Content uploaded by Philip George Desmet
Author content
All content in this area was uploaded by Philip George Desmet on May 04, 2018
Content may be subject to copyright.
Diversity and Distributions (1998) 4, 27–36
BIODIVERSITY RESEARCH
Extraordinary high regional-scale plant diversity in
southern African arid lands: subcontinental and global
comparisons
R. M. COWLING
1
∗, P. W. RUNDEL
2
,P.G.DESMET
1
and K. J. ESLER
31
Institute for Plant
Conservation, University of Cape Town, Rondebosch 7701, South Africa,
2
Department of Biology, University of
California, Los Angeles, CA 90024-1768, U.S.A.
3
Department of Botany, University of Stellenbosch, Stellenbosch
7600, South Africa.
Abstract.
The Karoo-Namib Region, comprising the Saharan Desert. On the other hand, the Nama
Karoo does not appear to be unusually rich at thethe arid lands of southern Africa, supports an
extraordinarily rich and compositionally unusual flora. regional scale.
The Succulent Karoo represents a major extra-This region includes the winter-rainfall Succulent
Karoo (part of the Greater Cape Flora), the summer- tropical centre of plant biodiversity. Reasons for its
high diversity include unusually predictable seasonalrainfall Nama Karoo, and incorporates the Namib
Desert. The Succulent Karoo is home to the world’s rainfall and relatively mild summers. The climatic
regime has selected for short-lived and drought-largest succulent flora and is very rich in geophytes.
Species-area patterns at the regional scale sensitive shrub lifestyles, largely associated with dwarf
to low leaf succulents. A combination of short(10
1
–10
6
km
2
) show that the Succulent Karoo had 2.6
times as many species per unit area than the Nama generation times and limited gene flow has resulted in
massive diversification within certain lineagesKaroo. The Succulent Karoo was also richer than other
winter-rainfall, semi-arid (100–400 mmyr
−1
) regions of (especially the Mesembryanthema), leading to the fine-
scale discrimination of habitat and geographic space.the world, having, for example, nearly four times as
many species than equivalent-sized North American Plant-pollinator coevolution appears to have played
some role in the diversification of the region’s largeregions. Similarly, southern African winter-rainfall
desert (<100 mmyr
−1
) regions were exceptionally geophyte flora.
species-rich: here 331 species have been recorded in an
area of 1.3 km
2
. The Namib Desert, which spans the
Key words.
Arid lands, diversification, global
comparisons, plant diversity, regional scale, southernSucculent and Nama Karoo regions, has up to 200
times the number of species that similar-sized areas of Africa, Succulent Karoo.
INTRODUCTION
Region (e.g. Cowling, Holmes & Rebelo, 1992). The
extraordinary plant diversity and floristic peculiarity
For a largely warm temperate area, southern Africa of southern Africa’s arid lands is less well known.
has a remarkably rich flora of vascular plants, with While Cape fynbos has its analogue in the kwongan
23,404 species and infraspecific taxa recorded from the of mediterranean southwestern Australia (Cowling et
region, of which some 80% are endemic (Cowling & al., 1996), the flora of southern Africa’s winter-rainfall
Hilton-Taylor, 1994). Most research has focused on karoo, with its unrivalled diversity of succulent species
describing and explaining patterns of diversity in the (Van Jaarsveld, 1987; Ihlenfeldt, 1994, Hilton-Taylor,
predominantly fynbos flora of the Cape Floristic 1996), is in a league of its own.
In this paper we describe and compare patterns of
regional plant diversity at a subcontinental (southern
Africa) and global scale. We also attempt to explain∗Corresponding author; e-mail: rmc@botzoo.uct.ac.za.
27
1998 Blackwell Science Ltd
28 R. M. Cowling et al.
Fig. 1.
Phytogrographical map of south-western Africa. Redrawn from Ju
¨rgens’ (1991).
these patterns in terms of differences in selective
STUDY AREA
regimes, plant life histories, and differential rates of
Biogeographical background
diversification (Rosenzweig, 1995).
It is necessary to clarify the use of some terms in Our area of interest is the Karoo-Namib Region of
this paper. Firstly, we use the term ‘Karoo-Namib southern Africa (Werger, 1978a,b), an arid to semi-
Region’ to describe, in a phytogeographical context, arid phytochorion distinct from other arid lands of
the semi-arid shrubland and desert areas of southern sub-Saharan Africa. Werger (1978a,b) followed White
Africa (Werger, 1978a). Although Ju
¨rgens (1991) (1976, 1983) in recognizing a Karoo-Namib
provides the necessary justification for dividing this phytochorion and agreed with earlier workers that this
region into two phytochoria of equal rank, namely the unit forms part of the Palaeotropical Kingdom as
Succulent Karoo Region and the Nama Karoo Region opposed to the Cape Kingdom. He did, however,
(Fig. 1), we retain the term as a matter of convenience. emphasize the transitional nature of the region’s flora,
Secondly, our species counts always refer to vascular with strong Sudano-Zambezian affinities towards its
plants only. Thirdly, we use the terms ‘species’ instead northern and eastern boundaries, and strong Cape
affinities towards the south-west.of the more long-winded ‘species and infraspecific taxa’.
