Phylogenetic analyses using mtDNA gene sequences revealed the presence of two genetic clades in the genus Leptobrachella (former Leptolalax) from northern Thailand. One formed a clade including L. ventripunctata and L. aerea, while another included L. bourrreti and L. laoi. The first clade contained L. cf. minima from Doi Suthep, while the second clade contained L. murphyi and two undescribed species, one from Doi Inthanon, at a high altitude of 2,300 m asl, and another from Mae Hong Son, at Myanmar border. These form a clade, that is sister to L. murphyi. Morphologically, the first species is generally similar to L. murphyi, while the second species resembles L. cf. minima. However, they are clearly distinguished from those described species by great genetic differences, and we describe them as L. korifi sp. nov. and L. sinorensis sp. nov., respectively.
Introduction
Medium- to small-sized, forest-floor dwelling megophryids from southern China and Vietnam through Myanmar and Thailand to Peninsular Malaysia and Borneo were formerly called Leptolalax Dubois, 1980, but are now synonymized with Bornean Leptobrachella (Smith, 1925; see Discussion for problems with this classification). Now the genus includes nearly 100 species, of which one-third have been described within the last half-decade (e.g., Shi et al., 2021; Luo et al., 2022). This is chiefly because many of the members of this genus are morphologically very similar and were thus difficult to identify. However, acoustic analyses of their calls have detected some cryptic members (Matsui, 1997, 2006), and more recently, molecular phylogenetic analyses rapidly clarified the high species diversity of this genus (e.g., Chen et al., 2018).
Compared with neighboring Vietnam (e.g., Nguyen et al., 2021) and China (e.g., Chen et al., 2021), known species diversity of this genus in Thailand is low. Historically, Taylor (1962, as Leptobrachium) listed L. pelodytoides (Boulenger, 1893) and described L. minima. Then, in a publication of an illustrated guidebook on the amphibian fauna, Chan-ard (2003) listed four species (as Leptolalax: L. bourreti (Dubois, 1983), L. gracilis (Günther, 1872), L. heteropus (Boulenger, 1900), and L. pelodytoides). However, these records are doubtful, without sources of data or the bases for identification (Matsui, 2006). Matsui (2006) added three species (as Leptolalax: L. melanoleuca, L. fuliginosa, and L. sola) from southern region, while Jiang e al. (2013) described L. zhangyapingi (as Leptolalax) from Doi Saket, northern region. More recently, Niyomwan et al. (2019) used the generic name Leptobrachella and listed L. cf. bourreti, L. fuliginosa, L. heteropus, L. melanoleuca, L. minima, L. cf. pelodytoides, L. sola, and L. zhangyapingi as the Thai members, and Chen et al. (2021) described L. murphyi from the northern region. Thus, only seven species have been recorded as valid from the country.
During our field survey in Thailand nearly 30 years ago, we collected Leptobrachella or ascertained its occurrence without collection in northeastern (Loei Province), northern (Mae Hong Son and Chiang Mai Provinces), southwestern (Kanchanaburi and Prachuap Khiri Khan Provinces), and southern (Surat Thani and Narathiwat Provinces) regions of that country between 1994 and 2004 (regional divisions following Nabhitabhata et al., 2000). We failed to find specimens of this genus in other parts of Thailand. Specimens from the southern region were already described as three distinct species, L. melanoleuca, L. fuliginosa, and L. sola (Matsui, 2006, all as Leptolalax), but details of the more northern samples have been left unreported. Recently, L. murphyi Chen, Suwannapoom, Wu, Poyarkov, Xu, Pawangkhanant, and Che, 2021 was described from Chiang Mai Province, northern region. We also had this species in our collection. We examined molecular variation of northern specimens and found two species that are significantly different genetically from L. cf. minima (see below) and L. murphyi. In this paper, we describe these two species as new to science.
Materials and Methods
We collected Leptobrachella during our field survey between 1994 and 1995. We collected specimens at the type locality of L. minima (Doi Suthep: Taylor, 1962), but our specimens differed slightly from the original description and the holotype now stored in the FMNH (M. Matsui, unpublished data). We thus call our specimens L. cf. minima in this paper. We took tissues for subsequent biochemical analysis, and fixed the specimens for vouchers. Specimens, fixed in 10% formalin and later preserved in 70% ethanol, are stored at Graduate School of Human and Environmental Studies, Kyoto University (KUHE).
