FULL PAPER Anatomy

Osteometrical Skull Character in the Four Species of Tree Shrew

Hideki ENDO, Isao NISHIUMI, Yoshihiro HAYASHI1), Worawut RERKAMNUAYCHOKE2), Yoshi KAWAMOTO3), Hirohisa HIRAI3), Junpei KIMURA4), Agustinus SUYANTO5), Jarujin NABHITABHATA6) and Junzo YAMADA7)

Department of Zoology, National Science Museum, Tokyo, Shinjuku-ku, Tokyo 169–0073, 1)Department of Veterinary Anatomy, Graduate School of Agricultural and Life Sciences, The University of Tokyo, Tokyo 113–8657, 2)Department of Veterinary Anatomy, Kasetsart University, Bangkok, Thailand, 3)Primate Research Institute, Kyoto University, Aichi 484–0081, 4)Department of Veterinary Anatomy, College of Bioresource Sciences, Nihon University, Kanagawa 252–0813, 5)Bogor Zoological Museum, Bogor, Indonesia, 6)Reference Collection Division, National Science Museum, Bangkok, Thailand, and 7)Department of Veterinary Anatomy, Faculty of Agriculture, Obihiro University of Agriculture and Veterinary Medicine, Hokkaido 080–8666, Japan

(Received 10 September 1999/Accepted 21 January 2000)

ABSTRACT. The skull size and shape were osteometrically examined in the four species of the tree shrews (Tupaia tana, T. javanica, T. minor and T. dorsalis). We suggest that the skull characters were affected by the species specific behavior and terrestrial, arbo-terrestrial and arboreal life, among the genus Tupaia. The neurocranium was laterally narrower in the braincase area, and the splanchnocranium was longer only on dorsal side in T. tana, and these characters were opposite to T. minor. The principal component analysis confirmed the obviously separated clusters among T. tana, T. javanica and T. minor, affected by the adaptation for each behavior. T. dorsalis was considered as terrestrial species from the results of proportion analysis and the principal component analysis.—KEY WORDS: skull, Tupaia dorsalis, Tupaia javanica, Tupaia minor, Tupaia tana. J. Vet. Med. Sci. 62(5): 517–520, 2000

The tree shrews consist of about 18 species and have nearest 0.01 mm. The measurement items are indicated in morphologically adapted for various environments and Table 2. The nomenclature of items was based on Driesch behaviors [2, 5, 8, 11, 19]. The argument, so called Tupai- [7] and their abbreviations were arranged in Table 2. Problem, that the tree shrews should taxonomically belong The mean values and standard deviations were calculated to the insectivore [2, 3], or primate [12–18], or the separate from the measurement data. Some measurement ratios to order [1, 6], has attracted many evolutionary the PL were statistically compared among the four species mammalogists. Many articles have considered them as by Student’s t-test to discuss the functional significance of separate order now [4, 5, 20], however, in the taxonomical the skull characters. The principal component analysis was controversy, the comparative morphology of skull has not carried out among four species from all measurement data. been undertaken among species of tree shrews, and the functional diversity in skull character of Tupaia has RESULTS disturbed taxonomists to recognize the evolutionary relationships among tree shrews, insectivores, and primates. The measurement values, measurement ratios and Therefore, we have osteometrically compared the skull standard deviations are shown in Tables 3 and 4. We used 8 specimens in the four shrews as follows: Tupaia tana as measurement ratios in t-test on inter-specific significant mainly terrestrial species, T. javanica as arbo-terrestrial, T. difference (Table 5). minor as mainly arboreal [2, 5, 8, 11, 19], and T. dorsalis as In exception of T. dorsalis, the three species are without behavior data. We examined the functional obviously different in skull size. T. tana is the largest, and significances in skull characters among these species of the T. minor the smallest in all measurements (Table 3). In genus Tupaia. Table 5, we recognize the similarity in many measurement ratios to PL between T. tana and T. dorsalis. However, the MATERIALS AND METHODS Table 1. Composition of species, locality and sex of skull speci- We used totally 71 skulls of four species of tree shrews mens (Tupaia tana, T. javanica, T. minor, and T. dorsalis). These Species Origin of specimens Sex Sample size specimens have been stored in Bogor Zoological Museum and Muséum National d’Histoire Naturelle (Paris, France). Tupaia tana Sumatra, Kalimantan male 10 The sex determination was dependent on the description female 13 of collection biological data. All specimens used in this Tupaia javanica Sumatra, Java male 15 female 13 study have already finished the molar eruption, so, they Tupaia minor Sumatra, Kalimantan male 9 were considered as adult. The composition of species, female 9 origin, and sex are shown in Table 1. The skull Tupaia dorsalis Kalimantan male 1 measurement was carried out with a vernier calliper to the female 1 518 H. ENDO ET AL.

