Comprehensive approaches reveal three cryptic species of genus Nidirana (Anura, Ranidae) from China

Abstract Three cryptic species, which were previously reported as Nidirana adenopleura, are revealed on the basis of comprehensive approaches. Nidirana guangdongensis Lyu, Wan, and YY Wang, sp. nov. is distributed in Nanling Mountains and southern Luoxiao Mountains, Nidirana mangveni Lyu, Qi, and YY Wang, sp. nov. is known from northern Zhejiang, and Nidirana xiangica Lyu and YY Wang, sp. nov. occurs in Xiangjiang River Basin, while the true Nidirana adenopleura is designated from Taiwan Island, northern Fujian, southern Zhejiang, and central Jiangxi. These three new species can be distinguished from all congeners by significant divergences in the mitochondrial 16S and CO1 genes, differences in advertisement calls, and the combination of multiple characteristics. This work indicates that the current records of Nidirana adenopleura should be of a species complex composed of multiple species and have clarified the true identity of N. adenopleura.

Among the species in genus Nidirana, N. adenopleura has the widest distribution area and has been reported from Taiwan, Fujian, Zhejiang, Anhui, Jiangxi, Guangdong, Guangxi, Hunan and Guizhou (Fei et al. 2009(Fei et al. , 2012. In the previous study (Lyu et al. 2017), the populations from Taiwan, northern Fujian, Jingning County of Zhejiang, and Mt Jinggang of Jiangxi were confirmed as the same species, which also synonymized N. caldwelli Schmidt, 1925 with N. adenopleura. Besides, it is worth noting that the frogs previously considered as N. adenopleura from Mt Dayao of Guangxi and Mt Leigong of Guizhou were respectively revealed as two new species, N. yaoica and N. leishanensis, most recently (Lyu et al. 2019;Li et al. 2019). Nevertheless, the exact taxonomic statuses of other N. adenopleura populations from China have not yet be tested.
Through our herpetological surveys throughout southeastern China, we have collected a series of Nidirana specimens which were previously reported as N. adenopleura (Fei et al. 2009(Fei et al. , 2012Li et al. 2011;Mo et al. 2014). However, comprehensive analyses of molecules, bioacoustics, and morphology have indicated that these specimens are distinctive from all known congeners including the true N. adenopleura (designated here as N. adenopleura sensu stricto), which suggests they should belong to three unnamed cryptic species. Therefore, based on the results of our present work, we herein describe them as three new species of the genus Nidirana.

Taxon sampling
For the molecular analysis, a total of 54 muscular samples of Nidirana were used, of which 41 are from the undescribed specimens, eight from the true N. adenopleura, two from N. hainanensis and three from N. leishanensis. All samples were attained from euthanatized specimens and then preserved in 95% ethanol and stored at -40 °C. In ad-dition, 43 sequences from all known Nidirana congeners and two sequences from the out-group Babina Thompson, 1912(following Lyu et al. 2017 were obtained from GenBank and incorporated into our dataset. Detailed information of these materials is shown in Table 1 and Fig. 1.

Phylogenetic analyses
DNA sequences were aligned by the Clustal W algorithm with default parameters (Thompson et al. 1997) and trimmed with the gaps partially deleted in MEGA 6 (Tamura et al. 2013). Two gene segments, 644 base pairs (bp) of CO1 and 1049 bp of16S, were concatenated seriatim into a 1693-bp sequence, and were further tested in jmodeltest v2.1.2 with Akaike and Bayesian information criteria, all resulting the best-fitting nucleotide substitution models of GTR+I+G. Sequenced data was analyzed using Bayesian inference (BI) in MrBayes 3.2.4 (Ronquist et al. 2012), and maximum likelihood (ML) in RaxmlGUI 1.3 (Silvestro and Michalak 2012). Two independent runs were conducted in a BI analysis, each of which was performed for 10,000,000 generations and sampled every 1000 generations with the first 25% samples were discarded as burn-in, resulting a potential scale reduction factor (PSRF) of < 0.005. In ML analysis, the bootstrap consensus tree inferred from 1000 replicates was used to represent the evolutionary history of the taxa analyzed. Mean genetic distances between and within species were calculated in MEGA 6 using the uncorrected p-distance model. window with no overlap). The following measurements were taken for each call: call duration (the time between onset of the first note and offset of the last note in a call), note duration (the time between onset and offset of a note), note rise time (the time between onset and max amplitude of a note), and note interval (the time between adjacent notes in a call). Comparison bioacoustics descriptions of known congeners were obtained from the literature (Matsui and Utsunomiya 1983;Chou 1999;Fei et al. 2007Fei et al. , 2009Chuaynkern et al. 2010;Lyu et al. 2017Lyu et al. , 2019Li et al. 2019).

