Tropical Natural History 12(2): 175-188 October 2012 2012 by Chulalongkorn University

Differences among Populations of the Mekong Mud Snake (Enhydris subtaeniata: Serpentes: Homalopsidae) in Indochina

HAROLD K. VORIS1*, JOHN C. MURPHY1, DARYL R. KARNS1,2,3, ERICA KREMER2 AND KATHARINE O’CONNELL2

1Department of Zoology, Field Museum of Natural History, 1400 South Lake Shore Drive, Chicago, IL 60605, USA 2Department of Biology and Rivers Institute, Hanover College, Hanover, IN 47243, USA 3Deceased * Corresponding author. E-mail: [email protected] Received: 29 January 2012; Accepted: 20 July 2012

ABSTRACT.– This study examines between-population variation of the Mekong Mud Snake, Enhydris subtaeniata (Bourret, 1934) in Thailand, Cambodia, Laos, and Viet Nam, and relates the observed patterns to previously published patterns based on molecular data. Enhydris subtaeniata were collected from six representative locations throughout its range that encompassed three important river drainage basins: the Chao Phraya, the Middle Mekong and the Lower Mekong. Data on size, sexual size dimorphism, scalation, reproductive biology and diet were used to explore biogeographical patterns revealed by a previously published phylogenetic and population genetic analysis. In several cases the size and reproductive characteristics mirrored the genetic differences and habitat differences associated with the sampled locations.

KEY WORDS: Freshwater snake, biogeography, sea levels, Mekong River, Chao Phraya River, Khorat Basin

INTRODUCTION We have been conducting studies investigating the effects of geographical Southeast Asia is a global bio-diversity features on evolutionary and ecological hotspot with 20 – 25% of the planet’s processes in this region using homalopsid animal and plant species (Woodruff, 2010). snakes, which dominate the semi-aquatic Previous studies have identified various snake communities of Southeast Asia (e.g., geographic features of the region that have Alfaro et al., 2004, 2008; Karns et al., 2005, likely influenced geographic variation in the 2010a; Murphy, 2007). The Homalopsidae distribution of terrestrial, freshwater and (47 species currently recognized) are marine species of this region, as well as the opisthoglyphous (rear-fanged), mildly patterns of intraspecific genetic diversity venomous (Fry et al., 2006), and distributed and population structure. In particular, from Pakistan and the Indian subcontinent Indochina has a complex geological history, across Southeast Asia to northern Australia with a number of potentially important (Murphy, 2007). Most homalopsids are physiographic events occurring in the semi-aquatic, primarily nocturnal, and Quaternary that certainly affected many taxa usually associated with mud substrates. (see reviews in Rainboth, 1996; Inger and Semi-aquatic snakes are particularly Voris, 2001; Glaubrecht and Köhler, 2004; interesting with respect to gene flow, Woodruff, 2010). dispersal, and speciation processes because of their intermediate ecological position

176 TROPICAL NATURAL HISTORY 12(2), OCTOBER 2012

FIGURE 1. Map showing the six E. subtaeniata collection locations in Indochina. Bung Boraphet is in the Chao Phraya basin, located in the Central Plain of Thailand. Adjacent to the Middle Mekong and in the Khorat Basin is Khon Kaen (Ban Mai and Ban Nong Pueng sites) in the Chi River catchment and Ban Badan in the Mun River catchment (the Chi and Mun Rivers are tributaries of the Mekong). Dong Khanthung is in Laos above Khone Falls on the southern edge of the Middle Mekong. Tonle Sap and U. Minh Thong are in the Lower Mekong (See Lukoschek et al. 2011 for GPS coordinates and other details on sampling locations). across the continuum of terrestrial-aquatic medium-sized, semi-aquatic piscivorous habitats. This study compares several snake found in a variety of freshwater populations of the Mekong Mud Snake, habitats with mud substrates, including rice Enhydris subtaeniata (Bourret, 1934), a paddies, streams, ponds, ditches, and canals species known only from lowland regions of (Karns et al., 2005, 2010a). It occurs central Indochina (Fig. 1). This is a primarily in the Middle and Lower Mekong

