 68: 497-517.

SIEBERT, S.J., VAN WYK. A.E., BREDENKAMP, G.J. & DU

PLESSIS. F. 2002d. The grasslands and wetlands of the Sekhu-

khuneland Centre of Plant Endemism, South Africa. Bothalia

32: 211-231.

SIEBERT, S.J., VICTOR, J.E.. VAN WYK. A.E. & BREDENKAMP.

G.J. 2002e. An assessment of threatened plants and conserva-

tion in Sekhukhuneland. PlantLife 26: 7-18.

SIEBERT. S.J.. MATTHEE, M. & VAN WYK, A.E. 2003. Semi-arid

savanna of the Potlake Nature Reserve and surrounding areas in

Sekhukhuneland, South Africa. Koedoe 46: 29-52.

SMIT, C.M., BREDENKAMP, G.J.. VAN ROOYEN, N., VAN WYK,

A.E. & COMBRINCK, J.M. 1997. Vegetation of the Witbank

Nature Reserve and its importance for conservation of threat-

ened Rocky Highveld Grassland. Koedoe 40: 85-104.

SOUTH AFRICAN WEATHER BUREAU. 1998. Unpublished long-term

weather statistics. Department of Environmental Affairs and

Tourism, Pretoria.

TYLER, G. 1996. Soil chemistry and plant distributions in rock habi-

tats of southern Sweden. Nordic Journal of Botany 16: 609-635.

VAN DER MEULEN, F. 1979. Plant sociology of the western Transvaal

Bushveld, South Africa. Dissertationes Botanicae 49: 1-191.

VAN WYK, A.E. & SMITH. G.F. 2001. Regions of floristic endemism

in southern Africa: a review with emphasis on succulents.

Umdaus Press, Pretoria.

VILJOEN, M.J. & SCHURMANN, L.W. 1998. Platinum-group metals.

In M.G.C. Wilson & C.R. Anhaeusser, The mineral resources of

South Africa : 532-568. Handbook No. 16, Council for Geo-

science, Cape Town.

VISSER, D.J.L., COERTZE, F.J. & WALRAVEN, F. 1989. Explana-

tion of the 1 : 1000000 geological map, fourth edition, 1984: the

geology of the Republics of South Africa, Transkei, Bophu-

thatswana, Venda and Ciskei and the Kingdoms of Lesotho and

Swaziland. The Government Printer, Pretoria.

WEBER. H.E., MORAVEC, J. & THEURILLAT, J-P. 2000. Inter-

national Code of Phytosociological Nomenclature, edn 3. Jour-

nal of Vegetation Science 11: 739-768.

WHITE, F. 1981. The history of the Afromontane archipelago and the

scientific need for its conservation. African Journal of Ecology

19: 33-54.

WHITTAKER. R.H. 1977. Evolution of species diversity in land com-

munities. Evolutionary Biology 10: 1-67.

WINTERBACH, R„ BREDENKAMP. G.J., DEUTSCHLANDER, M.S.

& MUCINA, L. 2000. Preliminary syntaxonomic scheme of veg-

etation classes for the central bushveld of South Africa. In PS.

White, L. Mucina, J.S. Leps & E. van der Maarel, Proceedings of

the 1AVS Symposium: 123-127. Opulus Press, Uppsala.

Bothalia 33,2: 229-243 (2003)

National Botanical Institute South Africa:

administration and research staff 31 March 2003,

publications 1 April 2002-31 March 2003

Compiler: B.A. Momberg

CHIEF DIRECTORATE (ADIR)

CAPE TOWN— PEARSON HOUSE

Huntley, Prof. B.J. M.Sc. Chief Executive Officer

Adams, Ms M.A. Principal Typist I. Receptionist

Laidler, Mrs S.A. B.Sc. (Agric.) (Hons) Senior Provisioning Admin. Officer. Personal Assistant

MARKETING AND COMMUNICATION SUBDIRECTORATE (AMAR)

CAPE TOWN

Director — Vacant

Van Aswegen, Ms C.E. H.E.D., B.A. (History & Psychology).

Senior Provisioning Admin. Clerk III. Admin, support

PRETORIA

Mthembi, Ms C. Dip. (Market. & Pub. Relations), Dip. (Market. Manag.).

Senior Communications Officer. Marketing & pub. relations

GRAPHIC SERVICES— CAPE TOWN (AMAR)

Loedolff, Mrs J. B.Sc.(Ind.Tech.). Chief Industrial Technician. Photographer/Illustrator (part time)

HUMAN RESOURCES DIRECTORATE (APER)

CAPE TOWN

Tlali, P.L.N. B.Iuris Lib. Director. Human Resources

Kriel, Mrs G.A. Dip. (Sec.). Senior Secretary IV

Finca, Ms N.F. Specialist Cleaner

Haupt, Mrs C.S. Specialist Cleaner. Guest house

Woodward, Ms Y.J. Provisioning Admin. Officer. Secretary to NBI Board

PERSONNEL— CAPE TOWN

Van Zyl. J.M. M.Econ.(Ind. Psych.). Assistant Director: Organizational Development & Training

Engelbrecht, B. N.Dip.(Hort.), N.Dip.(PRM), Dip. (Forestry). Assistant Director: Human resources

management and health & safety co-ordination

Staal, P.B. Dip. (Social Science), Industrial Relations Cert. Assistant Director: Staff relations

230

Bothalia 33,2 (2003)

Claassen, Ms G. Senior Telkom Operator III. Admin, support

Crowie, A.C. Senior Registry Clerk I

Dollie, Mrs N.J. Specialist Cleaner

Du Toit, Ms R. H.E.D. Senior Training Officer. Training

Co-ordinator

Fredericks, C.H. Driver II. Courier services

Mtyalela, D.M.S. Senior Training Officer

Nicholas, Mrs W.L. Senior Photocopy Machine Operator

Petersen, R.E. Senior Personnel Practitioner. Employment

equity

Sass, Ms D. Senior Provisioning Admin. Clerk I. Leave

& IOD admin.

Williams, Mrs L.R. Dip. (Sec.). Senior Personnel Officer

II. Recruitment

FINANCE DIRECTORATE (AFIN)

CAPE TOWN

Hughes, W.S.G. B.Com., C.A.(SA). Director. Chief Financial Officer

Rawoot, N.A. B.Acc.Sc. Deputy Director: Financial management

Maholwana, S. B.Com. Assistant Director. Financial management

Neuwirth, Ms E.V. Assistant Director. Staff benefits

Cassiem, Ms S. Senior Accounting Clerk III. Salaries

Cooper, S.K. Accounting Clerk II. Creditors

Du Plessis, Ms B.P Accounting Clerk II. Debtors

Goodman, Mrs I.W. State Accountant. Supervisor: Credi-

tors

Jacobs, F.H. Senior Accounting Clerk II. Staff benefits

Jacobs, Ms S. State Accountant. Financial admin.: General

ledger

Koyana, Ms Z. Dip. (Sec.), Cert.(Acc.). Senior Accounting Clerk

I. Salaries

Mcontsi, Ms N. N.Dip.(Acc.). Senior Accounting Clerk I. Creditors

Mirkin, Ms Y.A. Senior Secretary III

Ogunniyi, Mrs E. Internal Auditor (contract worker)

Paulse, Mrs D.W.S. Dip.(Bookkeep.), Dip. (Sec.). Senior

Accounting Clerk III. Creditors

September, Ms M. Senior Accounting Clerk II. Debtors

BIODIVERSITY POLICY AND PLANNING DIRECTORATE

Njobe, Ms K. M.Sc. Director. Biodiversity and Action Planning

Baloyi, Ms I. Senior Secretary IV

Golding, Ms J. Manager: Threatened Plants Programme

GARDENS DIRECTORATE— ADMIN STAFF (GDIR)

PRETORIA

Willis, C.K. M.Sc. (Cons. Biol.)- — Director: Gardens and Horticultural Services

Heilgendorff, J.P. N.H.Dip.(Hort.). Gardens IT Manager

URBAN CONSERVATION— CAPE TOWN (GDIR/UC)

Davis, G.W. Ph.D. Assistant Director: Project management and fund raising

Mkefe, T.J. S. PE. Dip. Principal Communications Officer. Public affairs & liaison

Peter, L.M. Dip.(Hort.). Agricultural Development Technician. Urban conservation

INTERPRETATION

Joffe, Mrs H. B.Sc. Chief Agricultural Development Technician. Interpretation (part time)

Roff, J. Cert.(Envir. Interpr. & Ed.). Communication Officer. Interpretation (Pietermaritzburg)

PLANNING, MAINTENANCE & DEVELOPMENT— CAPE TOWN (BPMD)

Linde, D.C. N.T.C.III(Civil & Structural: Building) N.T.C.III (Inspector of Works: Building).