Morphology
Comparison characters of all known congeners were obtained from the literature (Boettger 1895;Boulenger 1904Boulenger , 1909Schmidt 1925;Chang and Hsu 1932;Bourret 1937;Kuramoto 1985;Chou 1999;Fei et al. 2007Fei et al. , 2009Matsui 2007;Chuaynkern et al. 2010;Lyu et al. 2017Lyu et al. , 2019Li et al. 2019) and 74 examined museum specimens of seven species which are listed in the Appendix. All specimens were fixed in 10% buffered formalin and later transferred to 70% ethanol, and deposited in the Museum of Biology, Sun Yat-sen University (SYS), Institute of Herpetology, Shenyang Normal University (SYNU), Natural History Museum of Guangxi (NHMG), and Chengdu Institute of Biology, Chinese Academy of Sciences (CIB), China.
Morphological descriptions mainly follow Fei et al. (2009), Chuaynkern et al. (2010 and Lyu et al. (2017). Sex and age were determined by secondary sexual characters, i.e., the presence of suprabrachial glands in males. Webbing formula was written according to Savage (1975). External measurements were made for the unnamed Nidirana specimens and 18 specimens of N. adenopleura, with digital calipers (Neiko 01407A Stainless Steel 6-Inch Digital Caliper, USA) to the nearest 0.1 mm. These measurements were as follows: SVL snout-vent length (from tip of snout to posterior margin of vent); HDL head length (from tip of snout to the articulation of the jaw); HDW head width (head width at the commissure of the jaws); SNT snout length (from tip of snout to the anterior corner of the eye); IND internasal distance (distance between nares); IOD interorbital distance (minimum distance between upper eyelids); ED eye diameter (from the anterior corner of the eye to posterior corner of the eye); TD tympanum diameter (horizontal diameter of tympanum); TED tympanum-eye distance (from anterior edge of tympanum to posterior corner of the eye); HND hand length (from the proximal border of the outer palmar tubercle to the tip of digit III); RAD radio-ulna length (from the flexed elbow to the proximal border of the outer palmar tubercle); FTL foot length (from distal end of shank to the tip of digit IV); TIB tibial length (from the outer surface of the flexed knee to the heel).
Principal component analysis (PCA), one-way analysis of variance (ANOVA) and Tukey test for multiple comparisons, were performed on the adult male specimens, of which the morphometric measurements were ln-transformed in order to normalize the variables, to test the significance of differences on morphometric characters among different species, using R 3.3.2 (R Core Team 2016).

Phylogenetic analyses
The ML and BI analyses resulted in essentially identical topologies and were integrated in Fig. 2, in which the major nodes were sufficiently supported with the bootstrap supports (BS) for maximum likelihood analysis > 75 and the Bayesian posterior probabilities (BPP) > 0.95. Mean p-distance among all in-group and out-group species used in this study are given in Table 2. In the phylogenetic result, all samples of genus Nidirana formed a monophyletic group, which can be further divided into four highly supported clades A, B, C, and D (the names of clades follow Lyu et al. (2017)). This result is consistent with the phylogenic relationship in previous studies (Lyu et al. 2017(Lyu et al. , 2019Li et al. 2019). However, the relationship among clades B, C, and D remains unresolved due to the insignificant supported values among these clades.  Table 1. Within clade D, the samples from Taiwan, northern Fujian, southern Zhejiang and central Jiangxi are grouped in a distinct and substantial single lineage (red color in Figs. 1 and 2) that represent the Nidirana adenopleura s. s. whose type locality is in Taiwan Island. Nevertheless, the samples from northern Zhejiang (brown color in Figs. 1 and 2), which were previously recorded as N. adenopleura, are grouped in a distinct single lineage (designated here as the northern lineage), that is non-monophyletic with the lineage of N. adenopleura s. s. and has significant divergences against all congeners. In addition, the samples from Xiangjiang River Basin (dark green color in Figs. 1 and 2) and from Nanling Mountains and southern Luoxiao Mountains (bright green color in Figs. 1 and 2), which were also previously considered as N. adenopleura, form two distinct lineages with significant divergences respectively (designated here as the western and southern lineages respectively) within clade C, which are both significantly distant from the lineage of the N. adenopleura s. s. in clade D. This phylogenetic result indicates that the previous identifications for the populations from northern Zhejiang (northern lineage), from Xiangjiang River Basin (western lineage), and from Nanling Mountains and southern Luoxiao Mountains (southern lineage) are incorrect, and these three populations represent three separate evolutionary lineages within the genus Nidirana.

