Research Article |
Corresponding author: Kevin R. Messenger ( kevinrmessenger@gmail.com ) Academic editor: Angelica Crottini
© 2022 Kevin R. Messenger, Siti N. Othman, Ming-Feng Chuang, Yi Yang, Amaël Borzée.
This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Citation:
Messenger KR, Othman SN, Chuang M-F, Yang Y, Borzée A (2022) Description of a new Kurixalus species (Rhacophoridae, Anura) and a northwards range extension of the genus. ZooKeys 1108: 15-49. https://doi.org/10.3897/zookeys.1108.81725
|
Knowledge of biodiversity before species become extinct is paramount to conservation, especially when the relevant species are far from their expected distribution and, thus, likely overlooked. Here, we describe a new Kurixalus species corresponding to a range extension of Kurixalus on the Asian mainland, with the closest population in Taiwan. The species diverged from its closest relative during the Late Pliocene to Pleistocene, ca. 3.06 Mya (HPD 95%: 5.82-0.01), based on calibrations with a relaxed clock species tree of unlinked mtDNA 12S rRNA and nuclear DNA TYR. The status of the newly-described species is also supported by a divergence in call properties and morphometrics. We named the species described here as Kurixalus inexpectatus sp. nov. due to the nature of the discovery, as well as the adjunct distribution of the species relative to its closest congeners. The species was found in Zhejiang Province and it represents a range extension of 663 km for the Kurixalus genus.
Bush frog, China, East Asia, species description, taxonomy, Rhacophorid
The taxonomy of the genus Kurixalus Ye, Fei & Dubois in Fei (
Many anurans in Asia have undergone several taxonomic changes in the last decade and continue to undergo massive re-assignments at the generic level, such as: Adenopleura, Bufo, Hyla sensu lato (s.l.), Megophrys s.l., Polypedates s.l., Rana s.l. and Theloderma s.l., just to name a few (
Species in the genus Kurixalus are morphologically similar and species identification is difficult (
During herpetological surveys in April and July 2018, we found an unknown frog that could be allocated to family Rhacophoridae, subfamily Rhacophorinae, genus Kurixalus
We collected 12 Kurixalus samples in April and July 2018 in north-western Zhejiang Province, People’s Republic of China (Fig.
Specimens collected in July were humanely euthanised through cooling in line with
Samples and sequences used as taxa for the phylogenetic trees in this study.
Species | Sample voucher | GenBank accession number | Localities | Literature | |
---|---|---|---|---|---|
12S-tRNA val-16S | TYR | ||||
Kurixalus inexpectatus sp. nov. |
|
MW115094 | MW148393 | Huzhou, Zhejiang, China | Present study |
Kurixalus inexpectatus sp. nov. |
|
MW115093 | MW148394 | Huzhou, Zhejiang, China | Present study |
Kurixalus inexpectatus sp. nov. |
|
MW115095 | MW148395 | Huzhou, Zhejiang, China | Present study |
Kurixalus inexpectatus sp. nov. |
|
MW115092 | MW148396 | Huzhou, Zhejiang, China | Present study |
Kurixalus inexpectatus sp. nov. |
|
MW115090 | MW148397 | Huzhou, Zhejiang, China | Present study |
Kurixalus inexpectatus sp. nov. |
|
- | MW148398 | Huzhou, Zhejiang, China | Present study |
Kurixalus inexpectatus sp. nov. |
|
MW115088 | MW148400 | Huzhou, Zhejiang, China | Present study |
Kurixalus inexpectatus sp. nov. |
|
MW115091 | MW148401 | Huzhou, Zhejiang, China | Present study |
Kurixalus inexpectatus sp. nov. |
|
MW115096 | MW148402 | Huzhou, Zhejiang, China | Present study |
Kurixalus inexpectatus sp. nov. | - | - | MW148399 | Huzhou, Zhejiang, China | Present study |
Kurixalus inexpectatus sp. nov. |
|
MW115089 | MW148403 | Huzhou, Zhejiang, China | Present study |
Kurixalus baliogaster | ROM29862 | KX554476 | KX554740 | Krong Pa, Gia Lai, Vietnam | ( |
Kurixalus baliogaster | ROM29860 | KX554475 | KX554739 | Krong Pa, Gia Lai, Vietnam | ( |
Kurixalus baliogaster | ROM33963 | KX554474 | KX554738 | Krong Pa, Gia Lai, Vietnam | ( |
Kurixalus banaensis | ROM32986 | GQ285667 | GQ285799 | Krong Pa, Gia Lai, Vietnam | ( |
Kurixalus bisacculus | KUHE 19333 | KX554473 | KX554737 | Phu Luanag, Loei, Thailand | ( |
Kurixalus bisacculus | KUHE 19330 | KX554472 | KX554736 | Phu Luanag, Loei, Thailand | ( |
Kurixalus bisacculus | KUHE 35069 | AB933291 | KX554734 | Pilok, Kanchanaburi, Thailand | ( |
Kurixalus bisacculus | FMNH 261902 | KX554471 | KX554733 | Kampot Dist, Prov, Cambodia | ( |
Kurixalus bisacculus | FMNH 261901 | KX554470 | KX554732 | Kampot Dist, Prov, Cambodia | ( |
Kurixalus bisacculus | FMNH 261900 | KX554469 | KX554731 | Kampot Dist, Prov, Cambodia | ( |
Kurixalus bisacculus | FMNH 257903 | KX554458 | KX554699 | Pakxong Dist, Champasak, Laos | ( |
Kurixalus bisacculus | FMNH 256453 | KX554456 | KX554697 | Nakai Dist, Khammouan, Laos | ( |
Kurixalus bisacculus | FMNH 255656 | KX554453 | KX554694 | Con Cuong Dist, Nghe An, Vietnam | ( |
Kurixalus bisacculus | FMNH 255654 | KX554451 | KX554692 | Con Cuong Dist, Nghe An, Vietnam | ( |
Kurixalus bisacculus | FMNH 255661 | KX554450 | KX554691 | VietnamTuong Duong Dist, Nghe An, Vietnam | ( |
Kurixalus bisacculus | FMNH 255655 | KX554452 | KX554693 | Con Cuong Dist, Nghe An, Vietnam | ( |
Kurixalus bisacculus | FMNH 256452 | KX554455 | KX554696 | Nakai Dist, Khammouan, Laos | ( |
