A Study of Plant Species

Extinction in Singapore: Lessons

for the Conservation of
Tropical Biodiversity
I.M. TURNER
H. T. W. TAN
Y.C. WEE
Department of Botany
National University of Singapore
Lower Kent Ridge Road
0511 Singapore

ALI BIN IBRAHIM

P.T. CHEW
National Parks Board
Singapore Botanic Gardens
Cluny Road
1025 Singapore

R.T. CORLETT
Department of Ecology and Biodiversity
University of Hong Kong
Pokfulam Road
Hong Kong

Abstract: The native vascular p l a n t f l o r a o f the Republic o f Esmdio de la extinci6n de especies vegetales en Singapur:
Singapore has suffered the extfnctton o f 594 o u t o f a total Lecciones para la conservaci6n de la biodiversidad tropical
2277 specte~ These represent loca~ not globa~ species ex-
tinction.z Coastal habitat~ including m a n g r o t ~ have lost
39% of their specte~ while inland forests have last 29%. R e s u m e n : La flora de plantas vasculares nattvas de la Re-
Epiphytic species (62% loss) appear particularly p r o n e to p ~ l i c a de Stngapur ha sufrldo la extict6n de 594 especies de
extinctiorg which is reflected in a similar disposition exhib- un total de 2277. Estas representan exticiones de espectes a
ited by the Orchldacea¢ Deforestation and disturbance have ntvel local antes que global Hdbitats costero~ incluyendo
been the m a i n cause o f p l a n t species extinction in Singapore manglare~ ban perdido 39% de sus especies mientras que
The rich mangrove epiphyte f l o r a has been totally extermi- forestas de tierra adentro ban perdldo 29%. Especies epO~tas
nate~ a n d a n u m b e r o f tree species are reduced to popula- (62% de pddldas) parecen ser particularmente suceptibles a
tions o f a f e w mature indivldual~ M a n y more species con- la extinci6r~ 1o que es refl~ado en una disposici6n similar
tinue to survive than the species-area relationship would exhibida p o r Orchidicea~ La deforestaci6n y perturbaci6n ha
predict given the 99.896 loss o f p r i m a r y forest This is inter- sido la principal causa de la extinci6n de especies vegetales
en Stngapur. La rlca f l o r a de eptfitas de los manglares ha
sldo completamente exterminada y un nt~mero de especies
Paper submitted March 8, 1993; revised manuscript accepted August de drboles estdn reducldas a polaciones de unos pocos indi.
10, 1993 triduos maduro~ Dado el 99.896 de p&rdlda de la foresta

705

ConservationBioiosy, Pages 705-712

Volume 8, No. 3, September 1994
706 Plant Species ~ainction in S/ngapore Turner eta/.

preted as a result o f the failure o f equilibrium to be achieved primariag continuan sobreviviendo muchas mds especies
y e t in the remnant forest fragment~ even after more than a que las que sedan predictas p o r la relaci6n especies-arect Se
century o f isolatior~ Singapore's secondary forests appear to interpreta que esto es debido a que todavia no se ha alcan.
accrete p l a n t diversity very slowly, even i f contiguous with zado el equilibrio en los fragmentos remanentes de foresta
primary forest area.~ We conclude that remnant fragmenis o f tropical primari• atin luego de rods de un siglo de ais-
primary tropical foresg even o f very small size, can play a lamiento. A pesar de ser contiguas a dreas de forestas pri-
major role in the conservation o f tropical biodiversity. The maria~ las forestas secundarias de Singapur parecen acre.
patterns o f extinction observed in Singapore indicate that centar m u y lentamente la diversidad de planta~ Nosotros
coastal and estuarine sites are in greatest demand f o r devel- concluimos que a pesar de su pequef~o tamafzo, los fragmen-
opment and therefore must be given high priority f o r con- tos remanentes de foresta troptcal prlmariag pueden j u g a r un
servation despite their somewhat lower biodiversity. Epi- papel tmportante en la conservaci6n de la biodtversidad
phyte and orchid diversity appear to be very good indicators tropical Los patrones de extinci6n observados en Singapur
o f the degree o f disturbance suffered by a habitat in the indican que los sitios costeros y estuariales son los rods de-
h u m i d tropic~ mandados p o r el desarrollo y p o r consiguiente deben ser
considerados como primera prioridad para la conservaci6n
a pesar de que su diversidad es un tanto meno~. La diver-
sidad de eplfitas y/o de orquideas parece ser un buen indi-
cador del grado de perturbaciOn sufrido p o r un hdbitat en
los tr6picos ht~medos

