Frog Calls Reveal Hidden Species and a New Range in Guizhou

Deep in the subtropical forests of Nayong County, Guizhou Province, China, a small horned toad has been quietly singing a song that scientists had never systematically documented before. A new study published in the journal Herpetozoa reports the first distribution record of the recently described species Boulenophrys daxuemontis in Guizhou Province, along with the first complete description and quantification of its advertisement call. The findings matter far beyond one county border: they show how the sound of a frog’s voice can settle the identity of species that look almost identical, in one of the most rapidly growing genera in amphibian science.

Boulenophrys daxuemontis was only formally described in 2025, with its type locality in Junlian County, Sichuan Province. The genus Boulenophrys, the Chinese horned toads, is the largest clade of the family Megophryidae and currently comprises 81 species distributed mainly across southern China, extending southward into Vietnam, Laos, Thailand and Myanmar. The group is a textbook example of underestimated diversity: molecular phylogenetic frameworks have repeatedly proposed cryptic species, and in the past five years alone, 40 new species have been described in the genus, accounting for roughly half of all currently recognized members. Most of those descriptions, however, rest on morphology and genetics rather than on what the animals actually sound like.

That reliance on body form creates a problem. Many Boulenophrys species show only subtle interspecific morphological differences, and certain plastic traits such as body coloration can vary with environmental conditions, making accurate identification difficult, especially for cryptic or recently diverged taxa. Advertisement calls, the reproductive vocalizations male frogs use to attract mates, offer a way around this obstacle. Nearly all anuran species produce vocal signals, and because advertisement calls are species-specific, they serve as a reliable basis for the systematics and identification of cryptic species. In cases where morphological differentiation is limited or ambiguous, acoustic diagnostic features can be a valuable complementary tool to morphometric and molecular approaches.

During field surveys in 2025, the research team investigated Boulenophrys populations in Nayong County, Guizhou Province, at a site roughly 1760 meters above sea level, where the vegetation consisted of subtropical mixed coniferous and broad-leaved forest dominated by ferns and broadleaf shrubs under dense canopy cover. On the night of August 11, 2025, between 19:00 and 23:00 hours, with an air temperature of 20 degrees Celsius and 90 percent ambient humidity, the team recorded 297 advertisement calls from four calling males of B. daxuemontis. Each call was captured with a SONY ICD-PX470 digital recorder at a sampling rate of 44.1 kHz and 16-bit resolution, from a distance of just 0.5 meters, and only from isolated individuals without overlapping calls or mixed choruses. The recordings are publicly available on Figshare, a transparency measure that lets other researchers verify and reuse the acoustic data.

Species identification did not rest on appearance alone. The team measured snout-vent lengths of all recorded males in situ with a precision digital calliper, and collected one male of each species for molecular confirmation. Cytochrome oxidase subunit I (COI) gene fragments were amplified, purified, and sequenced, and the resulting sequences were deposited in GenBank, with accession number PZ142573 for the Nayong specimen. The molecular results confirmed that the animals calling in Nayong County were indeed B. daxuemontis, a species whose new Guizhou record lies approximately 140 kilometers from its Sichuan type locality, in a border region shared by Sichuan, Guizhou and Yunnan Provinces. The discovery raises the number of amphibian species in Guizhou Province and suggests that the province’s amphibian diversity remains underestimated, with the species potentially distributed in other nearby areas as well.

The acoustic analysis, performed with Raven Pro 1.6 using call-centred terminology, revealed that the advertisement call of B. daxuemontis consists of a series of repeated pulsative calls. The mean call duration was 144.91 ± 24.55 milliseconds, the mean call interval 373.36 ± 107.62 milliseconds, and the mean call repetition rate 154.69 ± 21.60 calls per minute. Each call contained a mean of 12.00 ± 1.00 pulses delivered at a mean pulse rate of 79.31 ± 16.70 pulses per second, and the mean dominant frequency was 3.56 ± 0.12 kHz. Structurally, call amplitude was consistent across call groups except for the initial one or two calls, which showed lower amplitude, and each call began with relatively low amplitude before suddenly rising to its peak and then gradually declining until the end. Males called sporadically rather than in continuous choruses, often responding to the calls of nearby males, and were highly vigilant, ceasing calling immediately at any disturbance.