Table 2. List of the measurements and their abbreviations since the binocular sense and recognition are essential for 1.Cranium their life on tree, while T. javanica as arbo-terrestrial Profile length PL animals may be moderate in the skull shape. Condylobasal length CL We quantitatively confirmed the relationship between Short lateral facial length SL skull proportion of the four species (Tables 4 and 5). The Zygomatic width ZW differences of skull proportion between T. javanica, T. Least breadth between the orbits LBO minor and T. tana are certainly in many items (Tables 4 and Greatest neurocranium breadth GNB Median palatal length MPL 5), however, the interspecies relationship is different from Length from Basion to Staphylion LBS that having been expected. The SL/PL is larger in T. tana, Dental length DL however, the LBO/PL is the smallest in this species. We Greatest palatal breadth GPB demonstrated that the distance between both eyes is not Greatest mastoid breadth GMB relatively large in this species, and that arboreal T. minor Height from Akrokranion to Basion HAB does not have a narrow breadth in interorbital space. While, the GNB/PL is much smaller in T. tana than the 2. Mandible Length from the condyle LC other two species. It suggests that T. tana may possess Length from the angle LA smaller neurocranium, however, the LBS/PL and the ZW/ Length from the Infradentale to LIA PL are obviously larger in this species. We can point out aboral border of the alveolus that the T. tana has laterally-narrower neurocranium in the Height of the vertical ramus HR braincase area among Tupaia species. Height of the mandible at M HM 1 The largest SL/PL and the smallest MPL/PL are shown The measurement items were based on Driesch [7]. in T. tana. It demonstrates that T. tana certainly possesses a longer nasal part, however, the slanchnocranium is not large in ventral and oral area. We suggest that the length from the ratios essentially show the significant differences among orbit to the nose is large in this terrestrial species, however, three species. we can conclude that the splanchnocranium is not totally The principal components analyses based on all elongated, but longer only in dorsal side on this species. measurement data are plotted in Fig. 1. The three clusters The principal component charts show the tendency of of T. tana, T. javanica and T. minor are obviously morphological adaptation in these species. Each separated distinguished in this chart. The two points of T. dorsalis are cluster confirms the terrestrial, arbo-terrestrial and mainly- plotted adjacent to the cluster of T. tana. arboreal behavior in the three species, T. tana, T. javanica and T. minor, respectively. In the rare species, T. dorsalis, DISCUSSION the behavior has not substantially examined yet. Our present data suggest that this species may live completely T. tana has been thought to act mainly on the ground on the ground as shown in T. tana, although we could [19]. So, we have expected that this species possesses the measure only two specimens in this species. wide interorbital space and the long facial length and The skull of the common tree shrew (T. glis) has already relatively small ratios in the neurocranium items. It also examined on geographical variation in Thailand and has been predicted that T. minor, belonging to the arboreal Malaysia regions [9]. It has been guessed that this species groups, has the opposite character to terrestrial T. tana, is arbo-terrestrial. However, as we indicated [9], the