Introduction land to a very densely populated, urban landscape in less

than 200 years has had a massive affect on the biota of
Biological diversity reaches its zenith in the natural Singapore (Corlett 1992a). We are fortunate, from a
communities of the humid tropics. H u m a n interference scientific point of view, that it is possible with s o m e
with tropical ecosystems almost inevitably reduces their confidence to say which species have b e e n lost. This is
species richness and is recognized as the major cause of because Singapore has a long history of biological col-
loss of global biodiversity (Reid & Miller 1989; W-nit- lection, which for plants has b e e n thorough and contin-
m o r e & Sayer 1992). While m u c h has b e e n written uous since the 1880s. Therefore, w e have b e e n able to
about the biodiversity crisis, detailed studies of authen- analyze the extinctions that have o c c u r r e d in the highly
ticated alterations in biodiversity due to h u m a n activity diverse and species-rich native flora and to search for
in the tropics remain scarce (Simon 1986; H e y w o o d & ecological and taxonomic patterns. In hindsight it is pos-
Stuart 1992). Those available are generally for mammals sible to see ways in which s o m e of the loss of biodiver-
or birds and are m o r e likely to be c o n c e r n e d with re- sity might have b e e n avoided. These retrospective les-
m o t e islands than regions on the major continents. If w e sons are lost on Singapore but could b e of value for
are to slow or halt the local losses of biodiversity, a conservation strategies in areas of the species-rich trop-
better understanding of the biology of local species ex- ics that have not yet suffered from such intense h u m a n
tinction is required. In particular, conservationists need disturbance.
to k n o w which communities and taxonomic/ecological
groups within communities are m o s t susceptible to ex-
Singapore
tinction in order to prioritize the d e p l o y m e n t of limited
resources. Lowland tropical rain forests are the most The Republic of Singapore (103 ° 50' E, 1° 20' N) con-
diverse plant communities on the planet ( G e n t r y 1988), sists of a group of m o r e than 50 islands, with the largest,
and it is probable that high plant diversity engenders Singapore, making up m o r e than 90% of the total land
high total biodiversity. Estimates of arthropod biodiver- area. The original area of Singapore Island was about
sity for tropical forests have b e e n made by extrapolating 544 km 2, to which another 30 km 2 have b e e n added by
from tree diversity (Erwin 1982; Stork 1988). There- land reclamation. The island lies less than 140 km from
fore, if w e understand the effects of h u m a n interference the equator and is separated from the Malay Peninsula
on botanical diversity, w e may also, within certain lim- by straits only 600 meters wide at their narrowest (Cor-
its, justifiably claim c o m p r e h e n s i o n of the total diversity lett 1992b). The highest point is the summit of Buldt
as well. Timah at 162 meters above sea level. The land is low-
Singapore is a small state ( n o w about 620 km 2) with lying, with a mean elevation of 15.1 m e t e r s ( T h o m a s
nearly three million residents. The c o v e r of primary for- 1991 ), gently rolling, and almost devoid of steep slopes.
est has d r o p p e d f r o m virtually 100% in 1819 at the The soils are relatively uniform, mostly consisting of
founding of m o d e r n Singapore to 0.2% today (Corlett infertile sandy-clay loams (Corlett 1992b).
1992a). This change from an almost uninhabitated is- Singapore's climate is a tropical diurnal one with high