Kurixalus bisacculus | KUHE:19428 | AB933290 | KX554735 | Nakon Sri Tamarat, Thailand | ( |
Kurixalus eiffingeri | UMFS 5969 | DQ283122 | DQ282931 | NanTou, Lu-Gu Chi-Tou, 900–1100 m, Taiwan | ( |
Kurixalus eiffingeri | AF458128 | ( |
|||
Kurixalus idiootocus | UMFS 5702 | DQ283054 | DQ282905 | NanTou, Tung Fu, 750 m, Taiwan | ( |
Kurixalus idiootocus | AF458129 | ( |
|||
Kurixalus idiootocus | SCUM 061107L | EU215547 | EU215607 | Lianhuachi, Taiwan | ( |
Kurixalus odontotarsus | YGH 090132 | GU227241 | KX554683 | Caiyanghe, Yunnan, China | ( |
Kurixalus odontotarsus | YGH090130 | GU227239 | KX554681 | Caiyanghe, Yunnan, China | ( |
Kurixalus odontotarsus | Rao 14111401 | KX554445 | KX554680 | Menglun, Yunnan, China | ( |
Kurixalus odontotarsus | KIZ060821122 | EF564456 | KX554679 | Menglun, Yunnan, China | ( |
Kurixalus odontotarsus | YGH090177 | GU227235 | KX554677 | Mengyang, Yunnan, China | ( |
Kurixalus odontotarsus | YGH090176 | GU227234 | KX554676 | Mengyang, Yunnan, China | ( |
Kurixalus odontotarsus | YGH090175 | GU227233 | KX554675 | Mengyang, Yunnan, China | ( |
Kurixalus odontotarsus | Rao 14111307 | KX554443 | KX554674 | Bada, Yunnan, China | ( |
Kurixalus odontotarsus | Rao 14001643 | KX554441 | KX554672 | Cangyuan, Yunnan, China | ( |
Kurixalus odontotarsus | YGH090179 | GU227236 | KX554678 | Mengyang, Yunnan, China | ( |
Kurixalus odontotarsus | Rao 14102907 | KX554442 | KX554673 | Cangyuan, Yunnan, China | ( |
Kurixalus verrucosus | Rao 14102913 | KX554440 | KX554671 | Yingjiang, Yunnan, China | ( |
Kurixalus verrucosus | Rao 14102912 | KX554439 | KX554670 | Yingjiang, Yunnan, China | ( |
Kurixalus verrucosus | Rao 06308 | KX554428 | KX554657 | Muotuo, Tibet, China | ( |
Kurixalus verrucosus | Rao 06306 | KX554427 | KX554656 | Muotuo, Tibet, China | ( |
Kurixalus verrucosus | Rao 06302 | KX554423 | KX554654 | Muotuo, Tibet, China | ( |
Kurixalus verrucosus | Rao 06301 | KX554422 | KX554653 | Muotuo, Tibet, China | ( |
Kurixalus verrucosus | Rao 06201 | KX554419 | KX554651 | Muotuo, Tibet, China | ( |
Kurixalus verrucosus | Rao 06194 | KX554416 | KX554650 | Muotuo, Tibet, China | ( |
Kurixalus verrucosus | Rao 06193 | KX554415 | KX554649 | Muotuo, Tibet, China | ( |
Kurixalus verrucosus | CAS225128 | GU227276 | JQ060918 | Nagmung, Kachin, Myanmar | ( |
Kurixalus verrucosus | CAS 224381 | GU227274 | JQ060917 | Nagmung, Kachin, Myanmar | ( |
Kurixalus verrucosus | Rao 06202 | KX554423 | KX554654 | Muotuo, Tibet, China | ( |
Kurixalus verrucosus | Rao 06305 | KX554426 | KX554655 | Muotuo, Tibet, China | ( |
Kurixalus verrucosus | Rao 14102902 | KX554430 | KX554661 | Muotuo, Tibet, China | ( |
Kurixalus verrucosus | Rao 14102904 | KX554432 | KX554663 | Nanjingli, Ruili, Yunnan, China | ( |
Kurixalus verrucosus | Rao 14102905 | KX554433 | KX554433 | Nanjingli, Ruili, Yunnan, China | ( |
Kurixalus verrucosus | Rao 14102906 | KX554434 | KX554665 | Nanjingli, Ruili, Yunnan, China | ( |
Kurixalus verrucosus | Rao 14102910 | KX554437 | KX554668 | Yingjiang, Yunnan, China | ( |
Kurixalus verrucosus | Rao 14102909 | KX554436 | KX554667 | Yingjiang, Yunnan, China | ( |
Kurixalus sp. | MVZ Herp 223856 | JQ060941 | JQ060904 | Tam Dao, Vinh Phu, Vietnam | ( |
Kurixalus sp. | MVZ Herp 223863 | JQ060943 | JQ060921 | Tam Dao, Vinh Phu, Vietnam | ( |
Kurixalus sp. | MVZ Herp 223864 | JQ060944 | JQ060922 | Tam Dao, Vinh Phu, Vietnam | ( |
Kurixalus sp. | MVZ Herp 223865 | JQ060945 | JQ060923 | Tam Dao, Vinh Phu, Vietnam | ( |
Kurixalus sp. | MVZ Herp 223867 | JQ060946 | JQ060924 | Tam Dao, Vinh Phu, Vietnam | ( |
Kurixalus sp. | MVZ Herp 223868 | JQ060947 | JQ060925 | Tam Dao, Vinh Phu, Vietnam | ( |
Kurixalus hainanus | HNNU A1180 | EU215608 | Mt. Diaoluo, Hainan, China | ( |
|
Orixalus carinensis | ROM39660 | GQ285670 | GQ285806 | Sa Pa, Lao Cai, Vietnam | ( |
Romerus ocellatus | HN0806045 | GQ285672 | GQ285802 | Mt. Wuzhi, Hainan, China | ( |
Romerus romeri | KIZ 061205YP | EU215528 | EU215589 | Mt. Shiwan, Guangxi, China | ( |
Zhangixalus appendiculatus | FMNH:267897 | ( |
|||
Zhangixalus appendiculatus | FMNH 267896 | JQ060926 | Bukit Sarang, Sarawak, Malaysia | Yu et al. (2013) | |
Zhangixalus nigropunctatus | - | EU215533 | EU924583 | - | ( |
For all 11 individuals from which we extracted tissues, we amplified one mitochondrial and one nuclear gene fragment. For the mtDNA, we sequenced 827 bp from a section of the genes 12S rRNA, the complete tRNA-Valine (Val) and 16S rRNA, using the primer pair F0001 (5’-AGA TAC CCC ACT ATG CCT ACC C-3’), R1169 (5’-GTG GCT GCT TTT AGG CCC ACT-3’) (
The Polymerase Chain Reactions (PCR) were carried out in 20 µl reaction with 50 to 100 ng of template DNA, with 1.0 µl of each primer (10 mM). The final concentration of each PCR reaction resulted to 1.5 µl of MgCl2 (25 mM), 1.6 µl of dNTP (2.5 mM), 2.0 µl of 10× Buffer and 0.1 µl of TaKaRa Taq DNA polymerase (5 unit/µl). PCR amplifications were performed under the following thermal profiles: initial denaturation at 95 °C for 5 min, followed by 35 cycles with denaturation at 94 °C for 1 min, annealing at 55 °C for the mtDNA genes fragment and 54 °C for TYR for 1 min and extension at 72 °C for 1 min. The cycles were followed by a 10 min final extension at 72 °C. The amplified PCR products were sent for purification and sequencing to Cosmo Genetech Co. (Cosmo Genetech, Republic of Korea) on an ABI platform.