Bioacoustics analysis
The call spectrograms of Nidirana adenopleura s. s. and the three unnamed lineages are shown in Fig. 3 and the measurement parameters are listed in Table 3.
The advertisement calls of the southern lineage is different from the congeners by (1) containing 2-4 (2.9 ± 0.7, N = 54) identical regular notes vs. containing 10-25 fast-repeated regular notes in Nidirana okinavana; containing 5-7 regular notes in N. lini; containing 4-7 regular notes in N. pleuraden; containing 2-4 fast-repeated double-notes in N. hainanensis; containing a significantly different first note in N. daunchina and N. nankunensis; containing a single note in N. leishanensis; (2) the call notes last 134.0-226.7 ms vs. the call notes last 30-54 ms in N. yaoica; (3) the calls of the southern lineage is similar to that of N. adenopleura s. s. but can be distinguished by the relative shorter note duration (164.3 ± 16.2 ms vs. 212.3 ± 33.0 ms) and shorter note rise time (28.7 ± 32.4 ms vs. 106.1 ± 70.7 ms).
The advertisement calls of the northern lineage is different from the congeners by (1) containing 2-7 (4.6 ± 1.2, N = 108) identical regular notes vs. containing 10-25 fast-repeated regular notes in Nidirana okinavana; containing 2-4 fast-repeated double-notes in N. hainanensis; containing a significantly different first note in N. daunchina and N. nankunensis; containing a single note in N. leishanensis; (2) the call notes last 89.0-203.0 ms vs. the call notes last 30-54 ms in N. yaoica; (3) the calls of the southern lineage is similar to that of N. adenopleura s. s. but can be distinguished by the relative shorter note duration (136.9 ± 23.2 ms vs. 212.3 ± 33.0 ms) and shorter note rise time (79.5 ± 26.9 ms vs. 106.1 ± 70.7 ms); (4) the calls of the southern lineage is similar to that of the southern lineage but can be distinguished by more note number in per call (2-7, 4.6 ± 1.2 vs. 2-4, 2.9 ± 0.7).
The advertisement calls of the western lineage is different from the congeners by (1) containing a significantly different first note vs. containing several identical regular notes in Nidirana adenopleura, southern lineage, northern lineage, N. yaoica, N. chapaensis, N. lini, and N. pleuraden; containing 2-4 fast-repeated double-notes in N. hainanensis; containing a single note in N. leishanensis; (2) containing 2-3 notes vs. containing 10-25 fast-repeated regular notes in N. okinavana; containing 13-15 fastrepeated notes in N. nankunensis; (3) the calls of the western lineage is similar to that of N. daunchina but can be distinguished by the relative shorter note intervals time (125.8 ± 17.8 ms vs. 193.6 ± 26.3 ms) and shorter duration of non-first notes (74.6 ± 11.8 ms vs. 140.6 ± 5.6 ms).

Morphology
The results of PCA based on morphometric measurements of the male specimens of Nidirana adenopleura s. s. and the three unnamed lineages are shown in Fig. 4. The extracted components PC1 eigenvectors accounted for 58.8% of the variance, PC2 for 18.6%, PC3 for 6.12%, and PC4 for 5.46%, which cumulate 88.98% of the variance. As shown on the scatter plot of PC1 and PC2, the specimens of N. adenopleura s. s., are significantly different from the specimens of the other three lineages. The specimens of the western lineage are also well separated from others. However, the specimens of the southern and northern lineages, which are significantly distant from each other in the phylogenetic tree, overlap with each other in the PCA result.
The results of one-way ANOVA and Tukey test for multiple comparisons are given in Table 4. The results indicate that all morphometric data are significantly different among Nidirana adenopleura s. s. and the three unnamed lineages (all p-values < 0.05). Specifically, for the specimens of N. adenopleura s. s. and the unnamed northern lineage which phylogenetically clustered within clade D together, they are significantly different in SVL, HDL, ED, RAD, FTL, and TIB. For the specimens of the unnamed western and southern lineages which phylogenetically clustered within clade C together, they are significantly different in HDW, SNT, IND, ED, TD, and RAD.
