Hidden treefrog species split by six million years of evolution in northeast Asia

One of northeast Asia’s most familiar frogs is not one species at all. A new integrative study of the Dryophytes japonicus complex, published in the journal Herpetozoa, has combined mitochondrial barcoding, genome-wide nuclear sequencing, morphometrics, and bioacoustics to show that the northeast Asian treefrog is actually two deeply diverged species, each with two candidate subspecies. The finding rewrites the evolutionary map of a frog that ranges from Mongolia across the Russian Far East, the Korean Peninsula, China, and most of the Japanese archipelago, and it carries immediate consequences for how conservationists should monitor its populations.

The research team, led by Amaël Borzée of Nanjing Forestry University with collaborators including Christophe Dufresnes and Yikweon Jang, sampled the complex between 2016 and 2019 in six countries: the Republic of Korea, the Democratic People’s Republic of Korea, China, Russia, Mongolia, and Japan. Fieldworkers located calling males, recorded their advertisement calls with linear PCM recorders and unidirectional microphones, caught and measured individuals with digital callipers to the nearest 0.1 millimetre, and collected genetic material through oral swabs stored at minus 20 degrees Celsius. In total, 294 new individuals were barcoded for four mitochondrial gene fragments, alongside call recordings from 194 individuals and morphometric data from 314 frogs.

The mitochondrial evidence was decisive. Sequences of roughly 900 base pairs of cytochrome-b, 450 base pairs of COI, about 1,000 base pairs of the 16S-ND1 region, and 900 base pairs of 12S rRNA, concatenated into a 3.2-kilobase alignment, revealed two major clades. Clade A occupies Sakhalin Island in Russia, northeastern Japan, and Central Japan, while clade B lives in southwestern Japan and across the Asian mainland. The two clades diverged during the Late Miocene, approximately 6.1 million years ago, with a 95 percent highest posterior density interval of 5.3 to 7.0 million years. Notably, the researchers barcoded both matrilines in Kyoto, pinpointing a contact zone in central Japan where the two species meet.

Genome-wide nuclear data independently confirmed the split. Using RAD-sequencing, the team assembled a 42.8-kilobase nuclear alignment containing 2,302 single nucleotide polymorphisms. Three phylogenetic reconstruction methods — maximum likelihood with PhyML, time-calibrated Bayesian inference with BEAST, and the coalescent-based species tree method SNAPP applied to variable sites only — produced identical topologies with fully supported internal nodes. The nuclear divergence was dated to 5.9 million years ago, closely matching the mitochondrial estimate, with a net genetic distance of 0.23 percent across the RAD-seq alignment. This convergence between mitochondrial and multilocus nuclear evidence rules out the possibility that the earlier split was an artefact of single-gene inference, a known pitfall in anuran phylogeography.

Within each species, the story is subtler. Clade B, now assigned the name Dryophytes japonicus, splits into Japanese populations (subclades B1, B2, and B3) and mainland populations (subclades B4 and B5), separated around 1.2 million years ago during the Pleistocene. The mainland lineage B4 is restricted to the southern Korean Peninsula and Jeju Island, while B5 extends along the Baekdudaegan mountain range and northwards into Mongolia, China, and Russia, with three haplogroups mapped within it. Clade A, provisionally labelled D. cf. japonicus because no available scientific name fits it, divides into a northern group (A1 and A2, spanning Sakhalin, Hokkaido, and northern Honshu) and a southern group (A3 and A4, in Central Japan and Sado Island), separated roughly 2.7 million years ago.