To reconstruct the independent and concantenated genes tree, we relied on two different datasets: (i) 827 bp-long fragments of mtDNA 12S rRNA, tRNA-Val and 16S rRNA (n taxa = 98), (ii) 486 bp-long fragments of sequences of protein-coding nuDNA Tyrosinase gene (TYR; n taxa = 110); and, (iii) 80 concatenated sequences of partial 12S rRNA (292 bp) and TYR (479 bp). We trimmed the sequences in each dataset manually and aligned the three sequences datasets indepedently using Clustal Omega (
We calculated sequences similarity and estimated the genetic distance (or net evolutionary divergence) on the datasets of mtDNA 12S rRNA-trNA-Val-16S rRNA (n sequences = 98) and nuDNA TYR (n sequence = 110) using MEGA X (
For subsequent phylogenetic analyses, we downloaded supplemental sequences data of 98 homologous sequences of Kurixalus and Zhangixalus and other Rhacophoridae genera from Genbank (
We used Partition Finder v. 2.1.1 (
We built phylogenetic trees for all three datasets: mtDNA 12S rRNA-tRNA-Val-16S rRNA, protein coding nuDNA TYR and concatenated 12S rRNA-TYR using Bayesian Inference methodologies with MrBayes v.3.2.6 (
To test the presence of population differentiation using the TYR marker, we ran an analysis of molecular variance (AMOVA;
Relying solely on a distance-based method is insufficient. The coalescent-based species delimitation was determined as the most efficient method for comparative study of species delimitation in genus of Kurixalus (
Additionally, the recent study on the phylogeography of Taiwanese Kurixalus showed that the genus colonised the Island attributes through a land-bridge during the last glacial maxima (
Finally, we projected the possible dispersal pathways, based on the molecular dating estimates focusing on the clade containing our focal species and Taiwanese Kurixalus on paleomaps using QGIS v.2.18.15. The oscillayers used to reconstruct the Plio-Pliocene maps was adapted from datasets provided in
The acoustic recordings of putative new Kurixalus species were recorded between April and July in 2018 at 24 °C with a linear PCM recorder (Tascam DR-40; California, USA) linked to a unidirectional microphone (Unidirectional electret condenser microphone HT-81, HTDZ; Xi’an, China). To determine the relationship with other species, we first compared the number of consecutive calls within a series of calls between the individuals recorded and K. idiootocus, K. eiffingeri, K. berylliniris and K. wangi. We then compared the call properties of the new population with that of K. idiootocus as it was the most closely-related species and the only species with the same number of calls within a series of consecutive calls (see results). The recordings of K. idiootocus were obtained in central Taiwan (23.9240 N, 120.8910 E) in July 2013, using a Tascam DR-70D digital recorder (TEAC Corporation, Tokyo, Japan) and a Sennheiser ME67/K6 directional microphone (Sennheiser Electronic GmbH & Co. KG, Hanover, Germany). All our recordings were recorded at a sampling rate of 44.1 kHz with 16-bit resolution. Temperature was recorded with a Tecpel DIT-517 infrared thermometer (between 22 and 25 °C; TECPEL Corporation, New Taipei, Taiwan). The genus emits a series of continuous notes, pooled in bouts of continuous calls. To compare K. idiootocus and the new Kurixalus population, we selected one entire series of consecutive calls for each individual and analysed 373 advertisement calls in total, including 238 calls for K. idiootocus (9 to 24 calls in a bout from each of 16 males) and 135 calls for the new population (9 to 21 calls in a bout for 9 males).
We used Raven Pro v.1.5 (
The call property measurements. This figure shows A the waveform of entire series of a consecutive call B the waveform C the spectrogram of two calls and D the spectral power distribution of a single call from the new Kurixalus population from Zhejiang, China. We extracted the number of calls in a bout, call interval, call duration (CD), rise time (RT), fall time (FT), dominant frequency (here also the max frequency), secondary peak frequency and the relative amplitude of two peaks.
Call duration refers to the time between the onset and offset of a call. Rise time refers to the time between the onset of call and the local maximum in the waveform. Fall time is the time between the local maximum in the waveform and the offset of a call (Fig.
We first corrected the calls for temperature variation by adjusting the value of each variable to the average temperature of all recordings using the equation originating from the linear regression of each focal variable in function of temperature. As the contribution of each call property for each individual is not independent of other call properties and consequently correlated, we used a principal component analysis (PCA) to convert those call properties into a set of values of linearly uncorrelated factors. The PCA provided four principal components with Eigenvalues larger than 0.5, explaining 95.7% of the total variance. We used a Discriminant Function Analysis (DFA) to classify the call properties and test for the correctness of group assignment. We then plotted the two significant PCs against each other to illustrate the divergence between the two species. Finally, to determine the differing call variables between the two clades, we used a Multivariate Analysis of Variance (MANOVA) to compare each call property between these two species.
We collected eighteen morphological measurements three times each and averaged the values for further analyses. Morphometric data were taken using digital calipers to the nearest 0.1 mm and included the following characters: snout-vent length (SVL), head width (HDW), distance between left and right articulations of jaw, head length (HDL), from the tip of the snout to the articulation of the jaw, snout length (SNT), from tip of snout to the anterior corner of the eye, horizontal eye diameter (EYE) from the anterior to the posterior corner of the eye, width of the upper eyelid (UEW), the horizontal length of the upper eyelid, internares distance (IND), the distance from nostril to eye (DNE), from the posterior border of nostril to anterior border of the eye, narrowest interorbital distance (IOD), greatest horizontal tympanum diameter (TMP), tympanum-eye distance (TEY) from anterior edge of tympanum to posterior corner of eye, hand length (HND) from distal end of radio-ulna to tip of finger III, radio-ulna length (RAD), forelimb length (FLL), distance from the proximate end of radio-ulna to distal end of finger III, thigh length (THL), distance from vent to distal end of femur, tibia length (TIB), foot length (FL) from proximal end of inner metatarsal tubercle to tip of toe IV and the length of the foot and tarsus (TFL), distance from tibio-tarsal joint to tip of toe IV. All specimens were measured by a single author (YY) to minimise sampling error. The dataset is available Suppl. materials.
To be able to compare with the morphometrics of other clades, we extracted data from the literature for all species available (Suppl. material
As the variables were strongly correlated (Pearson’s correlation; Table
Pearson correlation for all ten selected variables. We run a Pearson Correlation test (n = 68) to highlight the correlation between variables and highlight the need for a variable reduction analysis, such as a PCA. Cells in bold highlight significance.
