Three Spiny Frog Tadpoles Sharing One Stream Identified by DNA in Wuyishan

Deep in the mountain streams of Wuyishan, China, herpetologists have pulled off a small forensic triumph: they caught tadpoles of three different spiny frog species living side by side in the same water, and worked out exactly which baby belonged to which adult. The achievement matters because tadpoles of closely related frogs are notoriously difficult to tell apart, and one of the three species had never had its larval form described at all. The study, published in the journal Herpetozoa, combines DNA barcoding with detailed photographs of live specimens to give scientists a practical field guide to three of southern China’s most iconic stream frogs.

Spiny frogs of the genus Quasipaa are heavyweight residents of mountain torrents across southern China and the Indochina Peninsula. Their tadpoles are large, bottom-dwelling creatures that cling to rocks by day and emerge at night to feed opportunistically on whatever the stream offers. Many of these larvae are remarkable in their own right: most Quasipaa tadpoles overwinter, with some, such as those of Q. robertingeri, potentially spending two full seasons in the larval state before metamorphosing. The largest tadpole ever reported for the genus, a Q. verrucospinosa specimen at Gosner stage 40, stretched an astonishing 91 millimeters in total length. Yet despite their size and ecological importance, the larval forms of eight of the genus’s fifteen recognized species remain unknown to science.

Three Quasipaa species are known from Wuyishan: Q. exilispinosa, the little spiny frog; Q. spinosa, the giant spiny frog; and Q. jiulongensis, the Jiulong spiny frog. Tadpoles of the first two had been described decades ago, but Q. jiulongensis posed a stubborn problem. Its range overlaps extensively with those of its two relatives, so any unidentified tadpole found in their shared streams could not be confidently assigned to the species. As a result, the larva of Q. jiulongensis remained officially undescribed, a gap in knowledge that has persisted since the frog was first named in 1985.

The breakthrough came during a rapid herpetological survey on 26 July 2023, when researchers collected four tadpole specimens from Dazhulan, Guadun, in Wuyishan, Fujian Province, at roughly 950 meters elevation. The specimens were photographed alive in a transparent acrylic box, preserving coloration details that vanish in preservative, before being euthanized with buffered clove oil and fixed in 10 percent formalin. Tail fin tissue was preserved in 95 percent ethanol for molecular work, and all specimens were deposited at the Chengdu Institute of Biology of the Chinese Academy of Sciences under voucher numbers CIB T1024 through T1027.

Species identification rested on segments of the 16S ribosomal RNA gene, a workhorse marker in amphibian taxonomy. The team amplified the gene using established primers, sequenced the products on an automated DNA sequencer, and searched the resulting sequences against public databases, which returned approximate matches to the three target species. Because hybridization has been documented within Quasipaa, the researchers could not rely on a single match; they instead built a phylogenetic framework using 26 samples representing 11 known species and an undescribed lineage of the genus, with Fejervarya limnocharis as the outgroup. Sequences were aligned with the MUSCLE algorithm in MEGA 12, and a Bayesian inference tree was constructed in MrBayes 3.2.7a with two independent runs of 10 million generations each, sampled every 1000 generations, discarding the first 25 percent as burn-in and using a convergence threshold of PSRF no greater than 1.005.

The phylogeny delivered unambiguous answers. Specimen CIB T1027 nested firmly within a monophyletic Q. jiulongensis clade that included sequences from the same locality as the new samples. Specimens CIB T1024 and CIB T1025 clustered with 13 Q. exilispinosa samples from Wuyishan and one from Dehua, Fujian, with strong Bayesian posterior probability support of 1.00. The situation for Q. spinosa was more delicate, since samples from its type locality split into multiple lineages; the team treated the clade containing only Q. spinosa samples from Zhejiang, Fujian, and Hunan provinces, also supported at 1.00, as the true lineage, and their tadpole CIB T1026 fell squarely within it. With identities settled, the researchers turned to morphology, measuring the specimens from photographs taken beside a ruler using ImageJ software, staging them by the Gosner system, and applying the standard terminology of Altig and McDiarmid.

