A preliminary investigation of Sundarbans tiger morphology

Article in press - uncorrected proof Mammalia 74 (2010): xxx–xxx
2010 by Walter de Gruyter • Berlin • New York. DOI 10.1515/MAMM.2010.040 Short Note A preliminary investigation of Sundarbans tiger morphology Adam C.D. Barlow1,*, Ji Mazák2, Ishtiaq U. Ahmad1,3 and James L.D. Smith1 1 Department of Fisheries, Wildlife and Conservation Biology, 200 Hodson Hall, 1980 Folwell Avenue, University of Minnesota, Saint Paul, Minnesota 55108, USA, e-mail: [email protected] 2 Shanghai Science and Technology Museum, No. 2000 Century Avenue, Pudong New Area, Shanghai 200127, PR China 3 Forest Department of Bangladesh, Agargaon, Dhaka, Bangladesh *Corresponding author Keywords: adaptation; Bangladesh; mangroves; Panthera tigris. Introduction The tiger (Panthera tigris) probably originated in east Asia, and was already well established throughout its historical range by about two MYA ago (Hemmer 1987, Kitchener 1999). During the evolutionary history of the tiger, it has adapted to a wide range of ecological conditions, from temperate forests to mangroves (Kitchener 1999). Since the first formal description of tigers as Felis tigris in 1758 by Linnaeus, eight subspecies have been established, with the addition of a ninth recently proposed (Luo et al. 2004). However, these classifications are disputed on morphological, genetic, and biogeographical grounds (Cracraft et al. 1998, Kitchener and Dugmore 2000, Mazák 2010). Furthermore, considering the small sample sizes that current classification is based upon and the lack of agreement among studies, there seems considerable scope to improve understanding of tiger evolution and taxonomy. Totally lacking in previous work has been genetic or morphological representation of tigers from the Sundarbans of Bangladesh and India. These tigers are traditionally assigned to Panthera tigris tigris, but they might have been isolated long enough to become morphologically or genetically distinct. The objectives of this study were to investigate if Sundarbans tigers were morphologically distinct from other groups, by analyzing skull morphometrics and body weights. This information will help assess the conservation value of the Sundarbans tiger. The Sundarbans are currently isolated from the next nearest tiger habitat by 200–300 km of arable and urban land so there is no probability of genetic exchange through normal dispersal events with other tiger populations. The forest is made up of vegetated low lying islands interspersed with a maze of tidal waterways (Iftekhar and Islam 2004). Five adult tiger skulls were collected from the Bangladesh Sundarbans (three male and two females) and compared with a sample collection of 175 complete skulls (88 males and 87 females) representing nine tiger subspecies. Following Mazák (2010), 18 craniodental measurements were taken from each skull and size-adjusted by log transformation after they had been divided by the condobasal length. To test measurement errors between observers, a paired t-test was performed on six randomly selected male skulls (three Sundarbans and three Sumatran) (Yamaguchi et al. 2004). Errors between observers in all measurements were insignificant (p)0.05), but there is still the possibility of observer bias considering the small sample sizes involved. A one-way ANOVA indicated that means of 13 of 18 size adjusted skull variables differed significantly (Fs17.28, p-0.01) among the ten groups. The post-hoc Tukey’s HSD test showed that Sundarbans males differed significantly (p-0.01) from mainland subspecies; Sundarbans male tiger skulls were smaller in overall size (skull length, width, and jaw size), and had proportionally narrower muzzles and mastoid regions. The Sundarbans male skulls also differed significantly in shape from two Sunda Island subspecies (Panthera tigris sondaica, Panthera tigris balica), primarily by having a relatively broad occiput. A principal component analysis (PCA) of 18 size-adjusted variables resulted in two PCs with eigenvalues greater than one that accounted for 76% of the variance in shape of male tiger skulls (Table 1, Figure 1). The Sundarbans males were separated clearly from the mainland and Sunda Island groups on both PC1 and 2 (Figure 1). A stepwise DFA grouped all samples into two major clusters: Sunda Island cluster and mainland cluster with the Sundarbans group within the range of P. t. tigris (Figure 1). Mahalanobis D2 distance showed that the Sundarbans group differs in shape significantly (p-0.05) from all other groups, but differs most from P. t. sondaica, Panthera tigris altaica and Panthera tigris sumatrae. A jackknife classification correctly identified Sundarbans samples with 100% accuracy. Results of ANOVA and post-hoc Tukey’s HSD of females were similar to that of males, but with less marked PCA differences between Sundarbans and other groups (Table 1, Figure 1). Stepwise DFA clearly separated Sundarbans female skull shapes from all mainland groups (Figure 1). Mahalanobis D2 distance showed that Sundarbans females differed significantly (p-0.05) in shape from Panthera t. altaica, Panthera tigris jacksoni, P. t. tigris, P. t. balica and most markedly from P. t. sondaica. A jackknife classification correctly classified all Sundarbans females with 100% accuracy. 2010/031 Article in press - uncorrected proof 2 A.C.D. Barlow et al.: Sundarbans tiger morphology Table 1 Factor loadings for PCA for male and female tiger skulls. Measurement GLS CBL BL RB IFB IOB POB BZB MB SOB OH GLN ML MH P4L CP4L M1L CM1L Males Females PC 1 PC 2 PC 1 PC 2 0.827 0.796 0.813 0.69 0.58 0.542 0.277 0.798 0.782 0.222 0.722 0.792 0.82 0.882 0.543 0.523 0.526 0.664 0.513 0.526 0.504 0.517 0.596 0.658 0.645 0.466 0.546 0.948 0.557 0.286 0.49 0.185 0.281 0.675 0.255 0.621 0.889 0.864 0.859 0.802 – 0.735 0.462 0.853 0.704 0.266 0.761 0.778 – – 0.659 0.692 – – 0.355 0.343 0.35 0.315 – 0.303 0.427 0.296 0.443 0.963 0.445 0.253 – – 0.205 0.398 – – Note: Abbreviations correspond to measurement names taken from Mazák (2010). Adult tiger weights were taken in the field from two radiocollared female tigers and one female tiger killed by local people, and compared with weight data from other subspecies, compiled by Slaght et al. (2005). The mean weight of Sundarbans female tigers was 76.7 kg (SDs2.89, range 75–80) (Table 2). The weight of one of the two older females (75 kg) was slightly less than normal owing to her relatively poor condition at the time of capture. One-way ANOVA analysis of female weights (p-0.05) indicated a significant difference between groups (dfs7, fs17.26, p-0.001). A post-hoc Tukey’s HSD test showed that Sundarbans females were significantly different (p-0.05) in mean weight from Panthera t. tigris and P. t. altaica, and had the smallest mean weights of any group (Table 2). Although sample size was small, skulls of Sundarbans tigers were found to be significantly different from all other currently defined subspecies, both in terms of size and shape. This distinction was most notable for male tigers, which tend to be more variable than females (Mazák 2004). The female weights so far recorded suggest that Sundarbans tigers are the lightest throughout the range of the tiger, but occur at the same latitude as the largest subspecies, Panthera t. tigris (Slaght et al. 2005). Geographic variation in skull dimen- Figure 1 Principal component analysis (A) and discriminant function analysis (B) results of male and female tiger skull measurements. Sample sizes for skulls were P. t. altaika (11, 12), P. t. virgatta (5, 5), P. t. amoyensis (6, 4), P. t. jacksoni (3, 3), P. t. corbetti (10, 16), P. t. tigris (30, 21), P. t. Sumatra (7, 13), P. t. sondaica (14, 11), P. t. balica (2, 2), Sundarbans (3, 2). Article in press - uncorrected proof A.C.D. Barlow et al.: Sundarbans tiger morphology 3 Table 2 Male and female tiger weights. All weights taken from Slaght et al. (2005), plus the Sundarbans weights from this study. Group P. t. tigris P. t. altaica P. t. virgatta P. t. amoyensis P. t. corbetti P. t. sumatrae P. t. sondaica Sundarbans Males Females n Mean (SD) weight (kg) Range (kg) n Mean (SD) weight (kg) Range (kg) 3 44 2 15 6 22 1 – 212 (13.8) 173.7 (28) 156.5 (34.7) 134.9 (19.5) 120.6 (8.9) 110.8 (15.5) 110 – 200–227 118–248 132–181 96–174.6 109–132 91–140.2 – – 16 62 2 11 7 21 1 3 138.2 (20.5) 123 (18.9) 116 (28.9) 103.4 (18) 98.5 (8.7) 86.7 (12.7) 95 76.7 (2.9) 103–177 83.5–180 97–135 75–144.5 89.1–110.5 61.82–107.27 – 75–80 sions and body mass could potentially be explained by island/insular dwarfism, latitude, prey size, or some unidentified variable (Kitchener 1999, Mazák 2010). We suspect that the small skull and body size of Sundarbans tigers could be a consequence of having no sambar (Cervus unicolor) sized or larger prey available. Sundarbans tigers