ConscrvsUon Biology
Volume 8, No. 3, September 1994
Turner et al. Plato Species ~xtinaion in Sit~apore 707

rainfall and constantly warm temperatures. The mean 39% ), the forests 478 species (478/1674 = 29% ), and
daily m a x i m u m t e m p e r a t u r e remains around 30 ° C the open sites 19 species (19/352 = 5% ). The extinc-
throughout the year (Chia & Foong 1991), and most tions by life form are as follows: 159 tree species (159/
areas of the Republic receive more than 2000 mm of 848 = 19% ), 77 shrub species (77/224 = 34% ), 100
precipitation per annum (Chia & Foong 1991 ). All sites herb species (100/591 = 17% ), 73 species of climber
have monthly averages consistently over 100 ram. The (73/317 = 23% ), and 185 epiphytic species (185/297
rainfall is highest in December, but there is no dry sea- = 62% ). Epiphytes include stranglers and climbers
son. with adpressed adventitious roots (mostly aroids).
If not for human activity, Singapore would be almost Taxonomically, the largest family of the flora, the Or-
entirely covered with forest. The former forest was of chidaceae, seems to have been the heaviest sufferer. 173
three main types: primary lowland tropical rain forest, ( 8 8 % ) of the 196 native orchid species appear to have
f r e s h w a t e r s w a m p forest, and m a n g r o v e . C o r l e t t been exterminated. All 34 coastal orchid species have
( 1 9 9 1 a ) has estimated that these would have occupied disappeared, including the 32 epiphytic species that
82%, 5%, and 13% of the land area respectively (Fig. 1). were found on old trees in mangroves and at the mouths
At present, less than 5% of Singapore (see Fig. 1) re- of rivers. Nearly as dramatic, only 10 of the 110 inland
mains under natural or seminatural vegetation (Corlett epiphytic orchids can be found today, representing a
1992a), of which hardly 200 ha could be defined as 91% extinction rate. The loss of 39 of 52 terrestrial
primary (Corlett 1992b). orchid species ( 7 5 % ) is less extreme but still most
alarming. The other major group of epiphytes, the pteri-
dophytes, shows a better survival: 23 out of 62 epiphytic
Methods species ( 3 7 % ) have b e e n lost. Overall, the pterido-
phytes have suffered extinction at about the same rate as
As part of the compilation of a Red Data book for the
the whole flora, losing 49 out of 174 species (28%).
Singapore biota, w e have designated conservation status
Extinction has been patchy across other major taxa.
to each of the Republic's native vascular plant species.
Families with high losses include the Rubiaceae (56/140
The classification followed a slightly modified version of
:- 40%), the Meliaceae (14/36 = 39%), and the Myr-
the standard of the International Union for the Conser-
taceae (14/54 = 26% ). Other speciose families in the
vation of Nature. The categorization was based upon our
native flora have escaped more lightly: none of the 100
combined extensive knowledge of the flora and refer-
species of Gramineae and only two of the 91 Cyper-
ence to taxonomic works, particularly Ridley ( 1 9 0 0 )
aceae have suffered extinction. The Annonaceae (6/64
and Keng (1990). The list of 2277 native species is de-
= 9% ), the Moraceae (8/60 = 13% ), and the Euphor-
rived from the vascular plant checklist for Singapore
biaceae (18/116 = 16%) have undergone greater re-
(Turner et al. 1990), with some modifications. The re-
ductions in their diversity.
vised list was used by Turner ( 1 9 9 4 ) for a statistical
analysis of the taxonomy and ecology of the native flora,
which involved assigning each species to life-form and
Discussion
habitat classes. Three habitats w e r e recognized. These
w e r e coastal ecosystems, such as sandy shores, rocky
There are limitations to this study set by our under-
shores, mangroves, and river mouths; inland forests (un-
fortunately insufficient information is available to differ- standing of the past and present flora of Singapore. Prob-
entiate the swamp from the dry-land forest floras); and ably at least 90% of Singapore's forests w e r e cleared
those species of open habitats including weeds, grass- before serious plant collecting began in the 1880s (Cor-
land species, and waterside plants. Species w e r e desig- lett 1992a). Therefore it is probable that some species
nated coastal only if they are confined to that habitat in became extinct without being collected. It does not
Singapore. seem likely, however, that the n u m b e r of species in-
volved was very large. The species list w e are using is, in
comparison with other tropical regions, as long as if not
Results longer than areas of comparable size ( T u r n e r 1994).
Despite the extent of forest now representing under
Among the native flora, a total of 218 species were des- 5% of the land area, u n r e c o r d e d native species continue
ignated common, 957 rare, 391 vulnerable, 117 endan- to be found (Tan et al. 1992). We cannot claim our
gered, and 594 extinct. It is the extinctions (26% of the searches for the extinct species to have been exhaus-
native flora) that are the subjects of this paper. These tive, and some may still be extant within the Republic.
represent extinctions within the Republic of Singapore On the other hand, species previously recorded from
and are fortunately unlikely to represent global extinc- sites that remain physically intact, such as the Bukit Ti-
tions. mab Nature Reserve or Nee Soon s w a m p forest, have
The coastal habitats have lost 97 species (97/251 = usually been assumed to be still present even if w e have