VORIS ET AL. — POPULATION DIFFERENCES IN THE MEKONG MUD SNAKE 177

River basins from Laos, Thailand, Viet Nam, Thailand (Fig. 1). We evaluate our results in and Cambodia, although one population is the context of the known biogeographical known from Bung Boraphet in the Central history of the region and the results of Plain of Thailand (Karns et al., 2010a). previous molecular studies (Lukoschek et In this study we broaden our under- al., 2011). standing of the biogeography of semi- aquatic taxa in Indochina by comparing the MATERIALS AND METHODS morphology (size, sexual size dimorphism, and scalation), habitat utilization, Collecting.– Enhydris subtaeniata (Fig. 2) reproduction, and diet among populations of specimens used in this study came from E. subtaeniata from six geographic collections obtained in 2003, 2004 and 2007 localities in three major river basins in the (Karns et al., 2005, 2010a) and from region: the Middle Mekong (Mekong specimens collected by Stuart (Stuart et al., tributaries inside the Khorat Basin), the 2000; Stuart, 2004). We typically obtained Lower Mekong (outside the Khorat Basin), adult and large subadult snakes collected as and the Chao Phraya in the Central Plain of

FIGURE 2. Enhydris subtaeniata photographed in July of 2004 by J. C. Murphy at Bung Boraphet, Thailand.

178 TROPICAL NATURAL HISTORY 12(2), OCTOBER 2012 incidental by-catch from local fishers. morphology and the presence of hemipenes. Young and small subadult snakes were For all snakes, we measured snout-vent rarely collected and were not included in the length (SVL) and tail length to the nearest analysis. Dead snakes were quickly mm and weighed snakes to the nearest 0.1 g. processed or frozen. Some dead snakes were We also measured the girth at the neck and partially decomposed and were not retained. mid-body for preserved specimens. A sexual Live snakes were euthanized, processed, and dimorphism index (SDI) was calculated by preserved with formalin. The preserved dividing the mean SVL of the larger sex by collection of snakes was deposited with the the mean SVL of the smaller sex; a positive National Science Museum of Thailand. value is assigned if females are the larger sex and a negative value if the males are the Study sites.– We collected snakes from a larger sex (Gibbons and Lovich, 1990). We total of 12 sites from the following river also calculated sex ratios for adult E. basins: the Chao Phraya River basin, subtaeniata collected at the three Thailand Thailand; the Mun and Chi River basins in locations with sufficient sample sizes (Bung Thailand and Laos; and the Lower Mekong Boraphet, n = 54; Khon Kaen area, n = 50; in Cambodia and Vietnam (Fig. 1; see Ban Badan, n = 15). Lukoschek et al., 2011 for detailed information about collection sites). Snakes Scalation, reproduction and diet. – We from 12 sites were grouped into six documented standard scalation charac- locations (Fig. 1, Table 1) for analysis. Sites teristics for preserved specimens (nomen- were combined if they were within several clature used follows Lillywhite, 2008) kilometers of each other and at the same including the condition of the loreal and elevation. The six locations were clustered internasal, and the numbers of upper labials, into three geographic regions defined by lower labials, ventral, subcaudal, dorsal, and river basins as follows: Chao Phraya River ocular ring scales. For this aspect of the Basin – Bung Boraphet; Middle Mekong – study we examined 92 preserved specimens Khon Kaen (Ban Mai and Ban Nong (54 females, 38 males) of E. subtaeniata Pueng), Ban Badan, Dong Khanthung; (Khon Kaen area, n = 46; Ban Badan, n = 12; Lower Mekong – Tonle Sap, U. Minh Bung Boraphet, n = 14; Dong Khanthung, n = Thong. 9; U. Minh Thong, n = 8; Tonle Sap, n = 3). Statistics on three variable characters Size, sexual dimorphism and sex ratios.– We (number of ventral scales, number of examined sexual size dimorphism (SSD) subcaudal scales, and number of anterior using adult snakes obtained from gill nets dorsal scales) are provided in Table 2. that could be reliably sexed using tail