mainly prey on chital and wild boar (Reza et al. 2001). Elsewhere tiger prey always includes a large ungulate species, such as sambar, swamp deer (Cervus duvauceli), sika deer (Cervus Nippon) or banteng (Bos javanicus), that contribute a substantial component to tiger diets (Sunquist et al. 1999). The skull and weight differences observed in this study are substantial enough to highlight the Sundarbans tigers as potentially distinct from other groups. However, the small sample sizes of this study make any inferences tentative, until additional specimens can be obtained. Furthermore, genetic analysis is needed to confirm or not any conclusions based on skull and weight measurements. As a precautionary measure, it is recommended that no exchange of individuals or re-introduction should be allowed that mixes tigers from distinguishable populations, or when the characteristics of a population in question have not been ascertained (Moritz 1999). Acknowledgements We are grateful to the Bangladesh Ministry of Environment and Forest, and the Bangladesh Forest Department for permission to carry out this study. This study was a bi-product of a larger program of research funded by the United Fish and Wildlife Service. Guide Tours Ltd. gave some valuable logistic support. Earlier drafts of this document were improved by comments by Andrew Kitchener, Todd Arnold, Dave Garshelis, Don Siniff, Christina Greenwood, Samuel Purvey, Jon Slaght, and two anonymous reviewers., Md. Mozaharul Islam, Naser Hossain and Md. Abdur Rob assisted to locate and collect some of the specimens. Skulls were exhumed through the hard work of Mizan Rahman, Alam Howlader, and Tanjilur Rahman. References Cracraft, J., J. Feinstein, J. Vaughn and K. Helm-Bychowski. 1998. Sorting out tigers (Panthera tigris): mitochondrial sequences, nuclear inserts, systematics, and conservation genetics. Anim. Conserv. 1: 139–150. Iftekhar, M.S. and M.R. Islam. 2004. Degeneration of Bangladesh’s Sundarbans mangroves: a management issue. Int. For. Rev. 6: 123–135. Kitchener, A.C. 1999. Tiger distribution, phenotypic variation and conservation issues. In: (J. Seidensticker, S. Christie, and P. Jackson, eds.) Riding the tiger. Tiger conservation in human-dominated landscapes. Cambridge University Press, Cambridge, UK. pp. 19–39. Kitchener, A.C. and A.J. Dugmore. 2000. Biogeographical change in the tiger, Panthera tigris. Anim. Conserv. 3: 113–124. Luo, S.J., J.H. Kim, W.E. Johnson, J. Walt, J. Martenson, N. Yuhki, D.G. Miquelle, O. Uphyrkina, J.M. Goodrich, H.B. Quigley, R. Tilson, G. Brady, P. Martelli, V. Subramaniam, C. McDougal, S. Hean, S.Q. Huang, W. Pan, U.K. Karanth, M. Sunquist, J.L.D. Smith and S.J. Brien. 2004. Phylogeography and genetic ancestry of tigers (Panthera tigris). PLoS Biol. 2: e442. Mazák, J.H. 2004. On the sexual dimorphism in the skull of the tiger (Panthera tigris). Mammal. Biol. 69: 392–400. Mazák, J.H. 2010. Craniometric variation in the tiger (Panthera tigris): implications for patterns of diversity, taxonomy and conservation. Mammal. Biol. 75: 45–68. Moritz, C. 1999. Conservation units and translocations: strategies for conserving evolutionary processes. Hereditas 130: 217–228. Reza, A., M.M. Feeroz and M.A. Islam. 2001. Food habits of the Bengal tiger (Panthera tigris tigris) in the Sundarbans. Bang. J. Zool. 29: 173–180. Slaght, J.C., D.G. Miquelle, I.G. Nikolaev, J.M. Goodrich, E.N. Smirnov, K. Traylor-Holzer, S. Christie, T. Arjanova, J.L.D. Smith and K.U. Karanth. 2005. Chapter 6. Who’s king of the beasts? Historical and recent body weights of wild and captive Amur tigers, with comparisons to other subspecies. In: (D.G. Miquelle, E.N. Smirnov, and J.M. Goodrich, eds.) Tigers in Sikhote-Alin Zapovednik: ecology and conservation. PSP, Vladivostok, Russia (in Russian). pp. 25–35. Sunquist, M., K.U. Karanth and F. Sunquist. 1999. Ecology, behaviour and resilience of the tiger and its conservation needs. In: (J. Seidensticker, S. Christie, and P. Jackson, eds.) Riding the tiger. Tiger conservation in human-dominated landscapes. Cambridge University Press, Cambridge, UK. pp. 5–18. Yamaguchi, N., C.A. Driscoll, D.W. Macdonald, A.C. Kitchener and J.M. Ward. 2004. Craniological differentiation between European wildcats (Felis silvestris silvestris), African wildcats (F. s. lybica) and Asian wildcats (F. s. ornata): implications for their evolution and conservation. Biol. J. Linn. Soc. 83: 47–63.