Table 3. The measurement value in the four species of tree shrew PL CL SL ZW LBO GNB MPL LBS DL GPB GMB HAB LC LA LIA HR HM T. tana male 58.65 50.8 28.94 23.88 25.22 14.76 20.14 32.49 18.83 30.54 15.5 13 39.37 38.52 24.88 12.6 3.69 1.81 1.43 1.57 2.02 1.25 0.88 0.39 1.48 0.47 1.1 0.46 0.57 1.35 1.28 1.19 0.67 0.35 female 58.32 50.38 28.73 23.01 25.58 14.9 20.23 32.32 18.44 30.56 15.84 12.87 39.53 38.7 24.9 12.93 3.75 2 1.56 1.43 1.63 0.82 0.54 0.49 1.26 0.48 1.12 0.49 0.55 1.34 1.46 0.92 0.55 0.36 T. javanica male 42.31 36.81 16.91 13.57 22.07 12.58 17.35 21.91 15.17 21.35 13.38 10.21 28.19 28.23 17.08 11.83 3.11 1.01 0.79 0.43 0.82 0.58 0.47 0.58 0.6 0.58 0.46 0.35 0.34 0.97 0.81 0.4 0.54 0.22 female 42.83 37.04 17.17 13.7 21.96 12.4 17.5 22.07 15.17 21.42 13.62 10.44 28.25 28.43 17.28 12.1 3.2 1.16 0.71 0.59 0.67 0.84 0.43 0.59 0.5 0.53 0.73 0.49 0.36 0.83 0.87 0.58 0.59 0.21 T. minor male 36.21 31.01 13.38 10.36 19.98 11.73 16.16 17.62 13.36 17.64 11.89 9.18 24.04 23.89 14.34 10.56 2.74 0.67 0.61 0.44 0.58 0.65 0.6 0.22 0.56 0.49 0.42 0.29 0.36 0.58 0.66 0.51 0.37 0.17 female 36.44 31 13.43 10.25 19.7 11.46 16.28 17.61 13.26 17.68 11.9 9.19 24.02 23.84 14.32 10.36 2.7 0.67 0.75 0.3 0.27 0.51 0.47 0.5 0.39 0.42 0.47 0.23 0.29 0.43 0.74 0.36 0.44 0.19 T. dorsalis male 49.45 43 21 17.45 23.15 13.3 18.9 28.4 14.8 26.95 14.1 11.75 33.45 33.1 21.75 10.95 3.05 female 48.25 43.35 24.1 19.65 22.45 12.5 17.75 27.9 14.65 26.5 13.9 11.45 33.7 33.3 22.15 10.35 3.15 The mean values (mm) are shown in the upper rows, and the standard deviations in the lower rows. Only one specimen was used for each sex in T. dorsalis. SKULL OF TREE SHREWS 519

Table 4. The measurement ratios to the PL in the four species of tree shrews SL/PL ZW/PL LBO/PL GNB/PL MPL/PL LBS/PL LC/PL HR/PL T. tana male 49.35 40.71 43.00 25.17 34.34 55.40 67.12 21.49 1.39 2.78 1.42 1.20 0.77 0.95 1.30 0.80 female 49.25 39.45 43.86 25.55 34.69 55.42 67.78 22.17 0.96 2.62 1.20 1.05 1.05 0.64 1.33 1.00 T. javanica male 39.96 32.08 52.17 29.74 41.00 51.79 66.62 27.96 0.91 1.67 1.28 0.88 1.56 0.86 1.51 1.45 female 40.10 31.98 51.28 28.94 40.87 51.52 65.96 28.26 1.36 1.02 1.58 0.86 1.24 0.68 1.21 1.43 T. minor male 36.95 28.60 55.19 32.41 44.62 48.66 66.41 29.17 0.60 1.49 1.38 1.48 0.81 1.12 0.62 0.82 female 36.84 28.13 54.05 31.43 44.66 48.32 65.91 28.43 0.63 1.01 1.27 1.37 1.70 0.81 1.45 1.08 T. dorsalis male 42.47 35.29 46.81 26.90 38.22 57.43 67.64 22.14 female 49.95 40.73 46.53 25.91 36.79 57.82 69.84 21.45 The mean values are shown in the upper rows, and the standard deviations in the lower rows. Only one specimen was used for each sex in T. dorsalis.

Fig. 1. The principal component chart of skulls between the first and second transformed variables from 17 measurement items. The numbers plotted in the chart indicate the species and sex as follows: 1, male Tupaia tana, 2, female T. tana, 3, male T. javanica, 4, female T. javanica, 5, male T. minor, 6, female T. minor, 7, male T. dorsalis, 8, female T. dorsalis. PC1, the first principal component. PC2, the second principal component. The percentage of the variation explained by PC1 is 92.7, and that by PC2 is 2.9. The items, which largely contribute to the first principal component, are PL, LC and LBS in higher ranking, and to the second principal component, are HR, HM and LBO. common tree shrew shows a wide geographical variation in ACKNOWLEDGEMENTS. We thank Dr. Jacques Cuisin, the skull size and shape, so we think that the behavior Muséum National d'Histoire Naturelle (Paris, France) for should be observed in various geographical populations of his kind management in our examinations of Museum T. glis. We also suggested that the skeletal and muscular specimens, and the staffs of the Faculty of Veterinary systems of forelimbs have been adapted for arboreal life in Medicine, Kasetsart University, for their support and the northern smooth-tailed tree shrew (Dendrogale murina) encouragement throughout this work. We are also grateful [10]. Although the skull specimens are very rare in this to Dr. Eiichi Hondo, the Department of Veterinary species, we should morphologically compare the skull Anatomy, Obihiro University of Agriculture and Veterinary between D. murina and T. minor to clarify the arboreal Medicine. A part of this work was financially supported by adaptation for arboreal behavior in Tupaiidae in the future. Grant-in-Aids for Scientific Researches Nos. 05041105, 520 H. ENDO ET AL.