1998 Blackwell Science Ltd, Diversity and Distributions,4, 27–36
High plant diversity in the Karoo 29
Using distribution data for 1700 species, Ju
¨rgens partly Namib Subdomain of the Namaland Domain)
(Robinson, 1978). There are no reliable estimates of(1991) developed a new phytogeographical subdivision
of southern Africa’s arid lands (Fig. 1). Rather than the size of the Nama Karoo flora: Gibbs Russell (1987)
provides a value of 2147 species in a ‘core area’ oftinkering with boundaries, his proposals have far-
reaching implications for African biogeography. The 198,500km
2
, about 30% of the region.
Floras from the Succulent Karoo differfundamentals of his scheme are summarized below.
compositionally from Nama Karoo floras and those
•The division of the Karoo-Namib Region into two from other arid lands (Cowling & Hilton-Taylor, in
phytochoria of equal rank, namely the Succulent press). They are uniquely characterized by high
Karoo Region (more-or-less coincident with Werger’s numbers of Aizoaceae (especially Mesembryanthema)
(1978a) Western Cape Domain) and the Nama Karoo (Hartmann, 1991), and relatively high numbers of
Region.Iridaceae and Geraniaceae. The last two-mentioned
•The recognition that the Succulent Karoo Region families are well represented in all fynbos floras of the
forms part of a greater Cape Flora rather than part Cape Region, and Aizoaceae rank highly in lowland
of the Palaeotropical Kingdom (also see Hilton- fynbos and renosterveld floras (Cowling & Holmes,
Taylor, 1987). 1992). Succulents are extremely well represented in
•The recognition of a new system of domains in both Succulent Karoo floras and are associated with many
regions, delimited on the basis of the distribution of of the larger families (Aizoaceae, Asteraceae, Liliaceae,
species from zonal vegetation complexes. Thus, the Crassulaceae, Geraniaceae, Euphorbiaceae and
Namaqualand-Namib Domain (strongly winter Ascelepiadaceae). Indeed, the Succulent Karoo
rainfall) and the Southern Karoo Domain (winter probably harbours about one third of the world’s
to non-seasonal rainfall) are recognized for the approximately 10,000 succulent species (Van Jaarsveld,
Succulent Karoo Region; and the Namaland 1987; Smith et al., 1993).
Domain (including most of Werger’s (1987a) The Succulent Karoo flora includes 730 genera of
Namaqualand and Namib Domains), Damaraland- which sixty-seven (9.2%) are endemic (Hilton-Taylor,
Kaokoland Domain (see also Nordenstam, 1974; 1996). This is more than three times the number of
Hilton-Taylor, 1994), and the Eastern Karoo endemic genera in the Sahara–Arabian and North
Domain (corresponding to Werger’s (1978a) Karoo American arid lands (Shmida, 1985), which are orders
Domain) are recognized for the Nama Karoo of magnitude larger than the Succulent Karoo. Another
Region. unusual feature of the flora is the high species to genus
•The division of the Namib Desert (as a geographic ratio of 6.9. The region includes several unusually
entity) into two phytochoria: a south-central coastal large genera for an arid land flora, including Ruschia
portion as part of the Succulent Karoo (Aizoaceae: 136 spp.), Conophytum (Aizoaceae:
(Namaqualand-Namib Domain); and a northern and 116 spp.), Oxalis (Oxalidaceae: 114 spp.), Euphorbia
inland zone as part of the Nama Karoo (Namib (Euphorbiaceae: 77 spp.), Pelargonium (Geraniaceae:
Subdomain of the Namaland Domain) (see also 72 spp.), Senecio (Asteraceae: 72 spp.), Eriospermum
Robinson, 1978). (Liliaceae: 65 spp.), Othonna (Asteraceae: 61spp.), and
Drosanthemum (Aizoaceae: 55 spp.) (Hilton-Taylor,In this paper we refer exclusively to Ju
¨rgens’ (1991)
scheme of regions and domains. 1996). Most of these large genera are comprised entirely
of either succulent shrubs or geophytes.
The Nama Karoo flora is not as well known as the
Floras
flora of the Succulent Karoo. Generally, regional floras
are more typical of those from other arid landsThe flora of southern Africa’s arid lands is unusually
rich and compositionally interesting (Cowling & (Cowling & Hilton-Taylor, in press) where Asteraceae,
Poaceae and Fabaceae are the most frequent top-Hilton-Taylor, in press). The entire flora of the Karoo-
Namib Region probably exceeds 7000 species of which ranking families (Shmida, 1985). As with the Succulent
Karoo, Nama Karoo floras are unusual among aridup to 50% are endemic (Hilton-Taylor & Le Roux,
1989). There are 4849 species in the Succulent Karoo land floras in the low importance of Chenopodiaceae.
Interestingly, the Damaraland-Kaokoland Domain(40.3% endemic) (Hilton-Taylor, 1996) and 1427 species
(34.5% endemic) in the Namibian part of the Namib flora (Robinson, 1978; Nordenstam, 1974; Hilton-
Taylor, 1994) bears a strong resemblance to TharDesert (partly Namaqualand-Namib Domain and
1998 Blackwell Science Ltd, Diversity and Distributions,4, 27–36
30 R. M. Cowling et al.
(Indian) Desert floras in the relative importance of provide additional sources of moisture during the
dry summer months (Walker, 1986; Von Willert etAcanthaceae, Capparaceae and Cucurbitaceae
(Shmida, 1985). Although no data have been published, al., 1992; Desmet, 1996). In the Nama Karoo, violent
rainfall events are the principle source ofthere are certainly fewer genera endemic to the Nama
Karoo than the Succulent Karoo. However, the former precipitation (Werger, 1986).