ConservationBiology
Volume8, No. 5, Septembe~1994
 708 Rant Species b~inction in Singapore ~wner et al.

N1'20'

~IARYRA IN FOREST
.Lo~ ,
~ BB ~.GR°vE
.,= i~ " ~, ~FRESHWATER SWAMP
I ". "- • • IFOREST
j,, 0 5 10 krn
I I I

E103 =40' ~ E103 °50' E104 •

" m I

L
~"-Am,m
N1O20 '
!

::::::~CULTIYATED AND URBAN

AREAS
~ SECONDARV FOREST

6 • ~ (~'~ P PRIMARY RAIN FOREST

u ° .0~ 5 1Okra
F ~RES.WA,~RSWAMP
~" t t n FOREST
E103 o40' ~' E103 o50' E104 o
I z I

Figure 1. Maps showing the changes in the vegetation cover o f Singapore over the last 170 yearx The vegetation
o f Singapore in 1819 (A) is reconstructed f o l l o w i n g Corlett (1991a). The contemporary scene (B) is compiled
f r o m our o w n observation~

Conservation Biology
Volume 8, No. 3, Septembe~ 1994
Tamer et aft. PlantSpeciesEXtinctionin Singapore 709

not seen them. This may be unjustified but can only be lett 1992a). Many of the plants are long-lived, and
rectified by a detailed inventory of such sites, whiclv-- populations turn over so slowly that it takes decades
despite their small sizemis no easy undertaking~ It does (centuries?) for extinction finally to o c c u r in remnant
seem that some very clear results emerge that are not patches. High diversity remains only in areas that have
likely to be merely reflections of our o w n shortcomings never been cleared of forest cover, such as Buldt Timah
as field botanists. Nature Reserve, Nee Soon swamp forest, and tiny pri-
The magnitude of vascular plant extinctions ( 2 6 % ) mary patches around the reservoirs of the Central
appears similar to that reported for other taxonomic Catchment Area (Corlett 1990, 1992a¢ Ng & Lira 1992).
groups in Singapore. Corlett ( 1 9 9 2 a ) estimates a 28% The species poverty of the highly disturbed rural land-
loss of species for birds and one of 44% for freshwater scape of contemporary Singapore is emphasized by the
fish. results of a r e c e n t botanical survey of Pulau Ubin
A number of approaches have b e e n used to predict (Turner et al. 1992). The third largest island (about
the effects of deforestation o n tropical biodiversity 1000 ha) of the Republic was found to have only 261
(Lugo 1988; Reid & Miller 1989). The use of the spe- native plant species, probably only one-third the num-
cies-area relationship, an empirical rule of thumb relat- ber found at Buldt Timah Nature Reserve in more than
ing species n u m b e r to habitat area, has generally given 10 times the area.
some of the most conservative estimates of likely levels Extinction in disturbed landscapes will be a result of
of extinction. The Singapore data provide a good test of two main processes: the destruction of populations by
the utility of the relationship, which is summarized al- disturbance and the inviability of populations in small
gebraically as relict fragments. Thus far, Singapore appears mostly to
have suffered from the former, thoughmas mentioned
S = c A z,
above---our data may be biased against recording ex-
tinctions due to within-fragment processes.
w h e r e S is the n u m b e r of species occupying an area A; c
Land that has been cleared of forest cover at some
and z are constants. For the humid tropics a value of
time develops secondary vegetation, but this is gener-
0.26 appears valid for z (Simberloff 1992). Thus the
ally species-poor and, on more nutrient-rich sites, is fre-
