ENDANGERED SPECIES RESEARCH
Endang Species Res
Vol. 8: 15–23, 2009
doi: 10.3354/esr00157
Printed July 2009
Published online February 27, 2009
INTRODUCTION
South-East Asia includes 4 biodiversity hotspots,
each containing exceptional animal and plant species
diversity (Myers et al. 2000). Alarmingly, it is estimated
that 42% of these species, many of which are endemic
to those regions, could be extinct by 2100 (Sodhi et al.
2004, Sodhi & Brook 2006). This ‘worst case scenario’
has already started in Singapore, where 881 species
have been lost (28% of the 3196 species covering a
wide range of terrestrial and freshwater taxa), mainly
due to anthropogenic changes, including deforestation,
habitat degradation and overexploitation of plant and
animal populations (Brook et al. 2003). A group of par-
ticular concern is the order Chiroptera, as it has been
estimated that 67% of recently extant bat species have
become extinct in Singapore (Lane et al. 2006). Small
geographic ranges are considered the principal factor
influencing extinction risks in bats (Jones et al. 2003).
Therefore, species with a limited distribution are partic-
ularly vulnerable. With 121 bat species currently re-
corded in Thailand (Bumrungsri et al. 2006, Thong et
© Inter-Research 2009 · www.int-res.com*Corresponding authors. Email addresses:
s.puechmaille@gmail.com; emma.teeling@ucd.ie
Population size, distribution, threats and
conservation status of two endangered bat
species Craseonycteris thonglongyai and
Hipposideros turpis
Sébastien J. Puechmaille1,*, Pipat Soisook2, Medhi Yokubol2, Piyathip Piyapan2,
Meriadeg Ar Gouilh3, 4, Khin Mie Mie5, Khin Khin Kyaw5, Iain Mackie6,
Sara Bumrungsri2, Ariya Dejtaradol2, Tin Nwe5, Si Si Hla Bu7, Chutamas Satasook2,
Paul J. Bates8, Emma C. Teeling1,*
1School of Biology and Environmental Science, University College Dublin, Belfield, Dublin 4, Ireland
2Faculty of Science, Prince of Songkla University, Hat-Yai, 90112 Songkla, Thailand
3Muséum National d’Histoire Naturelle, rue Cuvier, 75005 Paris, France
4Faculty of Science, Mahidol University, Nakhon Pathom, 73170 Bangkok, Thailand
5Zoology Department, Yangon University, 11041 Kamayut, Yangon, Myanmar (Tin Nwe, retired)
6Department of Biology, University of Aberdeen, IBES, Aberdeen, UK
7Department of Zoology, Mandalay University, 05071 Mandalay, Myanmar
8Harrison Institute, Bowerwood House, St Botolph’s Road, Sevenoaks, Kent, UK
ABSTRACT: The population size, distribution, threats and conservation status of 2 bat species
included as ‘endangered’ in the IUCN Red List were assessed in Thailand and Myanmar. Craseonyc-
teris thonglongyai Hill, 1974 and Hipposideros turpis Bangs, 1901 are both rare and poorly known
cave-dwelling species, which have restricted distributions and small population sizes. Surveys of the
2 species were conducted in a large number of locations including previously reported sites and
potential new locations. Colony size was counted at currently known localities, and population size
was estimated. The known distribution and population are larger than in previous estimates, but both
species still remain threatened by roost disturbance and the destruction of foraging habitat. Clarifi-
cation of the taxonomic status of allopatric populations within each species is urgently required.
KEY WORDS: Craseonycteris thonglongyai · Hipposideros turpis · Chiroptera · Endangered species ·
Distribution · Population size · Thailand · Myanmar
Resale or republication not permitted without written consent of the publisher
OPEN
ACCESS
Contribution to the Theme Section ‘Bats: status, threats and conservation successes’

Endang Species Res 8: 15–23, 2009
al. 2006, Soisook et al. 2007), only 2 are classified as ‘en-
dangered’ by the IUCN: Craseonycteris thonglongyai
Hill, 1974 and Hipposideros turpis Bangs, 1901 (Chi-
roptera Specialist Group 1996), and both have a limited
distribution and small population size.
