We investigated genetic variation in mitochondrial cytochrome b within the long-tailed rice field mouse, Mus caroli Bonhote, 1902, across its entire geographic range in Southeast and East Asia with a view to: 1) assessing the pattern and causality of phylogeographic structure in a terrestrial small mammal from continental Southeast Asia; and 2) distinguishing genuine insular relics from cases of human-assisted translocation. We identified five main mtDNA lineages which show a similar level of differentiation as the subspecies of M. musculus and probably diverged during the Middle Pleistocene. Two of the lineages are restricted to large islands (Taiwan and Java) and their existence is explicable in terms of regional palaeogeographic factors including changes in sea level and climate. The remaining lineages are distributed in different regions on mainland Southeast Asia but vicariant explanations are inappropriate given the relatively short time frame. Dispersal across barriers followed by local differentiation probably explains the observed phylogeographic patterning on the mainland. A close genetic link between Okinawan M. caroli and populations in Laos confirms previous suggestions that people carried this species to the Ryukyu Archipelago. However, more intensive regional sampling is needed to identify a precise source area.

Random amplified polymorphic DNA (RAPD) markers were used to estimate the level and pattern of genetic diversity in Microtus fortis pelliceus populations from the mainland and islands of the Russian Far East. No markers were found both for individual populations and for the mainland or island voles as a whole. However, they appeared to be different in both their allele frequencies and microevolution mode based on correlation pleiad analysis. The island populations demonstrated a higher level of genetic differentiation among themselves than did mainland populations and each mainland population probably represents more genetic diversity than any population of islands. Overall, genetic variation between island and mainland M. f. pelliceus populations was low (Dst = 0.003), and can be explained by a low value of gene differentiation coefficient (Gst = 0.058). Nevertheless, exact test did not support the hypothesis that both united mainland and united island populations belong to the same genetic unit (p = 0.0025). RAPD data reliably distinguished voles of Matveeva Island, which are strongly differentiated by morphological traits, and this in turn did not exclude the island speciation event. Unlike UPGMA and NJ trees for individuals, population UPGMA strongly separated island voles from the mainland conspecifics, and MS reconctructions identified the Amur population as an ancestor.

To examine the origin of Callosciurus erythraeus introduced to Japan, we compared mitochondrial DNA control region sequences of eight haplotypes from Japan with those of 42 haplotypes from Taiwan. There were two distinct phylogroups in Japan. Six haplotypes from Japan were included in the Taiwan population, suggesting that they could have been introduced from Taiwan. Two haplotypes, however, were distantly related to the cluster consisting of all Taiwanese haplotypes and six Japanese haplotypes. The uncorrected genetic distances (5.2–5.7%) between these two haplotypes and the out-group (Callosciurus finlaysonii from Laos) were less than those (8.3–9.8%) between these two haplotypes and the cluster consisting of the other six Japanese haplotypes and all Taiwanese haplotypes. Both C. erythraeus and C. finlaysonii have variable external characteristics, such as pelage color, and include many variable subspecific forms, so that it is difficult to evidently identify both species on the basis of their external characteristics. Callosciurus finlaysonii may have been also introduced in Japan with both species now coexisting in Japan.

The bony labyrinth of specimens representing eight diprotodontian species were visualized by high-resolution computed tomography. Linear measurements of the labyrinth were taken, e.g., the height and width of the arc of each semicircular canal. The relative sizes and spatial arrangements of the semicircular canals were compared and some of the variation was atomized into 17 characters, which were then phylogenetically interpreted. There has been a change both in size and in relative arrangement of the semicircular canals that for some aspects maps onto the ecological change from arboreality to terrestriality. In particular, there are differences among diprotodontians in the height of the anterior semicircular canal in relation to the posterior one. In arboreal species, the lateral semicircular canal is relatively longer than the equivalent semicircular canals of terrestrial species. A rounder anterior semicircular canal is widespread for Diprotodontia with a shift in Pseudocheiridae, where it is more flattened. Dendrolagus shows features typical of terrestrial species in spite of its arboreal lifestyle. The fact that it shows the derived character state is congruent with the fact that it has secondarily and only recently evolved an arboreal lifestyle.

The manipulating action of the lesser panda (Ailurus fulgens) was quantitatively analyzed by the CT and three-dimensional digitalized techniques. The forearm, metacarpals, radial sesamoid bone and accessory carpal bone were considered as movable parts in the simulation of the dead body, and we examined the change of angles among these four segments in the grasping action. In the manipulation the radial sesamoid bone moved independently from the metacarpals in the lesser panda. It was not articulated to the first metacarpal bone, but supported by some muscle tendons and the other soft parts of the wrist and palm regions. Although the grasping apparatus consists of the functional units of the radial sesamoid bone and accessory carpal bone in the giant panda (Aliluropoda melanoleuca), the manipulating mechanism using the radial sesamoid bone in the lesser panda had no functional-morphological similarities to that in the giant panda. The radial sesamoid bone as a manipulating tool of the lesser panda has been developed in the separated evolutionary history of this species.