region is home to the only exclusively desert Thus, the Succulent Karoo has a warm temperate and
family—the Welwitschiaceae (Shmida, 1985). oceanic climate characterized by relatively mild winters
with low but reliable winter rainfall, and relatively mild
summers where drought is ameliorated by heavy dew
Selective regimes
and frequent fog. The Nama Karoo has a more
The Succulent Karoo and Nama Karoo differ continental climate, characterized by hot summers with
fundamentally in their climatic, and hence, selective low and unreliable rainfall, and very dry, mild to cool
regimes (Desmet & Cowling, in press). These winters with frequent night frosts. It is important to
differences, which have profound implications for emphasize that the environmental conditions in the
diversity and rates of diversification, are summarized Southern Karoo Domain of the Succulent Karoo are
as follows. in many respects transitional between those in the
Namaqualand-Namib Domain and the Nama Karoo.
•The Succulent Karoo derives most of its rainfall
from depressions associated with the circumpolar
westerly belt (Schulze & McGee, 1978). Most rain
METHODS
falls during winter (May–September) (Werger, 1978a,
1986; Rutherford & Westfall, 1986) and, even at We compiled species–area data at a regional scale
low annual totals, rainfall is remarkably predictable (10
1
–10
6
km
2
) for warm temperate and subtropical arid
when compared to other winter-rainfall arid lands lands from throughout the world. We divided our
(Desmet & Cowling, in press; Elser, Rundel & data set into different categories: Nama Karoo sites;
Cowling, in press). Succulent Karoo sites; summer rainfall, semi-arid
•Nama Karoo rainfall is derived from a number of (100–400 mmyr
−1
) sites; winter rainfall, semi-arid sites;
sources to the east and north of the region (Desmet and desert (<100 mmyr
−1
) sites. We had difficulty in
& Cowling, in press). Most rain falls during the late locating data from the New World, summer rainfall
summer months (or with strong autumn and weak regions and from Australia. The double logarithmic
spring peaks in the Eastern Karoo) and, for a given form of the species–area curve provided the best fit for
annual total, rainfall is much less predictable than all data sets (Williamson, 1988; Rosenzweig, 1995).
in the Succulent Karoo (Hoffman & Cowling 1987). We used Ftests to compare whether the slopes and
•In the Succulent Karoo, rainfall events are intercepts of species–area regressions for different data
widespread, of long duration and are mostly gentle sets were significantly different (Zar, 1984). In cases
showers; rainfall events in the Nama Karoo are where we could not reject the null hypothesis that
invariably highly localized, short, intense slopes were identical in the overall populations, i.e.
thundershowers (Desmet & Cowling, in press). they were homogeneous, we compared the intercept-
•Temperature extremes in the Succulent Karoo are or k-ratio of the two curves. This ratio provided an
amelioraed by proximity to the ocean and the high overall estimate of the difference in regional diversity
incidence, in the Namaqualand-Namib Domain, of between the two data sets (Gould, 1979).
fog during the summer months (Schulze & McGee,
1978; Olivier, 1995; Desmet, 1996). Frosts are rarely,
if ever experienced over most of the area (Rutherford
RESULTS
& Westfall, 1986; Werger, 1986). In the Nama Karoo,
continentality is more pronounced: temperature
Southern African comparisons
extremes (including freezing conditions in winter)
prevail. The log–log species–area curves were significant for
both the Succulent Karoo (P<0.001) and the Nama•In the Succulent Karoo, high humidity (especially
at night and in the early morning) and frequent Karoo (P<0.05) (Fig. 2). Since the slopes of these two
curves were homogeneous (F=0.026; df=18; P=0.87),fog (at least in the Namaqualand-Namib Domain)
1998 Blackwell Science Ltd, Diversity and Distributions,4, 27–36
High plant diversity in the Karoo 31
Fig. 2.
Species-area relations for Succulent Karoo and Nama Karoo sites in southern Africa. Y=log spp. no. and X=log area.
Data from Nordenstam (1974), Gibbs Russell (1987), Hilton-Taylor (1994, 1996), Desmet (1996), Cowling et al. (1997) and N.
Ju
¨rgens (pers. comm.).
and the intercepts were significantly different (F=9.709; Succulent Karoo has nearly four times as many species
df=19; P=0.006), we computed their k- or intercept- as North American winter-rainfall semi-arid regions.
ratio. This value was 2.65, indicating that for any given Regional diversity in the Nama Karoo did not appear
region within the range of areas in Fig. 2, Succulent to be substantially higher than that in other summer
Karoo has more than 2.5 times the number of species rainfall, semi-arid areas (Fig. 3b), although more data
as the Nama Karoo. would enable more meaningful comparisons. Certainly,
Australian regions were considerably poorer than those
from southern Africa, but the Sonoran Desert (the
Global comparisons
only North American site) had similar richness to an
When compared to other winter fainfall, semi-arid equivalent area of the Nama Karoo.
