FULL PAPER Anatomy Geographical Variation of Skull Size and Shape in Various Populations in the Black Giant Squirrel Hideki ENDO1), Junpei KIMURA2), Tatsuo OSHIDA3), Brian J. STAFFORD4,5), Worawut RERKAMNUAYCHOKE6), Takao NISHIDA6), Motoki SASAKI7), Akiko HAYASHIDA7) and Yoshihiro HAYASHI8) 1)Department of Zoology, National Science Museum, Tokyo, 3–23–1 Hyakunin-cho, Shinjuku-ku, Tokyo 169–0073, 2)Department of Veterinary Anatomy, College of Bioresource Sciences, Nihon University, Fujisawa, Kanagawa 252–8610, 3)Laboratory of Molecular Ecology, Department of Biology, Tunghai University, Taichung, Taiwan 407, R.O.C., 4)Mammal Division, National Museum of Natural History, Smithsonian Institution, Washington DC, 5)Deparmtent of Anatomy, Howard University College of Medicine, Washington DC, U.S.A., 6)Department of Veterinary Anatomy, Faculty of Veterinary Medicine, Kasetsart University, Bangkok, Thailand, 7)Department of Veterinary Anatomy, Obihiro University of Agriculture and Veterinary Medicine, Obihiro, Hokkaido 080–8555 and 8)Department of Global Agricultural Sciences, Graduate School of Agricultural and Life Sciences, The University of Tokyo, Tokyo 113–8657, Japan (Received 18 November 2003/Accepted 25 May 2004) ABSTRACT. We osteometrically examined the skulls of the black giant squirrel (Ratufa bicolor) from three mainland populations (M. Malayan Peninsula, V. South Vietnam, and B. Burma, India and North Thailand) and from two island populations (T. Tioman, and S. Sumatra Islands). The skull in the Malayan peninsula population was significantly smaller than that of the two other mainland popula- tions. It is consistent with Bergmann’s rule as shown in the gray-bellied squirrel. The two island populations did not show obvious differences in comparison with the Malayan population in many measurements. In the proportion analysis eliminating the size factor, the differences among populations were not easily confirmed and we concluded that the osteological characters peculiar to each popu- lation could not be shown in this species. The first and second principal component scores of M, S, and T populations were intermingled, whereas the V and B populations of V and B were not separated in the chart. We pointed out that the morphological differences were demonstrated between northern and southern groups of the Isthmus of Kra in the mainland populations, and that the two island popula- tions did not show the island-isolation effect in comparison with the M population. The adaptational variation related to feeding and locomotion could not be confirmed among populations of the black giant squirrel as shown in the proportion analysis. KEY WORDS: adaptation, black giant squirrel, geographical variation, Ratufa bicolor, skull. J. Vet. Med. Sci. 66(10): 1213–1218, 2004 The 30 species of the Family Sciuridae showing varia- tion of origin and sex, and population symbols are shown in tions in the geographical populations are distributed Table 1. Skull measurement was carried out with vernier throughout the Indomalayan Region [1, 2, 10, 11]. The calipers to the nearest 0.05 mm. Measurements are defined black giant squirrel (Ratufa bicolor) is widely distributed in in Table 2, and were based on Driesch [3]. Differences of the southern part of China, Burma, east Himalayan region, mean values among localities were examined by Student’s t- Thailand, Laos, Vietnam, Malayan Peninsula, Sumatra, Java test. We compared the three mainland populations, and the and some islands of Southeast Asia [1, 2]. Since the distri- mainland M population was compared with the two island bution area is so large, skeletal specimens from various geo- populations. Proportion indices were obtained for discus- graphical locations have been expected to osteometrically sion of the adaptation. The indices were the quotients of examine. The genus Ratufa consists of four species with a each measurement value divided by the geometric mean of large body size that is highly adapted to arboreal locomotion all measurement values. The significant differences of pro- and herbivorous feeding and nutrition [1, 2, 10]. The black portion indices were examined among localities by nonpara- giant squirrel has attracted anatomists as a model indicating metric U-test using software Statistica (Statsoft Inc., Tokyo, these adaptational patterns in the Family Sciuridae. Our aim Japan). In addition, principal component analysis was car- is to confirm the zoogeographical similarities and to eluci- ried out with all measurement data to clarify variations date the adaptational strategies in skull size and shape in this among geographical localities. A package software for mul- species. tivariate analysis (Shakai-Joho Service, Tokyo, Japan) added to Microsoft Excel 98 was used for this