Detail comparisons among specimens of the western, southern, and northern lineages and all recognized congeners are listed in Table 5. The populations of the southern, northern, and western lineages can be readily and consistently distinguished from all other species by a combination of characteristics (see Comparisons below).      Etymology. The species name guangdongensis refers to Guangdong (广东), also known as Yue (粤), which is the province where the type locality, Shimentai Nature Reserve, belongs to. Differential diagnosis. Nidirana guangdongensis sp. nov. is distinguished from its congeners by the following combination of the morphological characteristics: (1) body large and elongated, with SVL 50.0-58.4 (53.9 ± 3.3, N = 5) mm in adult males, and SVL 55.3-59.3 (57.0 ± 2.1, N = 3) mm in adult females; (2) disks of digits dilated, rounded; (3) lateroventral grooves present on every digit except finger I; (4) heels overlapping; (5) tibio-tarsal articulation reaching the nostril; (6) mid-dorsal stripe present on posterior dorsum; (7) week supernumerary tubercles below the base of each finger, palmar tubercles prominent and distinct; (8) supratympanic fold absent; (9) white horny spinules on the entirely dorsum, dorsolateral folds, flanks and dorsal hindlimbs, while absent on temporal regions in males; (10) a pair of subgular  (2) relative finger lengths II < I < IV < III vs. II < I = IV < III in N. chapaensis; vs. II < IV < I < III in N. leishanensis; (3) presence of lateroventral groove on every digit except finger I vs. absent on fingers and toes in N. pleuraden; vs. absent or barely visible on fingers in N. daunchina; vs. present on finger I in N. yaoica, N. leishanensis and N. hainanensis; (4) tibio-tarsal articulation reaches at the nostril vs. beyond the snout tip in N. lini; (5) white horny spinules on the entirely dorsum and flanks in males vs. absent on dorsum and flanks or few above vent in N. nankunensis, N. okinavana, N. daunchina, N. yaoica, N. chapaensis, N. leishanensis and N. hainanensis; vs. present on dorsum while absent on flanks in N. adenopleura, N. lini and N. pleuraden; (6) the presence of a single nuptial pad on finger I vs. absent in N. hainanensis; vs. divided into two parts in N. chapaensis; vs. two nuptial pads on fingers I and II respectively; (7) the presence of a pair of subgular vocal sacs vs. absent in N. okinavana.
Description of holotype. SYS a005767 (Figs 5, 6), adult male. Body large and elongated, SVL 55.2 mm; head longer than wide (HDW/HDL 0.90), flat above; snout rounded in dorsal and lateral views, slightly protruding beyond lower jaw, longer than horizontal diameter of eye (SNT/ED 1.30); canthus rostralis distinct, loreal region concave; nostril round, directed laterally, closer to the snout than to the eye; a longitudinal swollen mandibular ridge extending from below nostril through lower edges of eye and tympanum to above insertion of arm, where the ridge is intermittent, forming a maxillary gland and shoulder gland; supratympanic fold absent; interorbital space flat, narrower than internasal distance (IND/IOD 1.24); pupil elliptical, horizontal; tympanum distinct, round, TD/ED 0.86, and close to eye, TED/TD 0.32; pineal ocellus slightly visible; vomerine ridge present, bearing small teeth; tongue large, cordiform, notched behind; a pair of subgular vocal sacs present.
Forelimbs moderately robust, lower arm 0.17 of SVL and hand 0.27 of SVL; fingers thin, relative finger lengths II < I < IV < III; tip of each finger slightly dilated, forming rounded disks; lateroventral grooves on all fingers except finger I, not meeting at the tip of disks; fingers free of webbing; presence of distinct lateral fringes on inner and outer sides of fingers II, III and IV, and on outer side of finger I; subarticular tubercles prominent and rounded; week supernumerary tubercles below the base of each finger; three elliptic, large, prominent and very distinct palmar tubercles; a single nuptial pad on the dorsal surface of first finger, nuptial spinules invisible.