We obtained sequence data of mitochondrial DNA from muscle/liver samples preserved in 99% ethanol. We reconstructed phylogenetic trees from approximately 2,000 base pairs (bp) of the partial sequences of mitochondrial 12S and 16S rRNA genes to clarify the genetic structure of specimens representing Leptobrachella species and outgroup species, Leptobrachium hasseltii Tschudi, 1838 (Table 1). Methods for DNA extraction, and amplification and sequencing of the mtDNA fragments are the same as those reported by Hamidy et al. (2011). The resultant new sequences were deposited in GenBank (Accession numbers LC741025–741038; Table 1). The alignment matrices were subjected to estimate phylogenetic relationships using maximum likelihood (ML) and Bayesian inference (BI). Pairwise comparisons of uncorrected sequence divergences (p-distance) were also calculated. Details for these procedures are given in Matsui et al. (2010) and Hamidy et al. (2011). Pairwise comparisons of uncorrected sequence divergences (p-distance) were also calculated for 16S rRNA.
For morphological analyses, 19 body measurements were taken, mainly following Matsui (1984): (1) snout-vent length (SVL); (2) head length (HL); (3) nostril-eyelid length (N-EL); (4) snout length (SL); (5) eye length (EL, including eyelid); (6) tympanum-eye length (TEL); (7) tympanum diameter (TD); (8) head width (HW); (9) internarial distance (IND); (10) interorbital distance (IOD); (11) upper eyelid width (UEW); (12) forelimb length (FLL); (13) lower arm and hand length (LAL); (14) inner palmar tubercle length (IPTL); (15) hindlimb length (HLL); (16) tibia length (TL); (17) foot length (FL); (18) inner metatarsal tubercle length (IMTL); and (19) first toe length (1TOEL). We made all measurements to the nearest 0.1 mm with dial calipers under a binocular dissecting microscope.
Table 1.
Samples of Leptobrachella frogs and an outgroup species used for DNA analysis in this study, together with the information on voucher specimens, collection locality, and GenBank accession numbers. Voucher abbreviations; CIB: Chengdu Institute of Biology; KUHE: Kyoto University, Human and Environmental Studies; MNHN: Muséum National d'histoire Naturelle, Paris; VNMN: Vietnam National Museum of Nature.
In describing coloration, we adopted the terminology on dorsal pattern shown by Ohler et al. (2011). Sex and maturity of specimens were determined by observation of gonads and secondary sexual characters such as vocal opening. A single specimen from near the peak of Doi Inthanon proved to be a female, but we had no comparable females of L. murphyi at hand. We thus compared our data for L. murphyi with those shown in Table 3 of Chen et al. (2021). We chose 12 characters that could be reliably compared: SVL, HL (HDL in Chen et al., [2021]), N-EL (NEL in in Chen et al., 2021), SL (SNT in Chen et al. [2021]), T-EL (TEY in Chen et al. [2021]), TD (TMP in Chen et al. [2021]), HW (HDW in Chen et al. [2021]), IND, IOD, UEW, TL (TIB in Chen et al. [2021]), and IMTL (IMT in Chen et al. [2021]). We found that the two datasets for male L. murphyi overlap with each other in terms of the ranges of variation relative to SVL. In the data of Chen et al. (2021), sexual dimorphism in L. murphyi was found only in SVL (23.2–24.9 mm in males and 29.3–32.1 mm in females) and HL relative to SVL (RHL 35.3–39.1% in males and 31.5–34.5% in females). We thus compared the single female specimen from near the peak of Doi Inthanon with eight male specimens of L. murphyi at hand, excluding SVL and RHL for possible presence of sexual dimorphism.
Fig.1.
Bayesian tree from a ∼2,849 bp sequence of mitochondrial 12S and 16S rRNA genes for samples of Leptobrachella. Numbers above or below branches represent bootstrap supports for ML inference and Bayesian posterior probability (ML-BS/BPP).
For comparison, we also examined specimens of L. pelodytoides stored at the Museo Civico di Storia Naturale, Genova (MSNG 29845, designated lectotype); Natural History Museum, London (BM 1947.2.2514, syntype); and Zoologisches Museum, Universität Humboldt, Berlin (ZMB 11588), and the holotype of L. minima stored at the Field Museum of Natural History, Chicago (FMNH 178228).