VORIS ET AL. — POPULATION DIFFERENCES IN THE MEKONG MUD SNAKE 179

180 TROPICAL NATURAL HISTORY 12(2), OCTOBER 2012

VORIS ET AL. — POPULATION DIFFERENCES IN THE MEKONG MUD SNAKE 181

Snakes were dissected in the field prior significant differences in slope among to preservation and the reproductive locations for both males (F(2, 43) = 5.35, P < condition of females (n = 63) was 0.01) and females (F(2, 67) = 7.36, P < 0.01). documented. We recorded females as Tukey HSD for unequal sample sizes post reproductively mature when they contained hoc test showed that males and females enlarged vitellogenic follicles, oviducal from Bung Boraphet were significantly eggs, or exhibited an obviously thickened and longer in SVL and heavier than their muscular oviduct (indicating that they were counter-parts from Khon Kaen and Ban post-partum). We counted the number of Badan (Fig. 3A and 3B). vitellogenic follicles and oviducal eggs and The sampled populations of E. calculated the relative clutch mass (RCM) as subtaeniata also exhibited sexual size the ratio of the clutch mass to the maternal dimorphism. Snakes from Bung Boraphet and mass minus the clutch mass (Table 3). We Khon Kaen had an SDI with females also examined stomach contents in the field significantly longer than males (Table 1, prior to preservation to determine diet for a Fig. 3A) and heavier than males (Fig. 3B). total of 112 snakes, 66 females and 46 males Males and females from Ban Badan did not (Khon Kaen area, n = 46; Bung Boraphet, n differ in SVL or mass (Figs. 3A and 3B, = 55; Ban Badan, n = 11). Table 1), but sample sizes were smaller. Male E. subtaeniata from all three localities had significantly longer tails, relative to RESULTS SVL compared to females (Table 1).

Sex ratios (Table 1) varied among the Size, sexual dimorphism and sex ratios.– three main localities sampled (Bung Snout-vent length (SVL) differed among Boraphet, Khon Kaen and Ban Badan) with three E. subtaeniata populations (Fig. 3A) a significant deviation from a 1:1 sex ratio for both males (F = 67.45, P < 0.001) (2,44) at Khon Kaen (M:F 0.34, n = 50), but not at and females (F = 55.43, P < 0.001). (2,68) Bung Boraphet (M:F 0.50, n = 54) or Ban ANCOVAs on male and female SVL, using Badan (M:F 0.27, n = 15). mass as the covariate, also demonstrated

182 TROPICAL NATURAL HISTORY 12(2), OCTOBER 2012

Scalation.– Sixteen of the 20 scale ventrals, subcaudals (left and right), and characters examined were found to be anterior dorsal scales showed some invariant among locations for both males significant variation (Table 2). Specifi-cally, and females. However, the numbers of there was sexual dimorphism in ventral and

FIGURE 3. Geographic variation in snout-vent length and mass among three populations of E. subtaeniata. The mean SVL (± SE), mass (± SE), and 95 % confidence limits are shown for females and males from each population. The letters above the means indicate significant differences among means at P < 0.05 based on Tukey HSD unequal sample sizes post hoc test; letters in common indicate no difference.

VORIS ET AL. — POPULATION DIFFERENCES IN THE MEKONG MUD SNAKE 183 subcaudal scale counts, with males having geographic differences in the number of significantly (P < 0.05) more ventral scales subcaucals, with females from Bung than females at Bung Boraphet, Khon Kaen, Boraphet having significantly higher and Ban Badan (Table 2, Fig. 4A and 4B). numbers of subcaudals than females from In addition, there were sex-specific Khon Kaen (P < 0.05), but not from Ban

FIGURE 4. Geographic variation in the number of ventral and subcaudal scales among three populations of E. subtaeniata. The mean (± SE) and the 95 % confidence limits are shown for females and males from each population. The letters above the means indicate significant differences among means at P < 0.05 based on Tukey HSD unequal sample sizes post hoc test; letters in common indicate no difference.