effects of habitat depletion on species diversity can be
quently dominated by exotic species such as Acacia
predicted from the equation
auriculiformis Benth., Paraserianthesfalcatarta ( L ) I.
Sne~/Song~m = (AneJaortginat) °'26 Nielsen, and Spathodea campanulata P. Beauv. (Corlett
1988). Areas of secondary vegetation appear to accrete
The forest c o m p o n e n t of the Singapore flora provides new species very slowly, particularly if they are not con-
the best test. The 1674 forest species would have occu- tiguous with forest of a later successional stage (Corlett
pied a b o u t 87% o f Singapore's land area ( C o r l e t t 199ib; Sim et al. 1992). Lugo ( 1 9 8 8 ) p o i n t s out that in
1991a), which is some 49,600 ha. At a most generous Puerto Rico secondary forests are very important for the
estimate, primary forest currently occupies 200 ha. The conservation of indigenous w o o d y plants. It may be that
species-area relationship would predict a new species the relatively high f r e q u e n c y of hurricanes passing
complement of 24% of the original forest flora, as ex- through the Caribbean has selected for a forest commu-
tinction of 76%---clearly very different from our esti- nity far more resilient to disturbance than the lowland
mate of an actual 29% loss. Alternatively, knowing the dipterocarp forest of Singapore.
species loss r e c o r d e d w e can estimate the habitat area Deliberately or accidentally introduced exotic spe-
that must remain. This is calculated as 13,600 ha (a 73% cies of plants or animals have frequently b e e n impli-
loss of habitat area), which is far higher than the total cated in the extinction of endemic species, particularly
area of primary and secondary forest in Singapore cur- on remote islands (Vitousek 1988). In Singapore, how-
rently. There is clearly a large discrepancy between re- ever, exotics appear to have had a minimal impact on
ality and the prediction of the species-area relationship. the vegetation. The melastomataceous herb Clidemia
Some of this might be explained by pre-1880 extinc- hirta (L.) D.Don appears to be the only alien plant to
tions, but the original flora of Singapore would have had have penetrated into the interior of Bukit Timah Nature
to have been unbelievably diverse, with more species Reserve (Corlett 1988). The Botanic Gardens' Jungle, a
than Sri Lanka or Java. The main reason for the discrep- fragment of 4 ha, is in decline because of regeneration
ancy is that the species-area relationship predicts the being smothered by exotic lianas (Turner & Tan 1992),
equilibrium value for the diversity of the remaining frag- most notably Dtoscorea sansibarensis Pax, Tanacctum
ments. Equilibrium c a n n o t have b e e n r e a c h e d in jaroba Sw., and Thunbergia grandiflora (Rottl.) Roxb.
Singapore. This must indicate that plant communities of The huge numbers of alien propagules entering the sys-
the tropical rain forest take a long time to reach equi- tem may overwhelm the remnants of the native vegeta-
librium, because m u c h of the forest clearance responsi- tion (Simberloff 1992).
ble for the fragmentation took place last century (Cor- It is evident that three overlapping sets of species

Conservation Biology
Volume 8, No. 3, Septembe~ 1994
710 PlantSpeciesE~tincaonin Singapore Turner et al.