HDW | SNT | IND | IOD | UEW | EYE | TD | DNE | FLL | TFL | FL | ||
---|---|---|---|---|---|---|---|---|---|---|---|---|
HDL | r | 0.84 | 0.34 | 0.23 | 0.44 | 0.60 | -0.11 | -0.13 | 0.09 | 0.86 | 0.45 | 0.08 |
p | < 0.001 | 0.005 | 0.065 | < 0.001 | < 0.001 | 0.365 | 0.298 | 0.480 | < 0.001 | < 0.001 | 0.521 | |
HDW | r | 0.38 | 0.22 | 0.52 | 0.49 | -0.06 | -0.15 | 0.05 | 0.81 | 0.54 | 0.10 | |
p | 0.001 | 0.072 | < 0.001 | < 0.001 | 0.634 | 0.218 | 0.698 | < 0.001 | < 0.001 | 0.432 | ||
SNT | r | -0.11 | 0.04 | 0.38 | 0.11 | -0.25 | -0.09 | 0.34 | 0.61 | -0.05 | ||
p | 0.365 | 0.722 | 0.002 | 0.394 | 0.044 | 0.450 | 0.005 | < 0.001 | 0.698 | |||
IND | r | 0.01 | 0.48 | -0.26 | 0.04 | 0.15 | 0.06 | 0.10 | 0.16 | |||
p | 0.961 | < 0.001 | 0.029 | 0.773 | 0.230 | 0.644 | 0.439 | 0.195 | ||||
IOD | r | 0.07 | 0.11 | -0.04 | 0.06 | 0.40 | 0.15 | 0.05 | ||||
p | 0.565 | 0.383 | 0.731 | 0.648 | 0.001 | 0.209 | 0.695 | |||||
UEW | r | -0.04 | -0.16 | 0.09 | 0.46 | 0.48 | 0.06 | |||||
p | 0.754 | 0.181 | 0.455 | < 0.001 | < 0.001 | 0.625 | ||||||
EYE | r | 0.39 | -0.70 | -0.06 | 0.12 | -0.42 | ||||||
p | 0.001 | < 0.001 | 0.655 | 0.330 | < 0.001 | |||||||
TMP | r | -0.47 | -0.14 | -0.23 | -0.47 | |||||||
p | < 0.001 | 0.273 | 0.059 | < 0.001 | ||||||||
DNE | r | -0.06 | -0.17 | 0.73 | ||||||||
p | 0.628 | 0.163 | < 0.001 | |||||||||
FLL | r | 0.61 | 0.09 | |||||||||
p | < 0.001 | 0.471 | ||||||||||
TFL | r | 0.13 | ||||||||||
p | 0.285 |
Variables and results for the Principal Component Analysis and resulting ANOVA. Principal components were to be extracted if their eigenvalue > 1, under a varimax rotation. Variables were selected as loading into a PC if the value is > 0.58. In bold are variables retained as loading into one if the PCs. Based on the variables loading on to each of the PCs, we assigned PC1 to the general morphology and PC2 to the horizontal head structure. PC1 and PC2 were not significantly different between Kurixalus inexpectatus sp. nov. and other Kurixalus species under a one-way ANOVA, but they were significantly different between Kurixalus inexpectatus sp. nov. and K. idiootocus. The sample sizes used in the analysis were such as: K. inexpectatus sp. nov. n = 12, K. idiootocus n = 8; all n = 71; details in the Suppl. material
PC1 | PC2 | |
---|---|---|
SVL | 0.90 | 0.21 |
HDL | 0.92 | 0.14 |
HDW | 0.96 | 0.17 |
SNT | 0.84 | 0.26 |
IND | 0.74 | -0.08 |
IOD | 0.85 | 0.17 |
UEW | 0.84 | 0.09 |
EYE | 0.33 | 0.84 |
TMP | 0.40 | 0.59 |
DNE | 0.35 | -0.86 |
FLL | 0.90 | 0.21 |
TFL | 0.91 | 0.25 |
FL | 0.85 | -0.28 |
Eigen value | 8.25 | 1.92 |
Variance (%) | 63.49 | 14.74 |
ANOVA all clades | ||
χ2 | 0.43 | 0.69 |
F | 0.47 | 0.66 |
df1, df2 | 1,66 | 1,66 |
p | 0.494 | 0.419 |
ANOVA focal clade-K. idiootocus | ||
χ2 | 1.42 | 1.39 |
F | 13.35 | 14.56 |
df1, df2 | 1,66 | 1,66 |
p | 0.002 | 0.017 |
Our analyses resulted in minor differences in the evolutionary divergence between the 12S rRNA gene fragments of K. inexpectatus sp. nov. and K. idiootocus (mean = 0.0004 SD ± 0.0004). Similarly, the protein coding nuclear TYR between K. inexpectatus sp. nov. and K. idiootocus showed a comparatively smaller mean of substitution rate (mean = 0.0035 ± 0.0004; value marked with double asterisks (**) in Table
Matrix of genetic distances between all pairs of sequences of protein-coding nuclear TYR between groups of sequences of 16 species rhacophorids species (n = 110). The 16 groups of species consisted of Kurixalus and Rhacophorus genera. Values in bold in the bottom left of diagonal matrix represent the means of estimate for each species divergence using maximum composite likelihood. Values of the upper right of diagonal matrix represents the standard deviation of each mean of divergence. The mean of distance between our proposed species K. inexpectatus sp. nov. and K. idiootocus noted with (**), which was higher to mean genetic distance of other pairwise species (values are in bold and marked with *).
