An overview of the Cape geophytes

Abstract

The Cape Region (here treated as the winter rainfall region of southern Africa, thus including fynbos, renosterveld and succulent karoo vegetation) is the world's foremost centre of geophyte diversity. Some 2100 species in 20 families have been recorded from this area, 84% of them endemic. The most important families, with more than a hundred geophyte species each, are Iridaceae, Oxalidaceae, Hyacinthaceae, Orchidaceae, Amaryllidaceae and Asphodelaceae. Although southern Africa does not appear to have been the main diversification centre for the plant orders with highest geophyte representation (Asparagales and Liliales), it represents an active centre of transition to geophytism, such transitions having occurred independently in numerous plant groups, often followed by rapid speciation. Several Cape geophyte groups have consequently expanded across Africa to the Mediterranean Basin, and possibly to other winter rainfall regions. Remarkably high local species diversity in renosterveld vegetation, even in relatively homogeneous environments, suggests that pollinator specificity and phenology play an important role in niche partitioning. However, character diversity is also high in storage organs and leaves, and this could account for the high species diversity values recorded at larger spatial scales, especially across environmental gradients. Long-term climatic stability, combined with topoclimatic and edaphic diversity and regular fire occurrence, is likely to be responsible for the remarkable geophyte diversity of the Cape, as compared to other mediterranean-climate regions. © 2006 The Linnean Society of London, Biological Journal of the Linnean Society, 2006, 87, 27–43.

Full text

Biological Journal of the Linnean Society
, 2006,
87
, 27– 43. With 7 figures
© 2006 The Linnean Society of London,
Biological Journal of the Linnean Society,
2006,
87
, 27– 43
27
Blackwell Science, LtdOxford, UKBIJBiological Journal of the Linnean Society0024-4066The Linnean Society of London, 2005? 2005
871
2743
Original Article
AN OVERVIEW OF THE CAPE GEOPHYTES
s
. PROCHE
s
ET AL
.
*Corresponding author. Present Address: Centre for Invasion
Biology, Stellenbosch University, Private Bag X1, Matieland
7602, South Africa. E-mail: sproches@sun.ac.za
An overview of the Cape geophytes
S
ERBAN PROCHE
S
1*
, RICHARD M. COWLING
1
, PETER GOLDBLATT
2
,
JOHN C. MANNING
3
and DEIRDRÉ A. SNIJMAN
3
1
Department of Botany and Terrestrial Ecology Research Unit, University of Port Elizabeth 6031, South
Africa
2
A. Krukoff Curator of African Botany, Missouri Botanical Garden, PO Box 299, St. Louis, Missouri
63166–0299, USA
3
Compton Herbarium, South African National Biodiversity Institute, Private Bag X7, Claremont 7735,
South Africa
Received 7 May 2004; accepted for publication 1 February 2005
The Cape Region (here treated as the winter rainfall region of southern Africa, thus including fynbos, renosterveld
and succulent karoo vegetation) is the world’s foremost centre of geophyte diversity. Some 2100 species in 20 families
have been recorded from this area, 84% of them endemic. The most important families, with more than a hundred
geophyte species each, are Iridaceae, Oxalidaceae, Hyacinthaceae, Orchidaceae, Amaryllidaceae and Asphodelaceae.
Although southern Africa does not appear to have been the main diversification centre for the plant orders with high-
est geophyte representation (Asparagales and Liliales), it represents an active centre of transition to geophytism,
such transitions having occurred independently in numerous plant groups, often followed by rapid speciation. Sev-
eral Cape geophyte groups have consequently expanded across Africa to the Mediterranean Basin, and possibly to
other winter rainfall regions. Remarkably high local species diversity in renosterveld vegetation, even in relatively
homogeneous environments, suggests that pollinator specificity and phenology play an important role in niche par-
titioning. However, character diversity is also high in storage organs and leaves, and this could account for the high
species diversity values recorded at larger spatial scales, especially across environmental gradients. Long-term
climatic stability, combined with topoclimatic and edaphic diversity and regular fire occurrence, is likely to be
responsible for the remarkable geophyte diversity of the Cape, as compared to other mediterranean-climate
regions. © 2006 The Linnean Society of London,
Biological Journal of the Linnean Society
, 2006,
87
, 27–43.
ADDITIONAL KEYWORDS:
bulbs – corms – morphological plasticity – phylogenetic constraints – rhizomes
– tubers.
INTRODUCTION
Mediterranean climates are well known for their
remarkable plant diversity (Cowling
et al
., 1996), and
the high numbers of geophytes they harbour have long
been noted (Raunkiaer, 1934). The geophyte floras of
some of these regions have been the subject of taxo-
nomic and ecological reviews (e.g. south-western Aus-
tralia: Pate & Dixon, 1982; Parsons & Hopper, 2003;
California: Rundel, 1996; Chile: Hoffmann, Liberona
& Hoffmann, 1998). The Cape Flora of South Africa is,
however, by far the richest in geophyte species (Fig. 1),
and while Cape petaloid monocots have recently
received a popular review (Manning, Goldblatt & Snij-
man, 2002), a comprehensive taxonomic, ecological
and evolutionary assessment for the Cape geophytes
is still lacking.
A number of recent monographs (e.g. Goldblatt,
1986, 1989; Marais, 1994; Perry, 1994; Goldblatt &
Manning, 1998; Linder & Kurzweil, 1999) clearly indi-
cate that good taxonomic information has been accu-
mulated, at least as far as some large genera are
concerned. These are mainly genera of horticultural
importance, with many species now being cultivated
worldwide. However, significant discoveries continue
to be made even in such groups, including the recent
discovery of a new
Clivia
near Nieuwoudtville,
hundreds of kilometres from the closest previously

28
S
. PROCHE
S
ET AL
.
© 2006 The Linnean Society of London,
Biological Journal of the Linnean Society,
2006,
87
, 27– 43
recorded species (Rourke, 2002). The current state
of knowledge of other Cape geophytes (e.g.
Bulbine,
Ornithogalum
) is, however, comparatively poor.
Among the major Cape geophyte families, the
Iridaceae and Orchidaceae have probably received
the most extensive taxonomic coverage (Linder &
Kurzweil, 1999; Manning
et al
., 2002). In the
Hyacinthaceae, a new generic-level organization is
emerging from molecular studies, resulting in numer-
ous genera being merged under
Drimia, Daubenya,
Lachenalia
, and
Ornithogalum
(Goldblatt & Manning,
2000a; Manning & van der Merwe, 2002; Manning,
Goldblatt & Fay, 2004). The discovery of new species
continues at a significant rate in these families,
with 11 new species of
Lachenalia
(Hyacinthaceae)
described since 1993 (e.g. Duncan, 1998). For the Oxal-
idaceae, an extensive, if outdated, review is available
(Salter, 1944), and the geophytic Asphodelaceae are
likewise in need of modern taxonomic study. Phyloge-
netic studies are underway for numerous geophytic
taxa (Cape clades included), making use of the taxo-
nomic information accumulated for the better-known
groups and helping to clarify the status of the lesser-
known ones (e.g. Fay
et al
., 2000; Vinnersten &
Bremer, 2001; Manning
et al
., 2004).
One field in which geophyte research has been
particularly productive is pollination biology. It is
apparent that pollinator specificity plays an important
role in the maintenance of Cape geophyte diversity,
and the Cape has become an important area for polli-
nation research (Johnson, 1996; Goldblatt & Manning,
2000b; Johnson & Steiner, 2000). Progress has also
been made in geophyte ecophysiology, with an emp-
hasis on the importance of morphology and life cycles
for adaptation to local environmental conditions
(Johnson, 1992; Le Maitre & Brown, 1992; Ruiters &
McKenzie, 1994; Esler, Rundel & Vorster, 1999; Rossa
& von Willert, 1999).
Despite this impressive progress in our understand-
ing of Cape geophyte taxonomy and ecology, a compre-
hensive picture is still lacking concerning the actual
extent of geophytism in the Cape. In particular, there
is a need for integrating phylogenetic information in
the study of geophytism (Parsons & Hopper, 2003).
Here we review the literature on Cape geophytes in
order to determine the main factors responsible for
geophyte diversification in the Cape, from a macro-
phylogenetic as well as ecological perspective.
WHAT IS THE APPROPRIATE STUDY AREA?
So far, the best coverage of the Cape geophytes is that
provided by Goldblatt & Manning (2000a) for the Cape
Floristic Region (CFR), which coincides largely with
Figure 1.
Geophyte diversity in the five mediterranean-climate regions of the world. Data from Cowling
et al
. (1996) and
Parsons & Hopper (2003), with an additional estimate of 1335 species for the Mediterranean Basin (
S
. Proche
s
, unpubl.
data). Data for the Cape, as compiled in this paper.

AN OVERVIEW OF THE CAPE GEOPHYTES
29
© 2006 The Linnean Society of London,
Biological Journal of the Linnean Society,
2006,
87
, 27– 43
the fynbos biome (comprising the fire-prone fynbos
and renosterveld shrublands), but also includes
enclaves of other biomes, such as Afromontane forest,
succulent karoo and subtropical thicket. The review of
Manning
et al
. (2002) includes, in addition to the CFR,
the Roggeveld (a relatively small area of the winter-
rainfall part of the Great Escarpment) as well as the
extensive and arid basin of the Tanqua Karoo. It is,
however, clear that high geophyte diversity is not
strictly associated with typical CFR vegetation (fynbos
and renosterveld); the succulent karoo (a sparse, suc-
culent-rich, fire-free biome) also harbours a remark-
able number of geophytes (Cowling
et al
., 1998, 1999b;
Esler
et al
., 1999). Therefore, a review of Cape
geophytes should ideally include both the fynbos and
succulent karoo biomes. This area has already
been defined as one biogeographical unit, namely the
extended ‘Capensis’ region (Bayer, 1984; Jürgens,
1991) though its definition as a phytochorion has yet
to be upheld by a comprehensive analysis. Geophytes
could be an excellent group for studying ‘Capensis’ dis-
tributions, since they include fynbos-centred genera,
succulent karoo-centred genera, and genera equally
prevalent in both biomes.
For the present study, the Cape region of geophyte
endemism is therefore defined as that covered by the
fynbos and succulent karoo biomes, and broadly equiv-
alent to the winter rainfall region of southern Africa
(the area where at least 50% of the annual precipita-
tion falls in winter). This region stretches from Lüder-
itz on the Namibian coast to about 150 km east of Port
Elizabeth in the Eastern Cape, and 100–300 km
inland (Fig. 2). It has long been noted that the Drak-
ensberg grasslands are also rich in geophytes (Bews,
1925). However, the inclusion of summer-rainfall
areas would raise a problem regarding the northern
limits of the study area, with the east African high-
lands also being fairly species-rich. Therefore, all spe-
cies counts and estimates to follow refer to the area
illustrated in Figure 2.
DEFINING GEOPHYTISM
Geophytes were defined by Raunkiaer (1934) as plants
with an underground perennation organ (bulb, corm,
tuber, or rhizome) and leaves that die back annually.
Several changes have since been made to that defini-
tion. However, in most parts of the world, no evergreen
plants are considered to be geophytes (Parsons, 2000).
In southern Africa, the transition from evergreen to
seasonally green can be seen in numerous genera, and
it is sometimes difficult to draw a firm distinction
between the two, as even plants belonging to the same
species may behave differently depending on local cli-
matic conditions. Given this continuous range of phe-
nologies, the most convenient approach is to consider
as geophytes all plants with an underground perenna-
tion organ, as long as it also fulfils a storage function.
This definition includes plants with pachymorphic
(thick), but not leptomorphic (thin) rhizomes.
Nevertheless, one still has to explore the limits of
geophytism, in order to cover all types of transitions
Figure 2.
The study area, largely equivalent to the fynbos and succulent karoo biomes (the distribution at quarter-degree
scale, based on Low & Rebelo, 1998).