have been particularly p r o n e to extinction in Singapore. of some of the showier species, but it is clear that there
These consist of two ecological groups, coastal species is now nowhere for most of these epiphytes to survive
and epiphytes, and one taxonomic group, the Orchi- even if they had escaped being plundered by collectors.
daceae. Coastlines are prime sites for development. The situation is analogous to the extermination by hunt-
Ports, industrial complexes, and residential areas all ing of the indigenous large mammals. The collectors/
compete for sea frontage. Also, in overcrowded Singa- hunters simply delivered the coup de g#ace to species
pore reclamation of land from the sea adds valuable real d o o m e d to extinction because of habitat loss. A species
estate to the islands but usually changes diverse coast- for which collection probably has played a major role in
lines to monotonous concrete banks fronted by a thin extinction is Singapore's only cycad, Cycas rumphii
muddy or sandy beach. The barraging of river mouths Miq. Naturally occurring on rocky shores, it was much
for water storage (Corlett 1992a) floods the surround- collected as mature individuals for the horticultural
ing areas with fresh water~ permanently changing their trade.
ecology. Even w h e r e the coastline has escaped such re- While the coastal epiphytes have b e e n almost oblit-
structuring, the vegetation has usually been greatly dis- erated, the other maritime life forms are clearly also
turbed. Mangrove forests have been heavily exploited under threat. We know of a number of trees d o w n to
for a variety of resources, including timber, charcoal, their last few individuals, including Aegtceras corntcu-
and bark for tanning (Watson 1928), or they have been latum (L.) Blanco, Dolichandrone spathacea (L.f.) IC
cleared altogether to make way for prawn ponds (Cor- Schum., Sonneratia caseolaris (L.) Engl., and Sonnera-
lett 1992a). Watson ( 1 9 2 8 ) referred to the mangroves tia ovata Backer. This is particularly unfortunate be-
of Singapore as being subjected to " c h r o n i c over- cause Singapore lies within the region of the world's
exploitation." Such activity has led to the demise of all most diverse mangrove vegetation (Tomlinson 1986;
old-growth mangrove stands, even on the more remote Woodroffe & Grindrod 1991).
of the smaller islands, and has clearly resulted in the
obliteration of their epiphyte flora.
Inland, the epiphytes have also suffered a catastrophic Conclusions
loss in diversity, presumably associated with the reduc- Our conclusions can be summarized in a series of short
tion in the numbers of big old trees within the forest lessons to be learned from Singapore for the conserva-
that are the preferred hosts (Jacobs 1987). Epiphytes tion of biodiversity in the lowland humid tropics.
are perhaps the life form most susceptible to extinctions
through forest disturbance and, if naturally sparse, may (1) Forest clearance is the major cause o f local ex-
be rapidly reduced to very low population sizes in rem- tinctiort
nant fragments. This may also be exacerbated by the In Singapore only those areas that have kept a for-
changes in microclimate brought about by fragmenta- est cover remain species-rich. Extensive second-
tion (Saunders et al. 1991). The epiphytic pteridophytes ary areas may provide valuable habitat for certain
have been reduced in diversity ( d o w n 37% ) much less animal and plant species but have failed to in-
then the epiphytic orchids (93% loss). This may be due crease rapidly in plant diversity without direct in-
to the differing ecologies of these two groups. The or- tervention.
chids are mostly light-demanding c r o w n epiphytes, (2) Extinction o f p l a n t species through habitat frag-
whereas the pteridophytes are more generally shade- mentation takes m a n y years
tolerant species of the forest interior. The disappearance F r a g m e n t a t i o n r e d u c e s diversity, t h o u g h the
of the biggest trees from the landscape may affect the mechanisms of species loss remain unclear. In
photophytic epiphytes more adversely than the skio- Singapore, tropical rain forest fragments exhibit a
phytic species. Singapore has a relatively low comple- prolonged relaxation time to the levels of diversity
ment of epiphytic species compared to species-rich predicted by the species-area relationship. The en-
tropical areas with a similar climate (Turner 1994), couraging conclusion from this is that if small rem-
which may reflect the generally low soil fertility (Gen- nant fragments can be safeguarded they will pro-
try & Dodson 1987). This indicates that lowland rain vide refuge for many species for a considerable
forest fragmentation might have an even greater effect period without requiring complex and expensive
on total diversity in other tropical areas because they management. Time will be available to implement
would have a larger n u m b e r of susceptible epiphyte spe- strategies employing both in situ and e x situ tech-
cies. niques to improve the likelihood of the long-term
Orchids, the major casualty of this depletion of diver- survival of the species in question.
sity, must have also suffered because of collection of (3) The highest priority must be given to preserving
plants for the horticultural trade, or just from the idle respresentative areas o f all the major local hab-
curiosity of amateur gardeners in Singapore. Such col- itats in their primary conditiorL
lection was possibly responsible for the extermination Amelioration of disturbed sites may prove possi-