Species | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
1 K. ocellatus | 0.0071 | 0.0260 | 0.0287 | 0.0292 | 0.0290 | 0.0293 | 0.0292 | 0.0292 | 0.0302 | 0.0340 | 0.0286 | 0.0264 | 0.0206 | 0.0216 | 0.0275 | |
2 K. romeri | 0.0144 | 0.0254 | 0.0280 | 0.0290 | 0.0286 | 0.0290 | 0.0291 | 0.0290 | 0.0298 | 0.0338 | 0.0279 | 0.0258 | 0.0201 | 0.0218 | 0.0270 | |
3 K. inexpectatus | 0.0780 | 0.0752 | 0.0047 | 0.0057 | 0.0054 | 0.0056 | 0.0072 | 0.0057 | 0.0063 | 0.0104 | 0.0047 | 0.0025 | 0.0150 | 0.0147 | 0.0221 | |
4 K. banaensis | 0.0862 | 0.0832 | 0.0081 | 0.0058 | 0.0056 | 0.0057 | 0.0074 | 0.0059 | 0.0065 | 0.0104 | 0.0058 | 0.0043 | 0.0162 | 0.0147 | 0.0218 | |
5 K. baliogaster | 0.0885 | 0.0876 | 0.0113 | 0.0109 | 0.0006 | 0.0006 | 0.0035 | 0.0001 | 0.0028 | 0.0100 | 0.0063 | 0.0048 | 0.0171 | 0.0161 | 0.0229 | |
6 K. bisacculus | 0.0877 | 0.0859 | 0.0108 | 0.0105 | 0.0007 | 0.0001 | 0.0029 | 0.0004 | 0.0029 | 0.0102 | 0.0060 | 0.0044 | 0.0170 | 0.0157 | 0.0226 | |
7 Kurixalus sp. | 0.0888 | 0.0875 | 0.0113 | 0.0110 | 0.0005 | -0.0001 | 0.0030 | 0.0001 | 0.0029 | 0.0102 | 0.0062 | 0.0046 | 0.0172 | 0.0160 | 0.0229 | |
8 K. hainanus | 0.0884 | 0.0876 | 0.0161 | 0.0157 | 0.0046 | 0.0037 | 0.0036 | 0.0035 | 0.0048 | 0.0114 | 0.0078 | 0.0064 | 0.0186 | 0.0167 | 0.0235 | |
9 K. odontotarsus | 0.0887 | 0.0878 | 0.0114 | 0.0110 | 0.0001* | 0.0004* | 0.0001* | 0.0046 | 0.0029 | 0.0100 | 0.0063 | 0.0048 | 0.0171 | 0.0161 | 0.0229 | |
10 K. verrucosus | 0.0926 | 0.0906 | 0.0134 | 0.0130 | 0.0035 | 0.0036 | 0.0037 | 0.0081 | 0.0035 | 0.0110 | 0.0068 | 0.0054 | 0.0166 | 0.0162 | 0.0229 | |
11 Zhangixalus appendiculatus | 0.1019 | 0.1009 | 0.0260 | 0.0256 | 0.0240 | 0.0250 | 0.0247 | 0.0290 | 0.0241 | 0.0278 | 0.0109 | 0.0096 | 0.0224 | 0.0213 | 0.0283 | |
12 K. eiffingeri | 0.0858 | 0.0829 | 0.0080 | 0.0109 | 0.0125 | 0.0120 | 0.0126 | 0.0173 | 0.0126 | 0.0143 | 0.0272 | 0.0037 | 0.0157 | 0.0152 | 0.0227 | |
13 K. idiootocus | 0.0797 | 0.0769 | 0.0035#* | 0.0070 | 0.0086 | 0.0081 | 0.0086 | 0.0133 | 0.0086 | 0.0106 | 0.0231 | 0.0054 | 0.0144 | 0.0138 | 0.0212 | |
14 K. carinensis | 0.0605 | 0.0580 | 0.0421 | 0.0456 | 0.0494 | 0.0490 | 0.0497 | 0.0549 | 0.0496 | 0.0479 | 0.0666 | 0.0436 | 0.0401 | 0.0111 | 0.0167 | |
15 R. nigropunctatus | 0.0617 | 0.0613 | 0.0400 | 0.0397 | 0.0451 | 0.0438 | 0.0448 | 0.0469 | 0.0452 | 0.0457 | 0.0618 | 0.0412 | 0.0370 | 0.0274 | 0.0145 | |
16 Rhacophorus sp. | 0.0823 | 0.0794 | 0.0655 | 0.0632 | 0.0691 | 0.0678 | 0.0689 | 0.0711 | 0.0693 | 0.0693 | 0.0850 | 0.0668 | 0.0622 | 0.0460 | 0.0370 |
Overall, the Bayesian Inference (BI) trees inferred from both mtDNA 12S rRNA-tRNA-Val-16S rRNA and nuDNA TYR fragments showed strong patterns of genetic structures for the East Asian and Southeast Asian Kurixalus phylogeny relationship, recovering four strongly supported clades (Clades A, B, C and D; see the distributions of the clades and the phylogenetic tree in Suppl. material
The phylogenetic relationship of concatenated gene fragments of partial 12S rRNA and TYR gene fragments recovered the three major clades within the Kurixalus genus with a Bayesian Posterior (BP) support of 90% for clade A, 52% for clade B, 71% for clade C (Fig.
Bayesian Inference (BI) tree inferred from 79 sequences of concatenated 12S rRNA-TYR gene fragments. The three clades (Clades A, B and C) recovered in the phylogenetic tree are labelled accordingly. Clade B included the species K. inexpectatus sp. nov. described in the present study, indicated by the red box. The value of the node represents the Bayesian posterior probability (BPP) for each clade. The clades are marked with a solid bar and labelled in accordance with their specific name.
The AMOVA provided support to the genetic differentiation recorded while using the TYR marker as it identified 21.80% of variance within clades and 78.20% of variance between clades for the three main clades Kurixalus (n = 108; Fig.
Haplotype network inferred from 216 phased nuDNA TYR sequences data (486 sites). The haplotype group for the focal Clade A comprised six representative haplotypes of Kurixalus. Clade A included three K. inexpectatus sp. nov. haplotypes. The size of each haplotype marker matches the haplotype scales. The colour coding matches with the name of the taxa in the legend. The colours used for the boundaries of Clade A and Clade B are coded similarly to the colours of their clades in the phylogenetic tree (Suppl. material
The topology of the coalescent unlinked 12S rRNA and TYR tree supported the divergence of the focal Kurixalus clade from the most closely-related species, K. idiootocus. Nested sampling analyses on both species trees was favoured on the topology proposed by Model 2 (MLE = - 3688.252; Bayes factor: 651.011; Table
Nested sampling analysis results on two competing topology models for combined 12S rRNA and TYR using calibrated species trees. The values include summation of estimated Marginal L value with calculated Bayes factor for designated topology Model 1 and model 2. The positive value favoured the designated model. Topology Model 1 clumped Kurixalus idiootocus and K. inexpectatus sp. nov. as a single species. Topology Model 2 proposed K. inexpectatus sp. nov. as a new species and split from K. idiootocus. Bold values indicate the mean of nested sampling for each model.
Species tree topology | Nested sampling | Consensus | ||||
---|---|---|---|---|---|---|
Marginal likelihood estimate (MLE) | sqrt (H/N) | Standard deviation | Bayes factor (mean of MLE1-mean of MLE2) | |||
Model 1 (clumping) | 1 | -4339.441 | 6.099 | 6.008 | -651.001 | Model 1 is not favoured |
2 | -4339.114 | 6.093 | 5.889 | |||
3 | -4339.123 | 6.092 | 5.992 | |||
4 | -4339.334 | 6.094 | 5.969 | |||
Mean | -4339.253 | 6.095 | 5.965 | |||
Model 2 (splitting) | 1 | -3688.305 | 5.144 | 5.161 | 651.001 | Model 2 is favoured |
2 | -3688.131 | 5.143 | 5.458 | |||
3 | -3688.361 | 5.145 | 5.279 | |||
4 | -3688.211 | 5.143 | 4.925 | |||
Mean | -3688.252 | 5.144 | 5.206 |
Calibrated species tree of Rhacophoridae represented by Kurixalus, Orixalus, Romerus and Zhangixalus distributed over East Asia and Southeast Asia. The species tree reconstructed from unlinked 12S rRNA and TYR. The asterisk (*) symbol indicates Kurixalus inexpectatus sp. nov. The highlighted lineages divergence noted with (a–d) and the time estimates are synchronised with datation in Table
Our calibrated species tree of unlinked 12S rRNA and TYR gene fragments provided support on the earliest split between Asian lineages of Kurixalus and Zhangixalus to be dated in Mid-Miocene, ca. 11.17 Mya (Table
Molecular dating of 16 species of Asian rhacophorid frogs estimates the age of lineage separation between K. inexpectatus sp. nov. and K. idiootocus. The molecular dating estimation was using an uncorrelated lognormal relaxed clock with Yule prior on species tree inferred from unlinked 12S rRNA and TYR gene fragments of Kurixalus and rhacophorid taxa (n taxa = 79) distributed across Southeast Asia and East Asia.