30
S
. PROCHE
S
ET AL
.
© 2006 The Linnean Society of London,
Biological Journal of the Linnean Society,
2006,
87
, 27– 43
between geophytic and non-geophytic habits (see
below). First of all, it must be pointed out that some
Cape plants can be classified as belonging to several
morphological types, e.g. geophytes and climbers [e.g.
Kedrostis
(Cucurbitaceae)], or geophytes and succu-
lents. Another marginal category is represented by
aquatic plants with tubers or pachymorphic rhizomes,
termed helophytes by Raunkiaer (1934). For example
Zantedeschia
(Araceae) and
Kniphofia
(Asphodelac-
eae) typically grow in temporarily inundated lands,
and some species of
Aponogeton
(
Aponogeton
aceae) in
temporary pools. In these cases, tubers and rhizomes
fulfil a typical perennation/storage function during the
dry season. In a similar way,
Isoetes
(Isoetaceae) and
Triglochin
(Juncaginaceae) could be considered geo-
phytes.
It is difficult to classify as geophytes plants that
have root enlargements deep below the ground surface
[e.g.
Asparagus
(Asparagaceae)]. In some cases these
are not constant features in a given species, and for
other species in groups with root thickenings the
underground organs are altogether unknown [e.g. in
some
Annesorhiza
species (Apiaceae)].
The above transitional cases are not included on
our geophyte list (Table 1; largely consistent with
Germishuizen & Meyer, 2003), although they could be
considered to be geophytes depending on the exact cri-
teria used. Also excluded are several other genera of
Aizoaceae, Apiaceae, Apocynaceae, Asteraceae and
Campanulaceae, the 14 Cape species of
Drosera
(Dros-
eraceae), and a few species of Crassulaceae, Cyper-
aceae, Lamiaceae, Polygonaceae and Ranunculaceae.
It is worth noting that, even among the species
included in the list, the size of storage organs varies
tremendously. In orchids (e.g.
Disa
,
Ceratandra
), root
enlargements vary from clearly differentiated tubers
to barely noticeable thickenings. In
Bulbine
(Aspho-
delaceae), the amount of water stored above ground
(succulence) is often more important than that stored
below ground (geophytism). Rather uncritically, these
genera were included here in their entirety. However,
in the dicotyledonous genera listed, the distinction
between succulence and geophytism was made as far
as possible, and the counts for Cape species include
strictly geophytic species only.
METHODS
A geophyte list was compiled mainly from Goldblatt &
Manning (2000a), Leistner (2000), Germishuizen &
Meyer (2003) and Manning
et al
. (2002), while also
consulting numerous generic revisions (see References
in Leistner, 2000). Plant families were circumscribed
according to APG II (2003). Seven large genera, for
which recent and comprehensive revisions were avail-
able, were used for a few simple morphological and
phenological analyses. These were
Geissorhiza
(Gold-
blatt, 1985),
Moraea
(subg.
Moraea
; Goldblatt, 1986),
Watsonia
(Goldblatt, 1989), and
Gladiolus
(Goldblatt
& Manning, 1998) (all Iridaceae),
Eriospermum
(Rus-
caceae; Perry, 1994),
Haemanthus
(Snijman, 1984)
and
Cyrtanthus
(Reid & Dyer, 1984) (both Amarylli-
daceae). For these genera, the phenology and average
bulb size of each species were recorded. Additionally,
the presence/absence of each species was recorded for
five localities (one-degree squares, so selected as to
cover the rainfall gradients in the Cape as much as
possible; see Proche
s
, Cowling & du Preez, 2005), in
order to compare the phenologies of their floras.
The phylogenetic discussion is based on recently
published analyses, which were also used in conjunc-
tion with current distributions to infer historical dis-
tributions (Ronquist, 1997). In order to assess the
relative contribution of vegetative and reproductive
divergence to geophyte diversity in the Cape, charac-
ters referring to leaves/storage organs (as opposed to
flowers/bracts), that were listed in genus-level dichot-
omous keys by Manning
et al
. (2002) were counted.
For orchid genera, the natural keys in Linder & Kurz-
weil (1999) were used, excluding the summer-rainfall
species. Local geophyte diversity values were derived
from previously unpublished data of R. M. Cowling
and R. H. Whittaker. Most other available data sets
(based on once-off vegetation surveys) were not
usable, as geophytes can only be observed for a limited
period of time each year, when they are either flower-
ing or leafing, and in many cases it is only possible to
identify flowering specimens. Given temporal niche
separation in pollinator use (Goldblatt & Manning,
2000b) and phylogenetic constraints (Johnson, 1992),
geophytes co-occurring at one site will have staggered
flowering periods, thereby making it difficult to collect
comprehensive data during one visit.
THE NUMBERS
Some 2098 species of geophytes were recorded in the
study area, 84% of these being endemic (Table 1). The
species belonged to 11 orders, 20 families and 107 gen-
era. The great majority (83% of the species) were
monocots, and most of these (79%) were concentrated
in the largest monocot order, Asparagales. The larg-
est six families were the Iridaceae (767 species),
Hyacinthaceae (285), Orchidaceae (230), Oxalidaceae
(181), Amaryllidaceae (147), and Asphodelaceae (105);
all other families included less than 100 geophytic spe-
cies. The largest genus was
Oxalis
(181 species), fol-
lowed by
Moraea
(147),
Lachenalia
(121),
Gladiolus
(117), and
Ornithogalum
(102).
Shifting the borders of the study area north into
Namibia or arid central South Africa, even by 500 km,
does not result in a major increase in species numbers;

AN OVERVIEW OF THE CAPE GEOPHYTES
31
© 2006 The Linnean Society of London,
Biological Journal of the Linnean Society,
2006,
87
, 27– 43
Table 1.
Main groups of Cape geophytes. The number of species given in parentheses after each genus represents the
total (geophytic and non-geophytic) number of species worldwide; followed by the number of geophytic species occurring
in the Cape (endemics in parentheses). Due to the frequent changes in taxonomic status, all numbers must be regarded
as tentative
Asparagales
Orchidaceae
Bartholina
(2) 2 (0)
Bonatea
(20) 2 (0)
Brachycorythis
(33) 1 (1)
Brownleea
(7) 3 (0)
Ceratandra
(6) 6 (6)
Corycium
(14) 11 (9)
Disa
(162) 95 (83)
Disperis
(84) 15 (8)
Eulophia
(250) 16 (1)
Evotella
(1) 1 (1)
Gastrodia
(17) 1 (1)
Habenaria
(800) 7 (0)
Holothrix (55) 16 (7)
Liparis (250) 1 (0)
Pachites (2) 2 (2)
Pterygodium (18) 13 (13)
Satyrium (88) 31 (23)
Schizodium (6) 6 (4)
Stenoglottis (4) 1 (0)
Hypoxidaceae
Empodium (9) 5 (4)
Hypoxis (80) 7 (1)
Pauridia (2) 2 (2)
Spiloxene (25) 21 (20)
Tecophilaeaceae
Cyanella (8) 8 (6)
Walleria (3) 1 (1)
Iridaceae
Babiana (80) 78 (76)
Chasmanthe (3) 3 (2)
Crocosmia (8) 1 (1)
Devia (1) 1 (1)
Dierama (44) 1 (0)
Ferraria (11) 10 (9)
Freesia (14) 11 (10)
Geissorhiza (84) 84 (84)
Gladiolus (260) 117 (101)
Hesperantha (79) 41 (36)
Ixia (50) 50 (50)
Lapeirousia (40) 34 (26)
Melasphaerula (1) 1 (1)
Micranthus (3) 3 (3)
Moraea (195) 147 (135)
Pillansia (1) 1 (1)
Romulea (95) 70 (61)
Sparaxis (15) 15 (15)
Syringodea (8) 5 (4)
Thereianthus (8) 8 (8)
Tritonia (28) 26 (23)
Tritoniopsis (24) 24 (23)
Watsonia (52) 34 (31)
Xenoscapa (2) 2 (2)
Asphodelaceae
Bulbine (75) 51 (42)
Bulbinella (23) 17 (17)
Trachyandra (52) 37 (32)
Alliaceae
Allium (550) 1 (0)
Tulbaghia (20) 8 (5)
Agapanthaceae
Agapanthus (3) 3 (2)
Amaryllidaceae
Amaryllis (2) 2 (2)
Ammocharis (5) 1 (0)
Apodolirion (6) 3 (3)
Boophone (2) 2 (1)
Brunsvigia (20) 12 (10)
Clivia (5) 1 (1)
Crinum (65) 2 (1)
Crossyne (2) 2 (2)
Cybistetes (1) 1 (1)
Cyrtanthus (57) 27 (20)
Gethyllis (32) 32 (31)
Haemanthus (22) 17 (11)
Hessea (14) 14 (10)
Nerine (23) 4 (4)
Scadoxus (9) 2 (0)
Strumaria (24) 25 (23)
Ruscaceae
Eriospermum (102) 76 (49)
Hyacinthaceae
Bowiea (1) 1 (0)
Daubenya (8) 8 (7)
Drimia (131) 33 (27)
Eucomis (10) 3 (1)
Lachenalia (121) 121 (116)
Ledebouria (46) 5 (2)
Massonia (7) 7 (6)
Merwilla (5) 1 (1)
Ornithogalum (234) 102 (80)
Pseudoprospero (1) 1 (1)
Spetaea (1) 1 (1)
Veltheimia (2) 2 (1)
Dioscoreales
Dioscoreaceae
Dioscorea (400) 6 (2)
Liliales
Colchicaceae
Androcymbium (40) 26 (25)
Baeometra (1) 1 (1)
Neodregea (1) 1 (1)
Onixotis (2) 2 (2)
Ornithoglossum (8) 6 (5)
Wurmbea (37) 15 (14)
Commelinales
Haemodoraceae
Dilatris (4) 4 (4)
Wachendorfia (4) 4 (4)
Caryophyllales
Aizoaceae
Phyllobolus (31) 11 (10)
Saxifragales
Crassulaceae
Tylecodon (41) 5 (4)
Geraniales
Geraniaceae
Pelargonium (300) 98 (87)
Malpighiales
Euphorbiaceae
Euphorbia (1500) 1 (1)
Oxalidales
Oxalidaceae
Oxalis (500) 181 (166)
Gentianales
Apocynaceae
Brachystelma (110) 9 (3)
Ceropegia (160) 14 (3)
Eustegia (1) 1 (1)
Fockea (6) 5 (4)
Orbea (20) 4 (0)
Pachycarpus (30) 2 (0)
Woodia (3) 1 (0)
Xysmalobium (40) 3 (0)
Asterales
Asteraceae
Othonna (120) 30 (27)
Total 2096 (1756)