Craseonycteris thonglongyai (Fig. 1) is the only
known representative of the monospecific family,
Craseonycteridae, which was discovered in 1973 by
Kitti Thonglongya near Sai Yok, Kanchanaburi
Province, Thailand (Hill 1974). At that time, about 50
caves were surveyed, but C. thonglongyai was only
found in 2 of them (Duangkhae 1991). In 1980, the
Royal Forestry Department declared 500 km2 of the
area around the Sai Yok Waterfall as a national park.
In the early 1980s, cave surveys were carried out in
Kanchanaburi, but, unfortunately, the data concerning
C. thonglongyai are very difficult to access. In some
cases, the information has not been published; in oth-
ers, it has been published as reports, including a num-
ber that were classified as confidential (see references
in Duangkhae 1991). In 1991, Duangkhae reported the
presence of C. thonglongyai in 21 out of 51 caves sur-
veyed in Kanchanaburi and estimated the population
size to be about 2000 (Duangkhae 1991). Since there
were many unexplored caves at that time, Duangkhae
(1991) suggested that the population size of C. thong-
longyai could be as high as 300 000. Since this publica-
tion, no other study has been carried out relating to the
population size and distribution of C. thonglongyai in
Thailand. However, the discovery of a new population
of about 1500 ind. in 9 localities in the states of Mon
and Kayin, Myanmar, revealed that the C. thong-
longyai distribution was larger than previously
thought (Bates et al. 2001, Ramos Pereira et al. 2006).
Hipposideros turpis (Fig. 2) was described from Ishi-
gaki Island, Japan, and is currently known to be found
on 4 Yaeyama (Shakishima) Islands, Japan (Abe 2005),
in northern Vietnam and in peninsular Thailand (Sim-
mons 2005). Two subspecies have been described; the
nominate form H. t. turpis restricted to Japan, and H. t.
pendleburyi Chasen, 1936 from Vietnam and Thailand
(Corbet & Hill 1992, Francis 2008). The global distribu-
tion of this species is quite disjunct, and its taxonomic
status remains unclear (Borissenko & Kruskop 2003,
Francis 2008). In Thailand, its distribution was hitherto
known from Khao Ram (i.e. Khao Ram Rome in Khao
Luang National Park) in Nakhon Si Thammarat
Province, where the taxon H. t. pendleburyi was de-
scribed (Chasen 1936), and from Klong Wildlife Sanc-
tuary (TISTR 1995) and other sites in Krabi Province
(Lekagul & McNeely 1977).
We report on the current distribution, population
size and conservation status of Craseonycteris thong-
longyai in Thailand and Myanmar and Hipposideros
turpis in Thailand.
MATERIALS AND METHODS
Craseonycteris thonglongyai. Distribution: A series
of field studies were carried out between 1997 and
2008, in which 181 caves were checked for the pres-
ence of C. thonglongyai (Fig. 3). In Thailand, cave sur-
veys were focused mainly on Kanchanaburi Province
16
Fig. 1. Craseonycteris thonglongyai. Specimen from Myanmar
Fig. 2. Hipposideros turpis pendleburyi. Specimen from
peninsular Thailand

Puechmaille et al.: Endangered bat species in SE Asia
(112 localities), where the species was first discovered
(Hill 1974). Caves were also surveyed in other
provinces around the central plain (49 localities).
These provinces were chosen because they included
areas of limestone karst and climatic conditions (mean
annual temperature from 26 to 28°C and rainfall from
800 to 2000 mm) that were ‘comparable’ to those in
Kanchanaburi, where the species had previously been
described. In Myanmar, limestone outcrops around
Mawlamyine (Moulmein) in the state of Mon and Hpa-
an in the state of Kayin were surveyed (20 localities)
(Fig. 3). A more extensive survey outside the known
distribution of the species was not possible for reasons
of personal security.