Results
In the molecular phylogenetic analyses, all samples, including outgroups, yielded 2,849 bp of concatenated fragments of mtDNA genes. The best substitution model selected was GTR with gamma shape parameter (G) of 0.396 for ML and 0.416 for BI, with the likelihood values (-lnLs) of 17,469.2785 and 17,497.847, respectively. Two different optimality criteria yielded identical relationships in the phylogenetic analyses employed. As shown in the BI tree (Fig. 1), specimens of Leptobrachella, including those from northern Thailand (MLBS=100, BPP=1.00), were clearly split into two well-supported major clades, A and B.
Major Clade A (MLBS=82, Bayesian posterior probability [BPP]=1.00) included L. sungi and a clade (MLBS=100, BPP=1.00), in which a subclade of L. ventripunctata and L. aerea (MLBS=74, BPP=0.98) and a subclade of L. cf. minima from Doi Suthep and Chiang Dao (MLBS=100, BPP=1.00) were recognized. Major Clade B (MLBS=97, BPP=1.00) included L. laoi and a clade of L. bourreti and northern Thai samples (MLBS=100, BPP= 1.00). The latter (MLBS=95, BPP=1.00) was split into a subclade of L. murphyi (MLBS= 100, BPP=1.00), and another subclade (MLBS =100, BPP=1.00) including the lineage from near the peak of Doi Inthanon, and the lineage from Mae Hong Son.
Uncorrected p-distances for a short fragment of 16S rRNA (516 bp) between the two clades, A and B, were substantially large, ranging from 9.3–13.6%. Distances in Clade B ranged from 7.2% between the lineage from near the peak of Doi Inthanon and the lineage from Mae Hong Son, to 11.7% between the lineage from Mae Hong Son and L. laoi (Table 2).
The specimens of Leptobrachella from northern Thailand are similar to each other, and their morphological separation was not easy. However, the values for p-distances were sufficiently larger than the minimum values reported in allied species (e.g., 5.5%: Luo et al., 2020). This indicates that each of two lineages in the Clade B, one from the high site of Doi Inthanon and another from Mae Hong Son, is a distinct species. Because these two lineages are separated not only genetically, but also morphologically (see comparisons below) from other Thai species of Leptobrachella, we conclude that each of these lineages represents a distinct species, which we describe as follows:
Systematics
Leptobrachella korifi sp. nov.
(Fig. 2A, B)
Diagnosis
A small-sized form of the genus Leptobrachella; adult female 22.7 mm in SVL (n=1); ventral side dusty; tibiotarsal articulation reaching anterior corner of eye; pectoral gland very large; lateral fringe on toes weak.
Etymology
The specific epithet, korifi, is from the Greek “koreefee,” meaning summit, in reference to the high elevation of the habitat of the new species.
Holotype
KUHE 19134, an adult female from near the summit of Doi Inthanon, Chom Thong, Chiang Mai Province, Thailand (18°35′ N, 98°29′ E, 2,300 m asl), collected on 6 August 1994 by Kunio Araya.
Description of holotype (measurements in mm)
SVL 22.7; habitus moderately slender; head longer (HL 9.2, 40.5%SVL) than broad (HW 8.5, 37.4%SVL); snout round in dorsal view, rounded in profile, projecting beyond lower jaw; eye length (EL 3.7, 16.3%SVL) slightly larger than snout length (SL 3.6, 15.9%SVL); canthus distinct, slightly constricted; lore oblique, slightly concave; nostril lateral, below canthus, distinctly closer to tip of snout than to eye; internarial distance (IND 2.8, 12.3%SVL) equaling to interorbital distance (IOD 2.8, 12.3%SVL), both wider than upper eyelid (UEW 2.1, 9.3%SVL); pineal spot absent; tympanum visible, diameter (TD 1.7, 7.3%SVL) less than half that of eye, and separated from eye by one-third of tympanic diameter (T-EL 0.6, 2.6%SVL); vomerine teeth absent; tongue notched, without papillae (Table 3).
Forelimb long (FLL 16.7, 73.6%SVL, LAL 11.7, 51.5%SVL) and slender; fingers slender, unwebbed; finger length formula I<II<IV<III; tips slightly swollen; inner palmar tubercle large (IPTL 1.6, 6.9%SVL), not extending onto first metacarpal and much smaller than outer palmar tubercle; subarticular tubercles indistinct.