32 S. PROCHES ET AL.
© 2006 The Linnean Society of London, Biological Journal of the Linnean Society, 2006, 87, 27 –43
however, advancing the same distance into the grass-
lands of the Drakensberg Plateau and along the east
coast could add a few hundred species, mainly orchids
and irids from a variety of genera already well repre-
sented in the Cape. On the other hand the Hya-
cinthaceae flora of these areas is somewhat more
distinct. By including all the borderline geophytes
mentioned in the previous section, the numbers could
increase by up to 100 species, although most of these
are not endemics. Future changes to the data pre-
sented in Table 1 will be due largely to taxonomic
reassessment of the species status in species-rich gen-
era, and to the discovery of new species.
It is clear that the Cape has the most diverse geo-
phyte flora worldwide (Fig. 1). But at what spatial
scale do species numbers become unusually high? It is
unfortunately difficult to answer this question, given
the scarcity of relevé data that include reliable lists of
geophytes. Unpublished data of R. M. Cowling & R. H.
Whittaker from fynbos and renosterveld sites across
the Cape, which are at best under-estimates of actual
numbers, show that most often (with 95% confidence)
between one and four species of geophytes occur in a 1-
m2 plot, between two and ten in a 10-m2 plot, between
five and 13 in a 100-m2 plot, and between seven and 16
in a 1000-m2 plot (Table 2). In our comparisons, renos-
terveld plots had significantly (or near-significant)
higher values compared to fynbos plots at all four
scales considered, although one fynbos site (a Protea
repens-dominated site in Elim fynbos underlain by fer-
ricrete on the Agulhas Plain at Rietfontein) showed
values typical of renosterveld, rather than fynbos.
Additionally, the percentage of geophytes was higher
in renosterveld than in fynbos. Western sites were
Table 2. Number of geophyte species for 1, 10, 100 and 1000 m2 plots in fynbos and renosterveld vegetation in the Cape.
The numbers for the 100 and 1000 m2 scales are counts for single plots, at the 1 m2 scale, N = 10, while at the 10 m2,
N = 2 (R. M Cowling & R. H. Whittaker, unpubl. data). Differences represent P-values for unpaired t-tests (in the case
of percentages, arcsine-transformed data were used), with asterisks marking Bonferroni-corrected significance
1 m210 m2100 m21000 m2Plants total
Percent
geophytes
WESTERN FYNBOS
Rietfontein 5.1 ± 0.5 13.5 ± 0.5 23 27 112 24.1
Brandewynkop 2.4 ± 0.4 5.0 ± 1.0 7 9 74 12.2
Jakkalsrivier 2.3 ± 0.3 4.0 ± 0.0 6 6 56 10.7
Groot Hagelkraal 1.0 ± 0.2 1.5 ± 0.5 2 5 57 8.8
Soetanysberg 0.2 ± 0.1 2.0 ± 1.0 4 4 47 8.5
EASTERN FYNBOS
van Stadens Mountains 1.2 ± 0.2 2.0 ± 0.0 2 6 42 14.3
Hankey 1 1.3 ± 0.2 3.0 ± 1.0 7 10 83 12.0
Hankey 2 1.3 ± 0.2 2.5 ± 1.5 4 6 84 7.1
Humansdorp 1 0.5 ± 0.3 1.5 ± 1.5 3 4 45 8.9
Humansdorp 2 0.0 ± 0.0 0.0 ± 0.0 1 5 60 8.3
Humansdorp 3 1.4 ± 0.2 3.5 ± 0.5 4 6 88 6.8
WESTERN RENOSTERVELD
Tygerberg, Cape Town 6.4 ± 0.6 17.0 ± 0.0 27 38 105 36.2
Signal Hill, Cape Town 3.1 ± 0.3 7.5 ± 1.5 11 18 99 18.2
Bontebok NP 5.0 ± 0.5 10.5 ± 0.5 16 18 51 35.3
EASTERN RENOSTERVELD
Humansdorp 4 4.8 ± 0.4 12.0 ± 0.0 15 17 74 23.0
Humansdorp 5 3.9 ± 0.4 7.0 ± 0.0 9 11 59 18.6
Humansdorp 6 2.0 ± 0.4 6.0 ± 0.0 9 12 87 13.8
Humansdorp 7 2.1 ± 0.1 5.5 ± 1.5 11 11 85 12.9
Humansdorp 8 3.1 ± 0.3 6.0 ± 0.0 8 10 95 10.5
Grahamstown 2.5 ± 0.3 8.0 ± 1.0 13 15 95 15.8
DIFFERENCES
fynbos/renosterveld <0.0001* <0.0001* 0.0120* 0.0198 – 0.0101*
western/eastern fynbos <0.0001* 0.0387 0.1964 0.3366 – 0.2940
western/eastern renost. 0.0001* 0.3448 0.5354 0.3986 – 0.2639
west/east <0.0001* 0.0329 0.1834 0.1645 – 0.1168

AN OVERVIEW OF THE CAPE GEOPHYTES 33
© 2006 The Linnean Society of London, Biological Journal of the Linnean Society, 2006, 87, 27 –43
generally richer than eastern sites, although at some
scales (10–1000 m2) the differences were not signifi-
cant (and less so in renosterveld than in fynbos)
(Table 2). Similar local-scale numbers for western
renosterveld have been observed independently (J. S.
Donaldson, pers. comm.), while also showing a geology
effect (dolerites being significantly richer than tillite
sites) and geophyte cover values averaging 8–19%.
Amazing density values in certain small-scale habi-
tats (up to 20 000 corms/m2 for Syringodea longituba)
are reported from the same sites (J. S. Donaldson,
pers. comm.). Comparative local diversity and abun-
dance data for succulent karoo are needed.
The west is also richer in geophytes than the east
at larger scales (16th degree, quarter degree, one
degree), as shown in several surveyed genera (e.g. Pel-
argonium: van der Walt & Vorster 1983; Oxalis: Ober-
lander, Dreyer & Esler, 2002). A study documenting
geophyte diversity in the winter rainfall region of
southern Africa at the quarter degree scale (Proches et
al., 2005), indicates that this pattern is almost univer-
sal among Cape geophytes (see Fig. 3). As shown at
this scale, the south-western Cape (including the
north-western Cape Floristic Region, as delimited in
Goldblatt & Manning, 2000a) is definitely the world’s
geophyte capital, with more than 500 species poten-
tially coexisting in one quarter-degree square (Proches
et al., 2005). In several localities in this area (e.g.
Nieuwoodtville, Saldanha), geophytes are the domi-
nant growth form, representing c. 40% of the total
number of plant species (Snijman & Perry, 1987; Hop-
per & Manning, 2004).
THE BROAD PHYLOGENETIC PICTURE
Cape geophytes belong to a large variety of unrelated
groups, most of which have non-geophytic sister taxa.
This suggests that, at least in large (family-level) taxa,
geophytism must be regarded as a derived state. While
looking for further taxonomic and phylogenetic pat-
terns, three questions must be kept in mind. (1) How
many times did the geophytic habit originate indepen-
dently? (2) Were there cases where this was second-
arily lost? (3) How many of these transitions are likely
to have occurred in the Cape, or elsewhere in southern
Africa?
The first distinction that needs to be made is
between predominantly geophytic groups (most of
the monocot families listed, and the Oxalidaceae);
groups that have a significant geophytic component
(Orchidaceae, Pelargonium); and predominantly non-
geophytic groups with a few geophytic representa-
tives (most of the dicots listed in Table 1 and some
groups mentioned in the discussion over defining
geophytism).
Significantly, many of the geophytes in the third cat-
egory (genera of Apocynaceae, Aizoaceae, Asteraceae,
etc.) are endemic to, or centred in, southern Africa,
suggesting that their transition to geophytism hap-
pened in this region. This is also the case with Pelar-
gonium, where all the geophytic taxa are endemic to
southern Africa (Marais, 1994; Bakker et al., 2004). In
Oxalis, although geophytism is not restricted to the
region, it is much more prevalent here than in South
America, the other centre of diversity for this genus.
This suggests that the Cape may have been globally
important in the evolution of geophytism. In part-
icular, the apparently multiple transitions in genera
such as Bulbine (Asphodelaceae), Tylecodon (Cras-
sulaceae), Pelargonium (Geraniaceae) or Othonna
(Asteraceae), between no stored water, above-ground
stored water (succulence) and below-ground stored
water (geophytism) present a baffling show of morpho-
logical plasticity. The Cape is also a centre of diversity
and diversification for succulents (Van Wyk & Smith,
2001; Klak, Reeves & Hedderson, 2004); the bio-
geographical and ecological connections between
geophytism and succulence represent yet another
interesting field requiring research.
Figure 3. Geographic patterns of diversity in the Cape (at
the quarter-degree scale), in two major geophyte families:
A, Iridaceae; B, Orchidaceae (data from Proches et al.,
2005).