M.S.A.I.D. Cert. Estate Agency. Control Inspector of Works

Arendse, D.S. Senior Handyman. Building maintenance

Manasse, S.P Dip. (Masonry). Artisan Foreman. Building maintenance

Peck, W.I. Senior Handyman. Building maintenance

Bothalia 33,2 (2003)

231

CURATORS

Behr, Ms C.M. Curator: Pretoria NBG

Blitz, R.M. Curator: Lowveld NBG (Nelspruit)

Gavhi, M.P. Curator: Free State NBG (Bloemfontein)

Le Roux, PH. Deputy Director. Curator: Kirstenbosch NBG (Cape Town)

Oliver, I.B. Curator: Karoo Desert NBG (Worcester)

Shaide, Ms A.C. Curator: Harold Porter NBG (Betty's Bay)

Tarr, B.B. Curator: Natal NBG (Pietermaritzburg)

Turner, Ms S.L. Curator: Witwatersrand NBG (Roodepoort)

HAROLD PORTER NBG— BETTY'S BAY (GHPG)

Shaide, Ms A.C. N.Dip.(Hort.). Curator

Abrahamse, S.J. Principal Foreman. Garden

Arendse. Ms M. Auxiliary Services Officer II. Access control

Bezuidenhout, Mrs H.M. Provisioning Admin. Officer

Carolus, Ms B.I. N.H.Dip.(Hort.) Agricultural Development

Technician. Horticulture

Forrester, Ms J.A. N.T.C.III(Hort.). Chief Agricultural De-

velopment Technician. Horticulture

October, Ms R.P. Dip.(Education). Senior Auxiliary Services

Officer. Plant records and asset register

Samuels, Ms D.C. Cleaner II

Simpson, H.A.S. Auxiliary Services Officer II. Access control

Van Wyk, A.B. Artisan. Building maintenance

KAROO DESERT NBG— WORCESTER (GKAR)

Oliver. I.B. N.Dip.(Hort.), N. Dip. (Public Relations & Admin). Control Agricultural Technician. Curator

Ashworth, Mrs E.H. Senior Provisioning Admin. Clerk III.

Admin, support

Mtetwa, A.M. Artisan. General maintenance

Makubalo, F.N. Senior Foreman. Nursery

Mpeke, Ms E.N. Specialist Cleaner

Simani, D.K. Principal Foreman. Garden

Van Tonder, M. (volunteer)

Viljoen, D.M. N.Dip.(Hort.). Chief Agricultural Development

Technician. Collections

Voigt, W.E. N.Dip.fHort.). Senior Agricultural Development

Technician. Garden development

Williams, C. (student)

KIRSTENBOSCH NBG— CAPE TOWN (GKBC)

Le Roux, PH. Dip. (Forestry), N.Dip.(Hort-), N. Dip. (Parks & Recr.), Cert. (Turf Management).

Deputy Director: Garden Management. Curator

Goldschmidt, S.M. B.A.(Soc.)(Hons.). Assistant Director. Personnel Manager

Adams, T.D. B.Tech.(Hort.). Agricultural Development Technician. Supervisor: Greenhouse

Hitchcock, A.N. N.H.Dip.(Hort.). Control Agricultural Technician. Nursery Manager

Morkel, A.T. N. Dip. (Nature Cons.). Control Agricultural Technician. Estate Manager

Notten, Ms A.L. B.Sc., N.Dip.fHort.). Chief Agricultural Development Technician. Supervisor: Seed room

Trautman, C.E. Artisan. Supervisor: Workshop

Adonis, A. Principal Foreman. Dell & ericas

Adonis, S.J. Senior Foreman. Aliens vegetation control

Arends, Ms S.J. Principal Auxiliary Services Officer. Plant

records

Botha. PA. N.H.Dip.(Hort.). Chief Agricultural Development

Technician. Seed research

Brown. B.M. N.Dip.l I lort.). Agricultural Development Tech-

nician. Nursery

Crowie, R.W. Senior Foreman. General garden

Crowie, Mrs U.M. Accounting Clerk II. Seed room

Davids, M.I. Senior Collection Specialist. Bulbs

Difoloko, J.A. (student)

Dlela, Ms D.I. (student)

Duncan, G.D. N.Dip.(Hort.). Control Agricultural Develop-

ment Technician. Bulbs

Engelbrecht, F. Provisioning Admin. Clerk II. Stores

Engelbrecht. Mrs L.D. Control Auxiliary Services Officer.

Plant records

Grace. T. Senior Provisioning Admin. Clerk III. Stores &

admin, support

Hope, C.F. Senior Handyman. Construction

Jacobs, H.C. Senior Foreman. Plant production

Jaftha, Ms D.L. Accounting Clerk II. Seed room

Jamieson, Mrs H.G. N.Dip.(Parks & Rec.). Chief Agricultural

Development Technician. Restio/Asparagus (part time)

Jodamus, Ms N.L. N.Dip.(Hort.). Senior Agricultural De-

velopment Technician. Annuals

Kamalie, Ms S. Senior Typist. Receptionist

Kayster, G.J. Senior Foreman. Construction

Kuscus, G.W. Senior Foreman. General maintenance

Lawrence, E. N.Dip.fHort.). Senior Agricultural Develop-

ment Technician. Dell & ericas

Magija, Ms N.F. N.Dip.fHort.). Senior Agricultural Develop-

ment Technician. Outreach programme

232

Bothalia 33,2 (2003)

Manuel, I.P. Senior Foreman. Seed room

Mathys, Mrs S.S.B. Senior Accounting Clerk III. Revenue

and visitor statistics

Matthews, I.N. Senior Foreman. Estate & trails

Mbambezeli, N.G. N.Dip.(Hort). Agricultural Development

Technician. Trees & shrubs

Mitchells, G. Senior Foreman. Mowers

Morris, J.N.M. Senior Foreman. Proteas

Newman, W. Artisan. Mechanical workshop

Picane, Ms S. Auxiliary Services Officer II. Tissue culture

Prins, F.B. Security Officer III

Rudolph, A. Security Officer III

Smith, Mrs A. Senior Provisioning Admin. Clerk I. Admin,

support

Solomons, T.C. Senior Security Officer II

Tamboer, J.S. Senior Foreman. Nursery services

Van Gusling, E.J. Senior Foreman. Mowers

Van der Walt, Mrs L.E. N.Dip.(Hort.). Chief Agricultural

Development Technician. Herbaceous collections

(part time)

Van Jaarsveld, E.J. M.Sc., N.Dip.(Hort.). Control Agricultural

Technician. Succulents

Van Rooy, K. Senior Foreman. Annuals

Van Wyk, F. Principal Auxiliary Services Officer II. Lable

Maker

Viljoen, Ms C.C. N.Dip.fHort.). Chief Agricultural Develop-

ment Technician. Plant production

VISITORS CENTRE— CAPE TOWN (GKBC/VC)

Struys, Ms S. Dip. (Market. Manag.), B. A. (Directing). Principal Communications Officer. Centre Manager

August, T. Senior Auxiliary Services Officer. Information

services

Fredericks, Ms N.C.E. Senior Auxiliary Services Officer.

Visitors’ Centre. Information services

Jacobs, A.P. Chief Auxiliary Services Officer. Visitors’

Centre. Information services

Jacobs, Ms B.C. Dip. (Travel & Tourism), Dip. (Computer

Business). Senior Admin. Clerk II. Facilities Co-

ordinator

Malan, Ms C.E. B. Sc. (Hons). Principal Communication

Officer: Tour co-ordinator

Williams, G.C. Senior Auxiliary Services Officer. Infonnation

LOWVELD NBG — NELSPRUIT (GLOW)

Britz, R.M. N. Dip. (Forestry). Control Agricultural Technical!. Curator

Froneman, W.C.F. N.T.C.III(Hort.), N. Dip. (Nature Cons.

& Man.), N. Dip. (Parks & Rec. Admin.), N.T.C.III

(Hort.). Control Agricultural Technician. Nursery

management & garden development

Haynie, R.D. Senior Handyman

Hurter, P.J.H. B.Sc.(Hons). Control Agricultural Technician.

Cycad conservation

Le Roux, Ms L.N. N.H. Dip. (Nature Cons.). Principal

Auxiliary Services Officer II. Interpretation

Maqungo, Ms V.L.B. Auxiliary Services Officer. Kiosk

Mathebula, Ms N.R. Accounting Clerk II. Admin, support

Musweli, K.J. Senior Foreman. New development

Ndlovu, L.D. Senior Handyman. Maintenance

Ngwenya, P.S. Auxiliary Services Officer II. Kiosk

Sibanyoni, Ms S.M. Cleaner I

Van der Walt, Mrs G.A.M. Senior Provisioning Admin.

Clerk III

Xozumti, M.M. Principal Foreman. Supervisor

NATAL NBG— PIETERMARITZBURG (GNAT)

Tarr, B.B. N. Dip. (Parks & Rec. Admin.), Advanced Dip. (Adult Educ.).

Control Agricultural

Dlungwane, T.R. Senior Foreman. Garden maintenance

Johnson, Ms I. M.Sc., H.E.D. Senior Agricultural Develop-

ment Technician

Nonjinge, S.H.B. N.T.C.III(Hort.). Chief Agricultural De-

velopment Technician

Technician. Curator

Sibiya, Ms C.P.T. Cleaner I

Van der Merwe, Mrs M.E.H. Senior Provisioning Admin.

Clerk III

Xaba, P.A. (student)

Zimu, M.J. Principal Foreman. Garden

FREE STATE NBG— BLOEMFONTEIN (GFSG)

Gavhi, M.P. N. Dip. (Hort.). Curator

Eysele, Mrs J.P. Senior Provisioning Admin. Clerk III.