184 TROPICAL NATURAL HISTORY 12(2), OCTOBER 2012

Badan (Fig. 4B). By contrast, male 1.41) than the Khon Kaen females (mean = subcaudal counts from Bung Boraphet were 9.2 ± 0.83) (t = 3.11, df = 28, P = 0.004). significantly higher than males from Ban The RCM of nine E. subtaeniata ranged Badan (P < 0.05), but not Khon Kaen (Fig. from 0.12 – 0.21. Female E. subtaeniata 4B). There were no sex-specific geographic from Khon Kaen (Ban Mai + Ban Nong differences in the numbers of ventrals (Fig. Pueng) exhibited a positive relationship 4A). between clutch size and SVL (F(1,22) = 18.092, r2 = 0.451, P < 0.001) (Fig. 5), Reproduction and diet.– The percentage of however this was not the case for females 2 female E. subtaeniata in reproductive from Bung Boraphet (F(1,13) = 3.011, r = condition ranged from 37.5% to 93.1% in 0.19, P = 0.11). the three Thai samples; Bung Boraphet, The majority of E. subtaeniata (61.6%) Khon Kaen and Ban Badan (Table 3). The had stomach contents. There was a number of vitellogenic follicles ranged from significant difference in the frequency of 3 to 23 while the number of oviducal eggs feeding between female (48 of 66 snakes ranged from 6 to 15 (Table 3). The larger with food: 72.7%) and male E. subtaeniata Bung Boraphet females had significantly (21 of 46 with food: 45.6%) (x2 = 8.40, df = more vitellogenic follicles (mean = 14.1 ± 1, P = 0.004). Moreover, a large proportion

FIGURE 5. Comparison of clutch size as a function of snout-vent length for E. subtaeniata from Khon Kaen (Ban Mai + Ban Nong Pueng) vs. Bung Boraphet. Due to limited samples clutch size was calculated from snakes with either enlarged vitellogenic follicles or oviducal eggs. The upper dotted line (open circles) represents the linear regression for the Khon Kaen snakes (n=24, r2 = 0.4513, p < 0.001; y = 0.4407*x - 10.7410) while the lower solid line (filled triangles) represents the linear regression for the Bung Boraphet snakes (n=15, r2 = 0.1881, p = 0.106; y = 0.3730*x - 9.016).

VORIS ET AL. — POPULATION DIFFERENCES IN THE MEKONG MUD SNAKE 185 of females in reproductive condition (32 of to the fact that Bung Boraphet is an 43 snakes: 74.4%) contained food. Stomach extensive and highly productive wetland contents invariably consisted of fish prone to flooding, whereas the Khorat Basin remains, although in most cases (75.4%) is relatively dry, with lower soil fertility they were limited to unidentifiable scales, (Parnwell, 1988). Stuart et al. (2000) bone and undigested flesh. Identifiable fish suggested that human overfishing of larger included five Puntius (Cyprinidae), one fish species in Tonle Sap, and consequent Esomus mettallicus (Cyprinidae), and one increase in smaller prey fish species, may Trichopodus trichopterus (Osphronemidae). influence size and repro-ductive Measurable prey length ranged from 6.2 to characteristics by increasing the food supply 8.1 cm and prey mass from 5.0 to 16.2 g, of homalopsids that eat relatively small prey representing 2.1–17.1 % of snake body items. This mesopredator release (Ritchie mass. and Johnson, 2009) may also be the case in the densely populated Central Plain where Bung Boraphet supports an important DISCUSSION fishery (Sriwongsitanon et al. 2007). Similar