To test whether these acoustic signatures could distinguish B. daxuemontis from its closest relative, the team also recorded 223 calls from four males of the sister species Boulenophrys jiangi at its type locality in Suiyang County, Guizhou Province, at about 1430 meters elevation. The call structure of B. jiangi proved broadly similar, but the numbers told a different story. B. jiangi calls were shorter, with a mean duration of 113.27 ± 22.53 milliseconds, and contained roughly twice as many pulses, a mean of 24.00 ± 2.00, delivered at a dramatically higher mean pulse rate of 214.96 ± 48.99 pulses per second. The mean dominant frequency of B. jiangi was 3.42 ± 0.25 kHz, and its call groups contained far more calls, ranging from 5.00 to 15.00 per group compared with 2.00 to 5.00 for B. daxuemontis.

A Mann-Whitney U test, chosen because the acoustic data from at least one species did not conform to a normal distribution, revealed statistically significant differences between the two species in every call parameter except the call repetition rate. Call duration, call interval, dominant frequency, frequency 5 percent, and frequency 95 percent were all significantly higher in B. daxuemontis, while the number of calls per call group, number of pulses per call, and pulse rate were significantly lower. A principal component analysis then condensed the nine measured parameters into five principal components explaining a cumulative 95.77 percent of the variance. The first component, accounting for 45.12 percent of the variation, was driven primarily by temporal parameters, specifically the number of pulses per call and pulse rate, with factor loadings of -0.88 and -0.89 respectively. Scatter plots of the first two components showed complete separation between the two species along the PC1 axis, with considerable overlap along PC2, which was driven by spectral parameters such as dominant frequency and frequency 95 percent.

The practical implication is striking: although spectral parameters also differed significantly between species, pulse-related characteristics are the primary acoustic features separating the advertisement calls of these two morphologically similar sister species. In other words, the rhythm of a frog’s call, the sheer speed at which it packs sound pulses into each note, works as a fingerprint that body measurements cannot match. Compared with published calls of other Boulenophrys species in Guizhou Province, the call of B. daxuemontis differs considerably, with a longer call duration than B. chishuiensis and B. congjiangensis, and fewer calls per call group than B. anlongensis, B. jiangi, B. leishanensis, B. qianbeiensis and B. spinata, reinforcing the specific identity of the newly recorded population.

The authors are candid about the limits of their dataset. The advertisement call samples were not collected from the type locality of B. daxuemontis, and the acoustic analyses rest on recordings from a limited number of individuals, so future studies should collect calls from the type locality and incorporate larger sample sizes to investigate potential geographic variation and clarify intraspecific variation patterns. Even so, the study enriches the fundamental biological database for the genus Boulenophrys, now at 81 species and growing, and offers important reference data for biodiversity assessment, conservation, and the acoustic identification of cryptic amphibian species in a region where the next hidden species may be revealed not by what it looks like, but by what it sounds like.

Subject of Research: Bioacoustic analysis of the advertisement call of Boulenophrys daxuemontis and a new distribution record in Guizhou Province, China

Article Title: Advertisement calls of Boulenophrys daxuemontis Liu, Li, Cheng, Wei, Wang & Cheng, 2025 ( Anura , Megophryidae ) from a new distribution record in Nayong County, Guizhou Province

Article References: Shen, T., Mu, L., Li, S., Liu, J., Feng, C., Yang, X., Yuan, S., Diao, Y., & Su, H. (2026). Advertisement calls of Boulenophrys daxuemontis Liu, Li, Cheng, Wei, Wang & Cheng, 2025 (Anura, Megophryidae) from a new distribution record in Nayong County, Guizhou Province. Herpetozoa, 39, 81-89. https://doi.org/10.3897/herpetozoa.39.e187284

Image Credits: AI Generated

DOI: 10.3897/herpetozoa.39.e187284

Keywords: Boulenophrys daxuemontis, advertisement call, bioacoustics, cryptic species, horned toads, Megophryidae, Guizhou Province, new distribution record, principal component analysis, species identification, amphibian conservation, Herpetozoa