Hindlimb moderately long (HLL 37.0, 163.0%SVL); tibia relatively long (TL 11.6, 51.1%SVL), heels overlapping when limbs are held at right angles to body; tibiotarsal articulation of adpressed limb reaching anterior corner of eye; foot (FL 11.0, 48.5%SVL) slightly shorter than tibia; toe length formula I<II<V<III<IV; toe tips similar to those of finger; webbing confined to bases of toes; toes weakly fringed laterally; subarticular tubercles at base of each toe, distally replaced by elongate, low keratinous dermal ridges that are interrupted at the articulations; inner metatarsal tubercle low, length (IMTL 1.0, 4.4%SVL) about half of first toe (1TOEL 2.2, 10.0%SVL); no outer metatarsal tubercle.
Skin nearly smooth, with tubercles and glandular folds above; a low, supratympanic ridge from eye to axilla; sides more tuberculate than dorsum; chest and abdomen smooth; pectoral gland prominent and very large, femoral gland indistinct and relatively small, at middle of posterior thigh; a pair of larger glands below cloaca; a round, white gland just above axilla; ventrolateral glandular ridges distinct; skin of gular region not modified.
Color
In preservative, light brown dorsally on head and body, with clear dark spots: spot on snout, triangle between eyes, and spots on shoulder region continuous; spots on upper part of iliac region present, but triangle on lower part of iliac region absent; dorsal portion of tympanum included in brown mask; black band below canthus and supratympanic fold; lips barred with black; blackish spots on flanks small and scattered, restricted to anterior half; groin without dark spot; limbs marked dorsally with dark brown crossbars, especially wide on lower arm and tibia; elbow and upper arm without dorsal bars; posterior side of thigh dark brown; femoral and circum-cloacal glands forming elongated oval white spots; throat, chest, and abdomen dusted with fine brown network; ventral surfaces of legs dusky, dotted with light brown; color of iris in life not recorded.
Variation
Unknown.
Range
Known only from the type locality, near the summit of Doi Inthanon, Chiang Mai Province, northern Thailand.
Natural history
The holotype was found in the daytime, under a stone on a mountain trail in the pine forest, far from streams. No other frog species were observed. In this female, half-developed ova, uniformly cream in color and with a diameter of about 0.54 mm, were seen in ovaries. This indicates that the female was found during the interval between breeding seasons.
Call characteristics
Calls unknown.
Comparisons
Compared with female L. murphyi (Chen et al., 2021), L. korifi sp. nov. is much smaller in SVL (22.7 mm vs. 29.3–32.1 mm in L. murphyi). It has larger values relative to SVL than male L. murphy in head width (RHW 37.4% vs. 35.6–36.6%), internarial distance (RIND 12.3% vs. 11.1–12.2%), snout length (RSL15.9% vs. 14.8–15.8%), forelimb length (RFLL 73.6% vs. 67.4–73.5%) and smaller value in hindlimb length (RHLL 163.0% vs. 167.8–179.1%). Also, the new species has pectoral gland distinct and very large (vs. small and indistinct in L. murphyi), lateral fringes on toes poorly developed (vs. well developed in L. murphyi), and ventral surfaces dusted with light brown (vs. abdomen creamy white in L. murphyi).
Leptobrachella korifi sp. nov. is smaller than L. sinorensis sp. nov., described below, in SVL (22.7 mm vs. 26.6–27.1 mm in L. sinorensis sp. nov.), but has larger relative values than L. sinorensis sp. nov. in head length (RHL 40.5% vs. 35.8–39.0%), head width (RHW 37.4% vs. 33.9–35.6%), internarial distance (RIND 12.3% vs. 9.0–9.7%), interorbital distance (RIOD 12.3% vs. 10.0–11.7%), snout length (RSL15.9% vs. 14.4–15.1%), tympanum diameter (RTD 7.3% vs. 5.5–7.0%), forelimb length (RFLL 73.6% vs. 70.7–72.0%), inner palmer tubercle length (RIPTL 6.9% vs. 5.7–6.4%), tibia length (RTL 51.1% vs. 46.9–49.4%), foot length (RFL 48.5% vs. 43.2–46.1%), and first toe length (R1TOE 10.0% vs. 9.2–9.9%). The pectoral gland is distinct and very large (vs. indistinct in L. sinorensis sp. nov.). Toe fringes are less developed than in L. sinorensis sp. nov. In L. korifi sp. nov., ventral surfaces dusted with light brown (vs. abdomen creamy white in L. sinorensis sp. nov.). From L. cf. minima, L. korifi sp. nov. differs in smaller body size (SVL 22.7 mm vs. 36.4 mm in female L. cf. minima), with a relatively wider head (RHW 37.4% vs. 34.0–36.5%) and internarial distance (RIND 12.3% vs. 9.6–11.2%), and longer forelimb (RFLL 73.6% vs. 68.8–71.1%), inner palmer tubercle (RIPTL 6.9% vs. 2.6–5.7%), tibia (RTL 51.1% vs. 46.5–47.0%), foot (RFL 48.5% vs. 40.6–45.6%), and hindlimb (RHLL 163.0% vs. 148.3–159.0%). Pectoral gland and ventral coloration also differ between them (distinct and very large vs. small, and dusty vs. immaculate white, respectively).