Hindlimbs relatively robust, tibia 0.54 of SVL and foot 0.77 of SVL; heels overlapping when hindlimbs flexed at right angles to axis of body; tibio-tarsal articulation reaching the nostril when hindlimb is stretched along the side of the body; toes relatively long and thin, relative lengths I < II < V < III < IV; tip of each toe slightly dilated with remarkable elongated ventral callous pad, forming long and pointed disk; well-developed lateroventral grooves on toes , not meeting at the tip of disks; webbing moderate, webbing formula: I 1⅓ -2 II 1⅓ -2⅓ III 1⅔ -3 IV 3⅓ -1⅓ V; presence of lateral fringes on inner and outer sides of each toes, forming distinct dermal flap on the lateral edges of toes I and V; subarticular tubercles rounded, prominent; inner metatarsal tubercle elliptic, length triple the width; outer metatarsal tubercle indistinct, small and rounded; tarsal folds and tarsal tubercle absent.
Dorsal surface rough with dense horny spinules; developed dorsolateral fold with sparse horny spinules from posterior margin of upper eyelid to above groin but intermittent posteriorly; flank rough with dense tubercles and dense horny spinules; a large and smooth suprabrachial gland behind base of forelimb, prominent; dorsal surface of forelimb relatively smooth without horny spinules, weak longitudinal ridges on upper arms and slightly extending to lower arm; the dorsal surfaces of thigh and tibia rough with dese tubercles and dense horny spinules, forming several longitudinal ridges. Ventral surface of throat, body, and limbs smooth; large flattened tubercles densely arranged on the rear of thigh and around vent.
Coloration of holotype. In life (Fig. 5), dorsal surface reddish brown; horny spinules on the skin white; pineal ocellus yellowish; a yellowish mid-dorsal stripe on the posterior dorsum; dorsolateral fold dark brown; upper flank dark brown; lower flank light brown; suprabrachial gland light brown. Dorsal forelimbs light brown; a longitudinal black stripe on the anterior surface of the forelimb; dorsal hindlimbs dark brown, four dark crossbars on the thigh, three on the tibia and three on the tarsus. Loreal and temporal regions dark brown, tympanum light brown; upper ⅓ iris brownish white and lower ⅔ iris reddish brown; maxillary gland and shoulder gland white. Throat dark purplish brown; ventral surface of body and limbs creamy white; rear thigh tinged with pink; ventral hand white with large purplish brown patches; ventral foot purplish brown.
In preservative (Fig. 6), dorsal surface faded with the pineal ocellus and mid-dorsal stripe clearer; white spinules more distinct; dorsal limbs faded with the crossbars more distinct; ventral surface faded, throat grey.
Variations. Measurements of type series are given in Table 6. All specimens were similar in morphology. Females (57.0 ± 2.1 mm, N = 3) (Fig. 7A) are not significantly larger than males (53.9 ± 3.3 mm, N = 5), but relatively smooth than males, not bearing white horny spinules on the dorsum, dorsolateral folds, and flanks. Pineal ocellus invisible in SYS a005765 (Fig. 7B); numerous black spots on flanks in SYS a005766.
Distribution and ecology. Currently, Nidirana guangdongensis sp. nov. is known from northern Guangdong, southern Jiangxi and southeastern Hunan, indicating that this frog is distributed in the Nanling Mountains and southern Luoxiao Mountains of southern China. The frog inhabits in natural ponds. The adult males call at the water surface and the females oviposit directly into the water (Fig. 7C) from April to June. The tadpoles of this species remain unknown. In Mt Nankun, N. guangdongensis sp. nov. is sympatric with N. nankunensis in the same pond and is more abundant.  ern Zhejiang, who contributed mostly on Chinese herpetological taxonomy and natural history. He is also the author of Nidirana daunchina, a congener of this new species.
Comparison. Morphologically, Nidirana mangveni sp. nov. is unique when compared with all recognized congeners by the combination of the following characteristics: (1) large body size, SVL 53.6-59.7 mm in males and 59.7-65.1 mm in females vs. < 53.0 mm in males or females in N. nankunensis, N. okinavana, N. daunchina, N. yaoica, N. chapaensis and N. hainanensis; (2) relative finger lengths I < II < IV < III vs. II < I = IV < III in N. chapaensis; vs. II < IV < I < III in N. leishanensis; vs. II < I < IV < III in all other congeners; (3) absent of lateroventral groove on fingers I and II vs. absent on fingers and toes in N. pleuraden; vs. absent or barely visible on fingers in N. daunchina; vs. present on finger II in all other congeners; (4) tibio-tarsal articulation reaches at the anterior corner of eye vs. beyond the snout tip in N. lini; vs. at the nostril in N. guangdongensis, N. nankunensis, N. daunchina, N. yaoica, N. chapaensis and N. hainanensis; (5) week supratympanic fold present vs. absent in N. guangdongensis, N. adenopleura, N. nankunensis, N. daunchina, N. yaoica, N. hainanensis, and N. lini; (6)  white horny spinules on the posterior or entire dorsum in males vs. absent on dorsum or few above vent in N. nankunensis, N. okinavana, N. daunchina, N. yaoica, N. chapaensis, N. leishanensis and N. hainanensis; (7) the presence of a single nuptial pad on finger I vs. absent in N. hainanensis; vs. divided into two parts in N. chapaensis; vs. two nuptial pads on fingers I and II respectively; (8) the presence of a pair of subgular vocal sacs vs. absent in N. okinavana.