Admin, support

Harris, Ms S. N.Dip.lHort.). Senior Agricultural Develop-

ment Technician

Mbolekwa, L.M. Foreman. Garden

Nakanyane, R.B. Principal Foreman. Garden

Raditlhare, Mrs E.M. Cleaner II

Sebolai, R.P.A.N. Senior Handyman. General maintenance

Thaele, Mrs M.E. Cleaner II

Bothalia 33,2 (2003)

233

PRETORIA NBG (GPTA)

Behr, Ms C.M. B. Sc. (Hons). Control Agricultural Technician. Curator

Baloyi, K.J. Senior Auxiliary Services Officer II. Garden

records

Baloyi, M.S. Dip. (IBM), Dip.(PTM), Dip.(Payroll Admin.).

Senior Provisioning Admin. Clerk I. Personnel records

and H.R. support

Baloyi, S.J. Senior Handyman

Creighton. Ms D.D. Senior Provisioning Admin. Clerk III.

Admin, support

Key ter, B.A. Senior Security Officer II

Klapwijk, N.A. N.Dip.(Hort.), N. Dip. (Plant Prod.), N.Dip.

(Diesel Fitting). Control Agricultural Technician

Kutama, B.T. Senior Foreman. Garden development

Makgobola, Ms M.R. Auxiliary Services Officer II. Recep-

tion & admin, support

Mariri, J.N. Senior Foreman. Maintenance

Mariri, Ms M.A. Cleaner II

Matshika, S.P Groundsman II. Cook

Matthews, A. V. B.Tech.(Hort). Chief Agricultural Develop-

ment Technician

Modisha, M.D. Cleaner II

Mphaka, Ms N.F. Student Horticulturalist

Mutshinyalo, T.T. N.Dip.(Hort.). Senior Agricultural De-

velopment Technician

Noku, Y.A. Principal Foreman. Transport

Solomons, Ms C.V. Senior Auxiliary Services Officer. Records

clerk

Venter, W.A. N.T.C.II. Senior Artisan. Workshop and general

maintenance

WITWATERSRAND NBG— ROODEPOORT (GWIT)

Turner, Ms S.L. B. Sc. (Hons), N.Dip.(Hort.). Control Agricultural Technician. Curator

Aubrey, Mrs A.E. B.Tech.(Hort.). Chief Agricultural De-

velopment Technician. Plant records, interpreta-

tion, information (part time)

Baloyi, S.J. Handyman

Dlamini, M.D. N.Dip.(Hort.). Agricultural Development

Technician

Hankey, A.J. N.Dip.(Hort.). Control Agricultural Techni-

cian. Garden, estate, collections, nursery

Head. Mrs S.E. Dip. (Shorthand & Typing). Provisioning

Admin. Officer

Mamosebo, M.A. Factotum

Manyikana, T.M. Factotum

Mmola, Mrs B.E. Cleaner II

Ndou. A.P Senior Auxiliary Services Officer II. Garden

information

Ndzondo, Ms N.F. Senior Provisioning Admin. Clerk I

Ndzondo, Mrs P.G. Cleaner II

Nedambale, M.P. Senior Foreman. Garden

Nemalili, M.E. Senior Foreman. Machines and vehicles

Nenungwi, M.S. Senior Foreman. Nursery

Tebeile, Ms Z.M. Senior Provisioning Admin. Clerk I.

Receptionist

RESEARCH DIRECTORATE (RDIR)

PRETORIA

Smith, Prof. G.F. Ph.D., F.F.S. Chief Director: Research & Scientific Services

Rutherford, M.C. Ph.D.. Dip.(Datamet-). Deputy Director: Ecology and conservation (Cape Town)

Wolfson, Mrs M.M. Ph.D. Deputy Director: Environmental education and research support

Meyer, Mrs N.F. B. Sc. (Hons) Technician (contract worker)

Marais, Mrs A.C. Senior Secretary IV

Steyn, Mrs E.M.A. D.Sc. Specialist Scientist. Embryology, anatomy, taxonomy

EDUCATION AND RESEARCH SUPPORT— PRETORIA (EDIR)

Wolfson, Mrs M.M. Ph.D. Deputy Director. Physiology/Ecophysiology of Poaceae, carbon uptake metabolism,

allocation in response to environmental and management stress

Liebenberg, Mrs E.J.L. Head: Research Support Services, Publications

Naicker, K. Head: Admin, and Human Resources

Potgieter, Mrs E. Principal Librarian

Symonds, Ms A.M. Head: Environmental education

234

Bothalia 33,2 (2003)

ADMINISTRATION —PRETORIA (RPTA)

Naicker, K. Dip. (Bookkeeping), Cert. (Sales & Market. Manag.),

H.Cert.(Prac. Accounting). Assistant Director

Khumalo, N.P. Principal Foreman. Supervisor: Cleaning Phaahla, M.C. Cleaner II

services Pretorius, Ms M.A. Senior Provisioning Admin. Clerk II

Ledwaba, Mrs D.M. Senior Registry Clerk I Randima, Ms G. Cleaner II

Malefo, R.P. Cleaner II Rukazhanga-Leboho, Ms N. Senior Personnel Practitioner

Marule, PM. Senior Handyman Sithole, A.M. Cleaner II

Nkosi, Mrs M.P Specialist Cleaner Tloubatla, J.M. Driver II. Courier services

ENVIRONMENTAL EDUCATION— NORTHERN AND SOUTHERN GARDENS (EENT/HO)

PRETORIA

Symonds, Ms A.M. N. Dip. (Nature Cons.), H.E.D. Assistant Director. Environmental education

Adams, Ms E.M. Senior Provisioning Admin. Officer. Environmental Education Centre Manager

Maphuta, Mrs M.S. Cleaner II

GOLD FIELDS CENTRE— CAPE TOWN (EECT)

Gaffoor, Ms N. H.E.D. Principal Environmental Education Officer. Co-ordinator: Environmental education

September, Ms M. Senior Provisioning Admin. Clerk II. Admin, support

PRETORIA (EENT/GP)

De Bruyn, Ms A.J. B. Sc. (Zoo., Bot. & Mammalogy). Principal Communication Officer.

Co-ordinator: Environmental education

Mathaba, T.C. Environmental Education Officer

Novellie, Mrs E. H.E.D. B. Sc. (Hons) (Zool. & Mammology). Senior Environmental Education Officer.

Environmental education

WITWATERSRAND (EENT)

Van der Westhuizen, Mrs S. M. Sc. (Bot.). Principal Communications Officer.

Co-ordinator: Environmental education

Molefe, Ms K.E. Dip. (Nature Cons.). Senior Environmental Education Officer

Moore, Mrs J.M. N.H.Dip.(Sec.). Senior Provisioning Admin. Clerk II. Admin, support (part time)

RESEARCH SUPPORT SERVICES AND PUBLICATIONS— PRETORIA (RPUB)

Liebenberg, Mrs E.J.L. M.Sc. Control Agricultural Technician. Cytotaxonomy. Manager

Brink, Mrs S.S. Dip. (Typing). Chief Typesetter. Typeset-

ting, layout, word processing

Condy, Ms G.S. M.A. Chief Industrial Technician. Botanical

artist

Du Plessis, Mrs E. B.Sc.(Hons), S.E.D. Chief Language Prac-

titioner. Technical editor. Editing, translating, layout

Germishuizen, G. M.Sc. Assistant Director. Editor

Mapheza, T.P Senior Provisioning Admin. Clerk III. Book-

shop Manager

Momberg, Mrs B.A. B.Sc.(Entomology & Zoology). Princi-

pal Language Practitioner. Technical editor. Editing,

layout (part time)

Maree, Ms D.J. H.E.D. Senior Computer Operator.

Nkosi, PB. Senior Provisioning Admin. Clerk I. Bookstore

Turck, Mrs S. B.A. (Information Design). Chief Industrial

Technician. Graphic design

MARY GUNN LIBRARY— PRETORIA (RLBP)

Potgieter, Mrs E. B.Libr. Principal Librarian

Fourie, Mrs A.L. B.A., H.Dip.Libr.Sci. Principal Librarian (part time)

Shipalana, Ms K.M. N.Dip. Library and Information Studies (contract worker)

Bothalia 33,2 (2003)

PLANT SYSTEMATICS SUBDIRECTORATE

PRETORIA

235

Smith, Prof. G.F. Ph.D., F.L.S. Systematics of succulents and rosulate, petaloid monocots

Arnold, T.H. Head: Data Management (Pretoria)

Crouch, N.R. Head: Ethnobotany Unit (Durban)

Koekemoer, Ms M. Curator: National Herbarium (Pretoria)

Leistner, O.A. D.Sc. F.L.S. Scientist (contract worker)

Rourke, J.P. Curator: Compton Herbarium (Cape Town)

Siebert, S.J. Regional Project Co-ordinator: SABONET (Pretoria)

Singh, Ms Y. Curator: Natal Herbarium (Durban)

COMPTON HERBARIUM— CAPE TOWN (RHEC)

Rourke, J.P. Ph.D., F.M.L.S., F.R.S.S.Af. Assistant Director. Systematics of southern African

Proteaceae, Stilbaceae

Baatjes, Ms A. Data Capturer (SABONET contract worker)

Chesselet, Ms P.C.M. M.Sc. Senior Agricultural Scientist

Conrad, Ms C. M.Sc. Agricultural Scientist. Molecular

systematics laboratory

Cupido, C.N. M.Sc. Chief Agricultural Scientist

Cupido, Mrs C.S. Senior Auxiliary Services Officer II. Tech-

nical Assistant

Engelbrecht, Ms M. (SABONET contract worker)