Node | Clade (speciation event) | Node age (Mya) | |
---|---|---|---|
Mean | HPD 95% | ||
a | Emergence of stem clade of Kurixalus after split off from Zhangixalus | 10.48 | 8.16–12.98 |
b | Stem clade of south-eastern and eastern Asian mainland group of Kurixalus (K. verrucosus + K. baliogaster + K. odontotarsus + K. hainanus + K. bisacculus) | 9.14 | 6.87–11.50 |
c | Stem clade of Taiwanese Kurixalus group (isolation of K. eiffingeri) | 5.66 | 3.32–8.07 |
d | Split off between Chinese mainland K. inexpectatus sp. nov. of south-eastern mainland and K. idiootocus of Taiwan Island | 3.06 | 5.82–0.01 |
As the contribution of each call property for each individual is not independent of other call properties and consequently correlated, we used a principal component analysis (PCA) to convert those call properties into a set of values of linearly uncorrelated factors. We selected the principal components from the results of the PCA to cover as much as possible of the total variance, resulting in four PCs with Eigenvalues larger than 0.5 and explaining 95.7% of the total variance. We used a Discriminant Function Analysis (DFA) to classify the call properties and test for the correctness of group assignment. We then plotted the two significant PCs against each other to illustrate the divergence between the two species. Finally, to determine the differing call variables between the two clades, we used a Multivariate Analysis of Variance (MANOVA) to compare each call property between these two species.
Based on the descriptions of the advertisement call and the number of calls in a series of consecutive calls, we could first segregate the species into two groups matching with the phylogenetic clustering. We grouped the putative new Kurixalus species and K. idiootocus together, while K. eiffingeri, K. berylliniris and K. wangi were grouped together (Suppl. material
The DFA on the four resulting PCs highlighted that only two PC1and PC3 were significantly different between the two species (PC1: Wilks’ Lambda = 0.93, F(1,19) = 141.10, p < 0.001 ; PC3: Wilks’ Lambda = 0.17, F(1,19) = 9.75, p = 0.005) and PC2 (Wilks’ Lambda = 0.11, F(1,19) = 0.61, p = 0.442) and PC4 (Wilks’ Lambda = 0.12, F(1,19) = 0.93, p = 0.345) were not. When plotting PC1 and PC3 against each other, a clear segregation of data was visible (Fig.
The description results of advertisement call properties and the MANOVA test in Kurixalus inexpectatus sp. nov. and Kurixalus idiootocus. From the MANOVA test, the whole model Wilks’ value = 0.087, F9,14 = 16.27, p < 0.001. “*” indicate the data are not following the assumption of normal distribution (Shapiro-Wilk test, p < 0.05) and we transformed the data to their natural logarithm before doing statistical tests.
Call property | K. inexpectatus sp. nov. (n = 8) | K. idiootocus (n = 16) | F1,22 | p |
---|---|---|---|---|
# of call in a bout | 16.9 ± 3.9 (9–21) | 14.9 ± 4.2 (9–24) | 1.26 | 0.274 |
Bout length (s) | 7.3 ± 2.9 (3.4–11.2) | 3.6 ± 1.1 (1.9–5.5) | 21.60 | < 0.001 |
Call interval (ms) | 376 ± 157 (115–539)* | 211 ± 35 (159–278) | 9.10 | 0.006 |
Call length (ms) | 76.1 ± 11.1 (58–91) | 34.8 ± 4.6 (28–43) | 169.71 | < 0.001 |
Rise time (ms) | 38.1 ± 5.5 (29.5–46.0) | 17.5 ± 2.4 (14–22) | 167.48 | < 0.001 |
Fall time (ms) | 37.9 ± 5.5 (29.0–45.0) | 17.5 ± 2.4 (14–22) | 166.03 | < 0.001 |
Max frequency (kHz) | 2.30 ± 0.06 (2.20–2.39) | 2.54 ± 0.08 (2.35–2.68) | 54.20 | < 0.001 |
2nd frequency (kHz) | 4.59 ± 0.11 (4.41–4.74) | 5.05 ± 0.14 (4.71–5.33) | 61.20 | < 0.001 |
Relative amplitude (dB) | 39.7 ± 8.3 (25.3–52) | 35.5 ± 2.4 (30–39.7) | 3.73 | 0.067 |
The unique PCA, used to identify the independent dimensions of the morphological characters between the individuals collected in this study and other Kurixalus sp. individuals, resulted in two PCs, with eigenvalues of 1.92 and 8.25, explaining a cumulated variation of 78.23% (Table
The results of the one-way ANOVA showed that there was no significant difference between the focal and non-focal groups for either of the PCs (Table
K. inexpectatus is morphologically most similar to K. idiootocus, its closest relative and, after standardising measurements by SVL, K. inexpectatus differs by having a relatively longer head length (34% vs. 33%), significantly shorter snout (13% vs. 15%; p < 0.001), significantly greater internasal distance (11% vs. 10%; p < 0.001), significantly smaller eye diameter (13% vs. 16%; p < 0.001), nearly significant wider tympanum diameter (7% vs. 6%; p = 0.06), significantly greater distance between the eyes and nares (8% vs. 7%, p = 0.03), significantly longer forelimb length (50% vs. 48%; p = 0.03), shorter tibia length (44% vs. 45%) and longer foot length (42% vs. 40%). Additionally, K. inexpectatus is further distinguished from K. idiootocus in having a tibio-tarsal articulation that extends beyond the anterior corner of the eye (versus centre of eye). K. inexpectatus can be differentiated from K. bisacculus, K. hainanus, K. naso, K. odontotarsus, K. raoi, K. silvaenaias, K. verrucosus and K. yangi by having an average adult SVL of less than 30 mm (27.5 – 31.8, × = 29.2) (vs. larger) (
Five adult males,
Chuanbu Village (川步村), Changxing County, Huzhou City, Zhejiang Province, People’s Republic of China.