The species was reported to be present if it was cap-
tured, or visually and acoustically identified inside
the cave (Hill 1974, Surlykke et al. 1993). Acoustic
identification of Craseonycteris thonglongyai was
possible by distinguishing species-specific echolo-
cation calls (Surlykke et al. 1993) via a heterodyne bat
detector tuned at 75 kHz (Pettersson D230) or via
time-expanded recordings obtained with ultra-sound
bat detectors (D-240X, D-1000X; Pettersson Elektronik
AB).
Population size: Colony size was estimated by 3
different methods. Most estimates were made by
counting emerging individuals from cave entrances
at dusk. This counting was feasible because of the
early emergence of the species just after sunset and
its typical flight pattern. When no emergence count
was possible, C. thonglongyai were counted inside
the cave or from photographs of the roosting colony.
When >1 estimate was available for the same cave,
only the most recent data were included in the
analyses.
Counting bats, as outlined above, provides an esti-
mate of colony size for the surveyed caves but not for
the species as a whole, as many caves containing
roosts are still unknown. Indeed, western and southern
Thailand have the largest limestone area in the coun-
try, and there are many caves still to be surveyed
(Sidisunthorn et al. 2006). However, by estimating the
density of caves with roosts, the area of the species dis-
tribution and the average number of individuals per
cave, it is possible to estimate the total population
size of the species. As the precision of these
estimates is dependent on the amount of data available
(see ‘Results’), we could only obtain a reliable popula-
tion size estimate for Craseonycteris thonglongyai in
Thailand.
To estimate the number of caves with Craseonyc-
teris thonglongyai yet to be discovered in Kan-
chanaburi, we measured the area of the Sai Yok for-
mation (Permian limestone) in this province. To be
conservative, we excluded the limestone area around
Srinakarin Dam, where no C. thonglongyai were
found (see ‘Results’). Geological maps of Thailand
1:250 000 (first edition in 1976, Accepted Reference
Nos. ND47-6, ND47-11 and ND47-7) from the Geo-
logical Survey Division of the Department of Mineral
Resources were digitised and areas were measured
using tpsDig software V.2.10 (Rohlf 2004). The den-
sity of caves with C. thonglongyai was estimated
from 2 reference areas (see area location in ‘Results’;
see Fig. 5) that have been well surveyed by our team
(present study) and Duangkhae (1991). Next, the
expected number of caves with C. thonglongyai was
estimated by multiplying the total area by the cave
density. The population size was then estimated by
multiplying the estimated number of caves with
roosts by the average colony size (the median and
average colony size did not markedly differ).
Hipposideros turpis. Distribution: Between 2006
and 2008, a series of expeditions was conducted
throughout Thailand by staff and students of Prince of
Songkla University, Hat Yai. A total of 50 localities
were surveyed in 10 provinces of peninsular Thailand,
where H. turpis was first discovered. Another 67 local-
ities were surveyed in 17 provinces between peninsu-
lar Thailand and Vietnam, where the species is also
reported (Fig. 4).
Population size: Only direct counts within caves and
photographic counts were undertaken for Hip-
posideros turpis.
17
Fig. 3. Craseonycteris thonglongyai. Provinces and states
surveyed for the presence of C. thonglongyai (dark grey
areas) in Thailand and Myanmar

Endang Species Res 8: 15–23, 2009
RESULTS
Craseonycteris thonglongyai
Distribution
In total, 181 caves were surveyed, 20 in 2 Myanmar
states and 161 in 10 provinces of Thailand. Craseo-
nycteris thonglongyai was found in 49 caves, 44 in
Thailand and 5 in Myanmar. Within its actual known
distribution in Thailand, the species was present in
50% of the surveyed caves and in 25% in Myanmar. In
Thailand, it was only found in Kanchanaburi Province
from Kanchanaburi City in the southeastern part of the
province, along the Kwae Noi River, up to Vajira-
longkorn Dam (formerly known as Khao Laem Dam) in
the northwestern part (Fig. 5). Current studies in
Myanmar did not extend its known distribution, but
did confirm its presence in 2 caves. Caves were gener-
ally surrounded by bamboo/ deciduous forest, temples,
plantations (mainly Cassava; Manihot esculenta) in
Thailand and rice fields in Myanmar. Exact geographic
locations for the caves can be obtained upon request
from the authors (S.J.P. & E.C.T.) or the Prince of
Songkla University Natural History Museum.