zones of the world, the Succulent Karoo emerged as Species richness in desert regions, defined here as
having the highest regional diversity over the full range both summer and winter rainfall areas with an annual
of areas (Fig. 3a). There were sufficient and appropriate rainfall of less than 100 mm, showed a non-significant
data for comparative analyses of the species–area relationship with area (Fig. 3c). Nonetheless, desert
relations of the Succulent Karoo and North American regions in southern Africa generally had more species
winter-rainfall semi-arid regions. The regression when compared to similar-sized desert regions in North
statistics were as follows: Africa, including the topographically complex uplands
of the central Sahara. For example, the Namib Desert
Succulent Karoo: has between four and 200 times as many species as
Y=2.350+0.221X;r
2
=87.7%; P<0.001; n=9
equivalent-sized areas of the Saharan Desert. The desert
North America: zone of the Namaqualand-Namib domain of the
Y=1.770+0.301X;r
2
=75.3%; P<0.001; n=21, Succulent Karoo is especially rich: in the mountain
desert region of South Africa’s Richtersveld, 331 species
Where Y=log species number and X=log area. The
have been recorded in an area of 1.3 km
2
! (Von Willert
overall slopes of these curves were not significantly
et al., 1992).
different (F=2.506; df=26; P=0.13) but the intercepts
Interestingly, the trend for winter-rainfall arid lands
were (F=31.439; df=27; P<0.0001). The resultant
to be richer than those from summer rainfall regions
intercept-ratio was 3.79, indicating that within the
range of areas for the two data sets (Fig. 3a), the holds at a global scale. A comparison of the two
1998 Blackwell Science Ltd, Diversity and Distributions,4, 27–36
32 R. M. Cowling et al.
indicating that winter-rainfall regions were more than
one-and-a-half times richer than equivalent sized
summer-rainfall regions. The analysis was constrained
by the lack of data from summer-rainfall areas outside
southern Africa.
DISCUSSION
Southern African comparisons
When compared with the Nama Karoo, the Succulent
Karoo has higher levels of local endemism (Cowling
& Hilton-Taylor, in press), higher local diversity
(Cowling et al., 1989; 1994), and higher differentiation
(beta and gamma) diversity (Cowling et al., 1989;
Cowling & Hilton-Taylor, in press). Therefore, it is not
surprising that considerably more species are packed
into Succulent Karoo landscapes. What explains these
differences?
The answer may lie in the unique climatic regime of
the Succulent Karoo. Mild, oceanic and fog-
ameliorated conditions, combined with predictable
(albeit low) winter showers have selected for relatively
short-lived (3–10 yr) shrub lifestyles. The overwhelming
majority of these species are leaf succulents (largely
Mesembryanthema) with limited water storage capacity
and shallow root systems (Von Willert et al., 1992).
This relatively benign environment enables regular
germination and seedling establishment of perennials,
thereby obviating the need for great longevity. The
predictability of the moisture regime (Desmet &
Fig. 3.
Species-area relations for warm-temperate to
Cowling, in press) also enables succulents to maintain
subtropical semi-arid shrublands and deserts of the world: (a)
very small water storage capacity, leading to the
winter-rainfall, semi-arid; (b) summer rainfall, semi-arid
widespread development of minutism (Midgley & van
(100–400 mmyr
−1
); and (c) desert (<100 mmyr
−1
) regions. Y=
der Heyden, in press). The occasional years of lower-
log spp. no. and X=log area. Data for southern Africa (SAf)
than-average rainfall cause widespread mortality
from Robinson (1978), Nordenstam (1974), Gibbs Russell
(1987), Cowling & Hilton-Taylor (1994), Hilton-Taylor (1994;
(Ju
¨rgens et al., 1997; R.M. Cowling, unpublished data)
1996), Desmet (1996) and Cowling et al. (1997). Data for
of these drought-intolerant plants (Von Willert et al.,
Australia (Aus) from Williams & Calaby (1985) and J.S. Beard
1985), and ensure that there is usually sufficient space
(pers. comm.). Data for North America (NAm) from Raven &
for seedling recruitment. The result is rapid, at least
Axelrod (1978), Mulroy, Rundel & Bowler (1979), Rundel &
for an arid land, population turnover and weakly
Gibson (1996) and P.W. Rundel (unpublished data). Data for
structured, species-rich communities (Cowling et al.,
South America (SAm) from Armesto & Vidiella (1993). Data
for the Mediterranean Basin (MeB) from Le Houe
`rou & Boulos
1994). But what is the link between regional diversity
1991). Data for the Middle East (ME) from Danin (1983). Data
and these local-scale processes?
for North Africa (NAf) from Le Houe
`rou (1986, 1992).
A possible explanation invokes differential rates of
diversification. Lineages with the traits described above
(the majority of Succulent Karoo species), have highspecies–area curves (Fig. 3a and b) revealed that the
slopes were not significantly different (F=0.276; df=numbers of sexually produced generations—attributes
that would invariably lead to high rates of genetic48; P=0.60), but the intercepts were (F=3.061; df=
49; P=0.09). The corresponding k-ratio was 1.78, recombination and rapid speciation (Rosenzweig,
1998 Blackwell Science Ltd, Diversity and Distributions,4, 27–36
High plant diversity in the Karoo 33
1995). Moreover, for the Mesembryanthema at least, 1995). These factors would militate against rapid
diversification and habitat specialization. Even in thelimited gene flow (in terms of seed and pollen dispersal)
(Ihlenfeldt, 1994), peculiar genetic controls on environmentally heterogeneous escarpment zones, beta
diversity is very low (Palmer & Cowling, 1994).morphology (Ihlenfeldt, 1994), and weakly persistent
seed banks (Esler, in press), would enhance
diversification. The rapid accumulation of taxa would
Global comparisons
result in a very fine-scale discrimination of habitats,
such as Ihlenfeldt (1994) describes for Argyroderma Here we ask the question: is regional diversity and
composition of growth forms in the Succulent Karooand Hammer (1993) for Conophytum (both
Mesembryanthema) in the Namaqualand-Namib unique when compared to other winter rainfall arid
lands of the world? The following conclusions can beDomain. The result is very species-rich landscapes with
numerous highly localized habitat specialists, often made.