Foster, Mrs S.E. Senior Secretary IV

Kurzweil, H. Ph.D. Specialist Scientist. Systematics of south-

ern African terrestrial orchids

Leith, Mrs J. Senior Provisioning Admin. Clerk III

Manning, J.C. Ph.D. Senior Specialist Scientist. System-

atics of Iridaceae and Hyacinthaceae; anatomy

Marinus, Ms E.D.A. Chief Auxiliary Services Officer. Her-

barium Assistant

Oliver, E.G.H. Ph.D. Specialist Scientist. Taxonomy of the

Ericoideae (Ericaceae)

Parker, Ms F. M.Sc. (SABONET contract worker)

Paterson-Jones, Mrs D.A. (nee Snijman) Ph.D. Specialist

Scientist. Systematics of Hypoxidaceae; cladistics

Reeves, Ms G. Ph.D. Senior Agricultural Scientist. Molecu-

lar systematics (contract worker)

Roux, J.P. N.T.C.IIKHort.), F.L.S., Ph.D. Specialist Scien-

tist. Systematics of Pteridophyta

Williams, Mrs V.J. Data Capturer (SABONET contract worker)

NATAL HERBARIUM— DURBAN (RHED)

Singh, Ms Y. M.Sc., H.E.D. Senior Agricultural Scientist. Taxonomy of Araceae, Hypoxidaceae. Curator

Dimon, Ms Z.Y. B.Sc. Data Capturer (SABONET contract

worker)

Govender, Mrs N. B.Sc. (Hons) Data Capturer. Chironia

systematics (SABONET contract worker)

Hlongwane, Mrs N.C. Cleaner II & messenger

Mazibuko, J.V.G. Senior Auxiliary Services Officer. Her-

barium Assistant

Mbonambi. B.M. Groundsman II. Garden maintenance

Nathoo, Ms M. B.Sc. Data Capturer (SABONET contract

worker)

Ngwenya, A.M. Senior Agricultural Development Tech-

nician. Herbarium Officer. Plant identification and

information, Zulu Botanical Knowledge Project

Noble, Mrs H-E. Senior Provisioning Admin. Clerk III

Williams, Ms R. B.Sc. (Hons), H.E.D. Principal Agricul-

tural Scientist

ETHNOBOTANY UNIT— DURBAN (RETH)

Crouch. N.R. Ph.D. Assistant Director. Ethnobotany of southern African Bora

Grace, Ms O.M. M.Sc. Assistant Bioprospecting Investigator (contract worker)

NATIONAL HERBARIUM— PRETORIA (RHEN)

Koekemoer, Ms M. Ph.D. Deputy Director. Herbarium management.

Taxonomy of Asteraceae: Gnaphalieae

Bredenkamp, Mrs C.L. Ph.D. Principal Agricultural Scientist. Assistant Curator: Public relations.

Taxonomy of Vitex, Phylica, Rhamnaceae, Sterculiaceae and other related families

Herman. P.P.J. M.Sc. Principal Agricultural Scientist. Assistant Curator: Personnel.

Taxonomy of Asteraceae, Flora of Transvaal

Mothogoane, M.S. Assistant Curator: Herbarium assistants. Wing C

Sebothoma, P.N. Cert. (Sec.). Assistant Curator: Service room. Plant identifications co-ordinator

Van Rooy, J. Ph.D. Assistant Curator: Technical staff. Taxonomy and biogeography of mosses

236

Bothalia 33,2 (2003)

Anderson, J.M. Ph.D. Specialist Scientist. Palaeobotany,

palaeogeography

Archer Mrs C. M.Sc. Senior Agricultural Scientist. Taxon-

omy of Cyperaceae, Restionaceae, Orchidaceae

Archer, R.H. Ph.D. Senior Agricultural Scientist. Taxon-

omy of mainly Celastraceae, Euphorbiaceae

Burgoyne, Ms PM. M.Sc. Principal Agricultural Scientist.

Mesembryanthemaceae

Fish, Mrs L. B.Sc. Principal Agricultural Scientist. Taxon-

omy of Poaceae. Collections manager

Glen, H.F. Ph.D. Specialist Scientist. Taxonomy of trees, her-

barium for cultivated plants, and botanical collectors

Glen. Mrs R.P. M.Sc. Chief Agricultural Development Techni-

cian. Taxonomy of ferns, water plants

Gotzel, Ms A. Senior Provisioning Admin. Clerk III

Jordaan, Mrs M. M.Sc. Principal Agricultural Scientist. Taxon-

omy of Casuarinaceae-Connaraceae, Maytenus

Kgaditsi, T.W. Senior Auxiliary Services Officer. Specimen

mounter, general assistant

Klein, R.G. Plant Collector, Bioprospecting Project (contract

worker)

Makgakga, M.C. B.Sc. Agricultural Development Techni-

cian. Herbarium Assistant, laboratory. Wing B

Makgakga, K.S. Senior Auxiliary Services Officer. Herba-

rium Assistant. Encoding plant specimens

Manamela, Ms M.T. B.Sc. Agricultural Development Techni-

cian. Information Officer

Maserumule, M.K. Auxiliary Services Officer I. Wing B

Masombuka, Ms A.S. N.Dip.(Nature Cons.). Senior Auxiliary

Services Officer. Herbarium Assistant. Wing A

Meyer, J.J. H.E.D. Chief Agricultural Development Techni-

cian. Bioprospecting Project

Mmakola, E.K. Data Capturer (SABONET contract worker)

Mothapo, M.A. Data Capturer (SABONET contract worker)

Nkoane, Ms G.K. Senior Auxiliary Services Officer. Loans,

exchanges, parcelling, stores

Nkonki, Mrs T. B.Sc. Senior Agricultural Development

Technician. Fabaceae taxonomy. Wing B

Nkuna, L.A. Agricultural Development Technician. Kew

Millenium Seed Bank Project (contract worker)

Perold, Mrs S.M. Ph.D. Taxonomy of Hepaticae (contract

worker)

Phahla, T.J. Senior Auxiliary Services Officer. Specimen

mounter of cryptogams, packer

Ready, Mrs J.A. N.Dip.(Hort). Principal Auxiliary Services

Officer. Plant identifications, Helichrysum. Wing D

Rampho, Ms E.T. Data Quality Controller (SABONET

contract worker)

Retief, Miss E. M.Sc. Principal Agricultural Scientist. Taxon-

omy of Boraginaceae, Verbenaceae, Lamiaceae, Aster-

aceae, Rubiaceae

Smithies, Mrs S.J. M.Sc., Dip. Ed. (Moray House). Chief

Agricultural Development Technician. Taxonomy

of Scrophulariaceae, Selaginaceae, Lobeliaceae

Steyn, Ms C.C. Principal Auxiliary Services Officer. Label

typist, service room support

Van Wyk, E. M.Sc. Agricultural Development Technician.

Seedbank Manager, Kew Millenium Seedbank Project

(contract worker)

Victor, Ms J.E. M.Sc., H.Dip.(Joum.). Principal Agricultural

Scientist. Taxonomy of Rutaceae, Asclepiadaceae

Welman, Ms W.G. M.Sc. Principal Agricultural Scientist.

Taxonomy of Convolvulaceae, Solanaceae, Cucurbit-

aceae, Campanulaceae, Asteraceae, Acanthaceae

DATA MANAGEMENT— PRETORIA (RPDC)

Arnold, T.H. M.Sc. Assistant Director. Computer database application especially in taxonomy

Botha, Mrs A.G. Chief Auxiliary Services Officer. Admini-

strative Assistant

De Wet, Mrs B.C. B.Sc.(Computer Science), B.A., H.D.L.S.

Principal Agricultural Datametrician

Neveling, Mrs V.H. Principal Typist I

Smit, G.C. NT Workstation 4, NT Server 4. Principal Net-

work Controller

Snyman, Mrs E.E. B.Sc. N.Dip.(Comp. Data Proc.). Senior

Agricultural Development Technician

Steyn, Ms H.M. Botanical Information Officer (contract

worker)

Swelankomo, Ms N. Agricultural Development Techni-

cian. Quality control

SABONET— PRETORIA (YSGE/IS)

Siebert, S.J. Ph.D. Regional Co-ordinator (contract worker)

Davis, Ms L.F. Senior State Accountant (part time contract worker)

Klopper, Ms R. M.Sc. Technician (contract worker)

Mossmer, Ms M. B.Sc. (Hons). Publications and Website Management (contract worker)

Noko, Ms N.R. Admin. Officer (contract worker)

ECOLOGY AND CONSERVATION SUBDIRECTORATE (RREL)

CAPE TOWN

Rutherford, M.C. Ph.D., Dip.(Datamet.). Deputy Director: Research

Morkel, Ms L. N. Dip. (Office Admin.). Senior Provisioning Admin. Clerk III. Personal

Assistant to Deputy Director: Research

Parenzee, Ms H.A. Dip. (Ed.). Senior Provisioning Admin. Clerk III

Powrie, L.W. M.Sc. Chief Information Technology Advisor. Spatial modelling, databases

Bothalia 33,2 (2003)

237

CLIMATE CHANGE

Rutherford, M.C. Ph.D., Dip.(DatameL). Chief Specialist Scientist.