The epithet inexpectatus is Latin for “the unexpected.” This was chosen for several reasons. We selected this name because we had come to survey this region of China for different taxa. KRM came to this locale to survey for Megophrys. AB came to this locale to survey for Dryophytes. It was not only surprising to find this species while surveying for two other target genera, but upon realising the immense distance to the next closest population of Kurixalus, the discovery was even more unexpected. For an English and Chinese common name, we are recommending the name Changxing Treefrog (pronounced “Chang-shing” in English) 长兴原指树蛙 (cháng xīng yuán zhǐ shù wā).
The specimen matched the genus Kurixalus, based on the following characters: tips of digits enlarged to discs, with circum-marginal grooves; small-body size; pointed snout, forming a beak-like appearance; serrated dermal fringes along the outer edge of the forearm and leg; an inverted triangular-shaped dark brown mark between the eyes; dorsal “) (“ saddle-shaped marking; and a coarse dorsal and lateral surface with several small, irregular tubercles [7, 18, 29].
Kurixalus inexpectatus sp. nov. is characterised and distinct from the majority of its congeners (19) by having a combination of being: (1) a small-sized species with an average adult size below 30 mm (in males); and (2) having two dark symmetrical pectoral blotches.
Genetically, the species is most closely related to K. idiootocus and is morphologically distinguished from this species by the combination of features: (1) having a tibio-tarsal articulation that extends beyond the anterior corner of the eye (versus the centre of eye); (2) having a significantly shorter snout relative to SVL; (3) a significantly greater internasal distance relative to SVL; (4) a significantly smaller eye diameter relative to SVL; (5) a nearly significantly wider tympanum diameter relative to SVL; (6) having a significantly greater distance between the eyes and the nares; (7) and by having a significantly longer forelimb length.
Adult male (SVL 29.4 mm); head width about the same as body, its length 37.9% of SVL; head slightly longer than wide in the holotype (11.1 mm vs. 11.0, respectively); snout pointed and slightly turned down, forming a small “beak-like” appearance typical in many rhacophorids; eye large, protuberant, ED 36.3% of HDL, 13.8% of SVL; pupil horizontal; tympanum distinct in form, but not distinct in texture or colour, its diameter 6.8% of SVL; nostrils protuberant; closer to the tip of the snout than the eye; vomerine teeth absent; tongue notched posteriorly; single internal vocal sac.
Relative length of fingers I < II < IV < III. Tips of all four fingers form discs with circum-marginal and transverse ventral grooves; relative width of discs is IV > III > II > I; nuptial pads absent; fingers lacking webbing at base; subarticular tubercles prominent and rounded; series of tubercles forming serrated dermal fringe along outer edge of forearm.
Heels overlapping when legs at right angle to body; relative length of toes is I < II < III < V < IV; toes moderately webbed at base; tips of toes expand to form discs with circum-marginal and transverse ventral grooves; toes discs are smaller than finger discs; relative size of toe discs I < V < IV < III < III; subarticular tubercles present, but not as obvious as hand.
Body is covered in numerous tubercles and dermal ridges. Ridges are present on the dorsum, but absent from the flanks and venter; tympanum also covered in tubercles.
The average of three measurements for each character is as follows: SVL 29.4, HDL 11.1, HDW 11.0, SNL 3.8, IND 3.3, IOD 3.3, UEW 3.1, ED 4.0, TD 2.0, TEY 0.9, DNE 2.7, FLL 15.7, THL 13.4, TL 14.5, FL 12.9, HND 9.1, RAD 6.7.
Light brown dorsum with white patch in the sacral region and extending a bit on to the femurs. Darker brown “) (“ dorsal saddle. Ventrally, white chest with brown colouration in the pectoral and axillary region. Ventral side of forelimbs have streaks of white and brown, almost like a marbled appearance. Ventral side of hind-limbs orange in the thigh and tibia region has the same brown and white marbled appearance present in the forelimbs. Palm of hand primarily light brown; sole of feet slightly darker than hand.
In preservation, the orange and light brown colours have faded, the darker brown has darkened compared to life. Pattern same as in life. Iris clouded. Chest white, throat black. Ventral side of arms black and white marbled appearance. Ventral side of tibia black and white marbled, similar to ventral aspect of forelimbs.
As the holotype and paratypes of the new species are all male, sexual dimorphism cannot be ascertained. Aside from SVL, which is to be expected, the next characters which showed the greatest variation were FLL, TL, FL and TFL. Though the holotype has a head length longer than head width, most specimens had a head length shorter than head width. Colour varied between individuals, likely induced by temperature and/or time of day, as we observed this change first-hand. See Table
Variation in morphological measurements amongst the holotype* and paratypes. Each character was measured three times, the values in the table represent the average of the three measurements. *Denotes holotype.
Specimen | SVL | HL | HW | SL | IND | IOD | UEW | ED | TD | TEY | DNE | FLL | THL | TL | FL | TFL | HND | RAD |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
20180704001* | 29.4 | 11.1 | 11.0 | 3.8 | 3.3 | 3.3 | 3.1 | 4.0 | 2.0 | 0.9 | 2.7 | 15.7 | 13.4 | 14.5 | 12.9 | 19.9 | 9.1 | 6.7 |
20180704002 | 31.8 | 10.3 | 10.9 | 3.8 | 3.4 | 3.1 | 3.4 | 4.3 | 2.2 | 0.8 | 2.6 | 16.0 | 14.1 | 13.9 | 13.2 | 19.9 | 9.5 | 6.9 |
20180704003 | 29.7 | 10.2 | 10.1 | 4.0 | 3.3 | 3.0 | 3.1 | 3.9 | 2.1 | 1.1 | 2.7 | 14.7 | 13.1 | 12.8 | 12.5 | 18.7 | 8.6 | 7.0 |
20180704004 | 29.5 | 10.3 | 11.1 | 4.0 | 3.4 | 2.7 | 3.1 | 4.0 | 1.9 | 1.0 | 2.8 | 15.7 | 13.9 | 13.5 | 13.0 | 20.1 | 9.1 | 7.1 |
20180704005 | 29.4 | 9.6 | 10.7 | 3.4 | 3.1 | 3.0 | 2.8 | 4.0 | 1.8 | 1.0 | 2.5 | 15.2 | 13.5 | 13.6 | 12.7 | 19.6 | 8.6 | 6.8 |
20180704006 | 28.3 | 10.0 | 10.1 | 3.7 | 3.4 | 3.1 | 3.2 | 3.6 | 1.7 | 0.9 | 2.3 | 14.1 | 12.9 | 12.5 | 12.1 | 18.2 | 8.4 | 5.9 |
20180705001 | 28.6 | 9.7 | 10.7 | 4.2 | 3.3 | 3.1 | 2.9 | 3.2 | 1.8 | 1.2 | 2.6 | 14.2 | 13.3 | 12.9 | 10.9 | 17.6 | 8.0 | 6.4 |
20180706001 | 29.4 | 9.3 | 10.4 | 3.7 | 3.3 | 3.1 | 3.0 | 3.7 | 2.1 | 1.1 | 2.5 | 13.5 | 12.2 | 12.0 | 11.0 | 17.7 | 8.0 | 5.9 |
20180706002 | 29.8 | 10.1 | 10.9 | 3.5 | 3.4 | 3.4 | 3.0 | 4.0 | 2.0 | 0.8 | 2.3 | 15.1 | 13.9 | 13.4 | 12.9 | 19.4 | 8.8 | 6.3 |
20180706003 | 28.5 | 9.9 | 10.2 | 4.0 | 3.3 | 3.2 | 3.1 | 3.7 | 1.6 | 0.9 | 2.4 | 13.8 | 12.2 | 12.0 | 11.4 | 17.6 | 8.5 | 5.6 |
20180706004 | 29.0 | 9.9 | 10.4 | 3.7 | 3.4 | 3.2 | 2.9 | 3.8 | 1.9 | 0.9 | 2.5 | 14.5 | 12.9 | 12.9 | 12.0 | 18.6 | 8.7 | 6.3 |
20180706005 | 27.5 | 9.1 | 10.1 | 3.2 | 3.1 | 3.0 | 2.3 | 3.9 | 1.8 | 0.4 | 1.5 | 12.3 | 9.7 | 10.7 | 11.4 | 16.1 | 8.3 | 5.7 |
We did not find any eggs or tadpoles despite being present during the breeding season.