Description of holotype. SYS a006313 (Figs 8, 9), adult male. Body large and elongated, SVL 54.0 mm; head longer than wide (HDW/HDL 0.87), flat above; snout rounded in dorsal and lateral views, slightly protruding beyond lower jaw, longer than horizontal diameter of eye (SNT/ED 1.22); canthus rostralis distinct, loreal region concave; nostril round, directed laterally, closer to the snout than to the eye; a longitudinal swollen mandibular ridge extending from below nostril through lower edges of eye and tympanum to above insertion of arm, where the ridge is intermittent, forming a maxillary gland and shoulder gland; week supratympanic fold present; interorbital space flat, narrower than internasal distance (IND/IOD 1.25); pupil elliptical, horizontal; tympanum distinct, round, TD/ED 0.73, and close to eye, TED/TD 0.31; pineal ocellus present; vomerine ridge present, bearing small teeth; tongue large, cordiform, notched behind; a pair of subgular vocal sacs present.
Forelimbs moderately robust, lower arm 0.18 of SVL and hand 0.26 of SVL; fingers thin, relative finger lengths I < II < IV < III; tip of each finger slightly dilated, forming rounded disks; lateroventral grooves on fingers III and IV, not meeting at the tip of disks; fingers free of webbing; presence of distinct lateral fringes on inner and outer sides of fingers II, III and IV, absent on finger I; subarticular tubercles prominent and rounded; developed supernumerary tubercles below the base of each finger; three elliptic, large, prominent and very distinct palmar tubercles; a single nuptial pad on the dorsal surface of first finger, nuptial spinules invisible.
Hindlimbs relatively robust, tibia 0.52 of SVL and foot 0.76 of SVL; heels overlapping when hindlimbs flexed at right angles to axis of body; tibio-tarsal articulation reaching the anterior corner of eye when hindlimb is stretched along the side of the body; toes relatively long and thin, relative lengths I < II < V < III < IV; tip of each toe slightly dilated with remarkable elongated ventral callous pad, forming long and pointed disk; well-developed lateroventral grooves on toes , not meeting at the tip of disks; webbing moderate, webbing formula: I 1½ -2⅓ II 1⅓ -2⅓ III 1½ -3 IV 3⅓ -1⅔ V; presence of lateral fringes on inner and outer sides of each toes, forming distinct dermal flap on the lateral edges of toes I and V; subarticular tubercles rounded, prominent; inner metatarsal tubercle elliptic, length triple the width; outer metatarsal tubercle indistinct, small and rounded; tarsal folds and tarsal tubercle absent.
Dorsal skin of head and anterior body smooth, posterior dorsum of body rough with dense tubercles with horny spinules; week intermittent dorsolateral fold from posterior margin of upper eyelid to above groin ; upper flank with sparse tubercles; a large and smooth suprabrachial gland behind base of forelimb, not prominent; dorsal surface of upper arm smooth with sparse tubercles without spinules; the dorsal surfaces of thigh and tibia relatively rough with several weak longitudinal ridges and tubercles bearing spinules. Ventral surface of throat, body, and limbs smooth; large flattened tubercles densely arranged on the rear of thigh and around vent.
Coloration of holotype. In life (Fig. 8), dorsal surface brown; pineal ocellus yellowish; mid-dorsal stripe unclear; dorsolateral fold dark brown; upper flank olive brown; lower flank creamy white; suprabrachial gland white. Dorsal limbs brown; a longitudinal black stripe on the anterior surface of the forelimb; three dark crossbars on the thigh, three on the tibia and three on the tarsus. Loreal and temporal regions dark, tympanum light brown; upper ⅓ iris brownish white and lower ⅔ iris reddish brown; maxillary gland and shoulder gland white. Throat white tinged with pink, but two subgular vocal sacs flesh colored; ventral surface of body and limbs creamy white; rear thigh tinged with pink; ventral hand flesh colored; ventral foot brown.