to this study, Karns et al. (2010b) found that Morphology and ecology.– Variation in E. enhydris, a close relative of E. body size and reproductive characteristics subtaeniata with similar ecology, collected (clutch size, frequency of reproduction, and from Bung Boraphet and Bung Ka Lo size at first repro-duction) are well (another extensive wetland complex in the documented among squamates (Seigel and Chao Phraya basin) were significantly larger Ford, 1987; Zug et al., 2001) and may differ (SVL and mass) than other E. enhydris due to local genetic diversification or populations sampled in Thailand. phenotypic effects influenced by food Despite moderate levels of genetic availability and other environmental factors divergence observed among localities (1– (Madsen and Shine, 1993; Boback and 2%, Lukoschek, et al., 2011), we did not Guyer, 2003; Weatherhead et al., 1995; find a suite of diagnostic morphological Pearson et al., 2002). Males and females differences among populations using classical from Bung Boraphet were significantly scale characters. Enhydris subtaeniata larger in size (both SVL and mass) than E. exhibited typical homalopsid female-biased subtaeniata from the Khorat Basin sites and sexual size dimorphism (Karns et al. 2005, Bung Boraphet females had larger clutch 2010b) and males have longer tails and a sizes (Table 3). These observed size greater number of subcaudal scales. The differences may be a reflection of genetic ratio of tail length to SVL is a common differences shown in the maximum sexually dimorphic trait in snakes and male parsimony (MP) tree based on molecular snakes of some species with longer tails data (Figure 2 in Lukoschek, et al., 2011). In achieve greater reproductive success (Shine this MP tree, snakes from Bung Boraphet et al., 1999). were grouped with U. Minh Thong and Freshwater homalopsids are primarily Tonle Sap snakes from the Lower Mekong piscivorous (Voris and Murphy, 2002) and while the Middle Mekong snakes from the E. subtaeniata is no exception. Voris and Khorat Basin grouped together. The Murphy (2002) documented that observed size differences could also be due homalopsids frequently feed on numerous,

186 TROPICAL NATURAL HISTORY 12(2), OCTOBER 2012 small prey. Brooks et al., (2009) found the locations, further supporting a pattern of prey-to-predator mass ratios of homalopsids local differentiation. In addition, the to be lower (< 0.2) than that recorded for differences observed in snake size and Xenochrophis piscator, Cylindrophis ruffus scalation between the Chao Phraya River and other snakes reported in the literature Basin (Bung Boraphet) and the Middle (Rodríguez-Robles et al., 1999). We have Mekong (Khon Kaen and Ban Badan) lend limited data for E. subtaeniata, but we further support to the notion that a barrier to documented low prey-to-predator mass gene flow exists for homalopsid snake ratios (0.02 – 0.17) comparable to another species between the Khorat Basin and the small-headed homalopsid, Enhydris Chao Phraya Basin. enhydris (0.01 – 0.19) (Karns et al., 2010b). The high percentage of stomachs with ACKNOWLEDGEMENTS unidentifiable fish remains (scales, bone, tissue) also suggests a high rate of passage This work was supported by the Field of small prey items. Contrary to reports of Museum of Natural History, the Hanover anorexia within snakes in reproductive College Biology Department, and the Rivers condition (Lourdais et al., 2002), we found Institute at Hanover College. The fieldwork that 74.4% of female E. subtaeniata in was partially funded by a grant from the reproductive condition contained food, John D. and Catherine T. MacArthur whereas ~45% of reproductive E. enhydris Foundation to the Field Museum. Collecting contained food (Brooks et al., 2009; Karns and handling of the specimens was in et al., 2010b). accordance with The Institutional Animal A recent paper by Lukoschek et al. Care and Use Committee of the Field (2011) revealed strong population Museum (protocol no. FMNH 02–3). We subdivision for the Mekong Mud Snake, E. wish to thank our many colleagues and subtaeniata, throughout most of its range in sponsors in Thailand. In particular, we are Indochina, with each sampled location grateful to Jarujin Nabhitabhata (deceased), characterized by a unique suite of Tanya Chanard, and Yodchaiy Chuaynkern haplotypes, but with varying levels of at the National Science Museum; Lawan genetic diversity. Relationships among Chanhome at the Queen Saovabha phylogroups largely conformed to the Memorial Institute, Thai Red Cross Society; geographic proximity of sampled locations Chantip Inthara at the Department of and there was a strong relationship between Biology, Faculty of Science at Khon Kaen genetic distance and geographic distances University; and Kumthorn Thirakhupt, and along suitable aquatic habitats, suggesting his students, Anchalee Aowphol, that the geographical configuration of river Chattraphas Pongcharoen, and other drainages may have significantly influenced graduate students in the Department of the distribution of genetic diversity in this Biology at Chulalongkorn University. We region. also appreciate the hard labour of the many The findings of the present study show snake collectors who assisted us at Bung that several aspects of the snakes’ biology Boraphet including Ms. Somrung, Krairatt including size, sexual size dimorphism, Iumupai, Kuntida Ittipron, Mr. Earth, and scalation and female reproductive Mr. Champ. Special thanks are due to our characteristics also differed between