Conservation Biology
Volume 8, No. 3, September 1994
Turner et al. PlantSpecies Extinctionin Singapore 711

ble, but individuals for restocking an area have to Corlett, IZ T. 1992b. The Angiosperm Hora of Singapore 1.
c o m e from somewhere, preferably with local gen- Introduction. Gardens' Bulletin Singapore 44:3-21.
otypes. The experience in Singapore points to the
Erwin, T.L. 1982. Tropical forests: Their richness in Co-
importance of coastal habitats as a high priority for
leoptera and other arthropod species. Coleopterists' Bulletin
conservation. 36:74-75.
(4) Alien species do n o t always appear to have a
major negative i m p a c t on native h u m i d tropical Gentry, A. H. 1988. Changes in plant community diversity and
vegetatiorL fioristic composition on environmental and geographic gradi-
ents. Annals of the Missouri Botanical Garden 75:1-34.
The superior competitive abilities of naturalized
alien plants may hasten the decline of isolated frag- Gentry, A. H., and C. H. Dodson. 1987. Diversity and biogeog-
ments, but in Singapore they are generally unable raphy of Neotropical vascular epiphytes. Annals of the Mis-
to invade any but the tiniest patches. souri Botanical Garden 74:205-233.
(5) Secondary succession on disturbed sites appears
Heywood, V. H., and S. N. Smart. 1992. Species extinctions in
to accrete native p l a n t species extremely slowly. tropical forests. Pages 91-117 in T.C. Whitmore and J.A.
Areas already cleared of their natural vegetation Sayer, editors. Tropical deforestation and species extinction.
must be the preferred sites for any development in Chapman & Hall, London, England.
a humid tropical arem But w h e n the primary forest
in a region has been as reduced and fragmented as Jacobs, M. 1987. The tropical rain forest: A first encounter.
Springer-Verlag, Berlin, Germany.
it has in Singapore, it will be necessary to attempt
to rehabilitate secondary areas and reassemble the Keng, H. 1990. The concise flora of Singapore: Gymnosperms
lost biodiversity. This will undoubtedly be one of and dicotyledons. Singapore University Press, Singapore.
the major challenges facing conservation biolo-
gists in the next century. Lugo, A. E. 1988. Estimating reductions in the diversity of trop-
ical forest species. Pages 58--70 in E. O. Wilson, editor. Biodi-
(6) Orchids a n d epiphytes m a y be excellent indicator versity. National Academy Press, Washington, D.C.
groups o f habitat "health."
Epiphytes appear to be extremely sensitive to Ng, P. ILL., and IC K. P. Lira. 1992. The conservation status of
tropical forest disturbance and fragmentation. As the Nee Soon freshwater swamp forest of Singapore. Aquatic
such, they probably represent an excellent indica- Conservation: Marine and Freshwater Ecosystems 2:255-266.
tor group of biodiversity that can be monitored to Reid, W.V., and IC 17, Miller. 1989. Keeping options alive: The
assess the effects of forest disturbance. scientific basis for conserving biodiversity. World Resources
Institute, Washington, D.C.