Population size
The census population size was 6487 in Thailand and
3770 in Myanmar. In Thailand, the average number of
bats in each cave differed between counting methods.
Averages of 73 and 74 bats were counted per cave via
direct (n = 15) or photographic (n = 3) count, respec-
tively. The average number of bats counted at emer-
gence (n = 20) was 258. For 6 caves where Craseonyc-
teris thonglongyai were only heard via bat detectors,
no estimates were available.
The 2 reference areas (110 and 80 km2; Fig. 5) con-
tained 14 and 11 caves, respectively, sheltering
18
Fig. 4. Hipposideros turpis. Locations of the provinces sur-
veyed for the presence of H. turpis (dark grey areas)
throughout Thailand
Fig. 5. Craseonycteris thonglongyai. Known distribution (dark
grey areas) in Thailand and Myanmar. In Thailand, the 2
‘reference areas’ (see ‘Materials and methods’) are mapped
as rectangles

Puechmaille et al.: Endangered bat species in SE Asia
Craseonycteris thonglongyai, which corresponds to
an average of 1 cave every 7.6 km2. The total surface
of limestone (Sai Yok formation) within the known
distribution of C. thonglongyai is 1312 km2. The
number of caves sheltering the species is estimated
to be ca. 173 (1312/7.6) caves. As emergence counts
were the most reliable estimate of colony size (see
‘Discussion’), we only used these data to estimate the
total population size in Thailand (ca. 45 000 [173 ×
258] ind.).
Hipposideros turpis
Distribution
Of the 117 localities visited, the species was found at
only 8 localities in 3 provinces of peninsular Thailand:
Krabi, Trang and Nakhon Si Thammarat (Fig. 6). Six
were in 3 districts of Krabi Province, with only 1 local-
ity in each of the other 2 provinces. All localities were
in limestone outcrops, mainly surrounded by rubber
plantations (Hevea brasiliensis), oil palm plantations
(Elaeis guineensis), temples and villages (Table 1).
Population size
Currently, the known population size of Hipposi-
deros turpis in peninsular Thailand slightly exceeds
1400 ind. (Table 1). About 60% was found in Krabi
Province, where the species was observed in 4 caves
and captured at 2 sites in the forest using harp-traps.
The other 40% was found in Nakhon Si Thammarat
and Trang Provinces, with 350 ind. at the Khao Daeng
cave and 190 ind. at the Rad cave. Females were preg-
nant in late March at Thanbok Koranee National Park
and lactating in early May at the Khao Daeng cave, but
we only counted adults.
DISCUSSION
Craseonycteris thonglongyai
Distribution
For nearly 20 yr, Craseonycteris thonglongyai was
thought to be restricted to a few caves in and around
the Sai Yok National Park (Duangkhae 1991). In 2001,
the discovery of a population in Myanmar (Bates et al.
