Morphological adaptation of the skull for various behaviors in the tree shrews.

Skull size and shape were examined among 14 species of the tree shrews (Tupaia montana, T. picta, T. splendidula, T. mulleri, T. longipes, T. glis, T. javanica, T. minor, T. gracilis, T. dorsalis, T. tana, Dendrogale melanura, D. murina, and Ptilocercus lowii). The bones of face were rostro-caudally longer in T. tana and T. dorsalis, contrasting with T. minor and T. gracilis, D. melanura, D. murina and P. lowii which have smaller facial length ratios. The arbo-terrestrial species (T. longipes and T. glis) were similar to terrestrial species in length ratios of bones of face unlike the other arbo-terrestrial species (T. montana, T. picta, T. splendidula, and T. mulleri). We propose that T. longipes and T. glis have adapted to foraging for termites and ants as have T. tana and T. dorsalis. Additionally small body size in T. javanica may be the result of being isolated in Java. We separated the species into 5 groups from the measurment values of skulls: 1) Terrestrial species; T. tana and T. dorsalis, 2) Arboreal species; T. minor and T. gracilis, 3) Arbo-terrestrial species group 1: T. montana, T. splendidula, T. picta and T. mulleri, and T. javanica, 4) Arbo-terrestrial species group 2: T. glis and T. longipes, 5) Arboreal species of Dendrogale and Ptilocercus. Principal component analysis separated species into 8 clusters as follows: 1) T. tana, 2) T. dorsalis, 3) T. montana, T. splendidula, T. picta and T. mulleri, 4) T. glis and T. longipes, 5) T. javanica, 6) T. minor and T. gracilis, 7) D. melanura and D. murina, and 8) P. lowii. We suggest that these clusters correspond to behavioral strategies and peculiarities observed in foraging, feeding and locomotion in each species.