Modelling, global change

Arnolds, Ms J.L. Chief Auxiliary Services Officer. Laboratory

Kgope, B.S. M.Sc. Senior Agricultural Scientist. Plant ecophysiology

Midgley, G.F. Ph.D. Principal Specialist Scientist. Plant ecophysiology, modelling

Millar, Ms D.L. M.Sc. Research Co-ordinator (contract worker)

Motete, Ms N. B.Sc.(Bot. & Ed.), M.Sc.(Envir. Biol.), Dip. (Science Ed.).

Senior Agricultural Scientist

Musil, C.F. Ph.D. Senior Specialist Scientist. Ecophysiology, modelling

Snyders, S.G. Senior Auxiliary Services Officer II. Greenhouse, maintenance

CONSERVATION BIOLOGY

Donaldson, J.S. Ph.D. (Zoology) Principal Specialist Scientist.

Supervisor: Conservation farming. Cycad biology

Bosenberg, J. de Wet. B. Sc. (Hons). Chief Agricultural Development Technician.

Cycad biology. Conservation farming

Ebrahim, 1. N.Dip.(HorL). Assistant. Protea Atlas Project (contract worker)

Marinus, E.M. N. Cert. (Building & Structures). Chief Auxilliary Services Officer.

Conservation farming

Rebelo, A.G. Ph.D. (Zoology). Principal Agricultural Scientist. Protea Atlas Project

Smit, W.J. M.A. Geographic Information Assistant. Protea Atlas Project (contract worker)

CONSERVATION FARMING PROJECT

Nanni, Ms I. B.Sc., H.E.D. Control Agricultural Development Technician.

Project Co-ordinator

Breebaart, Ms L. M.Sc. (Range and Forage Resources). Researcher. Production benefits of different

grazing systems (Nama Karoo) (contract worker)

Leonhart. Ms A. N. Dip. (Nature Cons.). Research Assistant. Assessments of biodiversity and

ecosystem services (contract worker)

Segers, Ms A. Senior Provisioning Admin. Clerk III (contract worker)

Skowno, A. M.Sc. Research Officer. Biodiversity risk analysis in farm landscapes (contract worker)

Theron, L.J. M.Sc. (Zoology). Research Assistant (contract worker)

DESERTIFICATION

Petersen. Ms A. B.Sc. (Hons). Senior Agricultural Development Technician.

Land use and vegetation mapping

HORTICULTURAL RESEARCH

Brown, N.A.C. Ph.D. Specialist Scientist. Seed research

INFORMATION TECHNOLOGY (SOUTH) (RRIT)

Evans, N. Control Network Controller. Technical IT Manager

Pekeur. Ms B.L. IT Admin. Clerk (contract worker)

SUPPORT SERVICES

Bardien-Overmeyer, Ms S. B.A.(Pharm.). Senior Provisioning Admin. Officer. Admin. Manager

Bowler, Mrs M. Specialist Cleaner. Assistant: teas and functions

De Witt. D.M. Senior Artisan. Maintenance

HARRY MOLTENO LIBRARY (RRLC)

Jagger, B.W. B.A.(Soc. Sc.), PGDip.Lis. Librarian

Ovens, Ms C.S.H. Ph.D.(Inf.Sc.), Dip.Datametrics. (contract librarian)

NBI WEBSITES (AMWS)

Reynolds, Ms P.Y. M.A.(Inf.Sc.), B.Proc., Dip.Datametrics. Principal Librarian.

NBI Web Site Manager

238

PUBLICATIONS BY THE STAFF

1 April 2002-31 March 2003

Bothalia 33,2 (2003)

ARCHER. C. 2002. Review: Bulbs. Revised edition by John E. Bryan,

2002. Bothalia 32: 251, 252.

ARCHER, C. & VICTOR. J.E. 2002. Cyperaceae. In J.S. Golding, South-

ern African plant Red Data Lists: 100. Ill, 112, 119. Southern

African Botanical Diversity Network Report No. 14.

ARNOLD. T.H. & SIEBERT, S. 2002. Computerisation of southern African

Herbaria: regional update. SABONET News 7: 92-96.

AUBREY, A. 2003-01 . Pterocarpus angolensis DC. (Fabaceae). Internet

3 pp. http://www.plantzafrica.com/plantnop/pterocarpangol.htm.

AUBREY, A. & LETSELA, M. 2002-04. Dovyalis zeyheri (Sond.) Warb.

(Flacourtiaceae). Internet 3 pp.

http://www.plantzafrica.com/plantcd/dovyaliszeyher.htm.

BAIJNATH, H. & SINGH, Y. (eds). 2002. Rebirth of science in Africa. A

shared vision for life and environmental sciences. Umdaus Press,

Hatfield, Pretoria.

BARKER. N.P., WESTON, P.H., ROURKE, J.P & REEVES, G. 2002. The

relationships of the southern African Proteaceae as elucidated by

internal transcribed spacer (ITS) DNA sequence data. Kew Bulletin

57: 867-883.

BEHR, K. 2003-01. Leucospermum saxosum S. Moore (Proteaceae).

Internet 2 pp.

http://www.plantzafrica.com/plantklm/leucospermsax.htm.

BEYERS, J.B.P 2002a. A new species of Gnidia from the Knersvlakte,

Western Cape. South Africa (Thymelaeaceae). Bothalia 32: 79-81.

BEYERS, J.B.P. 2002b. Reduction of Foveolina albida to Foveolina

dichotoma (Asteraceae-Anthemideae). Bothalia 32: 185.

BEYERS, J.B.P. 2002c. A new species of Arctotheca from northern Cape,

South Africa (Asteraceae). Bothalia 32: 185-187.

BEYERS, J.B.P. BREDENKAMP, C.L. & VICTOR, J.E. 2002. Thyme-

laeaceae. In J.S. Golding, Southern African plant Red Data Lists:

105, 116, 117, 120. Southern African Botanical Diversity Net-

work Report No. 14.

BEZUIDENHOUT, K. & GLEN, H. 2002. Editorial introduction. De

Nummis 5: 5.

BREDENKAMP, C.L. & VAN WYK, A.E. 2002a. Abstract: Mono-

graph of the genus Passerina L. (Thymelaeaceae). 28th Annual

Congress of the SA Association of Botanists held at Rhodes

University, Grahamstown, 13-16 January 2002. South African

Journal of Botany 68: 245.

BREDENKAMP, C.L. & VAN WYK, A.E. 2002b. Passerina quadrifaria

(Thymelaeaceae): a new species from the southern Cape and

Little Karoo in Soulh Africa. South African Journal of Botany 68:

304-307.

BREDENKAMP, C.L. & VAN WYK, A.E. 2002c. Taxonomy of the Passe-

rina filiformis complex (Thymelaeaceae). Bothalia 32: 29-36.

BREDENKAMP, C.L. & VAN WYK, A.E. 2002d. Systematics of Passe-

rina truncata and a new subspecies monticola (Thymelaeaceae).

Bothalia 32: 65-7 1 .

BREDENKAMP, C.L. & VAN WYK, A.E. 2002e. A new species of Passe-

rina from Western Cape, South Africa (Thymelaeaceae). Bothalia

32: 76-79.

BROWN, N. & BOTHA, P. 2002. Smoking seeds. An updated list of fyn-

bos species with seeds that have a good germination response to

smoke. Veld & Flora 88: 68, 69.

BURGOYNE, P.M. & PILLAY, N. 2002. Abstract: Pollination of Cytinus

visserii — the northern-most species of this Cape-centred genus.

28th Annual Congress of the SA Association of Botanists held at

Rhodes University, Grahamstown, 13-16 January 2002. South

African Journal of Botany 68: 246.

BUYS, M.H., BURGOYNE, P. & T1EDT, L.R. 2002. Abstract: Paly-

nology and the Stomatium group in the Aizoaceae. 28th Annual

Congress of the SA Association of Botanists held at Rhodes

University, Grahamstown, 13-16 January 2002. South African

Journal of Botany 68: 246.

CAFFERTY, S„ OLIVER, E.G.H. & OLIVER, I. 2002. (1561-1563) Pro-

posals to conserve the names Erica calycina, E. corifolia, and E.

imbricata (Ericaceae) with conserved types. Taxon 51: 810-812.

CAROLUS, B. 2002. Threatened plants programmes. SABONET News

7: 131-133.

CAROLUS, B. 2002-1 1 Aloe plicatilis (L.) Mill. (Asphodelaceae). Inter-

net 2 pp. http://www.plantzafrica.com/plantab/aloeplicatilis.htm.

CAROLUS, B. 2002- 1 1 . Phylica buxifolia L. (Rhamnaceae). Internet 2

pp. htlp://www.planlzafrica.com/plantnop/phylicabux.htm.

CAROLUS, B. 2002-1 1 . Psoralea pinnata L. (Fabaceae). Internet 2 pp.

http://www.plantzafrica.com/plantklm/psoraleapin.htm.

CAROLUS, B. 2002-11. Struthiola myrsinites Lam. (Thymelaeaceae). Inter-

net 2 pp. http://www.plantzafrica.com/plantqrs/struthiolmyrsin.htm.

CHESSELET, P. & SMITH, G.F. 2002. The rediscovery of Ruschia

acutangula (Haw.) Schwantes (Mesembryanthemaceae Fenzl).

Bradleya 20: 17-22.