34 S. PROCHES ET AL.
© 2006 The Linnean Society of London, Biological Journal of the Linnean Society, 2006, 87, 27 –43
To ascertain the significance of southern Africa in
the global context of geophytism, one needs to consider
the two orders in which geophytism is best repre-
sented, namely Asparagales and Liliales. In the Aspar-
agales, out of the 24 recognized families (APG II,
2003), 14 occur in southern Africa. Two of these, Lanar-
iaceae and Agapanthaceae, are endemic to the region,
and another seven have their worldwide centre of
diversity here: Amaryllidaceae, Asparagaceae, Aspho-
delaceae, Hyacinthaceae, Hypoxidaceae, Iridaceae,
and Ruscaceae (with the inclusion of Eriospermum)
(see APG II, 2003). The Behnioideae in Agavaceae may
also deserve family-level recognition (Conran, 1999).
While peak species diversity in southern Africa does
not necessarily imply a southern African origin, this
appears to be the case in at least five families, repre-
senting two multifamily clades (Fig. 4). The most sug-
gestive southern African origin is indicated by the
Agapanthaceae/Amaryllidaceae/Alliaceae clade. The
monogeneric Agapanthaceae, endemic to southern
Africa, are rhizomatous, and rhizomes are also found
in the only southern African subfamily of Alliaceae,
Tulbaghioideae (Fay & Chase, 1996). If rhizomes are
plesiomorphic in the Amaryllidaceae (as implied by
Nordal & Duncan, 1984 and Müller-Doblies & Müller-
Doblies, 1996) then the bulbous habit may have
evolved at least three times in the family: in Amaryl-
lideae, in Haemantheae (both centred in the Cape),
and within the ancestral stock for the rest of the fam-
ily (although alternative scenarios are plausible; see
Meerow et al., 1999 for a familial phylogeny and dis-
cussion of character state evolution).
A second dichotomy, the one between Asparagaceae
and Ruscaceae (Chase et al., 2000a; Fay et al., 2000;
but not supported by Soltis et al., 2000), again points
to southern Africa, the largely African genera
Dracaena (including Sansevieria) and Eriospermum
having a basal position in Ruscaceae, while the Aspar-
agaceae are also most diverse in southern Africa. The
origins of two other important geophytic families,
Asphodelaceae and Hyacinthaceae, are less obvious.
Evidence suggests a South American origin for Hya-
cinthaceae, followed by massive radiation in southern
Africa (Manning et al., 2004), while the origins of
Asphodelaceae were most likely Eurasian (Chase
et al., 2000b).
In Liliales, too, the most plesiomorphic groups
(Alstroemeriaceae, Luzuriagaceae, Campynemata-
ceae) have limited distributions in mainly non-
mediterranean climates in the southern hemisphere,
while the major radiations happened in mediterra-
nean climates (the core Colchicaceae in the Cape,
Tulipa and other Liliaceae in the Mediterranean
Basin) (Vinnersten & Bremer, 2001).
Given the basal position of rhizomatous (geophytic
or non-geophytic) herbs in most families of the Aspar-
Figure 4. Phylogenetic relationships in the Asparagales (based on Chase et al., 2000a; Fay et al., 2000; Soltis et al., 2000;
APG II, 2003; Rudall, 2003; only recognizing clades with > 50% support), indicating geophytism (Stevens, 2001) and
southern African taxa (see Leistner, 2000). Numbers of genera/species indicated for each family, based on Stevens (2001).
Hyacinthaceae
Themidaceae
Aphyllanthaceae
Agavaceae
Asparagaceae
Ruscaceae
Laxmanniaceae
Amaryllidaceae
Agapanthaceae
Alliaceae
Asphodelaceae
Xanthorrhoeaceae
Hemerocallidaceae
Xeronemataceae
Iridaceae
Doryanthaceae
Tecophilaeaceae
Ixioliriaceae
Hypoxidaceae
Asteliaceae
Lanariaceae
Blandfordiaceae
Boryaceae
Orchidaceae
Behnioideae
Eriospermoideae

AN OVERVIEW OF THE CAPE GEOPHYTES 35
© 2006 The Linnean Society of London, Biological Journal of the Linnean Society, 2006, 87, 27 –43
agales and Liliales (see Fay et al., 2000; Stevens, 2001;
Vinnersten & Bremer, 2001), it can be assumed that
more efficient storage organs (tubers, corms, bulbs; see
Lewis, 1954) evolved independently in almost each
family, and more than once in some families (e.g. Iri-
daceae; see Reeves et al., 2001). Family-level differen-
tiation certainly occurred long before the onset of
mediterranean climates in the mid Pliocene, and most
of the isolated plesiomorphic lineages persist to the
present day in non-mediterranean regions (in Aus-
tralasia, South America and eastern South Africa).
Later, within most families, more advanced storage
organs evolved, and these transitions are likely to
have been a key factor in diversification.
Such transitions almost certainly occurred in south-
ern Africa in the cases of Iridaceae, Ruscaceae and
Amaryllidaceae, probably triggered by the climate
changes that started in the late Eocene, and became
dramatic during the Oligocene and Miocene (see Axel-
rod & Raven, 1978; Goldblatt, 1983; Zachos et al.,
2001). Unfortunately, relatively few phylogenies
are available for tracking this process. Speciation in
Androcymbium (Colchicaceae) appears to have started
in the late Eocene (when the genus was presumably
already geophytic) and intensified during the Oli-
gocene (Caujapé-Castells et al., 2002), while the sepa-
ration of Moraea and Ferraria happened during the
Oligocene, followed by fairly constant speciation in
Moraea during the Oligocene and Miocene (Goldblatt
et al., 2002; V. Savolainen & P. Goldblatt, unpubl.
data). However, in the most species-rich Cape geo-
phytic group, the Crocoideae (Iridaceae), the genera
separated more recently, during the Miocene and
Pliocene, and this radiation was followed by rapid spe-
ciation, probably triggered by the aridification and
increased rainfall seasonality in the mid Pliocene
(Reeves et al., 2001; Goldblatt et al., 2002, 2005).
Rapid speciation during the Pliocene may also have
occurred in many smaller groups, though more phy-
logenies are needed to confirm this hypothesis. In
addition to climate change, another factor that may
have contributed to this speciation process was the
rejuvenation of the Cape landscapes following a mas-
sive uplift in the east during the Pliocene (Deacon,
Jury & Ellis, 1992), which must have increased
habitat diversity. Under the new climatic condi-
tions geophytes were clearly favoured, as compared to
their non-geophytic relatives, and speciated more
extensively.
In several instances non-geophytic lineages are sig-
nificantly poorer in species than their geophytic sister
taxa. Thus there are ten (or less, depending on the spe-
cies concept used) species of Agapanthaceae (all rhi-
zomatous), compared with 230 species in the almost
entirely bulbous sister taxon, the Amaryllidaceae
(numbers for southern Africa only). This situation is
repeated twice in Iridaceae: subfamily Nivenioideae
(shrubby) has less than one hundred species, com-
pared to almost a thousand species in the sister
subfamily Crocoideae (cormous); in the Irideae, the
rhizomatous Dietes−Bobartia grade has less than 20
species in all, while the embedded clade Ferraria−
Moraea, made up entirely of cormous (and almost
entirely African) species, totals over 200 species (Gold-
blatt, 1990; see Reeves et al., 2001 for phylogeny,
Lewis, 1954 for morphology). Almost every other geo-
phytic family of Asparagales has a few rhizomatous
genera with significantly lower diversity values than
the geophytic sister clades (with corms, bulbs or
tubers) (Fig. 4).
The Orchidaceae, with a basal position in Aspar-
agales, do not appear to have had any association with
southern Africa during their early history. Several
orchid clades colonized the Cape area, but only one
diversified extensively here. This is the largely African
tribe Diseae, containing the spectacular and species-
rich genera Disa and Satyrium. Unlike in higher
Asparagales families, the geophytic condition appears
to be primitive in orchids, and a great proportion of the
orchid diversification happened secondarily in non-
geophytic clades (mainly epiphytes). Among the lin-
eages that diversified in the Cape, most are primarily
geophytic; however, in some cases this habit was sec-
ondarily lost (e.g. there are evergreen species of Disa,
Satyrium, and Coryciinae in which the storage organs
are reduced or absent) (Linder & Kurzweil, 1999).
Both in the core Asparagales and in Orchidaceae (and
also in Hypoxidaceae), important radiations have also
taken place in the summer rainfall region of southern
Africa, in particular in the moister parts of the grass-
land biome.
Since geophytism has evolved numerous times, in
different parts of the world, and flourished mainly in
semiarid regions, it is interesting to examine whether
there was any exchange of taxa between different
mediterranean-climate areas. This is obvious in the
northern hemisphere genus Allium, that has broad
radiations in both the Mediterranean Basin and
California, and which occurs in non-mediterranean
climates in between the two regions, but southern
hemisphere distributions are more ambiguous. Sev-
eral highly disjunct generic distributions are known
(e.g. Wurmbea and Bulbine: Africa and Australia; Spi-
loxene: Africa, Australia and New Zealand; Bulbinella:
Africa and New Zealand; Caesia: Africa, Australasia
and Madagascar; Trachyandra: Africa and Madagas-
car; Dietes: southern Africa and Lord Howe Island).
With the exception of Bulbinella and Spiloxene,
none of these are true ‘Cape’ (or mediterranean-
climate) genera and their species are distributed in
both mediterranean and non-mediterranean climates.
Comparable disjunct distributions are lacking within