capable of existing in small and isolated populations •In the Succulent Karoo, total diversity and the
(Hammer, 1993; Ihlenfeldt, 1994; Cowling & Hilton- diversity of leaf succulents shrubs at the regional
Taylor, in press). Clearly, landscapes with greater scale, is without parallel. The only other part of the
heterogeneity will support more species (Cowling et world with an even moderate concentration of non-
al., 1997), but the capacity for fine-scale habitat halophytic leaf succulents is the Canary Islands
discrimination of explosively diversifying lineages lends (Ju
¨rgens, 1986). Interestingly, these species are
supports to Rosenzweig’s (1992, 1995) argument that concentrated in foggy, coastal environments which
habitat diversity is a co-evolved property of species receive low but predictable rainfall (Shmida &
diversity. Werger, 1992).
Predictable winter rain of the Succulent Karoo would •The explosive diversification of certain Succulent
also favour seasonally active geophytes, dependent as Karoo genera, resulting in large species flocks, and
they are on regular moisture inputs to sustain storage the high level of generic endemism (Hilton-Taylor,
organs and promote flowering. Recent research 1996), is unique among the world’s arid lands.
suggests that diversification of Namaqualand-Namib
Domain geophytes has been promoted by pollinator- However, it is of interest that globally, winter-rainfall
semi-arid regions are richer than those that receiveflower coevolution, e.g. Iridaceae and Geraniaceae with
long-tongued flies, (Nemestrinidae) (Goldblatt, summer rainfall. Cowling et al. (1992) showed that
regions in the strongly winter-rainfall part of the fynbosManning & Bernhardt, 1995; Manning & Goldblatt,
1996), and Iridaceae, Liliaceae (sensu lato) and biome have more than double the number of species
than those that receive a substantial amount of summerOrchidaceae with monkey beetles (Scarabaeidae:
Hoplinii) (Picker & Midgley, 1996). Goldblatt & rain. A similar pattern exists for the heath (kwongan)
areas of southern Australia (Fox, 1994). In semi-aridManning (1996) suggest that edaphic specialization
provided the initial impetus for genetic differentiation areas, winter rainfall is more predictable and effective
than summer rainfall (Noy-Meir, 1985; Hoffman &between founder and parent populations of the
predominantly Succulent Karoo irid genus, Cowling, 1987). These conditions may favour the
persistence and diversification plants with short-livedLapeirousia. Specialization to different pollinators
(long-tongued flies) is invariably associated with lifestyles—leaf succulents in the Succulent Karoo, and
annuals elsewhere (Shmida & Whittaker, 1979;parapatric species pairs in this genus.
In the Nama Karoo, community structure, diversity Armesto & Vidiella, 1993; Rundel & Gibson, 1996).
The low diversity and low local endemism of annualspatterns and rates of diversification are not unusual
for an arid land (Cowling & Hilton-Taylor, in press). in the Succulent Karoo (Desmet, 1996; Hilton-Taylor,
1996) remains a puzzling enigma.Shrubs are mainly long-lived and well-dispersed and
the short-lived component is comprised almost entirely
of widespread, wind-pollinated, and wind- or animal-
Diversification in an arid world: is the
dispersed grasses with persistent seed banks (Hoffman
Succulent Karoo unique?
& Cowling, 1987; Esler, in press). Shrub community
structure is relatively stable and competition from this There are two extra-tropical areas that are recognized
as extraordinary centres of plant diversity andlong-lived component (Yeaton & Esler, 1990) limits
recruitment (Milton, 1995; Wiegand, Milton & Wissel, endemism: the mediterranean-climate regions of South
1998 Blackwell Science Ltd, Diversity and Distributions,4, 27–36
34 R. M. Cowling et al.
Africa and south-western Australia (Cowling et al.,
REFERENCES
1996). Clearly, the Succulent Karoo is a third such
Armesto, J.J. & Vidiella, P.E. (1993) Plant life-forms and
region. Interestingly, the patterns and determinants
biogeographic relations in the flora of Lagunillas (30°S)
of diversity in the Succulent Karoo bear a striking
in the fog-free Pacific coastal desert. Ann. Mo. Bot.
resemblance to those in the mediterranean-climate
Gdns. 80, 499–511.
regions, even though the predominant vegetation
Axelrod, D.I. (1972) Edaphic aridity as a factor in
(fynbos in South Africa and kwongan in Australia) is
angiosperm evolution. Am. Nat. 106, 311–320.
a dense, sclerophyllous and fire-prone shrubland. Both
Cowling, R.M., Esler, K.J., Midgley, G.F. & Honig, M.A.
(1994) Plant functional diversity, species diversity and
regions have very species-rich landscapes with high
climate in arid and semi-arid southern Africa. J. Arid
differentiation diversity and large numbers of habitat-
Environ. 27, 141–58.
specific, local endemics associated with a limited
Cowling, R.M., Gibbs Russell, G.E., Hoffman, M.T. &
number of genera (Cowling et al., 1996). Diversification
Hilton-Taylor, C. (1989) Patterns of plant species
is largely associated with fire-sensitive lineages (short
diversity in southern Africa. Biotic diversity in southern
generation times) that have limited gene flow. Thus,
Africa. Concepts and conservation (ed. by B.J. Huntley),
fire plays a key role in promoting diversification and
pp. 19–50. Oxford University Press, Cape Town.