The descriptions reveal three larvae with subtly different builds and mouths. The Q. exilispinosa tadpole at stage 25 measured 46.6 millimeters in total length with an oval body and a keratodont row formula of 1:4+4/1+1:2, meaning four complete tooth rows plus a split fourth on the upper labium and one complete plus a split second row on the lower. The Q. spinosa tadpole, at stage 37 and 72.6 millimeters total length, was substantially larger, with a stronger tail muscle and a spiracle opening closer to the snout than to the anal tube. The newly described Q. jiulongensis tadpole, at stage 25 and 33.2 millimeters, had a relatively weak tail muscle and a distinctive lower labium in which the inner papilla row was only slightly longer than the outer, whereas in the other two species the inner row was roughly twice as long. All three shared keratinized jaw sheaths with fine serrations and a thin upper sheath covering the lower.

Live coloration proved to be the most striking and field-friendly discriminator. The Q. jiulongensis tadpole lacks large spots on the upper tail musculature, its tail instead bearing cloud-form dark speckles with a darkened middle area at the body-tail junction when viewed from above. The Q. spinosa tadpole displays a dark stripe at the body-tail junction in dorsal view but lacks large spots on the upper tail muscle. The Q. exilispinosa tadpole, by contrast, carries large spots on the upper tail musculature while lacking the dark stripe at the junction. All three show goldish lateral lines, bronze irises with black pupils, and semi-translucent undersides, with pink chests in the early-stage larvae and goldish pigments around the mouth corners.

The authors note that previous descriptions of Quasipaa tadpoles relied heavily on line drawings, which often omit the color and pigmentation details that are critical for field identification. Their documentation of live coloration, following recent practice in the field, should make in situ tadpole monitoring far more practical, particularly for Q. exilispinosa, whose dark dorsolateral spots on the tail muscle are reportedly most distinct in early-stage larvae. Coloration patterns on the tail and fins typically become more prominent as tadpoles develop, so the team cautions that their observations, based on single specimens of each species, represent initial snapshots rather than definitive diagnoses; unobserved variation in color patterns may exist and should be the focus of further investigation.

The comparative diagnosis also highlights how much remains to be learned about the genus. The keratodont row formula of the newly described Q. jiulongensis tadpole overlaps with those of Q. exilispinosa, Q. spinosa, Q. boulengeri, and Q. robertingeri, so mouthparts alone cannot separate these species; the papilla proportions and coloration carry the diagnostic weight. The team also found that their Q. spinosa specimen, at 27.3 millimeters snout-vent length, was larger than the 18.7 to 22.1 millimeter range reported by earlier authors, potentially indicating developed pigmentation associated with preparation for metamorphosis, and that earlier accounts of transparent abdomens with visible gut coils did not match their specimens. With three sympatric species co-occurring in the same stream, the study provides the first field-usable framework for distinguishing their larvae, and the authors suggest that future work on internal morphology, together with careful analysis of samples from hybrid populations of the phylogenetically complex Q. spinosa, will be essential to avoid misidentification as the genus’s larval diversity continues to be uncovered.

Subject of Research: Tadpoles of three sympatric spiny frogs (Anura, Dicroglossidae, Quasipaa) from Wuyishan, China

Article Title: Tadpoles of three sympatric spiny frogs ( Anura , Dicroglossidae , Quasipaa ) from Wuyishan, China

Article References: Qian, T., Li, C., Huang, S., Chen, B., Guo, Y., & Jiang, J. (2025). Tadpoles of three sympatric spiny frogs (Anura, Dicroglossidae, Quasipaa) from Wuyishan, China. Herpetozoa, 38, 311-320. https://doi.org/10.3897/herpetozoa.38.e162906

Image Credits: AI Generated

DOI: 10.3897/herpetozoa.38.e162906

Keywords: Tadpoles, three, sympatric, spiny, frogs, Anura, Dicroglossidae, Quasipaa, Wuyishan, China, scientific research