36 S. PROCHES ET AL.
© 2006 The Linnean Society of London, Biological Journal of the Linnean Society, 2006, 87, 27 –43
the ‘core’ Cape geophytic clades (the Crocoideae and
core Iridoideae of Iridaceae, Amaryllidaceae and Hya-
cinthaceae). The age of these clades is generally too
young to represent Gondwanan relicts (Donato, Leach
& Conran, 2000; Vinnersten & Bremer, 2001). At the
same time, advanced geophytism is too old to have
originally evolved in mediterranean climates (which
are a Late Pliocene-Quaternary feature), and in some
cases it may have coevolved with the pre-mediterra-
nean sclerophyllous vegetation that was widespread
in mediterranean-climate areas during the Tertiary
(Axelrod, 1975; Herrera, 1992; Verdú et al., 2003).
Despite this remarkable antiquity, several geophyte
lineages are relatively range-restricted, suggesting
limited dispersal abilities or habitat specialization
(see Conran, 1995). At least in Dietes and Wurmbea,
however, the disjunct distributions appear to be the
result of single long-distance dispersal events, sug-
gesting that, occasionally, such events do happen.
Among the ‘Cape’ genera it is therefore very likely that
the lack of suitable habitats across the southern
Hemisphere was important in preventing the spread
of taxa among the isolated mediterranean-climate
regions.
In contrast, continuous distributions from the Cape
to the Mediterranean Basin (Androcymbium, Chloro-
phytum, Drimia, Gladiolus, Hesperantha, Moraea sect.
Gynandriris, Ornithogalum and Romulea) suggest
that geophyte exchanges were more frequent between
these two regions. More often, it is the Cape taxa that
have invaded northwards, as suggested by biogeo-
graphical patterns and recent phylogenies for both geo-
phytes and other plants (Quézel, 1978; Cowling,
Proches & Vlok, 2005b). For example, in Androcym-
bium, most of the species (the most plesiomorphic ones
included) occur in southern Africa (Caujapé-Castells
et al., 2002), while some derived species, as well as the
embedded clade Colchicum−Merendera−Bulbocodium
(Vinnersten & Reeves, 2003), are found in the Medi-
terranean Basin. A similar picture is starting to take
shape in the Hyacinthaceae, with the Mediterranean
species of Drimia and Ornithogalum embedded within
southern African or sub-Saharan groups of species
(Pfosser et al., 2003; Manning et al., 2004), and in Orni-
thogalum it appears that more than one invasion into
the Mediterranean must have occurred. Overall, how-
ever, it is likely that the Cape has contributed signif-
icantly to the geophyte flora of the Mediterranean
Basin, and possibly that of other regions.
MORPHOLOGICAL DIVERSITY AND SPECIATION
Cape geophytes have gained their reputation among
horticulturists because of their spectacular flower dis-
plays, and this amazing floral diversity has been
linked to pollinator specificity. Indeed, a variety of pol-
lination systems have been described for each of sev-
eral genera (e.g. Lapeirousia: Goldblatt, Manning &
Bernhardt, 1995; Satyrium: Johnson, 1997; Gladiolus:
Goldblatt, Manning & Bernhardt, 1998; Sparaxis:
Goldblatt, Manning & Bernhardt, 2000). These stud-
ies show that long-tongued flies, bees, butterflies,
moths, and birds have each driven the evolution of
their own guild of plants containing, in each case, sev-
eral unrelated taxa. It can hardly be contested that
pollinator specificity accounts for a great proportion of
the character diversity in Cape geophytes. However, it
has not been conclusively demonstrated that pollina-
tor specificity alone can trigger speciation. In most
instances other factors are also involved, such as
edaphic (e.g. in Iridaceae: Goldblatt, 1983; Goldblatt &
Manning, 1996a, b), and climatic specialization (e.g.
Moraea sect. Galaxia: Goldblatt, 1979, and Disa sect.
Herschelia: Linder, 1995). Furthermore, the morpho-
logical diversity displayed by the vegetative parts
(both above-ground and below-ground) has largely
been neglected.
Our analysis (Fig. 5) showed that while reproduc-
tive features do indeed account for the most obvious
species-specific characteristics in the majority of Iri-
daceae (Moraea, Gladiolus, Lapeirousia), and even
more so in Orchidaceae (Disa, Satyrium), the repro-
Figure 5. Diversity of reproductive and vegetative charac-
ters in some speciose geophytic genera in the Cape. Based
on the characters cited in naural dichotomous keys from
Linder & Kurzweil (1999) and Manning et al. (2002). Since
generic-level keys were used, the ratio between vegetative
and reproductive characters in each genus is likely to rep-
resent a good approximation of the reality; however, inter-
generic comparisons may not be strictly valid. Phylogenetic
relationships based on Meerow et al. (1999), Chase et al.
(2000a), Reeves et al. (2001), and Goldblatt et al. (2005).
Vegetative
Reproductive
150
100
50
0
Characters
Androcymbium
Disa
Satyrium
Gethyllis
Cyrtanthus
Strumaria
Eriospermum
Lachenalia
Drimia
Ornithogalum
Moraea
Tritoniopsis
Gladiolus
Lapeirousia
Watsonia
Romulea
Geissorhiza
Babiana
Tritonia
Ixia

AN OVERVIEW OF THE CAPE GEOPHYTES 37
© 2006 The Linnean Society of London, Biological Journal of the Linnean Society, 2006, 87, 27 –43
ductive component accounts for less than 50% of the
character diversity in Eriospermum (Ruscaceae), and
in some Hyacinthaceae (Ornithogalum) and Amarylli-
daceae (Gethyllis). In other genera such as Romulea
(Iridaceae), Strumaria (Amaryllidaceae) and Lachena-
lia (Hyacinthaceae), both vegetative and reproductive
biology have contributed significantly to character
diversity. While the proportions of characters recorded
in this analysis cannot be automatically equated to
the factors driving speciation, these results neverthe-
less suggest that reducing geophyte diversification in
the Cape to its pollination component is simplistic. It
must also be noted that not all the differences in the
morphology of reproductive parts have to do with
pollination (e.g. the shape of the inflorescence in
many Amaryllideae enhances wind dispersal in the
infructescence; Snijman & Linder, 1996), and not all
those affecting pollination directly relate to pollinator
specificity.
Most of the vegetative characters listed in geophyte
keys referred to leaf shape, size and texture, indicating
adaptations to local climatic conditions. Several char-
acteristic leaf shapes recurring in unrelated groups of
Cape geophytes represent a spectacular indication of
geophyte morphological plasticity. The best known
such case is the prostrate leaf syndrome, characterized
by few (often two) flattened leaves, pressed to the
ground, as observed in Amaryllidaceae (Brunsvigia,
Crossyne, Haemanthus, Strumaria), Colchicaceae
(Androcymbium), Hyacinthaceae (Daubenya, Lache-
nalia, Massonia, Ornithogalum), Ruscaceae (Eriosper-
mum), Orchidaceae (Bartholina, Holothrix, Satyrium),
and Iridaceae (Freesia, Sparaxis, Xenoscapa) (Esler
et al., 1999; Manning et al., 2002). This strategy has
been interpreted as representing an advantage in
avoiding herbivory or competition from neighbours, in
creating a CO2-rich environment underneath the
leaves, or in regulating evapotranspiration (Esler
et al., 1999). Coiled, wiry leaves are equally spectacu-
lar, and even less investigated, occurring in Ama-
ryllidaceae (Cyrtanthus, Gethyllis) Colchicaceae
(Androcymbium), Hyacinthaceae (Ornithogalum, Dri-
mia), Iridaceae (Moraea, Geissorhiza) and Orchi-
daceae (Disa). Inspection of monographs and our own
observations indicates that this trait is almost exclu-
sively associated with sparse vegetation on rocky sites,
especially gravel plains. Coiling may well be an adap-
tation to reduce wind damage to the relatively large
photosynthetic organs required to maximize photosyn-
thetic activity in the cool growing season.
Special mention must be made here of leaf-shape
diversity in Eriospermum, whose current position is in
Ruscaceae (APG II, 2003) but which has been consid-
ered to represent a family on its own (Eriospermaceae;
see Perry, 1994). This remarkable genus contains over
100 species, most of which occur in succulent karoo,
but with numerous fynbos, thicket, grassland, and
savanna species. Although floral characters are used
in its subgeneric classification, these are far less spec-
tacular than the leaves. In this genus, the number of
leaves has been reduced − generally to one − and this
can take a variety of shapes, from linear or terete to
heart-shaped. However, the most interesting evolu-
tionary process in Eriospermum is one through which
the reduction in photosynthetic surface area brought
about by the reduced number of leaves is reversed by
the development of unusual outgrowths (‘enations’) on
the leaf surface. These take the form of solitary or
numerous, often tree-shaped excrescences, with vary-
ing degrees of hairiness, representing a whole new
class of photosynthetic organs (see Perry, 1994). This
feature − associated with an enlargement of the pho-
tosynthetic organ − may have evolved in response to
the development of winter-wet climates, characterized
by low-energy growth conditions for shallow-rooted
plants.
Herbivory has also been invoked as a selective force
driving plant speciation (Westoby, 1989). However,
leaf characteristics associated with antiherbivore spe-
cialization, such as hair cover, are uncommon in Cape
geophytes, suggesting that above-ground herbivory
was probably less important in species differentiation
(but see Amaryllidaceae: Brunsvigia, Crossyne, Geth-
yllis, Haemanthus, Strumaria, and Hypoxidaceae:
Hypoxis). More important appears to be below-ground
herbivory. Underground organs, unlike leaves and
flowers, are present throughout the year, offering
greater opportunities for herbivore specialization. At
least two species-rich groups of herbivores have diver-
sified in the Cape, mainly in association with
geophytes: the Brachycerus weevils (Coleoptera: Cur-
culionoidea), of which there are more than a hundred
species in the region, and the mole rats (Rodentia:
Bathyergidae), with at least five species. Up to three
species of mole rat can co-occur in the same habitat
(Lovegrove & Jarvis, 1986). Numerous geophytes are
restricted to rocky areas, where mole-rat activity is
limited (cf. Thomson et al., 1996), while other species
survive in the protection offered by Restionaceae tus-
socks. Special adaptations, including storage and
perennation organs situated at different depths in the
soil, have developed in geophytes that occur in mole-
rat populated habitats (Lovegrove & Jarvis, 1986),
and these may be a significant force in speciation, hav-
ing to do with both predation avoidance and dispersal
(e.g. in Micranthus). Both below-ground and above-
ground predation may also be avoided by the accumu-
lation of highly toxic substances (especially in
Amaryllidaceae, Colchicaceae and Hyacinthaceae).
The potential role that this chemical evolution may
have played in Cape geophyte diversification is not
clear and requires further study.

