Frogs of the genus Microhyla include some of the world’s smallest amphibians and represent the largest radiation of Asian microhylids, currently encompassing 50 species, distributed across the Oriental biogeographic region. The genus Microhyla remains one of the taxonomically most challenging groups of Asian frogs and was found to be paraphyletic with respect to large-sized fossorial Glyphoglossus. In this study we present a time-calibrated phylogeny for frogs in the genus Microhyla, and discuss taxonomy, historical biogeography, and morphological evolution of these frogs. Our updated phylogeny of the genus with nearly complete taxon sampling includes 48 nominal Microhyla species and several undescribed candidate species. Phylogenetic analyses of 3,207 bp of combined mtDNA and nuDNA data recovered three well-supported groups: the Glyphoglossus clade, Southeast Asian Microhyla II clade (includes M. annectens species group), and a diverse Microhyla I clade including all other species. Within the largest major clade of Microhyla are seven well-supported subclades that we identify as the M. achatina, M. fissipes, M. berdmorei, M. superciliaris, M. ornata, M. butleri, and M. palmipes species groups. The phylogenetic position of 12 poorly known Microhyla species is clarified for the first time. These phylogenetic results, along with molecular clock and ancestral area analyses, show the Microhyla—Glyphoglossus assemblage to have originated in Southeast Asia in the middle Eocene just after the first hypothesized land connections between the Indian Plate and the Asian mainland. While Glyphoglossus and Microhyla II remained within their ancestral ranges, Microhyla I expanded its distribution generally east to west, colonizing and diversifying through the Cenozoic. The Indian Subcontinent was colonized by members of five Microhyla species groups independently, starting with the end Oligocene—early Miocene that coincides with an onset of seasonally dry climates in South Asia. Body size evolution modeling suggests that four groups of Microhyla have independently achieved extreme miniaturization with adult body size below 15 mm. Three of the five smallest Microhyla species are obligate phytotelm-breeders and we argue that their peculiar reproductive biology may be a factor involved in miniaturization. Body size increases in Microhyla—Glyphoglossus seem to be associated with a burrowing adaptation to seasonally dry habitats. Species delimitation analyses suggest a vast underestimation of species richness and diversity in Microhyla and reveal 15–33 undescribed species. We revalidate M. nepenthicola, synonymize M. pulverata with M. marmorata, and provide insights on taxonomic statuses of a number of poorly known species. Further integrative studies, combining evidence from phylogeny, morphology, advertisement calls, and behavior will result in a better systematic understanding of this morphologically cryptic radiation of Asian frogs.

We examine the phylogenetic structure and morphological differentiation within the Hynobius kimurae–H. boulengeri species complex—a widely-distributed group of stream-breeding hynobiid salamanders, inhabiting montane areas of western, central and eastern parts of Honshu Island, Japan. Phylogenetic relationships were assessed based on analyses of mitochondrial (mtDNA) and nuclear (nuDNA) gene fragments for a total of 51 samples representing 23 localities covering the entire range of the species complex. Morphological study included one-way analysis of variance (ANOVA) and principal components analysis (PCA) analyses of 26 morphometric and six meristic characters for 38 adult specimens of H. kimurae and three adult specimens of H. boulengeri. MtDNA genealogy supported monophyly of the H. kimurae–H. boulengeri complex, which is sister to all other Hynobius except H. retardatus. The complex is subdivided into three major clades, corresponding to the Eastern (Clade I) and Western (Clade II) populations of H. kimurae sensu lato, and to H. boulengeri (Clade III). Monophyly of H. kimurae sensu lato is only moderately supported by mtDNA, while nuDNA suggested that the Western form of H. kimurae is closer to H. boulengeri than to the eastern form. The time of the split of the H. kimurae–H. boulengeri complex is estimated as late Miocene and coincides with intensive crust movement in western Japan. Divergence between Clades I and II took place in early Pliocene and was likely influenced by the uplift of Central Japanese Highlands. All three clades were found to be different in a number of morphological characters, allowing us to describe the eastern form of H. kimurae as a new species, Hynobius fossigenus sp. nov.

Bufonid toads generally possess cardiotoxic steroids called bufadienolides as defensive chemicals. Although knowledge of the life stages at which the toad species possess the poison is important for our understanding of diversity of toxicity among bufonid toads, this knowledge is limited. In the present study, we revealed that the Japanese common toad, Bufo japonicus formosus, possesses toxins at the unfertilized egg stage by conducting a bioassay experiment. Recent studies documented that hatchlings of B. j. formosus have lethal toxic effects on native frog tadpoles (Rana pirica) in the invasive area of the toad (Hokkaido). In our bioassay experiment using R. pirica tadpole as a predator, no tadpoles died when they did not consume any prey item during two-days experimental period. However, approximately 90% of R. pirica tadpoles immediately died when they consumed an unfertilized egg of B. j. formosus. These results suggest that the toxin at the early life stages of B. j. formosus is, at least partly, provided from female parent.

Biological invasions adversely affect the survival of many native species, but long-term consequences of the novel pressures imposed by invaders on natives are less clear. If natives can adapt to the presence of the invader, the severity of impact will decline with time. On Hokkaido, Japan, alien toads (Bufo japonicus formosus) are highly toxic to native frog tadpoles (Rana pirica) that attempt to eat their hatchlings. Therefore, the arrival of toads potentially imposes selection pressure on native frogs in the context of behavioral response (feeding aversion) and physiological resistance to toad toxins. Here, we compared the consumption tendency and toxin resistance of R. pirica from several sites with different histories of toad invasion (0 to >100 years exposure). We found no evidence of adaptive response to invasive toads. The vast majority of R. pirica tadpoles from all sites readily consumed toad hatchlings and died, regardless of time since invasion or whether hatchlings were from sympatric versus allopatric toad populations. In contrast, tadpoles of a closely related species (Rana ornativentris), which co-occurs naturally with B. j. formosus on Honshu, readily consumed toad hatchlings without ill effect. Our results indicate that the lethal toxic effect of the alien toads on native frogs on Hokkaido is maintained over long time periods without mitigation through adaptation.

Toads of the Family Bufonidae possess neurotoxins (bufadienolides) which are generally considered to be synthesized de novo. Thus, invasive toad species pose a threat to native predators via toxic effects. However, the influence of diet on toad toxicity is poorly understood. We evaluated the effect of diet on toxicity of embryos of the invasive toad Bufo japonicus formosus in Japan, using native tadpoles (Rana pirica) as a bioassay. Specifically, we compared the toxicity of embryos spawned by captive toads reared entirely on non-toxic prey versus embryos spawned by wild toads. Embryos of captive-reared toads and wild toads were comparable in terms of toxic effects: all tadpoles that consumed a toad embryo died, irrespective of origin of embryos. Embryo toxicity appears to be via parental (likely maternal) provisioning. Our study indicates that adult B. j. formosus produce toxins (likely bufadienolides) in amounts sufficient for embryos to be fatal to native predator tadpoles irrespective of whether the adult diet contains toxins. Thus, the toxic effects of invasive B. j. formosus on R. pirica, and possibly on other sensitive native predator species, are likely to be widespread and occur regardless of variation in prey biota in invaded areas in Japan.