CHESSELET, P. & VAN WYK, A.E. 2002. Mesembs with nut-like schizo-

carpic fruit and Ruschianthemum Friedrich sunk under Stoebe-

ria Dinter & Schwantes (Mesembryanthemaceae). Bothalia 32:

187-190.

CHESSELET. P„ SMITH, G.F. & VAN WYK, A.E. 2002. A new tribal

classification of Mesembryanthemaceae: evidence from floral

nectaries. Taxon 51: 295-308.

CLEEVELY, R.J. & OLIVER, E.G.H. 2002. A preliminary note on the

publication dates of H.C. Andrews’ Coloured engravings of

heaths (1794-1830). Archives of natural history 29: 245-264.

The Society for the History of Natural History.

CONRAD, F. & REEVES, G. 2002. Molecular systematics of the genus

Clivia. Clivia Four: 20-23.

CONRAD. F. & SNIJMAN, D. 2002. Abstract: Molecular systematics of

the tribe Haemantheae (Amaryllidaceae) based on plastid DNA

sequence data. 28th Annual Congress of the SA Association of

Botanists held at Rhodes University, Grahamstown, 13-16

January 2002. South African Journal of Botany 68: 247.

CROUCH, N.R. 2002. Krantzkloof Nature Reserve: plant refuge and

botanists’ delight. Palmnut Post 5: 10-13.

CROUCH, N.R., CHETTY, J„ MULHOLLAND, D.A. & NDLOVU, E.

2002. Bulb alkaloids of the reputedly psychoactive Brunsvigia

radulosa (Amaryllidaceae). South African Journal of Botany 68:

86-89.

CROUCH, N.R., SMITH, G.F., LOTTER, M. & SYMMONDS, R.

2003. Encephalartos woodii: the first historically documented

mutlti casualty in southern Africa. Encephalartos 73: 22, 25-28.

CROUCH, N.R., SMITH, G.F., SYMMONDS, R. & CROUCH, T.E. 2002.

Batting curses and beating depression: Adenia gummifera in

Zulu traditional medicine. Bradleya 20: 67-78.

CROUCH, N.R. & STYLES, D. 2002. A new locality for Impatiens flana-

ganiae (Balsantinaceae) in the Transkei. PlantLife 27: 10, 11.

CUPIDO, C.N. 2002. A new species of Merciera from Western Cape,

South Africa (Campanulaceae). Bothalia 32: 74-76.

DE BRUYN, A. 2002. Environmental education in grasslands.

SABONET News 7 (Poaceae special edition): 200-204.

DLAMINI, M.D. 2002-05. Svzygiuin pondoense Engl. (Myrtaceae). Inter-

net 3 pp. http://www.plantzafrica.com/plantqrs/syzigpond.htm.

DLAMINI, M.D. 2002-07. Dimorphotheca sinuata DC. (Asteraceae). Inter-

net 2 pp. http://www.plantzaffica.com/plantccl/dimorphsinuata.htm.

DLAMINI, M.D. 2002-07. Jamesbrittenia grandiflora (Galpin) Hilliard

(Scrophulariaceae). Internet 2 pp.

http://www.plantzafrica.com/planthij/jamesbritgrand.htm.

DLAMINI, M.D. & AUBREY, A. 2002-05. Diascia rigescens Benth. (Scro-

phulariaceae). Internet 3 pp.

http://www.plantzafrica.com/planlcd/diasciarig.hlm.

DLAMINI, M.D. & HANKEY, A. 2002-06. Heteropyxis natalensis Harv.

(Heteropyxidaceae). Internet 2 pp.

http://www.plantzafrica.com/planthij/heteropnatal.htm.

DLAMINI, M.D. & TURNER, S. 2002-05. Buxus macowanii Oliv. (Bux-

aceae). Internet 3 pp.

http://www.plantzafrica.com/plantab/buxusmacowen.htm.

DONALDSON. J„ MILLS, A., O’FARRELL, P„ TODD, S„ BOSEN-

BERG, J.D. & NANNI, I. 2002. Abstract: Can plant diversity

save itself? Preliminary results from the study of conservation

farming and its benefits to farmers and rural communities on

the Bokkeveld Plateau. 28th Annual Congress of the SA As-

sociation of Botanists held at Rhodes University, Grahamstown,

13-16 January 2002. South African Journal of Botany 68: 249.

DONALDSON, J„ NANNI, I., ZACHARIADES, C. & KEMPER. J. 2002.

Effects of habitat fragmentation on pollinator diversity and

plant reproductive success in Renosterveld shrublands of South

Africa. Conservation Biology 16: 1267-1276.

DUNCAN, G. 2002a. 448. Nerine gaberonensis (Amaryllidaceae).

Curtis's Botanical Magazine 19: 173-177.

DUNCAN, G. 2002b. Brunsvigia pulchra. The spectacular Namaqua-

land tulip. Veld & Flora 88: 60-62.

DUNCAN, G. 2002c. Dwarf watsonias. Ten of the best for containers

and rock gardens. Veld & Flora 88: 94—98.

DUNCAN, G.D. 2002d. The genus Nerine. Bulbs 4,1: 9-15.

Bothalia 33,2 (2003)

239

DUNCAN, G.D. 2002e. Review: Gardening with Keith Kirsten. Veld &

Flora 88: 74, 75.

DUNCAN, G.D. 2002f. Just holding on — spectacular geophytes in peril.

Veld & Flora 88: 142-147.

DUNCAN, G.D. 2002g. Graham Dugald Duncan, 2001 Herbert Medalist

(autobiography). Herbenia 56: 12-17.

DUNCAN, G. 2002-04. Strumaria truncata Jacq. ( Amaryllidaceae). In-

ternet 3 pp.

http://www.plantzafrica.com/plantqrs/stRirnartruncata.htm.

DUNCAN, G. 2002-06. Haemanthus albiflos Jacq. (Amaryllidaceae).

Internet 3 pp.

http://www.plantzafrica.com/planthij/haemanalbif.htm.

DUNCAN, G. 2002-07. Cyrtanthus mackenii Hook.f. (Amaryllida-

ceae). Internet 3 pp.

http://www.plantzafrica.com/plantcd/cyrtanthmack.htm.

DUNCAN, G.D. 2003a. Polyxenas. Veld & Flora 89: 22-26.

DUNCAN. G.D. 2003b. Review: Clivias, by Harold Koopowitz, 2002.

Veld & Flora 89: 36.

DUNCAN, G.D. 2003-03. Nerine masoniorum L. Bol. (Amaryllida-

ceae). Internet approx. 3 pages, http://www.plantzafrica.com.

DUNCAN. G.D. & EDWARDS. T.J. 2002. A new species of Lache-

nalia from Namaqualand, South Africa (Hyacinthaceae: Mas-

sonieae). Bothalia 32: 190-192.

EGGLI, U. & VAN JAARSVELD. E. 2002. Convolvulaceae. In U. Eggli.

Sukkulenten-Lexikon Band 2: Zweikeimblattrige Pflanzen (Di-

cotyledonen ): 69. Eugen Ulmer, Stuttgart.

FISH, L. 2002a. Did you know? SABONET News 7 (Poaceae special

edition): 205, 206.

FISH, L. 2002b. Preparing Poaceae herbarium specimens. SABONET

News 7 (Poaceae special edition): 183-185.

FISH. L. 2002c. Useful Poaceae literature. SABONET News 7 (Poaceae

special edition): 219, 220.

FISH. L. & STEYN, H. 2002. A single map — much information and many

uses. SABONET News 7 (Poaceae special edition): 186-188.

FORSTER. P.I. & VAN JAARSVELD, E. 2002. Plectranthus (Lami-

aceae). In U. Eggli, Sukkulenten-Lexikon Band 2: Zweikeimblat-

trige Pflanzen ( Dicotyledonen ): 310-323. Eugen Ulmer, Stuttgart.

GAVHI, P. with additions by Yvonne Reynolds. 2002-10. Rhus erosa

Thunb. (Anacardiaceae). Internet 2 pp.

http://www.plantzafrica.com/plantqrs/rhuserosa.htm.

GAVHI. P. & HARRIS, S. with additions by Yvonne Reynolds. 2002-10.

Cussonia paniculata Eckl. & Zeyh. (Araliaceae). Internet 3 pp.

http://www.plantzafrica.com/plantcd/cussoniapan.htm.

GERMISHUIZEN. G. 2002. Fabaceae; Loranthaceae; Viscaceae. In U.

Eggli, Sukkulenten-Lexikon Band 2: Zweikeimblattrige Pflanzen

( Dicotyledonen ): 244—252; 328, 329; 481, 482. Eugen Ulmer,

Stuttgart.

GLEN, H.F. 2002a. A collection of Wanderers Club membership badges.

De Nummis 5: 24—31.

GLEN. H.F. 2002b. Notes on trade dollars of the Indo-Pacific Region,

with particular reference to the Spanish, Maria Theresa and

British dollars. De Nummis 5: 51-54.

GLEN, H.F. 2002c. Notes towards a southern African garden flora 4:

Ulmaceae, Part 1. Trees in South Africa: 2-5. The Tree Society

of Southern Africa, Johannesburg.

GLEN. H.F. 2002d. Obituary: Anna Amelia Obermeyer-Mauve ( 1 907—

2001). Bothalia 32: 127-130.