19
Sites No. of Surrounding habitat Threats
individuals
Khao Kanab Nam, Krabi 4 Mangrove forest –
Khao Phanombenja NP, Krabi 350 Oil palm plantation –
Sang Phet cave, Krabi 100 Rubber plantation, monastery, villages Habitat degradation
Khao Rang, Krabi 40 Oil palm plantation Habitat degradation, roost
disturbance
Khao Phra cave, Krabi 350 Oil palm plantation, temple Habitat degradation
Nature trail (Thanbok Koranee NP), Krabi 18 Evergreen forest –
Rad cave, Trang 190 Rubber plantation Habitat degradation
Khao Daeng cave, Nakhon Si Thammarat 350 Rubber plantation, temple, villages Habitat degradation, roost
disturbance
Table 1. Hipposideros turpis. Number of individuals, surrounding habitats and threats to known sites of H. turpis in peninsular
Thailand. NP: National Park
Fig. 6. Hipposideros turpis. Localisation of the sites (d) where
the species has been found in peninsular Thailand. In Krabi
Province, the northernmost dot represents 3 sites in Thanbok
Koranee National Park

Endang Species Res 8: 15–23, 2009
2001) suggested that its very limited known distribu-
tion might also, in part, be a reflection of a lack of
detailed survey work.
In the present study, we surveyed 161 caves in 10
different provinces of Thailand, but were unable to
find Craseonycteris thonglongyai outside the Kancha-
naburi Province. It is very difficult to prove the absence
of a species in a particular location, but our surveys
show that, if the species is present outside of Kan-
chanaburi, it is far less abundant. The species is pre-
sent in over half of surveyed caves around the Kwae
Noi River valley, Kanchanaburi. In addition, other
researchers have surveyed many caves throughout
Thailand (Bumrungsri et al. 2006, S. Bumrungsri et al.
unpubl. data), and none have reported the presence of
the species outside Kanchanaburi, suggesting that its
presence there is unlikely. For example, 18 caves were
surveyed in Ratchaburi Province, 30 to 50 km south of
the nearest known locality of C. thonglongyai, but,
despite the geographic proximity, the species was
not found.
Duangkhae (1991) reported 21 caves with Craseo-
nycteris thonglongyai, 13 of which were not visited
during our survey. Therefore, the total number of caves
where the species has been found in Thailand is 57. In
Kanchanaburi Province, the actual distribution of C.
thonglongyai is about 3 times larger in area (145 km
long and 10 to 15 km wide) than the assumed distribu-
tion based on previous published data (Duangkhae
1991). The species appears to be absent from the area
surrounding the Srinakarin Dam, where >15 caves
were surveyed. This absence is surprising as this lime-
stone area is only a few kilometres from the main lime-
stone block. However, the Sai Yok formation is divided
into 2 categories: ‘limestone, massive, with chert nod-
ules and fusulinids’, where C. thonglongyai was found,
and ‘limestone, gray to light gray, bedded, locally reef,
with chert nodules; and sandstone’, where the species
was not found (see Geological Map ND47-6). These
geological differences might explain the absence of C.
thonglongyai. The same is to be reported around Vaji-
ralongkorn Dam in the north. C. thonglongyai is pre-
sent 2 km south of the dam in the Sai Yok formation,
but was not found by Duangkhae (1991), who sur-
veyed caves from a different geological period (Juras-
sic-Triassic vs. Permian) just a few kilometres north of
the dam. Caves in different rock categories could have
a different microclimate and/or configuration due to
the influence of rock properties on cave formation
(Palmer 2007). A further study on climatic variation
between caves from different rock categories within
the distribution range of this species is recommended.
The influences of climate and other environmental or
biotic factors on the presence or absence of C. thong-
longyai need to be further investigated.