Cowling, R.M. & Hilton-Taylor, C. (1994) Patterns of
the fine-scale partitioning of habitat space. We suggest
plant diversity and endemism in southern Africa: an
that occasional droughts that interrupt the otherwise
overview. Strelitzia,1, 31–52.
remarkably predictable rainfall regimes of the
Cowling, R.M. & Hilton-Taylor, C. (in press) Plant
Succulent Karoo (Esler, Rundel & Cowling, in press),
biogeography, endemism and diversity. The Karoo:
are analogous to fire in these mediterranean-climate
ecological patterns and processes (ed. by W.R.J. Dean and
shrublands in that they also promote generation
S.J. Milton). Cambridge University Pess, Cambridge.
turnover and population fragmentation.
Cowling, R.M. & Holmes, P.M. (1992) Flora and
vegetation. The ecology of fynbos. Nutrients, fire and
There is no evidence that the massive diversification
diversity (ed. by R.M. Cowling), pp. 23–61. Oxford
of the Succulent Karoo flora is the result of an unusually
University Press, Cape Town.
long and stable history. Like mediterranean shrublands,
Cowling, R.M., Holmes, P.M. & Rebelo, A.G. (1992) Plant
karroid vegetation became a widespread formation in
diversity and endemism. The ecology of fynbos. Nutrients,
south-western Africa only fairly recently, probably in
fire and diversity (ed. by R.M. Cowling), pp. 62–112.
the Pliocene (Deacon, Hendey & Lamprechts, 1983;
Oxford University Press, Cape Town.
Scott, Anderson & Anderson, 1997). As was the case
Cowling, R.M., Richardson, D.M., Schulze, R.E.,
Hoffman, M.T., Midgley, J.J. & Hilton-Taylor, C. (1997)
globally, arid-adapted lineages occupied azonally dry
Species diversity at the regional scale. Vegetation of
sites since the late Cretaceous, and especially after the
southern Africa (ed. by R.M. Cowling, D.M. Richardson
Oligocene (Axelrod, 1972; Smida, 1985), and many
and S.M. Pierce), pp. 447–473. Cambridge University
have persisted as paleoendemics (e.g. Welwitschia
Press, Cambridge.
mirabilis). However, the explosive speciation of the
Cowling, R.M., Rundel, P.W., Lamont, B.B., Arroyo,
Mesembryanthema in southern Africa (2000 species,
M.K. & Arianoutsou, M. (1966) Plant diversity in
116 genera), largely concentrated in the Succulent
mediterranean-climate regions. Trends Ecol. Evol. 11,
362–366.
Karoo, is a relatively recent phenomenon, and is
Danin, A. (1983) Desert vegetation of Israel and Sinai.
probably unrivalled in angiosperms (Ihlenfeldt, 1994).
Cana Publishing House, Jerusalem.
Deacon, H.J., Hendey, Q.B. & Lamprechts, J.J.N. (eds)
(1983) Fynbos palaeoecology: a preliminary synthesis.
South African National Scientific Programmes Report
75. CSIR, Pretoria.
ACKNOWLEDGMENTS
Desmet, P.G. (1996) Vegetation and restoration potential of
the arid coastal belt between Port Nolloth and Alexander
Bay, Namaqualand, South Africa. MSc thesis, University
Aspects of the research for this paper were funded by
of Cape Town.
the Foundation for Research Development (Pretoria)
Desmet, P.G. & Cowling, R.M. (in press) The climate of the
and the Institute for Plant Conservation. We thank
Karoo—a functional approach. The Karoo: ecological
Steve Hammer, Craig Hilton-Taylor, Timm Hoffman,
patterns and processes (ed by W.R.J. Dean and S.J.
Norbert Ju
¨rgens, Sue Milton, Shirley Pierce, Mike
Milton). Cambridge University Press, Cambridge.
Rosenzweig, Ernst van Jaarsveld, Jan Vlok and
Esler, K.J. (in press) Plant reproductive ecology. The
Karoo: ecological patterns and processes (ed. by W.R.J.
Graham Williamson for valuable discussion.
1998 Blackwell Science Ltd, Diversity and Distributions,4, 27–36
High plant diversity in the Karoo 35
Dean and S.J. Milton). Cambridge University Press, Ihlenfeldt, H-D. (1994) Diversification in an arid world:
the Mesembryanthemaceae. Ann. Rev. Ecol. Syst. 25,Cambridge.
Esler, K.J., Rundel, P.W., and Cowling, R.M. (in press) 521–546.
Ju
¨rgens, N. (1986) Untersuchungen zur okologieThe succulent karoo in a global context: plant structural
and functional comparison with North American winter- sukkulenter pflanzen des sudlichen Afrika. Mitt. Inst.
Allgem. Bot., Hamb. 21, 139–365.rainfall deserts. The Karoo: ecological patterns and
processes. (ed. by W.R.J. Dean and S.J. Milton). Ju
¨rgens, N. (1991) A new approach to the Namib Region.
I: Phytogeographic subdivision. Vegetatio,97, 21–38.Cambridge University Press, Cambridge.