Table 2.
Uncorrected p-distances (%) among samples of Leptobrachella frogs and other species for fragments of 16S rRNA.
Table 3.
Measurements of Leptobrachella. SVL (mean±SD, in mm) and medians of ratios (R) of other characters relative to SVL, followed by ranges in parentheses. See text for character abbreviations.
Fig.2.
Dorsal (A) and ventral (B) views of female holotype of Leptobrachella korifi sp. nov. (KUHE 19134) in preservative. Scale bar=10 mm.
Fig.4.
Dorsal (A) and ventral (B) views of male holotype of Leptobrachella sinorensis sp. nov. (KUHE 19816). Scale bar=10 mm.
Leptobrachella sinorensis sp. nov.
(Figs. 3, 4A, B)
Diagnosis
A medium-sized form, adult males 26.6–27.1 mm in SVL (n=3); tibiotarsal articulation reaching center of eye; pectoral gland indistinct; ventral side of head and body creamy white.
Etymology
The specific epithet, sinorensis, is from the Greek “seenoro,” meaning border, in reference to the type locality, which lies near the Myanmar border of Thailand.
Holotype
KUHE 19816, an adult male from Huay Sa Kud, Mae Hong Son Province, northern Thailand (19°13′ N, 97°57′ E, 270 m asl), collected on 9 September 1995 by Masafumi Matsui.
Paratypes
KUHE 19809 and 19815, two adult males, data same as the holotype.
Description of holotype (measurements in mm)
SVL 26.6; habitus moderately stocky; head longer (HL 10.3, 38.7%SVL) than broad (HW 9.2, 34.6%SVL); snout round in dorsal view, rounded in profile, projecting beyond lower jaw; eye length (EL 4.5, 16.8%SVL) larger than snout length (SL 4.0, 15.0%SVL); canthus distinct, very slightly constricted; lore oblique, slightly concave; nostril lateral, below canthus, distinctly closer to tip of snout (NEL 1.7, 6.4%SVL) than to eye; internarial distance (IND 2.6, 9.7%SVL), narrower than interorbital distance (IOD 3.1, 11.7%SVL), latter wider than upper eyelid (UEW 2.4, 9.0%SVL); pineal spot indistinct, but present; tympanum visible, diameter (TD 1.5, 5.5%SVL) one-third that of eye, and separated from eye by two-fifths of tympanic diameter (T-EL 0.6, 2.3%SVL); vomerine teeth absent; tongue notched, without papillae; a median, subgular vocal sac in the mid-ventral line; vocal slits posterior to rictus.
Forelimb long (FLL 18.8, 70.7%SVL, LAL 13.5, 50.8%SVL) and slender; fingers slender, unwebbed, finger length formula I<II<IV<III; tips slightly swollen; inner palmar tubercle large (IPTL 1.7, 6.4%SVL), not extending onto first metacarpal and smaller outer palmar tubercle (OPTL 1.0, 3.8%SVL); subarticular tubercles indistinct; nuptial pads absent.
Hindlimb moderately long (HLL 41.2, 154.9%SVL); tibia relatively long (TL 13.1, 49.2%SVL), heels overlapping when limbs are held at right angles to body; tibiotarsal articulation of adpressed limb reaching center of eye; foot (FL 11.5, 43.2%SVL) shorter than tibia; third toe longer than fifth; toe tips less rounded than those of fingers; webbing confined to bases of toes; toes weakly fringed laterally on the second and third toes; subarticular tubercles obscure under base of toes, distally replaced by low keratinous dermal ridges; inner metatarsal tubercle low, length (IMTL 1.4, 5.4%SVL) about half of first toe (1TOEL 2.6, 9.7%SVL); no outer metatarsal tubercle.