Population size
Different methods were used to count Craseonyc-
teris thonglongyai, and as the counting method was
not chosen according to the colony size (which was not
known a priori), the average number of bats per cave
should theoretically be comparable for the different
methods. It is clear that emergence counts were higher
than photographic and direct counts. This difference
could either reflect an overestimate of the emergence
count method or an underestimate of the photographic
and direct count methods or both. For some caves,
emergence counts were carried out independently by
2 different people at the same time and furnished sim-
ilar results (data not shown). Additionally, colony size
counted when bats were returning to the cave in the
early morning furnished similar results to emergence
counts carried out in the previous or following evening
(data not shown). For these reasons, emergence counts
are believed to furnish reliable estimates of C. thong-
longyai colony size. Since C. thonglongyai often roosts
in small chambers, small crevices, or small holes that
are difficult to access, photographic and direct counts
usually underestimate colony size. As an example, in 4
out of the 5 caves sheltering C. thonglongyai in Myan-
mar, not a single individual was seen or recorded
inside the cave during the day, but they were observed
flying out at emergence time, emphasizing the validity
of emergence counts compared to direct or photo-
graphic counts. As a result, photographic and direct
counts can be regarded as minimum estimates in a
rapid assessment of bat populations, whereas emer-
gence counts can be considered the most reliable
colony size estimate.
Considering only emergence counts, 80% of the
colonies in Thailand included between 100 and 400
ind. (n = 20, range = 11 to 856). In Myanmar, emer-
gence counts at 4 caves furnished estimates of 40, 600,
1100 and 2000 ind. Even if the number of caves sur-
veyed is low, it seems that colony size is much larger in
Myanmar than in Thailand.
The known number of Craseonycteris thonglongyai,
6487 in Thailand and 3770 in Myanmar, probably rep-
resents only a fraction of the actual species population
size, as it is likely that many caves containing roosts
have not been surveyed. Considering the density of
known caves with the species present (1 every 7.6
km2), the average number of individuals per cave (258)
and the area of the limestone block where the species
is present (1312 km2), the total population size of the
species in Kanchanaburi, Thailand, can be estimated
at around 45 000 ind. However, this figure probably
underestimates the true population size for 2 reasons:
(1) because the cave density is underestimated, since
there are difficulties in finding caves owing to the
20

Puechmaille et al.: Endangered bat species in SE Asia
scarp topography and dense vegetation; and, (2) only
the area of limestone in the Sai Yok formation was con-
sidered, whereas some caves with C. thonglongyai
have been found in zones described as Quaternary
deposits. In these latter areas, the deposit layer is thin
(see the geological cross section on Map ND47-7) and
limestone pockets with caves are present, but are not
represented on the map due to their small size. Thus,
the estimate is conservative and should be regarded as
a minimum population size.
The same estimates could be calculated for Myan-
mar, but we would need more data regarding the
Craseonycteris thonglongyai distribution, the density
of caves with roosts and the average colony size. Con-
sidering the large area of Permian limestone in the
states of Mon and Kayin (1684 km2 with 137 outcrops;
calculated after Earth Sciences Research Division
1977) and the relatively large average colony size (935
ind. cave–1), it would not be surprising if the Myanmar
population was in excess of 100 000 ind. However,
more surveys are essential to obtain a reliable estimate
for the Myanmar population.
Taxonomic status
The taxonomic status of the 2 populations in Thai-
land and Myanmar has yet to be resolved as the large
difference in peak frequency of the echolocation calls
suggests that each population may have been isolated
for some time (Surlykke et al. 1993, Ramos Pereira et
al. 2006). A thorough study of the genetic diversity,
divergence and isolation of these populations is critical
to establish the taxonomic status of each population.
Threats, conservation status and recommendations
Human disturbance by tourists and pilgrims is a con-
siderable threat to Craseonycteris thonglongyai popu-
lations in Thailand (Yokubol 2000) and may become a
problem in Myanmar if the caves begin to attract more
people. Studying the roosting selection of the species is
required as it would enable us to predict suitable roost-
ing sites for C. thonglongyai. This would serve to
establish conservation priorities in order to inform cave
management guidelines in relation to human
access/disturbance. In Thailand, another major threat
is habitat destruction through deforestation and forest
burning, which reduces suitable foraging habitats for
the species (Yokubol 2000). A detailed study on forag-
ing habitat quality is immediately needed to investi-
gate what factors determine a good foraging site for
the species. Another concern is the killing of individu-
als through hunting as a source of food (Yokubol 2000).