Variations. Measurements of type series are given in Table 7. All specimens were similar in morphology. Females (62.4 ± 3.8 mm, N = 2) (Fig. 10) are relatively larger than males (56.2 ± 2.5 mm, N = 7), and more smooth than males. Pineal ocellus invisible in SYS a006310, 6311; dorsal surface light brown in SYS a006310, 6311; a short mid-dorsal stripe on the posterior dorsum in SYS a006311, 6416; spinules on the entire dorsum in SYS a006413.
Distribution and ecology. Currently, Nidirana mangveni sp. nov. is known from Mt Dapan, Mt Longmen, and Hangzhou Botanical Garden, all situated in northern Zhejiang, suggesting the Nidirana populations in northern Zhejiang might belong to this species. This frog inhabits natural or artificial swamps, ponds, and paddy fields. The adult males do not construct nests and calls at the water surface or the bank from May to August. The male individual SYNU12050569 which was found in early May bears indistinct nuptial pads but processes the suprabrachial gland, indicating the breeding season of this species begins from early May. The tadpoles of this species remain unknown.
Description of holotype. SYS a006492 (Figs 11, 12), adult male. Body large and elongated, SVL 56.3 mm; head slightly longer than wide (HDW/HDL 0.99), flat above; snout rounded in dorsal and lateral views, slightly protruding beyond lower jaw, longer than horizontal diameter of eye (SNT/ED 1.27); canthus rostralis distinct; loreal region concave, bearing horny spinules; nostril round, directed laterally, closer to the snout than to the eye; a longitudinal swollen mandibular ridge extending from below nostril through lower edges of eye and tympanum to above insertion of arm, forming a maxillary gland and shoulder gland; supratympanic fold absent; interorbital space flat, narrower than internasal distance (IND/IOD 1.27); pupil elliptical, horizontal; temporal region including tympanum with horny spinules, tympanum distinct, round, TD/ED 0.81, and close to eye, TED/TD 0.29; pineal ocellus distinct; vomerine ridge present, bearing small teeth; tongue large, cordiform, notched behind; a pair of subgular vocal sacs present.
Forelimbs moderately robust, lower arm 0.20 of SVL and hand 0.26 of SVL; fingers thin, relative finger lengths II < I < IV < III; tip of each finger slightly dilated, forming rounded disks; lateroventral grooves on all fingers, not meeting at the tip of disks; fingers free of webbing; presence of distinct lateral fringes on inner and outer sides of fingers II, III, and IV, and on outer side of finger I; subarticular tubercles prominent and rounded; developed supernumerary tubercles below the base of each finger; three elliptic, large, prominent and very distinct palmar tubercles; a single nuptial pad on the dorsal surface of first finger, nuptial spinules invisible.
Hindlimbs relatively robust, tibia 0.50 of SVL and foot 0.74 of SVL; heels just meeting when hindlimbs flexed at right angles to axis of body; tibio-tarsal articulation reaching the loreal region when hindlimb is stretched along the side of the body; toes relatively long and thin, relative lengths I < II < V < III < IV; tip of each toe slightly dilated with remarkable elongated ventral callous pad, forming long and pointed disk; well-developed lateroventral grooves on toes , not meeting at the tip of disks; webbing moderate, webbing formula: I 1½ -2 II 1⅓ -2⅓ III 1⅔ -3 IV 3⅓ -1⅔ V; presence of lateral fringes on inner and outer sides of each toes, forming distinct dermal flap on the lateral edges of toes I and V; subarticular tubercles rounded, prominent; inner metatarsal tubercle elliptic, length triple the width; outer metatarsal tubercle indistinct, small and rounded; tarsal folds and tarsal tubercle absent.
Dorsal surface very rough with dese tubercles and dense horny spinules; developed dorsolateral fold with sparse horny spinules from posterior margin of upper eyelid to above groin but intermittent posteriorly ; flank very rough with sparse warts, dense tubercles and dense horny spinules; a large and rough suprabrachial gland behind base of forelimb, distinctly prominent; dorsal surface of forelimb rough with dense horny spinules, two weak longitudinal ridges on upper arms and slightly extending to lower arm; the dorsal surfaces of thigh and tibia rough with dese tubercles and dense horny spinules, forming several longitudinal ridges. Ventral surface of throat, body, and limbs smooth; large flattened tubercles densely arranged on the rear of thigh and around vent.