GLEN, H.F. 2002e. The synonymy of Ceraria namaquensis (Sond.)

H.Pearson & E.L.Stephens (Portulacaceae). Bothalia 32: 196, 197.

GLEN, H.F. 2002f. Cultivated plants of southern Africa. Jacana Educa-

tion, Johannesburg.

GLEN, H.F. 2002g. News from South Africa. SABONET News 7: 245.

GLEN. H.F. 2002h. Review: Eucalypts of South-eastern Australia CD-

ROM, by M.I.H. Brooker, J.R. Connors, & A.V. Slee, 2002.

Plant Systematics and Evolution 233: 127-129.

GLEN, H.F. 2002i. Review: Rebirth of science in Africa. A shared

vision for life and environmental sciences, edited by H.

Baijnath & Y. Singh. 2002. SABONET News 7: 243.

GLEN, H.F. 2002j. Taxonomic notes. In E. Schmidt. M. Lotter & W.

McCleland. Trees and shrubs of Mpumalanga and Kruger

National Park. Jacana Education, Johannesburg.

GLEN, H.F & SIEBERT, S. 2002a. The paper chase. SABONET News

7: 138-143.

GLEN. H.F. & SIEBERT. S. 2002b. The paper chase. SABONET News

7: 233-238.

GOLDBLATT. P.. BERNHARDT, P. & MANNING, J.C. 2002. Floral

biology of Romulea (Iridaceae: Crocoideae): a progression

from a generalist to a specialist pollination system. Adansonia,

sen 3: 243-262.

GOLDBLATT, P. & MANNING, J.C. 2002a. Notes and new species of

Moraea (Iridaceae: Iridoideae) from the southern African win-

ter rainfall zone. Novon 12: 352-359.

GOLDBLATT, P. & MANNING. J.C. 2002b. Evidence for moth and

butterfly pollination in Gladiolus (Iridaceae-Crocoideae).

Annals of the Missouri Botanical Garden 89: 110-124.

GOLDBLATT, P. & MANNING, J.C. 2002c. Plant diversity of the Cape

region of southern Africa. Annals of the Missouri Botanical

Garden 89: 281-302.

GOLDBLATT. P. & MANNING, J.C. 2002d. Rediscovered after 200

years, Moraea ovata Thunberg is a rare Ferraria (Iridaceae: Iri-

doideae) from Namaqualand. South Africa. Novon 12: 460-464.

GOLDING, J. 2002. Threatened plants of southern Africa: trends and

general recommendations. In G.H. Verdoorn & J. le Roux, The

state of South Africa 's species: 51-63. Proceedings of a confer-

ence held at the Rosebank Hotel, Johannesburg, 4—7 September

2001. Endangered Wildlife Trust. Johannesburg.

GOLDING, J. & HURTER. P.J.H. 2003. A Red List account of Africa’s

cycads and implications of considering life history and threats.

Biodiversity and Conser\>ation 12: 507-528.

GOLDING, J. & SIEBERT, S. 2002. New Red Data books. Threatened

plants of southern Africa. Plant Talk 30: 40, 41 .

HAMMER, S.A. & VICTOR. J.E. 2002. In J.S. Golding, Southern Afri-

can plant Red Data Lists: 95, 107, 108, 118. Southern African

Botanical Diversity Network Report No. 14.

HANKEY, A. 2002. Vegetative key to the genus Eedebouria (Hya-

cinthaceae) in South Africa. PlantLife 27: 16-19.

HANKEY, A. & MASHININI, L. 2002-04. Setaria megaphylla (Steud.)

T.Durand & Schinz (Poaceae). Internet 2 pp.

http://www.plantzafrica.com/plantqrs/setariamegaphyl.htm.

HERMAN, P.P.J. 2002a. Die plantfamilie Asteraceae: 3. Die vrug. SA

Tydskrif vir Natuurwetenskap en Tegnologie 21: 19-21.

HERMAN, P.P.J. 2002b. Die plantfamilie Asteraceae: 4. Interessante

groeivorme en ekonomies belangrike soorte. SA Tydskrif vir

Natuurwetenskap en Tegnologie 21: 108-110.

HERMAN, P.P.J. 2002c. Revision of the Tarchonanthus camphoratus

complex (Asteraceae-Tarchonantheae) in southern Africa. Botha-

lia 32: 21-28.

HERMAN, P.P.J. & RETIEF. E. 2002. New combinations in the genus

Rotheca in southern Africa (Lamiaceae). Bothalia 32: 81.

HERMAN, P.P.J., WELMAN, W.G. & VICTOR, J.E. 2002. Asteraceae.

In J. Golding, Southern African plant Red Data Lists: 98-100,

109-111, 118, 119. Southern African Botanical Diversity Net-

work Report No. 14.

HITCHCOCK, A. 2002. Erica Gengold. the golden hybrid. Veld & Flora

88: 89.

HITCHCOCK. A. 2002-04. Mimetes chrysanthus Rourke (Proteaceae).

Internet 4 pp.

http://www.plantzafrica.com/plantklm/mimeteschrysan.htm.

HITCHCOCK, A. 2002-05. Erica urna-viridis Bolus (Ericaceae). Inter-

net 4 pp.

http://www.plantzafrica.com/plantefg/ericaumavirid.htm.

HITCHCOCK, A. 2002-06. Diastella proteoides (L.) Druce (Protea-

ceae). Internet 3 pp.

http://www.plantzafrica.com/plantcd/diastelprot.htm.

HITCHCOCK, A. 2002-11. Mimetes hottentoticus E. Phillips & Hutch.

(Proteaceae). Internet 4 pp.

http://www.plantzafrica.com/plantklm/mimeteshot.htm.

HITCHCOCK, A. & ADAMS, T. 2002-11 . Serruria aemula Salisb. ex

Knight (Proteaceae). Internet 3 pp.

http://www.plantzafrica.com/plantqrs/serruriaaemula.htm.

HOFFMAN. T„ MIDGLEY, G. & POWRIE, L. 2002. Abstract: The inter-

relationships of Acocks’ Veld Types as portrayed in an unpub-

lished 1 95 1 diagram. 28th Annual Congress of the S A Association

of Botanists held at Rhodes University, Grahamstown, 13-16

January 2002. South African Journal of Botany 68: 253.

HUNTLEY, B.J. 2002. South Africa. The flowering of the Rainbow Nation.

Plant Talk 29: 34-39.

HUNTLEY, B.J.. WILLIS, C.K., SMITH, G.F. & SIEBERT, S.J. 2002.

SABONET: its history and success in southern Africa. In H.

Baijnath & Y. Singh, Rebirth of science in Africa: 231-246.

Umdaus Press. Pretoria.

HURTER, J. 2002a. An introduction to the African rainforest at the

Lowveld National Botanical Garden. PlantLife 27: 12-15.

HURTER. J. 2002b. Threatened plants programmes. SABONET News

7: 222, 223.

JAMIESON. H.G. 2002-06. Protea repens (L.) L. (Proteaceae). Inter-

net 4 pp.

http://www.plantzafrica.com/plantnop/protearepens.htm.

240

Bothalia 33,2 (2003)

JAMIESON, H.G, 2002-08. Restio festuciformis Nees ex Mast. (Res-

tionaceae). Internet 2 pp.

http://www.plantzafrica.com/plantqrs/restiofest.htm.

JAMIESON, H.G. 2002-09. Leucadendron macowanii E. Phillips (Pro-

teaceae). Internet 3 pp.

http://www.plantzafrica.com/plantklm/leucadenmacowan.htm.

JAMIESON, H.G. 2002-10. Restio quadratus Mast. (Restionaceae). Inter-

net 2 pp. http://www.plantzafrica.com/plantqrs/restioquad.htm.

JAMIESON, H.G. 2002-1 1 . Asparagus densiflorus (Kunth) Jessop (Aspa-

ragaceae). Internet 3 pp.

http://www.plantzafrica.com/plantab/asparagdens.htm.

JODAMUS, N. 2002-09. Acmadenia mundiana Eckl. & Zeyh. (Rut-

aceae). Internet 2 pp.

http://www.plantzafrica.com/plantab/acmadenmund.htm.

JODAMUS, N. 2003-01. Brachylaena discolor DC. (Asteraceae). In-

ternet 2 pp.

http://www.plantzafrica.com/plantab/brachylaendiscol.htm.

JODAMUS, N. & NOTTEN, A. 2002-04. Plectranthus fruticosus L’Her.

‘James’ (Lamiaceae). Internet 3 pp.

http://www.plantzafrica.com/plantnop/plectfrutjames.htm.

JODAMUS, N. & NOTTEN, A. 2002-07. Agathosma mucronulata

Sond. (Rutaceae). Internet 3 pp.

http://www.plantzafrica.com/plantab/agathosmucro.htm.

JOFFE, P. 2002-10. Freylinia lanceolata (L.f.) G.Don (Scrophula-

riaceae). Internet 2 pp.

http://www.plantzafrica.com/plantefg/freylinlan.htm.

JOFFE, P. 2002-1 1 . Buddleja auriculata Benth. (Loganiaceae). Internet

2 pp. http://www.plantzafrica.com/plantab/buddlejauricul.htm.

JOFFE, P. 2002-11. Kiggelaria africana L. (Flacourtiaceae). Internet 2

pp. http://www.plantzafrica.com/plantklm/kiggelarafric.htm.