In Myanmar, the construction of a dam on the Salween
River and the development of a limestone mining
industry could rapidly become major threats to C.
thonglongyai (Ramos Pereira et al. 2006, Ngyeh 2007).
The status of Craseonycteris thonglongyai was last as-
sessed as ‘endangered’ (EN B1 + 2c, C2b; Ver. 2.3) by the
IUCN in 1996. This is now considered to be out of date
(Chiroptera Specialist Group 1996). Since the 1996 as-
sessment, a new population has been found in Myanmar
(Bates et al. 2001), and the population size and distribu-
tion has been extended in Thailand (present study).
These new data will provide important additional infor-
mation for a revision of its status. However, there are
some aspects that still require further investigation.
Habitat loss/degradation through deforestation near
caves directly affects Craseonycteris thonglongyai by
reducing its foraging sites (Yokubol 2000), but the
long-term impact of this habitat change in terms of
population fragmentation is not known and would
deserve particular attention.
Despite our efforts, the distribution and population
size of Craseonycteris thonglongyai in Myanmar are
still poorly known. Cave surveys need to be carried out
more extensively in the states of Mon and Kayin, espe-
cially in the area of the dam construction where C.
thonglongyai could be present.
For Thailand and Myanmar, the population trend of
Craseonycteris thonglongyai is difficult to assess at the
moment because of the limited data available and the
difficulty in comparing estimates carried out using dif-
ferent methods or at different periods of the year
(Yokubol 2000, Yokubol et al. 2006, S. Puechmaille
unpubl. data). Monitoring schemes need to be
launched, possibly including emergence counts and/or
walking or driving transects similar to the iBat monitor-
ing program (see www.ibats.org.uk/).
Hipposideros turpis
Distribution and population size
The species distribution in Thailand is quite small
and restricted to 3 provinces, Krabi, Nakhon Si Tham-
marat and Trang, where its presence is restricted to
limestone outcrops. Within this distribution, it appears
to be restricted to 4 main areas, which are located
between 30 and 130 km from each other. This suggests
that either the species is very rare or many caves with
roosts are still unknown. The actual known population
size is quite small, so each cave and its surroundings
are important. This is especially the case in Khao
Phanombenja National Park and in Khao Phra and
Khao Daeng caves, which are the roosts for three-
fourths of the known population.
21

Endang Species Res 8: 15–23, 2009
Taxonomic status
Hipposideros turpis pendleburyi is very similar to H.
armiger, but differs in its smaller size and the presence
of only 3 supplementary leaflets in the nose-leaf
(Chasen 1936, Lekagul & McNeely 1977) (see Fig. 2).
The 2 species were found sympatrically in Khao Phra
cave, Krabi. The specimens of H. turpis from peninsu-
lar Thailand, referred to H. t. pendleburyi, are consid-
erably larger than the nominate subspecies (Lekagul &
McNeely 1977). Borissenko & Kruskop (2003) reported
that the Vietnamese form seems to correspond with the
description of H. t. pendleburyi from peninsular Thai-
land. Specimens from Vietnam referred to H. turpis
were initially described as a subspecies of H. larvatus,
namely H. l. alongensis Bourret, 1942. Based on cranial
measurements only, H. l. alongensis was reclassified as
H. turpis alongensis (Topál 1993), although externally,
individuals look like H. larvatus. Photographs of H.
turpis from Vietnam (Hendrichsen et al. 2001, Hutson
et al. 2001, S. Puechmaille pers. obs.) clearly resemble
H. larvatus and not H. turpis. According to the average
forearm length (71.7 mm) (Hendrichsen et al. 2001),
individuals from Vietnam are smaller than those in
peninsular Thailand (mean = 77.8 mm, 75.2 to 80.8
mm, SD = 1.7, n = 11). Moreover, their echolocation call
frequency is much lower at around 72 kHz (n = 1; S.