Kurixalus inexpectatus sp. nov. has been found calling as early as 26 April. Males would call from shrubs approximately 20 to 160 cm above temporary pools in and along roadside ditches. Temporary pools were 15 cm deep and up to 8 m long. In April, only sparse numbers of individuals were found calling. In July, full choruses could be heard, yet no individuals were found engaged in amplexus. No females have been found.
The vegetation primarily consisted of shrubs and secondary broad-leaved forest. No specimens were found in the adjacent bamboo forest.
Currently, the species is only known from the type locality, on the outskirts of the Wizard of Oz resort in Chuanbu Village, Changxing County, Huzhou City, Zhejiang Province, China. Surveys were made in the surrounding mountains for additional populations without success, including mountain ranges in Anhui and Jiangsu Provinces. The resort is situated at the southeast edge, in a northwest-to-southeast valley lower than 100 m in elevation. A creek comes from the hills, into a reservoir, which then flows about 2 km along the valley through the extent of the resort. The area was intended to be a plantation (unconfirmed, but suspected to be bamboo, based on the number of surrounding bamboo plantations), but in 2013, the land was set aside for the resort (pers. comm.). Now the resort consists of tea plantations, peach orchards, well-manicured grasses, a bamboo forest and miscellaneous shrubbery.
The electronic edition of this article conforms to the requirements of the amended International Code of Zoological Nomenclature and, hence, the new names contained herein are available under that Code from the electronic edition of this article. This published work and the nomenclatural acts it contains have been registered in ZooBank, the online registration system for the ICZN. The ZooBank LSIDs (Life Science Identifiers) can be resolved and the associated information viewed through any standard web browser by appending the LSID to the prefix “http://zoobank.org/”. The LSID for this publication is: urn:lsid:zoobank.org:pub: 3CCB356B-F075-4EE5-8366-FE96B855F884. The new species name Kurixalus inexpectatus sp. nov. has been registered under LSID: urn:lsid:zoobank.org:act: 02D394DE-BB1C-4C17-BB70-656D68814C8F.
The molecular data and phylogeographic patterns presented here are supported by both call properties and morphological data, highlighting a significant segregation between K. inexpectatus and other species. The morphological analysis is robust in that K. inexpectatus is significantly different from closely-related clades in terms of calls and morphology and it has diverged from the most closely related species ca. 3.06 Mya.
In spite of the high genetic homogeneity between the 12S rRNA gene sequences of K. inexpectatus and its homologous species K. idiootocus, the haplotype distributions and phylogeny inferred from the nuDNA TYR gene fragment showed a distinction between the two clades. The incongruence in pattern of sequence divergences between 12S rRNA gene and TYR sequences may result from dissimilarities in the rate of evolution between mitochondrial and nuclear loci in the Kurixalus lineage. Accordingly, the phylogenetic tree inferred from nuDNA TYR gene and concatenating gene fragments of 12S rRNA and TYR also supported the sister species relationship between Kurixalus inexpectatus and K. idiootocus (subclade B2, BP = 97%) and recovered the monophyly of K. inexpectatus (Fig.
Our haplotype network inferred from the nuDNA TYR gene sequences demonstrated the absence of identical haplotypes between K. inexpectatus and K. idiootocus (Clade A; Fig.
Our results highlight the importance of advanced genetic analyses to support the conventional distance-based genetic divergence analysis and especially analyses on species delimitation (Table
The lack of clear morphological characteristics is not unexpected for cryptic species and especially in treefrogs. However, identification based on range seems to be a reliable criterion. It is interesting that we did not find any individual in the bamboo forest while the genus is generally associated with this type of vegetation and further surveys may provide a different point of view. We recommend surveys on the contiguous mountain chain to determine the range of the species and the potential connectivity with other geographically related mountain ranges.
Our work revealed a previous undescribed species of Kurixalus that was disjunct from the next closest population of the genus by nearly 700 km. The population was found in a highly developed region of northern China, yet surprisingly has gone unnoticed. This discovery reiterates the need to survey regions of the countryside that have been poorly studied. Such efforts should be especially considered in regions of high development, to ensure that potentially critically endangered species, previously unknown to science are not lost.
We would like to thank our driver Mr. Peng and the reviewers for this manuscript. This work was supported by the Foreign Youth Talent Program (QN2021014013L) from the Ministry of Science and Technology of the People’s Republic of China to AB.
Tables S1–S4, Figures S1, S2
Data type: Docx file.
Explanation note: Table S1. The description of advertisement calls of five Kurixalus species. Table S2. The factor loading of principal analysis on call properties in Kurixalus inexpectatus sp. nov. and Kurixalus idiootocus. We listed the coefficients of correlation between call properties and their corresponding factors and listed out the Eigen values and accumulate explained variance of each Factor. Table S3. The factor scores of each individual for Kurixalus inexpectatus sp. nov. and Kurixalus idiootocus and the results of Discriminant Function Analysis (Wilks’ Lambda = 0.11, F (4.19) = 38.1, p < 0.001). This table shows the statistics and coefficient of determination of each factor and the square of Mahalanobis distance from each group centre, the post hoc probability (in brackets) to a group and the assigned group for each observation (individual). Table S4. Morphological data for Kurixalus inexpectatus sp. nov. used in this analysis. For the analyses, all measurements were adjusted for variations in body size, i.e. each value was divided by the SVL of the individual. The data presented here are not corrected for size. Data extracted from our samples and from the literature ananjevae (