Fox, M.D. (1994) Australian mediterranean vegetation: Ju
¨rgens, N., Gunster, A., Seely, M.K. & Jacobsen, K.M.
(1997) Desert. Vegetation of southern Africa. (ed. byintra- and intercontinental comparisons. Ecology and
biogeography of mediterranean ecosystems in Chile, R.M. Cowling, D.M. Richardson and S.M. Pierce), pp.
189–215 Cambridge University Press, Cambridge.California and Astralia (ed. by M.T.K. Arroyo, M.D.
Fox and P.H. Zedler), pp. 137–159. Springer-Verlag, Le Houe
`rou, H.J. (1986) The desert and arid zones of
North Africa. Hot deserts and arid shrublands. Vol. B.Heidelberg.
Gibbs Russell, G.E. (1987) Preliminary floristic analysis (ed. by M. Evenari, I. Noy-Meir and D.W. Goodall),
pp. 101–148. Elsevier, Amsterdam.of the major biomes in southern Africa. Bothalia,17,
213–227. Le Houe
`rou, H.N. (1992) Outline of the biological history
of the Sahara. J. Arid Environ. 22, 3–30.Goldblatt, P., Manning, J.C. & Bernhardt, P. (1995)
Pollination biology of Lapeirousia subgenus Lapeirousia Le Houe
`rou, H.N. & Boulos, L. (1991) Bioclimatic and
phytogeographic characteristics of the arid rangelands(Iridaceae) in southern Africa; floral divergence and
adaptation for long-tongue fly pollination. Ann. Mo. of northern Africa and the Near East. Proceedings of
the IVth International Rangeland Congress, Montpellier,Bot. Gdns. 82, 517–534.
Goldblatt, P. & Manning, J.C. (1996) Phylogeny and France, pp. 536–543.
Manning, J.C. & Goldblatt, P. (1996) The Prosecaspeciation in Lapeirousia subgenus Lapeirousia
(Iridaceae: Izioideae). Ann. Mo. Bot. Gdns. 83, 346–361. peringuyei (Diptera: Nemenstrinidae) pollination guild
in southern Africa: long-tongued flies and their tubularGould, S.J. (1979) An allometric interpretation of
species–area curves: the meaning of the coefficient. Am. flowers. Ann. Mo. Bot. Gdns. 83, 67–86.
Midgley, G. & van der Heyden, F. (in press) Form andNat. 114 , 335–43.
Hammer, S.A. (1993) The genus Conophytum. Succulent function in perennial plants. The Karoo: ecological
patterns and processes (ed. by W.R.J. Dean and S.J.Plant Publications, Pretoria.
Hartmann, H. (1991) Mesembryanthema. Contrib. Bol. Milton). Cambridge University Press, Cambridge.
Milton, S.J. (1995) Spatial and temporal patterns in theHerb. 13, 75–157.
Hilton-Taylor, C. (1987) Phytogeography and origins of emergence and survival of seedlings in arid Karoo
shrubland. J. appl. Ecol. 32, 145–156.the Karoo flora. The karoo biome: a preliminary synthesis.
Part 2. Vegetatin and history (ed. by R.M. Cowling Mulroy, T.W., Rundel, P.W. & Bowler, P.A. (1979) The
vascular flora of Punta Banda, Baja California Norte,and P.W. Roux), pp. 70–95. South African National
Programmes Report 142. CSIR, Pretoria. Mexico. Madrono,26, 69–90.
Nordenstam, B. (1974) The flora of the Brandberg.Hilton-Taylor, C. (1994) The Kaokoveld. Namibia and
Angola. Centres of plant diversity. A guide and strategy Dinteria,11, 3–67.
Noy-Meir, I. (1985) Desert ecosystem structure andfor their conservation. Vol. 1: Europe, Africa, South West
Asia and the Middle East. (ed. by S.D. Davis, V.H. function. Hot deserts and arid shrublands. Vol. A (ed. by
M. Evenari, I. Noy-Meir and D.W. Goodall), pp. 93–104Heywood and A.C. Hamilton), pp. 201–203. IUCN
Publications Unit, Cambridge. Elsevier, Amsterdam.
Olivier, J. (1995) Spatial distribution of fog in the Namib.Hilton-Taylor, C. (1996) Patterns and characteristics of
the flora of the Succulent Karoo Biome, southern Africa. J. Arid Environ. 29, 129–138.
Palmer, A.R. & Cowling, R.M. (1994) An investigation ofThe biodiversity of African plants. (ed. by L.J.E. van der
Maesen, X.M. van der Burght and J.M. van Medenbach topo-moisture gradients in the eastern Karoo, South
Africa, and the identification of factors responsible forde Rooy), pp. 58–72. Kluwer, Dordrecht.
Hilton-Taylor, C. & Le Roux, A. (1989) Conservation species turnover. J. Arid Environ. 26, 135–147.
Picker, M.D. & Midgley, J.J. (1996) Pollination by monkeystatus of the fynbos and karoo biomes. Biotic diversity
in southern Africa. Concepts and conservation (ed. by beetles (Coleoptera: Scarabaeidae: Hopliini): flower and
colour preferences. Afr. Entomol. 4, 7–14.B.J. Huntley), pp. 202–223. Oxford University Press,
Cape Town. Raven, P.H. & Axelrod, D.I. (1978) Origin and
relationships of the California flora. Univ. Calif. Publ.Hoffman, M.T. & Cowling, R.M. (1987) Plant
physiognomy, phenology and demography. The karoo Bot. 72, 1–134.