VORIS ET AL. — POPULATION DIFFERENCES IN THE MEKONG MUD SNAKE 187

driver-translator-research colleague, Prapon Karns, D.R., Lukoschek, V., Osterhage, J., Murphy, (Pon) Noiwan. At the Field Museum we J.C. and Voris, H.K. 2010a. Phylogeny and biogeography of the Enhydris clade (Serpentes: thank Alan Resetar, Kathleen Kelly, Susan Homalopsidae). Zootaxa. 2452: 18-30. Hamnik, and Clara Richardson for Karns, D.R., Murphy, J.C., Voris, H.K. and Suddeth, assistance to our project. We also thank J.S. 2005. Comparison of semi-aquatic snake Darrin Rubino for assistance with the communities associated with the Khorat Basin, Thailand. The Natural History Journal of statistical analysis of the morphometric data. Chulalongkorn University. 5: 73-90. Karns, D.R., Murphy, J.C. and Voris, H.K. 2010b. Semi-aquatic snake communities of the central ITERATURE ITED L C plain region of Thailand. Tropical Natural History. 10: 1-25. Alfaro, M.E., Karns, D.R., Voris, H.K., Abernathy, E. Lillywhite, H.B. 2008. Dictionary of Herpetology. and Sellins, S.L. 2004. Phylogeny of Cerberus Krieger Publishing Inc., Malabar, Florida. 376 (Serpentes: Homalopsinae) and phylogeography pp. of Cerberus rynchops: Diversification of a coastal Lourdais, O., Bonnet, X. and Doughty, P. 2002. marine snake in Southeast Asia. Journal of Costs of anorexia during pregnancy in a Biogeography. 31:1277-1292. viviparous snake (Vipera aspis). Journal of Alfaro, M.E., Karns, D.R., Voris, H.K., Brock, C.D. Experimental Zoology. 292: 487–493. doi:10.1002 and Stuart, B.L. 2008. Phylogeny, evolutionary /jez.10065. history, and biogeography of Oriental-Australian Lukoschek, V., Osterhage, J.L., Karns, D.R., Murphy, rear-fanged water snakes (Colubroidea: J.C., and Voris, H.K. 2011. Phylogeography of Homalopsidae) inferred from mitochondrial and the Mekong mud snake (Enhydris subtaeniata): nuclear DNA sequences. Molecular Phylogenetics the biogeographic importance of dynamic river and Evolution. 46: 576-593. drainages and fluctuating sea levels for Boback, S.M. and Guyer, C. 2003. Empirical evidence semiaquatic taxa in Indochina. Ecology and for an optimal body size in snakes. Evolution. 57: Evolution. doi: 10.1002/ece3.29. 345-351. Madsen, T. and Shine, R. 1993. Phenotypic plasticity Brooks, S.E., Allison, E.H., Gill, J.A. and Reynolds, in body sizes and sexual size dimorphism in J.D. 2009. Reproductive and trophic ecology of an European grass snakes. Evolution. 47: 321-325. assemblage of aquatic and semi-aquatic snakes in Murphy, J.C. 2007. Homalopsid Snakes: Evolution in Tonle Sap, Cambodia. Copeia. 2009: 7-20. the Mud. Krieger Publishing Company, Malabar, Fry, B.G., Vidal, N., Norman, J.A., Vonk, F.J., Florida. 249 pp. Scheib, H., Ramjan, S.F.R., Kuruppu, S., Fung, Parnwell, M.J.G. 1988. Rural poverty, development K., Hedges, S.B., Richardson, M.K., Hodgson, and the environment: the case of North-East W.C., Ignjatovic, V., Summerhayes, R. and Thailand. Journal of Biogeography. 15: 199-208. Kochva, E. 2006. Early evolution of the venom Pearson, D., Shine, R. and Williams, A. 2002. system in lizards and snakes. Nature. 439: 584- Geographic variation in sexual size dimorphism 588. within a single snake species (Morelia spilota, Gibbons, J.W. and Lovich, J.E. 1990. Sexual Pythonidae). Oecologia. 131: 418-426. dimorphism in turtles with emphasis on the slider Rainboth, W.J. 1996. The taxonomy, systematics and turtle (Trachemys scripta). Herpetological zoogeography of Hypsibarbus, a new genus of Monographs. 4: 1-29. large barbs (Pisces, Cyprinidae) from the rivers of Glaubrecht, M. and Köhler, F. 2004. Radiating in a southeastern Asia. University of California river: systematics, molecular genetics and Publications in Zoology, Berkeley, California. morphological differentiation of viviparous 199 pp. freshwater gastropods endemic to the Kaek River, Ritchie, E.G. and Johnson, C.N. 2009. Predator central Thailand (Cerithioidea, Pachychilidae). interactions, mesopredator release and biodiversity Biological Journal of the Linnean Society. 82: conservation. Ecology Letters. 12: 982–998. 275-311. doi:10.1111/j.1461-0248.2009.01347.x Inger, R.F. and Voris, H.K. 2001. The biogeographical Rodríguez-Robles, J.A., Mulcahy, D.G. and Greene, relations of the frogs and snakes of Sundaland. H.W. 1999. Feeding ecology of the desert Journal of Biogeography. 28: 863-891.