Phenotypic analyses added an unexpected twist. Principal component analysis of call properties — high and dominant frequencies, note duration, effective amplitude, and fundamental frequency — showed significant differences among the four subclades, with all MANOVA results significant at p below 0.001. Morphometric PCA, built from twelve characters including head width, hind-limb length, and tympanum diameter, also yielded significant differentiation. Yet phenotypic divergence did not track molecular divergence cleanly. The deeply diverged northern Japanese group A1&A2 and the mainland group B4&B5, separated by sea straits and belonging to distinct species, remain phenotypically similar, possibly reflecting convergence towards similar ecological optima. Conversely, parapatric subclades such as A3&A4 and B1&B2&B3 show significant phenotypic differences, a pattern the authors suggest could fit a hypothesis of reinforcement, where character shifts in mating traits reduce the costs of interbreeding.

Taxonomic decisions hinged on a historical puzzle. The oldest name for the complex, Hyla arborea var. japonica, was described by Günther in 1858 from syntypes collected by Siebold and Bürger with the vague type locality of Japan. Because both clades inhabit the country, the name could apply to either species. The researchers relied on two anecdotal clues: Siebold was posted to Dejima Island off Kyushu, the only area easily accessible to westerners in mid-nineteenth-century Japan, and Boulenger, a contemporary of Günther, explicitly distinguished Hyla stepheni from the Japanese H. arborea var. japonica by referring to Kyushu. The team therefore assigned the name D. japonicus to the southern Japanese and mainland species, recommending Northern Asian treefrog as its common English name, and provisionally called the mainland subspecies D. j. cf. stepheni, with Geomun’s treefrog as a common name chosen to avoid colonial heritage. The northern species remains unnamed as D. cf. japonicus.

Conservation assessments followed the IUCN Red List categories and criteria for each delimited lineage. Despite relatively narrow ranges for several lineages, none currently crosses the threshold for listing as threatened. Population sizes are likely above 10,000 individuals, declines are unlikely to reach the 30 percent threshold under criterion A, and geographic ranges are unlikely to fall below 20,000 square kilometres under criterion D. However, the authors caution that narrow mitochondrial lineages such as A3 on Sado Island, and B1, B2, and B3 in Japan, may fit threatened criteria if assessed separately, and observed declines in northern China, eastern Mongolia, and adjacent Russia warrant further surveys. Climate change poses a moderate-to-high risk under the 1.5 degree Celsius scenario, with air temperatures in northeast Asia predicted to rise by 2.7 degrees Celsius and humidity by 3.3 percent.

The broader lesson is one that echoes far beyond treefrogs. Species complexes listed as abundant can mask threats to individual lineages, much as tigers are divided into six evolutionary clades with wildly different conservation situations, from increasing populations in Nepal to the extinct Caspian tiger. The authors argue that conserving the full diversity of the D. japonicus complex will require proper population monitoring and additional genomic investigations to determine whether the unnamed delimited lineages merit formal taxonomic descriptions. Half of the phylogeographic diversity in this complex has now received taxonomic attention; the other half, and the evolutionary history it represents, still hangs in the balance.

Subject of Research: Evolutionary biology and conservation of the Dryophytes japonicus treefrog complex

Article Title: Integrating phylogeographic and phenotypic evidence to delimit deep evolutionary lineages in the Dryophytes japonicus species complex, with an assessment of their conservation needs

Article References: Borzée, A., Jang, Y., Othman, S. N., Groffen, J., Maslova, I. V., Purevdorj, Z., Yasumiba, K., Shimada, T., Yi, Y., Schepina, N. A., Andersen, D., Bae, Y., Miura, I., & Dufresnes, C. (2025). Integrating phylogeographic and phenotypic evidence to delimit deep evolutionary lineages in the Dryophytes japonicus species complex, with an assessment of their conservation needs. Herpetozoa, 38, 25-42. https://doi.org/10.3897/herpetozoa.38.e137747

Image Credits: AI Generated

DOI: 10.3897/herpetozoa.38.e137747

Keywords: Dryophytes japonicus, species delimitation, phylogeography, RAD-sequencing, mitochondrial barcoding, bioacoustics, morphometrics, integrative taxonomy, IUCN Red List, amphibian conservation, northeast Asia, cryptic species