JOFFE, P. 2002-1 1. Olea europaea L. subsp. africana (Mill.) P.S. Green

(Oleaceae). Internet 2 pp.

http://www.plantzafrica.com/plantnop/oleaeurop.htm.

JOFFE. P. 2002-11. Pavetta gardeniifolia A. Rich var. gardeniifolia

(Rubiaceae). Internet 2 pp.

http://www.plantzafrica.com/plantnop/pavetgarden.htm.

JOHNSON, I. 2002-08. Aloe tenuior Haw. (Asphodelaceae). Internet 2

pp. http://www.plantzafrica.com/plantab/aloetenuior.htm.

JOHNSON, I. 2002-09. Dermatobotrys saundersii Bolus (Scrophula-

riaceae). Internet 2 pp.

http://www.plantzafrica.com/plantcd/dermatobotsaund.htm.

JOHNSON, I. 2002-09. Gerbera jamesonii Bolus ex Adlam (Aster-

aceae). Internet 2 pp.

http://www.plantzafrica.com/plantefg/gerberajames.htm.

JORDAAN, M. 2002. Abstract: The infraspecific classification of Salix

mucronata Thunb. (Salicaceae) in southern Africa. 28th Annual

Congress of the SA Association of Botanists held at Rhodes

University, Grahamstown, 13-16 January 2002. South African

Journal of Botany 68: 255.

JOSSELOWITZ! L. & GLEN, H.F. 2002. Setting the time. De Nummis

5: 47-50.

KGOPE, B.S. & MIDGLEY, G.F. 2002. Abstract: Photosynthetic tem-

perature and light response of dominant Savanna tree function-

al types under wet and dry conditions. 28th Annual Congress of

the SA Association of Botanists held at Rhodes University,

Grahamstown, 13-16 January 2002. South African Journal of

Botany 68: 256.

KGOPE, B.S.' & MIDGLEY, G.F. 2002. Abstract: Sub-ambient C02, an

under-appreciated factor in understanding growth-form coexis-

tence in savannas? 28th Annual Congress oflhe SA Association

of Botanists held at Rhodes University, Grahamstown, 13-16

January 2002. South African Journal of Botany 68: 256.

KLOPPER, R.R., SMITH, g’f. & CHIKUNI, A.C. 2002. The Global

Taxonomy Initiative in Africa. Taxon 51: 159-165.

KLOPPER, R.R., SMITH, G.F. & VAN ROOY, J. 2002. The biodiver-

sity of Africa. In H. Baijnath & Y. Singh, Rebirth of science in

Africa: 60-86. Umdaus Press, Hatield, Pretoria.

KOF.KEMOER, M. 2002. News from South Africa. Lesotho to Ceder-

berg. SABONETNews 7: 246-249.

KOEKEMOER, M. & BOTHA, F.C. 2002. Abstract: Partial purifica-

tion and characterisation of sugarcane neutral invertase. 28th

Annual Congress of the SA Association of Botanists held at

Rhodes University, Grahamstown, 13-16 January 2002. South

African Journal of Botany 68: 257.

KURZWE1L, H. 2002a. Notes on the genus Machairophyllum (Mesem-

bryanthemaceae) — its taxonomy, distribution and ecology. Aloe

39: 8-13.

KURZWEII .,11. 2002b. Tire structure of the orchid flower. Malayan Orchid

Review 37: 56-60, 87. Orchid Society of SE Asia, Singapore.

KURZWEIL, H. & KOCYAN, A. 2002. Morphology. Ontogeny of

orchid flowers. Orchid biology: reviews and perspectives , VIII:

83-138. Kluwer Academic Publishers. The Netherlands.

KURZWEIL, H„ LILTVED, B. & VICTOR, J.E. 2002. Orchidaceae.

In J. Golding, Southern African plant Red Data Lists: 101, 102,

112-115, 120. Southern African Botanical Diversity Network

Report No. 14.

LAWRENCE. E. 2002-09. Pelargonium tomentosum Jacq. (Geraniaceae).

Internet 2 pp.

http://www.plantzafrica.com/plantnop/pelargtomento.htm.

LAWRENCE, E. 2002-11. Pelargonium grcweolens L'Her. (Geraniaceae). In-

ternet 2 pp. http://www.plantzafrica.com/plantnop/pelarggrav.htm.

LAWRENCE, E. 2002-11 . Pelargonium zonale (L.) L'Her. (Geraniaceae).

Internet 2 pp.

http://www.plantzafrica.com/plantnop/pelargzonale.htm.

LETSELA, M.S. 2002-06. Strelitzia nicolai Regel & Koern. (Strelitzia-

ceae). Internet 2 pp.

http://www.plantzafrica.com/plantqrs/strelitznichol.htm.

LETSELA, M.S. 2002-07. Dorotheanthus bellidifonnis (Burm.f) N.E.Br.

(Mesembryanthemaceae). Internet 2 pp.

http://www.plantzafrica.com/plantcd/dorotheanthusbell.htm.

LETSELA, M.S. with additions by REYNOLDS, Y. 2002-07. Nemesia

fruticans (Thunb.) Benth. (Scrophulariaceae). Internet 3 pp.

http://www.plantzafrica.com/plantnop/nemesiafrut.htm.

LETSELA, M.S. & HANKEY, A. with additions by REYNOLDS, Y.

2002-05. Tarchonanthus camphoratus L. (Asteraceae). Internet

2 pp. http://www.plantzafrica.com/planttuv/tarchoncamphor.htm.

MAKWARELA, A.M., MARAIS, E.M.. DREYER, L.L. & BELLST-

EDT, D.U. 2002. Abstract: Phylogenetic analysis of the genus

Zygophyllum (Zygophyllaceae) in southern Africa. 28th Annual

Congress of the SA Association of Botanists held at Rhodes

University, Grahamstown, 13-16 January 2002. South African

Journal of Botany 68: 259.

MALAN. C. & NOTTEN, A. 2002-11. Athanasia dentata (L.) L. (Aster-

aceae). Internet 2 pp.

http://www.plantzafrica.com/plantab/athanasiadent.htm.

MANNING, J.C. & FAY, M. 2002. Abstract: Molecular systematics of

Hyacinthaceae. 28th Annual Congress of the S A Association of

Botanists held at Rhodes University. Grahamstown, 13-16

January 2002. South African Journal of Botany 68: 259.

MANNING, J.C. & GOLDBLATT, P. 2002a. The pollination of Trito-

niopsis parviflora (Iridaceae) by the oil-collecting bee Rediviva

gigas (Hymenoptera: Melittidae): the first record of oil secretion in

African Iridaceae. South African Journal of Botany 68: 171-176.

MANNING, J.C. & GOLDBLATT, P. 2002b. A distinctive new species

of Felicia (Astereae) from Western Cape, South Africa (Aster-

aceae). Bothalia 32: 193-195.

MANNING, J.C., GOLDBLATT, P. & SNIJMAN, D. 2002. The color

encyclopedia of Cape bulbs. Timber Press, Portland, USA &

Cambridge, UK.

MANNING, J.C. & SNIJMAN. D. 2002. Hawkmoth pollination in

Crinum variabile (Amaryllidaceae) and the biogeography of

sphingophily in southern African Amaryllidaceae. South

African Journal of Botany 68: 212-216.

MANNING, J.C. & VAN DER MERWE, A.M. 2002a. A new combi-

nation in Daubenya (Hyacinthaceae). Bothalia 32: 63-65.

MANNING, J.C. & VAN DER MERWE, A.M. 2002b. Systematics of

the genus Daubenya (Hyacinthaceae: Massonieae). Bothalia

32: 133-150.

MBAMBEZELI, G. 2002-10. Greyia flanaganii Bolus (Greyiaceae). Inter-

net 3 pp. http://www.plantzafrica.com/plantefg/greyiaflan.htm.

MBAMBEZELI, G. & NOTTEN, A. 2002-08. Erythrina lysistemon Hutch.

(Fabaceae/Leguminosae). Internet 4 pp.

http://www.plantzafrica.com/plantefg/erythrinlysist.htm.

MBAMBEZELI, G. & NOTTEN, A. 2002-12. Ochna serrulata (Hochst.)

Walp. (Ochnaceae). Internet 3 pp.

http://www.plantzafrica.com/plantnop/ochnaserrulat.htm.

MBAMBEZELI, G. & NOTTEN, A. 2003-01. Celtis africana Burm.f.

(Ulmaceae). Internet 3 pp.

http://www.plantzafrica.com/plantcd/celtisafrican.htm.

MBAMBEZELI, G. & REYNOLDS, Y. 2002-05. Podocarpus henkelii

Stapf ex Dallim. & Jacks. (Podocarpaceae). Internet 4 pp.

http://www.plantzafrica.com/plantnop/podocarphenk.htm.

MBAMBEZELI, G. with additions by REYNOLDS, Y. 2002-06. Ster-

culia murex Hemsl. (Sterculiaceae). Internet 3 pp.

http://www.plantzafrica.com/plantqrs/sterculmurex.htm.

MEYER, N.L. & SMITH, G.F. 2002. Aloaceae. Astroloba , xAstrowor-

thia. In U. Eggli. Sukkulenten-Lexikon Band I : Einkeimbldttrige

Pflanzen (Monocotyledonen): 194-196. Eugen Ulmer, Stuttgart.

Bothalia 33,2 (2003)