188 TROPICAL NATURAL HISTORY 12(2), OCTOBER 2012

nightsnake, Hypsiglena torquata (Colubridae). Voris, H.K. and Murphy, J.C. 2002. The prey and Copeia. 1999: 93-100. predators of homalopsine snakes. Journal of Seigel, R.A. and Ford, N.B. 1987. Reproductive Natural History. 36: 1621-1632. ecology. pp. 210-252. In: Snakes, Ecology and Weatherhead, P.J., Barry, F.E., Brown, G.P. and Evolutionary Biology. (Seigel et al., eds.) Forbes, M.R.L. 1995. Sex ratios, mating behavior McGraw-Hill Publishing Co., New York. 529 pp. and sexual size dimorphism of the northern water Shine, R., Olsson, M.M., Moore, I.T., Lemaster, M.P. snake, Nerodia sipedon. Behavioral Ecology and and Mason. R.T. 1999. Why do male snakes have Sociobiology. 36: 301-311. longer tails than females? Proceedings of the Woodruff, D.S. 2010. Biogeography and conservation Royal Society of London. 266: 2147-2151. in Southeast Asia: how 2.7 million years of Sriwongsitanon, N., Surakit, K., Hawkins, P.R. and repeated environmental fluctuations affect today's Chandrasena, N. 2007. Decision support tools for patterns and the future of the remaining refugial- water resource management: a case study of Bung phase biodiversity. Biodiversity and Conservation. Boraphet Wetland, Thailand. Journal of 19: 919-941. Developments in Sustainable Agriculture. 2: 17- Zug, G.R., Vitt, L.J. and Caldwell, J.P. 2001. 26. Herpetology: An Introductory Biology of Stuart, B.L. 2004. The harvest and trade of reptiles at Amphibians and Reptiles, Second ed. Academic U Minh Thuong National Park, southern Viet Press, San Diego. Nam. Traffic. 20: 25-34. Stuart, B.L., Smith, J., Davey, K., Dim, P. and Platt, S.G. 2000. Homalopsine Watersnakes: the harvest and trade from Tonle Sap, Cambodia. Traffic. 18: 115-124.