Robinson, E.R. (1978) Phytogeography ot the Namibboime: a preliminary synthesis. Part 2. Vegetation and
history (ed. by R.M. Cowling and P.W. Roux), pp. 1–34. Desert of South West Africa (Namibia) and its
significance to discussions of the age and uniqueness ofSouth African National Programmes Report,142. CSIR,
Pretoria. this desert. Palaeoecology of Africa,10, 67–74.
1998 Blackwell Science Ltd, Diversity and Distributions,4, 27–36
36 R. M. Cowling et al.
Rosenzweig, M.L. (1992) Species diversity gradients: we acid metabolism in succulents of the Richtersveld
(Namib desert, South Africa). Planta,164, 45–55.know more and less than we thought. J. Mammology,
73, 715–30. Von Willert, D.J., Werger, M.J.A., Brinckmann, E.,
Ihlenfeldt, H-D. & Eller, B.M. (1992) Life strategies ofRosenzweig, M.L. (1995) Species diversity in space and
time. Cambridge University Press, Cambridge. succulents in deserts: with special reference to the Namib
Desert. Cambridge University Press, Cambridge.Rundel, P.W. & Gibson, A.C. (1996) Ecological
communities and processes in a Mohave Desert ecosystem: Walter, H. (1986) The Namib Desert. Hot deserts and arid
shrublands. Vol. B. (ed. by M. Evenari, I. Noy-Meir andRock Valley, Nevada. Cambridge University Press,
Cambridge. D.W. Goodall), pp. 245–282. Elsevier, Amsterdam.
Werger, M.J.A. (1978a) The Karoo-Namib Region.Rutherford, M.C. & Westfall, R.H. (1986) Biomes of
southern Africa—an objecive categorization. Mem. Bot. Biogeography and ecology of southern Afrida (ed. by
M.J.A. Werger), pp. 231–299. Junk, The Hague.Surv. S. Afr. 54, 1–98.
Schulze, R.E. & McGee, O.S. (1978) Climatic indices Werger, M.J.A. (1978b) Biogeographical divisions of
southern Africa. Biogeography and ecology of southernand classifications in relation to the biogeography of
southern Africa. Biogeography and ecology of southern Africa (ed. by M.J.A. Werger) pp. 145–170. Junk, The
Hague.Africa (ed. by M.J.A. Werger), pp. 11–52. Junk, The
Hague. Werger, M.J.A. (1986) The Karoo and Southern Kalahari.
Hot deserts and arid shrublands. Vol. B (ed. by M.Scott, L., Anderson, H.M. & Anderson, J.M. (1997)
Vegetation history. Vegetation of southern Africa. (ed. Evenari, I. Noy-Meir and D.W. Goodall), pp. 283–360.
Elsevier, Amsterdam.by R.M. Cowling, D.M. Richardson and S.M. Pierce),
pp. 62–84. Cambridge University Press, Cambridge. White, F. (1976) The vegetation map of Africa—the history
of a completed project. Boissiera,24, 659–666.Shmida, A. (1985) Biogeography of the desert flora. Hot
deserts and arid shrublands. Vol. A (ed. by M. Evenari, White, F. (1983) The vegetation of Africa. UNESCO, Paris.
Wiegand, T., Milton, S.J. & Wissel, C. (1995) A simulationI. Noy-Meir and D.W. Goodall), pp. 23–78. Elsevier,
Amsterdam. model for a shrub ecosystem in the semiarid Karoo,
South Africa. Ecology,76, 2005–2221.Shmida, A. & Whittaker, R.H. (1979) Convergent
evolution of deserts in the Old and New Worlds. Werden Williams, O.B. & Calaby, J.H. (1985) The hot deserts of
Australia. Hot deserts and arid shrublands. Vol. A (ed.und Vergehen von Pflanzengesellschaften (ed. by O.
Wilmanns & R. Tuxen), pp. 437–450. J. Kramer, Vaduz. by M. Evenari, I. Noy-Meir and D.W. Goodall), pp.
269–312. Elsevier, Amsterdam.Shmida, A. & Werger, M.J.A. (1992) Growth form diversity
on the Canary Islands. Vegetatio,102, 183–199. Williamson, M. (1988) Relationship of species number to
area, distance and other variables. AnalyticalSmith, G.F., Hobson, S.R., Meyer, N.L., Chesselet, P.,
Archer, R.H., Burgoyne, P.M., Glen, H.F., Herman, biogeography. An integrated approach to the study of
animal and plant distributions (ed. by A.A. Myers andP.P.J., Retief, E., Smithies, S.J., van Jaarsveld, E.J.
& Welman, W.G. (1993) Southern African succulent P.S. Giller), pp. 91–115. Chapman & Hall, London.
Yeaton, R.I. & Esler, K.J. (1990) The dynamics of aplants—an updated synopsis. Aloe,30, 32–74.
Van Jaarsveld, E. (1987) The succulent riches of South succulent karoo vegetation, a studyofspeciesassociation
and recruitment. Vegetatio,88, 103–113.Africa and Namibia. Aloe,24, 45–92.
Von Willert, D.J., Brinckmann, E., Schleiter, B. & Eller, Zar, J.H. (1984) Biostatistical analysis. Prentice-Hall, New
York.M.B. (1985) Availability of water controls crassulacean
1998 Blackwell Science Ltd, Diversity and Distributions,4, 27–36