Table 5. Significant differences between species in some measure- REFERENCES ment ratios SL/PL MPL/PL 1. Adkins, R. M. and Honeycutt, R. L. 1991. Molecular phylog- male male eny of the superorder Archonta. Proc. Natl. Acad. Sci. U.S.A. tjm t jm 88: 10317–10321. j+ j + 2. Boonsong, L. and McNeely, J. A. 1988. Mammals of Thai- m++m ++ land, 2nd ed., Saha Karn Bhaet, Bangkok. d+++d+–+ 3. Campbell, C. B. G. 1974. On the phyletic relationship of the female female tree shrews. Mamm. Rev. 4: 125–143. tjm t jm 4. Corbet, G. B. and Hill, J. E. 1991. A World List of Mamma- j+ j + lian Species. 3rd ed., p. 39, Oxford Univ. Press, Oxford. m++m ++ 5. Corbet, G. B. and Hill, J. E. 1992. The Mammals of the d–++d–++ Indomalayan Region: A Systematic Review. Oxford Univ. ZW/PL LBS/PL Press, Oxford. male male 6. Dene, H., Goodman, M. and Prychodko, W. 1978. An immu- tjm t jm nological examination of the systematics of Tupaioidea. J. j+ j + Mamm. 59: 697–706. m++m ++ 7. Driesch, A. 1976. A Guide to the Measurement of Animal d–++d–++ Bones from Archaeological Sites. Harvard Univ. Press, Cam- female female bridge. tjm t jm 8. Ellerman, J. R. and Morrison-Scott, T. C. S. 1951. Checklist j+ j + of Palaearctic Region. Oxford Univ. Press, Oxford. m++m ++ 9. Endo, H., Cuisin, J., Nadee, N., Nabhitabhata, J., Suyanto, A., d–++d+++ Kawamoto, Y., Nishida, T. and Yamada, J. 1999. Geographi- LBO/PL LC/PL cal variation of the skull morphology of the common tree male male shrew (Tupaia glis). J. Vet. Med. Sci. 61: 1027–1031. tjm t jm 10. Endo, H., Rerkamnuaychoke, W., Kimura, J., Sasaki, M., j+ j – Kurohmaru, M. and Yamada, J. 1999. Functional morphology m++m –– of the locomotor system in the northern smooth-tailed tree d+++d––– shrew (Dendrogale murina). Ann. Anat. 181: 397–402. female female tjm t jm 11. Hill, J. E.1960. The Robinson collection of Malaysian mam- j+ j + mals. Bull. Raffles Mus. 29 : 1–112. m++m +– 12. Le Gros Clark, W. E. 1924. The myology of the tree shrew d–++d–++ (Tupaia minor). Proc. Zool. Soc. Lond. 1924: 461–496. 13. Le Gros Clark, W. E. 1924. On the brain of the tree shrew GNB/PL HR/PL male male (Tupaia minor). Proc. Zool. Soc. Lond. 1924: 1053–1074. tjm t jm 14. Le Gros Clark, W. E. 1925. On the skull of Tupaia. Proc. j+ j + Zool. Soc. Lond. 1925: 559–567. m++m ++ 15. Le Gros Clark, W. E. 1959. The Antecedents of Man. 1st ed., d–++d–++ Edinburgh Univ. Press, Edinburgh. female female 16. Le Gros Clark, W. E. 1971. The Antecedents of Man. 3rd ed., tjm t jm Quadrangle, Chicago. j+ j + 17. Simpson, G. G. 1945. The principles of classification and a m++m +– classification of mammals. Bull. Am. Mus. Natl. Hist. 85: 1– d–++d–++ 360. + P<0.05. – No significant difference. t, T. tana; j, T. javanica; m, T. 18. Sorenson, M. W. 1970. Behavior in the tree shrews. Primate minor; d, T. dorsalis. Behav. 1: 141–194. 19. Walker, E. P. 1983. pp. 354-356. Walker’s Mammals of the 08041132 and 11833017 from the Ministry of Education, World. 4th ed. (Nowak, R. M. and Paradiso, J. L. eds.), Johns Science and Culture, Japan, and by the Asia-Pacific Hopkins Univ. Press, London. cooperative research program of National Science Museum, 20. Wilson, D. E. and Reeder D. A. 1993. Mammal Species of the World; A Taxonomic and Geographic Reference. 2nd ed., Tokyo. Smithsonian Institution Press, Washington.