analysis. MATERIALS AND METHODS RESULTS We examined a total of 70 skulls of the black giant squir- rel that have been stored in the Mammal Division of the Mean values and standard deviations of the 23 measure- National Museum of Natural History at the Smithsonian ments are given in Table 3. Statistical differences in these Institution (Table 1). Sex determination was dependent on measurements between combinations of various populations the description of biological data of specimens. Only spec- are shown in Table 4. All mean values were significantly imens with fully erupted molars were used. The composi- smaller in M population than in V and B populations except 1214 H. ENDO ET AL. Table 1. Locality and sex composition of the specimens Subspecies Localities Symbols Male Female Ratufa bicolor peninsulae Malayan Peninsula (Surat Thani, Trang, Selangor) M 12 6 Ratufa bicolor simithi Vietnam (South District) V 9 4 Ratufa bicolor gigantea Burma, India and North Thailand" B 6 3 Ratufa bicolor palliata Sumatra Id. S 9 10 Ratufa bicolor tiomanensis Tioman Id. T 3 8 Total 39 31 Table 2. List of skull and mandibular measurements and their abbreviations Cranium Profile length PL Maximum length ML Condylobasal length CL Short lateral facial length SL Maximum width of the nasal bone MWN Least breadth between the orbits LBO Least breadth of caudal point of zygomatic process of frontal bone LBC Greatest neurocranium breadth GNB Greatest occipital breadth GOB Median palatal length MPL Dental length DL Length of molar row LMR1 Greatest palatal breadth GPB Length from Basion to Staphylion LBS Length from Basion to the most rostral point of zygomatic arch LBR Length from Basion to the most medial point of orbit LBM Height from Akrokranion to Basion HAB Mandible Length from the condyle LC Length of molar row LMR2 Thickness of mandible at middle point of M1 TM Height of the mandible at M1 HM Aboral height of the vertical ramus AHR Oral height of the vertical ramus OHR for the LMR2 in males. In contrast, many measurements did population was obviously different from S population in not show significant differences between the V and B popu- GNB, HM and AHR in male, and in LBO, LBC, GNB, lations. GOB, LBS and LBR in female. In the other comparisons In the length measurements representing the skull size, between the localities, however, the significant differences the mean values of PL and ML were smaller in the T popu- were found only in GOB and HM between the M and B lations than in the mainland M population in both sexes. In male populations, in PL, GOB and LC between the S and T ML, however, the male is larger in the M population than in male populations, in HAB between the M and T male popu- the S population, and the female is smaller in the M popula- lations, in AHR and OHR in the M and B female popula- tion than in the S population. Although the measurements tions and in LC between the V and B female populations. of GNB, GOB and LBM among the M, S and T populations The principal component charts are shown in Fig. 1, and were significantly different in both sexes (Table 4), we loading factors are arranged in each measurement (Table 7). could not establish a rule in items showing obviously differ- The plots were obviously separated into the two major ent values. The sexual dimorphism in size tendency was not groups: 1) M, T and S populations, 2) V and B populations. recognized in various populations. The plots were intermingled among M, S and T populations, The proportion indices are arranged in Table 5, and their and among V and B populations in their distribution. significant differences confirmed by nonparametric U-test among localities are shown in Table 6. The significant dif- DISCUSSION ferences were not confirmed in many comparisons in Table 6. The comparison in male between the M and V popula- Our two studies on the variation in geographical popula- tions showed significant differences in CL, SL, GOB, tions showed the different results between Dremomys rufi- LMR1, GPB, LBS, LBM, HAB, LC and LMR2. The M genis and Callosciurus caniceps. We have pointed out that SKULL VARIATION OF BLACK GIANT SQUIRREL 1215 Table 3. Mean values (mm) and standard deviations for craniometric measurement in various populations PL ML CL SL MWN LBO LBC GNB GOB MPL DL LMR1 GPB LBS LBR LBM HAB LC LMR2 TM HM AHR OHR Malayan Peninsula Male 66.46 66.30 60.02 25.66 13.41 27.85 23.34 27.66 32.34 31.76 33.42 13.85 18.21 28.48 52.53 45.03 17.83 45.22 14.77 5.58 12.10 28.51 25.62 1.81 1.67 1.86 1.05 0.86 0.96 0.65 0.64 0.93 1.16 0.79 0.28 0.44 1.12 1.38 1.08 0.45 1.23 0.55 0.24 0.52 1.21 1.28 Female 65.40 65.20 58.33 25.03 12.88 26.42 23.21 27.26 31.05 30.95 32.97 13.80 17.82 27.62 51.27 43.72 17.33 44.40 14.70 5.47 11.88 27.28 24.48 1.90 1.79 2.30 1.14 1.23