Coloration of holotype. In life (Fig. 11), dorsal surface greenish brown; horny spinules on the skin white; pineal ocellus yellowish; absence of mid-dorsal stripe; dorsolateral fold greenish brown; upper flank greenish brown, warts on flank yellowish; lower flank yellowish white with black stripe; suprabrachial gland yellowish white with black stripe. Dorsal limbs brown; two greenish crossbars on the thigh, two on the tibia and three on the tarsus. Loreal and temporal regions greenish brown, tympanum light brown; upper ⅓ iris brownish white and lower ⅔ iris reddish brown; maxillary gland and shoulder gland white. Throat and anterior chest dark purplish brown; ventral surface of body and limbs creamy white; rear thigh tinged with pink; ventral hand white with large brown patches; ventral foot purplish brown.
In preservative (Fig. 12), surface of dorsum and dorsal limbs changed as dark brown; white spinules significantly distinct; crossbars on limbs clearer; ventral surface faded, throat and anterior chest dark grey.
Variations. Measurements of type series are given in Table 8. All specimens were similar in morphology. Females (58.3 ± 4.0 mm, N = 4) (Fig. 13A, B) are not significantly larger than males (58.0 ± 2.2 mm, N = 6), but relatively smooth than males, not bearing white horny spinules on the dorsum, dorsolateral folds, flanks, and temporal region. Pineal ocellus invisible in SYS a006493 (Fig. 13C); dorsal surface reddish brown in SYS a006491 and greenish in SYS a007269 (Fig. 13D); numerous black spots on dorsum and flanks in SYS a007273; lateroventral grooves poorly developed on fingers I and II in SYS a002591.
Distribution and ecology. Currently, Nidirana xiangica sp. nov. is known from Mt Dawei and Mt Yangming of Hunan, Mt Wugong of western Jiangxi, and Mt Du-pangling of northeastern Guangxi, indicating its potential distribution area is in the Xiangjiang River Basin. The frog inhabits natural or artificial ponds and paddy fields. This species has no behavior of nest construction, and the adult males call at the water surface from May to August. The tadpoles of this species remain unknown.

Discussion
In morphology, most anuran species seem slightly similar to each other, and within several particular species, the coloration patterns are variable among individuals. These interspecific similarities and intraspecific variabilities have caused numerous misidentifications and synonymies, and calls for comprehensive approaches in the taxonomic research on anuran frogs. For instance, the species Nidirana guangdongensis sp. nov. and N. mangveni sp. nov. overlap with each other in the morphometric comparisons, while detailed morphological comparison, phylogenetic relationships, and bioacoustics analysis reveal their differences. The species N. xiangica sp. nov. is significantly different from N. adenopleura s. s. in morphology, phylogeny, and bioacoustics, but it was previously misidentified as N. adenopleura possibly to deficiencies in earlier research.
The species Nidirana adenopleura was originally described based on several specimens from Fuhacho Village (= Maobu or Wucheng, Nantou County), central Taiwan (Boulenger 1909;Jang-Liaw and Chou 2015), and was subsequently recorded over a wide area from southern China to northern Indochina (Chan-ard et al. 1999;Fei et al. 2009Fei et al. , 2012. Fei et al. (2007) described the reported population of N. adenopleura from Mt Diaoluo, Hainan as the new species N. hainanensis. Chuaynkern et al. (2010) re-allocated the specimens previously identified as N. adenopleura from Thailand to N. lini. Lyu et al. (2019) recognized the population from Mt Dayao, Guangxi as the new species N. yaoica. These taxonomic works have indicated that the current records of N. adenopleura represent a species complex. In the present study, based on a comprehensive molecular, morphological, and bioacoustics analysis, the recorded populations of N. adenopleura from Nanling Mountains and southern Luoxiao Mountains (southern lineage), northern Zhejiang (northern lineage), and Xiangjiang River Basin (western lineage), are revealed as the new species N. guangdongensis sp. nov., N. mangveni sp. nov., and N. xiangica sp. nov. Currently, the recognized distribution area of the true N. adenopleura covers the Taiwan Island, northern Fujian, southern Zhejiang and central Jiangxi, and other reported populations beyond these areas need further study.