38 S. PROCHES ET AL.
© 2006 The Linnean Society of London, Biological Journal of the Linnean Society, 2006, 87, 27 –43
Very little is known of the possible effect of storage
organs size on habitat specialization. It has been
shown (Proches et al., 2005) that larger storage
organs are more likely to occur where rainfall is less
abundant or less predictable, although storage organ
size may also have to do with microhabitats, possibly
with the local drainage conditions. Within the seven
genera analysed here (Geissorhiza, Moraea, Watso-
nia, Gladiolus, Eriospermum, Haemanthus and Cyr-
tanthus), a remarkable variation was observed in the
size of the storage organs (Fig. 6). Some genera, such
as Haemanthus, are characterized by relatively large
bulbs; others, like Geissorhiza and Gladiolus, by rela-
tively small corms. In most other genera, however,
the size of the storage organs varies from species to
species, even by a few orders of magnitude. Figure 6
does not reflect the complete range of sizes, the corms
in minute geophytes like Romulea being smaller than
in any of the included taxa, while the (arguably geo-
phytic) tubers of Dioscorea and some Apocynaceae
can reach 1 m in diameter. The possibility that under-
ground organs may actually fulfil several basic
functions could add new dimensions to habitat
specialization. Thus, while the main function of
geophytism is currently understood as that of water
storage (Rees, 1989), it is clear that geophytism also
presents an advantage in mineral ion storage
(Ruiters & McKenzie, 1994). The different types and
sizes of underground organs could, interpreted in
this light, yield much more information relating to
habitat (and especially edaphic) specialization.
In addition, little is known of whether certain types
of storage organ are more appropriate than others in
specific habitats. According to Ruiters (2001), the four
basic types of storage organ showed similar diversity
patterns in the south-western Cape (41% of the spe-
cies had corms, 32% tubers, 17% bulbs, and 10% rhi-
zomes). Similar patterns were found by Hoffmann
et al. (1998) for the geophyte flora of Chile. This is sur-
prising, considering that most corms and tubers are
replaced annually, thereby imposing a size limit on
these types of organ, while bulbs, rhizomes and hypo-
cotyledonary tubers are not. However, no clear size dif-
ferences were observed between bulbous and cormous
organs in the genera examined here (Fig. 6), although
the largest storage organs were indeed found in the
last three genera, where continuous growth occurs.
While the corms of Colchicaceae, Hypoxidaceae or Iri-
daceae never approach the maximum dimensions
achieved by the largest bulbs in certain genera of
Amaryllidaceae (e.g. Boophone, Brunsvigia and Cri-
num), in several other genera of Amaryllidaceae (Hes-
sea, Nerine and Strumaria) the bulbs of many species
are no larger than most corms.
Another set of interspecific differences refer to phe-
nology. Our review of patterns in seven genera shows
Cape geophyte phenology to be very variable and char-
acterized by a considerable percentage of evergreen
species. Among the deciduous species, both synant-
hous and hysteranthous species are represented, the
former leafing and flowering at the same time, while
in the latter the flowers are produced in the dry season
prior to or after the leaves (Dafni, Cohen & Noy-Meir,
1981; Ruiters & McKenzie, 1994). Among the seven
genera analysed, 69% of the species were synanthous,
22% hysteranthous, and 9% evergreen. Including
orchid genera in this analysis would undoubtedly
increase the proportion of evergreen species. The per-
centage of evergreen species in the local geophyte flo-
ras increased from zero (in the dry areas of
Namaqualand) to 20% in moister, less seasonal areas,
such as Knysna, in the southern Cape. In typical
deciduous species, the coupling between flower emer-
gence and leaf emergence also varied geographically,
with hysteranthous species being more numerous in
Namaqualand (40%) and rare in the eastern part of
the Cape (14% Knysna, 12% Humansdorp) (Fig. 7).
The differences between synanthous and hysteran-
thous geophytes have been attributed to resource allo-
cation, as optimized for different environments (Dafni
et al., 1981; Gutterman & Boeken, 1988), and this
seems to correspond to the results presented here,
hysteranthous species occurring mainly in areas of
scarce and more strictly seasonal rainfall. A different
pattern characterizes Gladiolus, where hysteranthy is
more common in the south-west (see Goldblatt & Man-
ning, 1998). It has been shown that there is sometimes
Figure 6. Storage organ size for seven large Cape geo-
phyte genera. Median, central 50% of the values, and
ranges are given (data from Reid & Dyer, 1984; Snijman,
1984; Goldblatt, 1985, 1986, 1989; Perry, 1994; Goldblatt
& Manning, 1998). The storage organs are annually re-
placed in the first four genera, but not in the last three, in
which they have continuous growth.
1 m3
1 dm3
1 cm3
1 mm3
Storage organ size
Geissorhiza
Gladiolus
Watsonia
Eriospermum
Cyrtanthus
Haemanthus
Moraea

AN OVERVIEW OF THE CAPE GEOPHYTES 39
© 2006 The Linnean Society of London, Biological Journal of the Linnean Society, 2006, 87, 27 –43
also a phylogenetic component to phenology, with cer-
tain groups that originated in summer-rainfall areas
(e.g. some Amaryllidaceae) adhering to their summer-
flowering patterns even in the winter-rainfall region
(Johnson, 1992; Snijman, 1999).
FINAL LINES
What are the mechanisms making it possible for so
many geophyte species to coexist at different spatial
scales in the Cape? The importance of topographic het-
erogeneity in the maintenance of species diversity in
the Cape has been shown for plants in general (Cowl-
ing et al., 1997) and specifically for geophytes (Ruiters,
2001), at scales close to quarter-degree. At this scale,
which encompasses entire landscapes and much het-
erogeneity, it is likely that non-reproductive adapta-
tions are the main explanation for geophyte species
diversity. The high species replacement rates within
landscapes that characterize much of the fynbos biome
(Cowling et al., 1997) indicate that climatic or edaphic
specialization may hold the key to coexistence.
At the local (community) scale, it appears that envi-
ronmental heterogeneity alone cannot alone explain
richness, since highest diversity is recorded in renos-
terveld, even in relatively homogeneous environ-
ments. Clearly, other factors such as pollination
biology have an important role to play. Given the rel-
atively small number of pollinator species (Johnson &
Bond, 1997), additional specializations such as stag-
gered flowering season (Gleeson, 1981), nototribic vs.
sternotribic pollen deposition (e.g. Goldblatt & Man-
ning, 2000b) or sterility barriers (e.g. Johnson et al.,
1998), are certainly of importance in enabling several
different plant species to coexist without competing
for pollinators. Frost-free sites at low altitudes in the
Cape can benefit from pollinator activity year-round,
and indeed these have remarkable geophyte diversi-
ties, but comparisons with high altitudes are not pos-
sible, as comprehensive records of geophyte diversity
are lacking in this latter environment. Additionally,
phenological differences − some directed by phyloge-
netic constraints (Snijman, 1999) − may add extra
niche space.
However, these different explanations for different
scales need not always imply different speciation his-
tories. In some instances, pollinator and edaphic spec-
ificity go together, and it has been suggested that the
main factor leading to divergence may be the adapta-
tion to a new substratum, while pollination serves to
seal the speciation process (Goldblatt & Manning,
1996a). More two-species case studies, ideally involv-
ing plant translocation, would be needed to partition
the contribution of pollinators, substratum, and pos-
sibly other factors to present-day diversity.
At the scale of the entire Cape region, the western
vs. eastern contrasts are unlikely to be explained by
differences in edaphic or topographic heterogeneity,
since these do not vary significantly at the local and
landscape scales along this gradient (Campbell, 1983;
Deacon et al., 1992; Cowling & Lombard, 2002). The
more heterogeneous rainfall seasonality in the east
should, if anything, produce higher diversity here, by
enabling summer-active bulbs to invade and persist
(e.g. Drakensberg species). While eastern landscapes
do include many summer rainfall species, especially in
the orchids but also other groups (e.g. Ledebouria, Tri-
tonia), this does not seem to have had an effect on rich-
ness at all scales. For a given rock type, eastern soils
are marginally more fertile than western ones (Camp-
bell, 1983). However, our renosterveld−fynbos and
west−east comparisons are not consistent with the
hypothesis that lower soil fertility should result in
higher richness via finer niche partitioning (Tilman,
1986). With regard to biological factors, pollinator
diversity is probably greater in the east (S.Proches &
R.M. Cowling, unpubl. data). Most likely, the greater
diversity in the west has to do with long-term climatic
stability, which reduced extinction rates and allowed
diversification processes to continue uninterrupted
throughout the Quaternary, whereas in the eastern
fynbos, climate change appears to have caused large-
scale vegetation disruption, thereby enhancing extinc-
tion rates (Cowling et al., 1999a). On shorter time
scales, the less predictable rainfall regime in the east
(Cowling et al., 2005a) may limit geophyte diversity by
filtering out those species that require predictable
rains in order to amass adequate resources for repro-
duction. Competition from grasses may also limit the
spread of certain geophyte groups in the east, where
Figure 7. Representation of hysteranthous, synanthous,
and evergreen geophytes at five localities in the Cape. Pie
size is proportional to the total numbers of species (data
from Reid & Dyer, 1984; Snijman, 1984; Goldblatt, 1985,
1986, 1989; Perry, 1994; Goldblatt & Manning, 1998).

40 S. PROCHES ET AL.
© 2006 The Linnean Society of London, Biological Journal of the Linnean Society, 2006, 87, 27 –43
grass cover is greater, although this hypothesis, too,
would need proper testing.
On a global scale, mediterranean-type ecosystems
are likely to owe their geophyte richness to pro-
nounced rainfall seasonality and to the occurrence of
fire, which creates a second (multiyear) periodicity,
and temporarily reduces competition (see e.g. Hoff-
man, Moll & Boucher, 1987; Schwilk, Keeley & Bond,
1997). In geophytes, fire does not create a reproductive
barrier between generations, speeding up evolution,
as has been suggested in the case of plants that are
killed by fire (Cowling, 1987). However, the increased
patchiness of fire-prone landscapes is likely to
increase speciation in geophytes too (cf. Verboom,
Linder & Stock, 2003, where the case of some bulbous
grasses is discussed). Fire also creates special niches
for geophytes. In the Cape, there are numerous species
that only flower or germinate after burning (in Ama-
ryllidaceae, Asphodelaceae, Geraniaceae, Haemodora-
ceae, Hyacinthaceae, and Iridaceae; Le Maitre &
Brown, 1992; Brown, van Staden & Johnson, 2003).
Fire and seasonality concur to temporarily eliminate
competition from shrubs and grasses, and encourage
the geophytic lifestyle. In comparison to other medi-
terranean-type ecosystems, the key to the Cape’s
unequalled geophyte richness may again have to do
with climatic stability. During the Quaternary, south-
ern Africa is likely to have been one of the world’s most
climatically stable land masses (Weaver et al., 1998).
In terms of short-term rainfall reliability, it comes sec-
ond only to south-western Australia (Cowling et al.,
2005a), which, on the other hand, has much less to
offer in terms of topographical diversity.
A stable climate with reliable seasonal rainfall over
a relatively long period of geological time, in combina-
tion with topographically diverse landscapes, was per-
haps the key factor determining the amazing diversity
of geophytes in the Cape. While the examined land-
scapes enabled population isolation on different sub-
strata and the associated diversification processes
(augmented by pollinator specialization), climatic sta-
bility minimized extinction, thus leading to the unin-
terrupted accumulation of species over the last few
millions of years. Within the huge array of resulting
taxa, it was more likely here than anywhere else in the
world that unusual pollination syndromes and leaf
shape adaptations should recur in numerous unre-
lated taxa.
ACKNOWLEDGEMENTS
This study was partly supported by a Claude Leon fel-
lowship to SP, and a National Research Foundation
grant to RMC. Mark Difford, John Donaldson, and Syd
Ramdhani and an anonymous referee are thanked for
their comments on earlier versions. Peter Linder,
Vincent Savolainen, and the participants of the
Hyacinthaceae guild symposium (Kirstenbosch,
September 2003) contributed valuable discussions.
REFERENCES
APG II. 2003. An update of the Angiosperm Phylogeny Group
classification for the orders and families of flowering plants:
APG II. Botanical Journal of the Linnean Society 141: 399–
436.
Axelrod DI. 1975. Evolution and biogeography of Madrean-
Tethyan sclerophyll vegetation. Annals of the Missouri
Botanical Garden 62: 280–334.
Axelrod DI, Raven PH. 1978. Late Cretaceous and Tertiary
vegetation history of Africa. In: Werger MJA, ed. Biogeogra-
phy and ecology of southern Africa. The Hague: Junk, 77–
130.
Bakker FT, Culham A, Hettiarachi P, Touloumenidou T,
Gibby M. 2004. Phylogeny of Pelargonium (Geraniaceae)
based on DNA sequences from three genomes. Taxon 53: 17–
28.
Bayer MB. 1984. The Capensis Flora and the Karoo – a winter
rainfall biome versus a fynbos biome. Veld and Flora 70: 17–
19.
Bews JW. 1925. Plant forms and their evolution in South
Africa. New York: Longman, Green.
Brown NAC, van Staden J, Johnson T. 2003. Patterns in
the seed germination response to smoke in plants from the
Cape Floristic Region, South Africa. South African Journal
of Botany 69: 514–525.
Campbell BM. 1983. Montane plant environments in the Fyn-
bos Biome. Bothalia 14: 283–298.
Caujapé-Castells J, Jansen RK, Pedrola-Monfort J,
Membrives N. 2002. Space-time diversification of Andro-
cymbium Willd. (Colchicaceae) in western South Africa.
Plant Systematics and Evolution 232: 73–88.
Chase MW, de Bruijn AY, Cox AV, Reeves G, Rudall PJ,
Johnson MAT, Eguiarte LE. 2000b. Phylogenetics of
Asphodelaceae: an analysis of plastid rbcL and trnL-F DNA
sequences. Annals of Botany 86: 935–951.
Chase MW, Soltis DE, Soltis PS, Rudall P, Fay MF, Hahn
WJ, Sullivan S, Joseph J, Molvray M, Kores PJ,
Givnish TJ, Sytsma KJ, Pires JC. 2000a. Higher-level
systematics of the monocotyledons. An assessment of current
knowledge and a new classification. In: Wilson KL, Morrison
DA, eds. Monocots: systematics and evolution. Melbourne:
CSIRO, 3–16.
Conran JG. 1995. Family distributions in the Liliiflorae and
their biogeographical implications. Journal of Biogeography
22: 1023–1034.
Conran JG. 1999. Anatomy and morphology of Behnia (Beh-
niaceae) and its relationships within Lilianae, Asparagales.
Botanical Journal of the Linnean Society 131: 115–129.
Cowling RM. 1987. Fire and its role in coexistence and spe-
ciation in Gondwanan shrublands. South African Journal of
Science 83: 106–111.
Cowling RM, Cartwright CR, Parkington JE, Allsopp J.
1999a. Fossil wood charcoal assemblages from Elands Bay