Acknowledgments Ridley, H.N. 1900. The flora of Singapore. Journal of the

Straits' Branch of the Royal Asiatic Society 33:27-196.
We are extremely grateful to Jean Yong for all his hard
work in producing the figure. Saunders, D.A., R.J. Hobbs, and C.R. Margules. 1991. Biolog-
ical consequences of ecosystem fragmentation: A review. Con-
servation Biology 5:18-32.
Literature Cited
Sire, J. w. S., H. T. W. Tan, and I. M. Turner. 1992. Adinandra
Chia, L S., and S. F. Foong 1991. Climate and weather. Pages belukar: An anthropogenic heath forest in Singapore. Vegeta-
13-49 in L S. Chia, A. Rahman, and D. B. H. Tay, editors. The tio 102:125-137.
biophysical environment of Singapore. Singapore University
Press, Singapore.
Simberloff, D. 1992. Species-area relationships, fragmentation,
and extinction in tropical forests. Malayan Nature Journal
Corlett, 1ZT. 1988. The naturalized flora of Singapore. Journal 45:398-413.
of Biogeography 15:657--663.
Simon, J. L 1986. Disappearing species, deforestation and data.
Corlett, R.T. 1990. Flora and reproductive phenology of the New Scientist 110:60--63.
rain forest at Buldt Timah, Singapore. Journal of Tropical Ecol-
ogy 6:55--63. Stork, N.E. 1988. Insect diversity: Facts, fiction and specula-
tion. Biological Journal of the Linnean Society 35:321-327.
Corlett, 1~ T. 1991¢t VegetatiorL Pages 134-154 in L. S. Chia,
A. Rahman, and D. B. H. Tay, editors. The biophysical environ- Tan, H. T. W., Ali bin Ibrahim, K.S. Chum, I.M. Turner, Y.C.
ment of Singapore. Singapore University Press, Singapore. Wee, and P. T. Chew. 1992. Additions to the flora of Singapore,
I. Gardens' Bulletin, Singapore 44:127-133.
Corlett, IZ T. 1991b. Plant succession on degraded land in
Singapore. Journal of Tropical Forest Science 4:151-161. Thomas, G.S.P. 1991. Geology and geomorphology. Pages
50-88 in L S. Chia, A. Rahman, and D. B. H. Tay, editors. The
Corlett, R.T. 19926 The ecological transformation of Singa- biophysical environment of Singapore. Singapore University
pore, 1819-1990. Journal of Biogeography 19:411-420. Press, Singapore.

Conservation Biology
Volume 8, No. 3, September 1994
712 Plant Species Ex~inc~on in Singapore Turner et al.

Tomlin.con, P.B. 1986. The botany of mangroves. Cambridge Gardens' Bulletin, Singapore 44:51-71.
University Press, Cambridge, England.
Vitouselg P. M. 1988. Diversity and biological invasions of oce-
Turner, I. M. 1994. The taxonomy and ecology of the vascular anic islands. Pages 181-189 in E. O. Wilson, editor. Biodiver-
plant flora of Singapore: A statistical analysis. Botanical Journal sity. National Academy Press, Washington, D.C.
of the Linnean Society 114:215-227.
Watson, J.G. 1928. Mangrove forests of the Malay Peninsula.
Turner, I.M., and H.T.W. Tan. 1992. Ecological impact of Malayan Forest Records No. 6.
alien plant species in Singapore. Pacific Science 46:389-390.
Whitmore, T. C., and J. A. Sayer. 1992. Deforestation and spe-
Turner, I. M., K. S. Chua, and H. T. W. Tan. 1990. A checklist of cies extinction in tropical moist forests. Pages 1-14 in T. C.
the native and naturalized vascular plant species of the Repub- Whitmore and J. A. Sayer, editors. Tropical deforestation and
lic of Singapore. Journal of the Singapore National Academy of species extinction. Chapman & Hall, London, England.
Science 18 & 19:58-88.
Woodroffe, C. D., and J. Grindrod. 1991. Mangrove biogeogra-
Turner, I.M., H. T.W. Tan, I~ S. Chua, Haji Samsuri bin Haji phy: The role of quaternary environmental and sea-level
Abroad, and Y. C. Wee. 1992. A botanical survey of Pulau Ubin. change. Journal of Biogeography 18:479 A-92.

ConservationBiology
Volume 8, No. 3, September 1994