Puechmaille unpubl. data) compared to 85.9 kHz (83.8
to 87.2 kHz, SD = 1.2, n = 23) in peninsular Thailand (P.
Soisook unpubl. data). The systematic status of H.
turpis and the different subspecies remains very
unclear in view of the unusual distribution (Hutson et
al. 2001), the differences in nose-leaf morphology, size
and echolocation call frequency. Further taxonomic
studies of the disjunct populations of H. turpis are a
priority. Echolocation and molecular data may prove
that H. turpis populations from Japan, Vietnam and
Thailand are not conspecific, and the taxa H. t. pendle-
buryi and H. t. alongensis may prove to be specifically
distinct and endemic to peninsular Thailand and Viet-
nam, respectively.
Threats, conservation status and recommendations
Most of the known localities for the species are out-
side protected areas, and here their foraging habitats
have been severely degraded. There has been a recent
and rapid conversion of land to agricultural use. In the
Thanbok Koranee National Park, where the largest
population of Hipposideros turpis is known to exist,
only the limestone outcrops are included in the park
area. The surrounding habitats have been converted to
oil palm and rubber plantations and to urban areas.
Therefore, the habitat around the roosting sites of
these populations is now fragmented. The future of the
population at Khao Daeng cave, another large colony,
is uncertain, as the Khao Daeng outcrop is being con-
sidered for limestone mining and quarrying. H. turpis
should be monitored by directly counting the number
of individuals in every cave where the species is
known to roost.
The status of Hipposideros turpis was last assessed
as ‘endangered’ (EN A2c; Ver. 2.3) by the IUCN in
1996 (Chiroptera Specialist Group 1996). This listing is
now considered out of date. When the species was pre-
viously assessed, there was no information on its popu-
lation status or the threats to it in Thailand and Viet-
nam (Hutson et al. 2001). The data presented here will
help to re-evaluate the status of the species for the
population present in peninsular Thailand.
CONCLUSIONS
Craseonycteris thonglongyai and Hipposideros
turpis are 2 cave-dwelling species with limited distrib-
utions in limestone areas. Their conservation requires
the protection of their roosts and foraging habitats. The
main threats for caves are quarrying, which would
destroy roosts, and human disturbance. Foraging habi-
tats are threatened by fragmentation and destruction.
For both species, the taxonomic status of each popu-
lation should be clarified: Craseonycteris thong-
longyai, with populations in Thailand and Myanmar,
and Hipposideros turpis, with populations in Japan,
Vietnam and Thailand. This should be done in connec-
tion with more cave surveys, allowing more precise
estimations of distribution and population size for each
species.
Acknowledgements. We are grateful to employees at Sai Yok
and Erawan National Parks, Thailand, for assisting us during
the Craseonycteris survey and at Thanbok Koranee National
Park for helping us during the survey of the Andaman World
Heritage Project. Thanks to Tuanjit Srithongchuay, Bounsa-
vane Douangboubpha, Phansamai Phommexay and all staff
and students in the PSU Bat Research Unit for their help in the
field. Thanks to the Biodiversity Research and Training Pro-
gram, Thailand, and Bat Conservation International for their
financial support. We also thank the Wildlife Fund Thailand
for their support during our project. Thanks to the IRD UR
178, Mahidol University, for their logistic support during field-
work in Thailand. Thanks to Nu Nu Aye, Wai Wai Myint, Aye
Thida and Thida Tin, who greatly helped in the field in Myan-
mar. We are indebted to Vu Dinh Thong and Neil Furey, who
co-ordinated the surveys in Cat Ba and Ba Be National Parks,
Vietnam. We thank Sumiko Matsumura (Japan) for the loan of
specimens of Hipposideros turpis. We thank the staff at the
World Bat library (Geneva, Switzerland) for kindly providing
us with references. Thanks to Fabio Bontadina and 3 anony-
mous reviewers for comments on the manuscript. We thank
the Darwin Initiative (DEFRA, UK) The Royal Society (Inter-
22