AN OVERVIEW OF THE CAPE GEOPHYTES 41
© 2006 The Linnean Society of London, Biological Journal of the Linnean Society, 2006, 87, 27 –43
Cave, South Africa: implications for Late Quaternary
vegetation and climates in the winter-rainfall fynbos biome.
Journal of Biogeography 26: 367–378.
Cowling RM, Esler KJ, Rundel PW. 1999b. Namaqualand,
South Africa – an overview of a unique winter-rainfall desert
ecosystem. Plant Ecology 142: 3–21.
Cowling RM, Lombard AT. 2002. Heterogeneity, speciation/
extinction history and climate: explaining regional plant
diversity in the Cape Floristic Region. Diversity and Distri-
butions 8: 163–179.
Cowling RM, Ojeda F, Lamont BB, Rundel PW, Lech-
mere-Oertel R. 2005a. Rainfall reliability, a neglected fac-
tor in explaining convergence and divergence of plant traits
in fire-prone mediterranean-climate ecosystems. Global
Ecology and Biogeography 14: 509–519.
Cowling RM, Prochess
ss SS
SS, Vlok JHJ. 2005b. On the origin of
southern African subtropical thicket vegetation. South Afri-
can Journal of Botany 71: 1–23.
Cowling RM, Richardson DM, Schultze RE, Hoffman
MT, Midgley JJ, Hilton-Taylor C. 1997. Species diversity
at the regional scale. In: Cowling RM, Richardson DM,
Pierce SM, eds. Vegetation of southern Africa. Cambridge:
Cambridge University Press, 447–473.
Cowling RM, Rundel PW, Desmet PG, Esler KJ. 1998.
Extraordinary high regional-scale plant diversity in south-
ern African arid lands: subcontinental and global compari-
sons. Diversity and Distributions 4: 27–36.
Cowling RM, Rundel PW, Lamont BB, Arroyo MK, Ari-
anoutsou M. 1996. Plant diversity in mediterranean-climate
regions. Trends in Ecology and Evolution 11: 362–366.
Dafni A, Cohen D, Noy-Meir I. 1981. Life-cycle variation in
geophytes. Annals of the Missouri Botanical Gardens 68:
652–660.
Deacon HJ, Jury MR, Ellis F. 1992. Selective regime and
time. In: Cowling RM, ed. The ecology of fynbos: nutrients,
fire and diversity. Cape Town: Oxford University Press, 6–22.
Donato R, Leach C, Conran JG. 2000. Relationships of
Dietes (Iridaceae) inferred from ITS2 sequences. In: Wilson
KL, Morrison DA, eds. Monocots: systematics and evolution.
Melbourne: CSIRO, 407–413.
Duncan GD. 1998. Five new species of Lachenalia (Hya-
cinthaceae) from arid areas of Namibia and South Africa.
Bothalia 28: 131–139.
Esler KJ, Rundel PW, Vorster P. 1999. Biogeography
of prostrate-leaved geophytes in semi-arid South Africa:
hypotheses on functionality. Plant Ecology 142: 105–120.
Fay MF, Chase MW. 1996. Resurrection of Themidaceae for
the Brodiaea alliance, and recircumscription of Alliaceae,
Amaryllidaceae and Agapanthaceae. Taxon 45: 441–451.
Fay MF, Rudall P, Sullivan S, Stobart KL, de Bruijn AY,
Reeves G, Qamaruz-Zaman F, Hong W-P, Joseph J,
Hahn WJ, Conran JP, Chase MW. 2000. Phylogenetic
studies of Asparagales based on four plastid DNA regions.
In: Wilson KL, Morrison DA, eds. Monocots: systematics and
evolution. Melbourne: CSIRO, 360–371.
Germishuizen G, Meyer NL, eds. 2003. Plants of southern
Africa: an annotated checklist. Strelitzia 14: 1–1231.
Gleeson SK. 1981. Character displacement in flowering phe-
nologies. Oecologia 51: 294–295.
Goldblatt P. 1979. Biology and systematics of Galaxia
(Iridaceae). Journal of South African Botany 45: 385–423.
Goldblatt P. 1983. Geography of Iridaceae in Africa. Bothalia
14: 559–564.
Goldblatt P. 1985. Revision of the southern African genus
Geissorhiza (Iridaceae: Ixioideae). Annals of the Missouri
Botanical Garden 72: 277–447.
Goldblatt P. 1986. The moraeas of southern Africa. Annals of
Kirstenbosch Botanical Gardens 14.
Goldblatt P. 1989. The genus Watsonia. A systematic mono-
graph. Annals of Kirstenbosch Botanical Gardens 19.
Goldblatt P. 1990. Phylogeny and classification of Iridaceae.
Annals of the Missouri Botanical Garden 77: 607–627.
Goldblatt P, Davies TJ, Savolainen V, Manning JC, van
der Bank M. 2005. Phylogeny of Iridaceae subfamily Cro-
coideae based on plastid DNAs. Monocots III. Proceedings of
the 3rd Monocot Meeting, Rancho Santa Ana Botanical Gar-
dens, March 2003.
Goldblatt P, Manning JC. 1996a. Phylogeny and speciation
in Lapeirousia subgenus Lapeirousia (Iridaceae: Ixioideae).
Annals of the Missouri Botanical Garden 83: 346–361.
Goldblatt P, Manning JC. 1996b. Two new edaphic endemic
species and taxonomic changes in Gladiolus (Iridaceae)
of southern Africa, with notes on Iridaceae restricted to
unusual substrates. Novon 6: 172–180.
Goldblatt P, Manning JC. 1998. Gladiolus in southern
Africa. Cape Town: Fernwood Press.
Goldblatt P, Manning JC. 2000a. Cape plants: a conspectus
of the Cape flora of South Africa. Strelitzia 7: 1–743.
Goldblatt P, Manning JC. 2000b. The long-proboscid fly pol-
lination system in southern Africa. Annals of the Missouri
Botanical Garden 87: 147–170.
Goldblatt P, Manning JC, Bernhardt P. 1995. Pollination
biology of Lapeirousia subgenus Lapeirousia (Iridaceae) in
southern Africa; adaptation for long-tongued fly pollination.
Annals of the Missouri Botanical Garden 82: 517–534.
Goldblatt P, Manning JC, Bernhardt P. 1998. Adaptive
radiation of bee-pollinated Gladiolus species (Iridaceae) in
southern Africa. Annals of the Missouri Botanical Garden
85: 492–517.
Goldblatt P, Manning JC, Bernhardt P. 2000. Adaptive
radiation of pollination mechanisms in Sparaxis (Iridaceae:
Ixioideae). Adansonia 22: 57–70.
Goldblatt P, Savolainen V, Porteous O, Sostaric I, Powell
M, Reeves G, Manning JC, Barraclough TG, Chase
MW. 2002. Radiation in the Cape flora and the phylogeny of
peacock irises Moraea (Iridaceae) based on four plastid DNA
regions. Molecular Phylogeny and Evolution 25: 341–360.
Gutterman Y, Boeken B. 1988. Flowering affected by
daylength and temperature in the leafless flowering desert
geophyte Colchicum tunicatum, its annual life cycle and veg-
etative propagation. Botanical Gazette 149: 382–390.
Herrera CM. 1992. Historical effects and sorting processes as
explanations for contemporary ecological patterns: character
syndromes in mediterranean woody plants. American Natu-
ralist 140: 421–446.
Hoffman MT, Moll EJ, Boucher C. 1987. Post-fire succes-
sion at Pella, a South African lowland fynbos site. South Afri-
can Journal of Botany 53: 370–374.

42 S. PROCHES ET AL.
© 2006 The Linnean Society of London, Biological Journal of the Linnean Society, 2006, 87, 27 –43
Hoffmann AJ, Liberona F, Hoffmann AE. 1998. Distribu-
tion and ecology of geophytes in Chile. Conservation threats
to geophytes in mediterranean-type regions. In: Rundel PW,
Montenegro G, Jaksic FM, eds. Landscape degradation
and biodiversity in mediterranean-type ecosystems. Berlin:
Springer, 231–253.
Hopper SD, Manning J. 2004. Granite outcrop floras of the
winter rainfall Greater Cape and South West Australia – a
comparative study. Presentation at the 4th Southern Connec-
tions Conference, 19–23 January 2004. University of Cape
Town.
Johnson SD. 1992. Climatic and phylogenetic determinants
of flowering seasonality in the Cape Flora. Journal of Ecol-
ogy 81: 567–572.
Johnson SD. 1996. Pollination, adaptation and speciation
models in the Cape Flora of South Africa. Taxon 45: 59–66.
Johnson SD. 1997. Insect pollination and floral mechanisms
in South African species of Satyrium (Orchidaceae). Plant
Systematics and Evolution 204: 195–206.
Johnson SD, Bond WJ. 1997. Evidence for widespread pollen
limitation of fruiting success in Cape wildflowers. Oecologia
109: 530–534.
Johnson SD, Steiner KE. 2000. Generalization versus spe-
cialization in plant pollination systems. Trends in Ecology
and Evolution 15: 140–143.
Johnson SD, Steiner KE, Whitehead VB, Vogelpoel L.
1998. Pollination ecology and maintenance of species integ-
rity in co-occurring Disa racemosa L.F. and Disa venosa Sw.
(Orchidaceae) in South Africa. Annals of the Missouri Botan-
ical Gardens 85: 231–241.
Jürgens N. 1991. A new approach to the Namib Region. I:
Phytogeographic subdivision. Vegetatio 97: 21–38.
Klak C, Reeves G, Hedderson T. 2004. Unmatched tempo of
evolution in southern African semi-desert ice plants. Nature
427: 63–65.
Le Maitre DC, Brown PJ. 1992. Life cycles and fire-stimu-
lated flowering in geophytes. In: van Wilgen BW, Richardson
DM, Kruger FJ, van Hensenbergen HJ, eds. Fire in South
African mountain fynbos. Berlin: Springer, 145–160.
Leistner OA, ed. 2000. Seed plants of southern Africa: fami-
lies and genera. Strelitzia 10: 1–775.
Lewis GJ. 1954. Some aspects of the morphology, phylogeny
and taxonomy of the South African Iridaceae. Annals of the
South African Museum 40: 15–113.
Linder HP. 1995. Setting conservation priorities: the
importance of endemism and phylogeny in the southern
African orchid genus Herschelia. Conservation Biology 9:
585–595.
Linder HP, Kurzweil H. 1999. Orchids of southern Africa.
Rotterdam: AA Balkema.
Lovegrove BG, Jarvis JUM. 1986. Coevolution between
mole-rats (Bathyergidae) and a geophyte, Micranthus (Iri-
daceae). Cimbebasia 8: 79–85.
Low AB, Rebelo AG. 1998. Vegetation of South Africa,
Lesotho and Swaziland. Pretoria: Department of Environ-
mental Affairs and Tourism.
Manning JC, Goldblatt P, Fay M. 2004. A revised generic
synopsis of Hyacinthaceae in sub-Saharan Africa based on
molecular evidence, including new combinations and the
new tribe Pseudoprospereae. Edinburgh Journal of Botany
60: 533–568.
Manning JC, Goldblatt P, Snijman D. 2002. The color ency-
clopedia of Cape Bulbs. Portland, OR: Timber Press.
Manning JC, van der Merwe AM. 2002. Systematics of the
genus Daubenya (Hyacinthaceae: Massonieae). Bothalia 32:
133–150.
Marais EM. 1994. Taxonomic studies in Pelargonium section
Hoarea (Geraniaceae). PhD Thesis, University of Stel-
lenbosch.
Meerow AW, Fay MF, Guy CL, Li Q-B, Qamaruz-Zaman F,
Chase MW. 1999. Systematics of Amaryllidaceae based on
cladistic analysis of plastid rbcL and trnL-F sequence data.
American Journal of Botany 86: 1325–1345.
Müller-Doblies D, Müller-Doblies U. 1996. Tribes and sub-
tribes and some species combinations in Amaryllidaceae J.
St.-Hil. Emend R. Dahlgren et al. 1985. Feddes Repertorium
107 (5–6): S.c. 1–9.
Nordal I, Duncan T. 1984. A cladistic analysis of Haeman-
thus and Scadoxus. Nordic Journal of Botany 4: 145–153.
Oberlander KC, Dreyer LL, Esler KJ. 2002. Biogeography
of Oxalis (Oxalidaceae) in South Africa: a preliminary study.
Bothalia 32: 97–100.
Parsons RF. 2000. Monocotyledonous geophytes: comparison
of California with Victoria, Australia. Australian Journal of
Botany 48: 39–43.
Parsons RF, Hopper SD. 2003. Monocotyledonous geo-
phytes: a comparison of south-western Australia with other
areas of mediterranean climate. Australian Journal of Bot-
any 51: 129–133.
Pate JS, Dixon KW. 1982. Tuberous, cormous and bulbous
plants. Perth: University of Western Australia Press.
Perry PL. 1994. A revision of the genus Eriospermum (Eri-
ospermaceae). Contributions from the Bolus Herbarium 17:
1–320.
Pfosser M, Wetschnig W, Ungar S, Prenner G. 2003. Phy-
logenetic relationships among genera of Massonieae (Hya-
cinthaceae) inferred from plastid DNA and seed morphology.
Journal of Plant Research 116: 115–132.
Prochess
ss SS
SS, Cowling RM, du Preez DR. 2005. Patterns of
geophyte diversity and storage organ size in the winter rain-
fall region of southern Africa. Diversity and Distributions 11:
101–109.
Quézel P. 1978. Analysis of the Flora of Mediterranean and
Saharan Africa. Annals of the Missouri Botanical Garden 65:
479–534.
Raunkiaer C. 1934. The life forms of plants. Oxford: Claren-
don Press.
Rees A. 1989. Evolution of the geophytic habit and its physi-
ological advantages. Herbertia 45: 104–110.
Reeves G, Chase MW, Goldblatt P, Rudall P, Fay MF, Cox
AV, Lejeune B, Souza-Chies T. 2001. Molecular system-
atics of Iridaceae: evidence from four plastid DNA regions.
American Journal of Botany 88: 2074–2087.
Reid C, Dyer RA. 1984. A review of the southern African spe-
cies of Cyrtanthus. La Jolla, CA: American Plant Life Society.
Ronquist F. 1997. Dispersal-vicariance analysis: a new
approach to the quantification of historical biogeography.
Systematic Biology 46: 195–203.

AN OVERVIEW OF THE CAPE GEOPHYTES 43
© 2006 The Linnean Society of London, Biological Journal of the Linnean Society, 2006, 87, 27 –43
Rossa B, von Willert DJ. 1999. Physiological characteristics
of geophytes in semi-arid Namaqualand, South Africa. Plant
Ecology 142: 121–132.
Rourke JP. 2002. Clivia mirabilis (Amaryllidaceae: Hae-
mantheae) a new species from Northern Cape, South Africa.
Bothalia 32: 1–7.
Rudall PJ. 2003. Unique floral structures and iterative evo-
lutionary themes in Asparagales: insights from a morpholog-
ical cladistic analysis. The Botanical Review 68: 488–509.
Ruiters C. 2001. Biogeographical and conservation analysis of
geophytes in the winter rainfall region of the Fynbos Biome,
a Mediterranean-type ecosystem of South Africa. PhD The-
sis, University of the Western Cape.
Ruiters C, McKenzie B. 1994. Seasonal allocation and effi-
ciency patterns of biomass and resources in the perennial
geophyte Sparaxis grandiflora subspecies fimbriata (Iri-
daceae) in lowland coastal fynbos, South Africa. Annals of
Botany 72: 633–646.
Rundel PW. 1996. Monocotyledonous geophytes in the Cali-
fornia flora. Madroño 43: 355–368.
Salter TM. 1944. The genus Oxalis in South Africa: a taxo-
nomic revision. Journal of South African Botany (supple-
ment) 1: 1–355.
Schwilk DW, Keeley JE, Bond WJ. 1997. The intermediate
disturbance hypothesis does not explain fire and diversity
pattern in fynbos. Plant Ecology 132: 77–84.
Snijman DA. 1984. A revision of the genus Haemanthus
(Amaryllidacae). Journal of South African Botany (supple-
ment) 12: 1–139.
Snijman DA. 1999. Growth periodicity, flowering, and phy-
logeny of the African Amaryllideae (Amaryllidaceae). In:
Timberlake J, Kativu S, eds. African plants: biodiversity, tax-
onomy and uses. Kew: Royal Botanic Gardens, 389–404.
Snijman DA, Linder PH. 1996. Phylogenetic relationships,
seed characters, and dispersal system evolution in Amaryl-
lideae (Amaryllidaceae). Annals of the Missouri Botanical
Gardens 83: 362–386.
Snijman DA, Perry PL. 1987. A floristic analysis of the Nieu-
woudtville Wild Flower Reserve, north-western Cape. South
African Journal of Botany 53: 445–454.
Soltis DE, Soltis PS, Chase MW, Mort ME, Albach DC,
Zanis M, Savolainen V, Hahn WH, Hoot SB, Fay MF,
Axtell M, Swensen SM, Prince LM, Kress WJ, Nixon
KC, Farris JS. 2000. Angiosperm phylogeny inferred from
18S rDNA, rbcL, and atpB sequences. Botanical Journal of
the Linnean Society 133: 381–461.
Stevens PF. 2001. Angiosperm Phylogeny Website. http://
www.mobot.org/MOBOT/research/APweb/.
Thomson JD, Weiblen G, Thomson BA, Alfaro S, Leg-
endre P. 1996. Untangling multiple factors in spatial dis-
tributions: lilies, gophers, and rocks. Ecology 77: 1698–1715.
Tilman D. 1986. Evolution and differentiation in plant com-
munities: the importance of the soil resource. Princeton, NJ:
Princeton University Press.
Van der Walt JJA, Vorster PJ. 1983. Phytogeography of
Pelargonium. Bothalia 14: 517–523.
Van Wyk AE, Smith GF. 2001. Regions of floristic endemism
in southern Africa. Pretoria: Umdaus Press.
Verboom GA, Linder HP, Stock WD. 2003. Phylogenetics of
the grass genus Ehrharta: evidence for radiation in the sum-
mer-arid zone of the South African Cape. Evolution 57:
1008–1021.
Verdú M, Dávila P, García-Fayos P, Flores-Hernández N,
Valiente-Banuet A. 2003. ‘Convergent’ traits of Mediterra-
nean woody plants belong to pre-Mediterranean lineages.
Biological Journal of the Linnean Society 78: 415–427.
Vinnersten A, Bremer K. 2001. Age and biogeography of
major clades in Liliales. American Journal of Botany 88:
1695–1703.
Vinnersten A, Reeves G. 2003. Phylogenetic relationships
within Colchicaceae. American Journal of Botany 90: 1455–
1462.
Weaver AJ, Eby M, Fanning AF, Wiebe EC. 1998. Simu-
lated influence of carbon dioxide, orbital forcing and ice
sheets on the climate of the Last Glacial Maximum. Nature
394: 847–853.
Westoby M. 1989. Selective forces exerted by vertebrate her-
bivores on plants. Trends in Ecology and Evolution 4: 115–
117.
Zachos J, Pagani M, Sloan L, Thomas E, Billups K. 2001.
Trends, rhythms, and aberrations in global climate 65 Ma to
present. Science 292: 686–693.