Skin nearly smooth, with a few tubercles of varying sizes above; a low, supratympanic ridge from eye to axilla; sides more tuberculate than dorsum; chest and abdomen smooth; pectoral gland inconspicuous, femoral gland distinct on the rear surface of thigh at distal one-third; a pair of small glands ventral to cloaca; a small, round, white gland just above axilla; ventrolateral glandular ridges distinct but regularly interrupted; skin of gular region not modified.
Color
In life, light brown dorsally on head and body, with dark spots, partly dotted with pink; spots on snout and shoulder region faint, but triangle between eyes clear on upper eyelids; spots on upper part of iliac region faint, merging with brownish ground color, and triangle on lower part of iliac region clear posteriorly; tympanum largely masked black; supratympanic fold brown spotted with red dorsally, and edged with black ventrally; black band below canthus; lips barred with black; blackish spots on flanks large; groin with dark spot not extending onto thigh; limbs marked dorsally with narrow dark brown crossbars, narrow on dorsal surfaces of thigh; elbow and upper arm without bar dorsally; sides of tibia with black spots, not continuous as bars across dorsal surface; posterior side of thigh dark brown; femoral and circum-cloacal glands forming elongated white oval spots; throat purplish; chest and abdomen milky white; ventral surfaces of legs purplish, dotted with white; iris reddish orange in upper half, copper reticulated with black in lower half. In preservative, the dorsal coloration has faded to olive brown. Eye color has faded and lower halves of iris have become indistinguishable.
Variation
Individuals of the type series are nearly uniform in morphology, but their coloration is variable. In the two paratypes, dorsal dark spots are clearer than in the holotype. In KUHE 19815, the dark spot on snout, triangle between eyes, and spots on shoulder region are separated, and the spots on upper part of iliac region and triangle on lower part of iliac region are continuous. Blackish spots on flanks are large, and only the upper portion of tympanum is contained within the black mask. In KUHE 19809, the dark spot on snout is faint, and the triangle between eyes and spots on shoulder region continuous. Spots on upper part of iliac region and triangle on lower part of iliac region are present, and blackish spots on flanks are medium-sized, with pinkish tips (Fig. 4).
Range
Known only from the type locality, Huay Sa Kud, 270 m asl, Mae Hong Son Province, northern Thailand, near the Thai-Myanmar border.
Natural history
In early September, males were found at night among litter on the floor of a forest far from a stream. Associated species observed were Xenophrys major (Boulenger, 1908), Duttaphrynus melanostictus (Schneider, 1799), Kaloula pulchra Gray, 1831, Fejervarya limnocharis (Gravenhorst, 1829), Occidozyga martensii (Peters, 1867), Sylvirana nigrovittata (Blyth, 1856), and Amolops marmoratus (Blyth, 1855).
Call characteristics
Calls not known.
Comparisons
Comparisons with L. korifi sp. nov. are described above, under that species. In males, L. sinorensis sp. nov. with SVL of 26.6–27.1 mm is smaller than L. cf. minima (27.7–28.9 mm), but for the first toe length relative to SVL, the new species is larger than L. cf. minima (R1TOE 9.2–9.9% vs. 7.0–8.8% in L. cf. minima). In males, L. sinorensis sp. nov. has smaller values relative to SVL than L. pelodytoides in internarial distance (RIND 9.0–9.7% vs. 10.1–10.5% in L. pelodytoides), upper eyelid width (RUEW 8.9–9.9% vs. 10.1–10.5%), snout length (RSL 14.4–15.1% vs. 15.4%), eye to tympanum length (RE-TL 2.3–3.7% vs. 4.0%), tibia length (RTL 46.9–49.4% vs. 49.7–49.8%), foot length (RFL 43.2–46.1% vs. 49.4–50.3%), and hindlimb length (RHLL 154.9–164.8% vs. 165.0–165.6%). The new species (26.6–27.1 mm) is larger than L. murphyi (22.0–24.8 mm) in male SVL and had much less well-developed toe fringes than does L. murphyi (Fig. 5).
Discussion
Doi Inthanon is the highest mountain in Thailand, with the peak at 2,565 m asl, and L. korifi sp. nov. was found 265 m below the peak. At lower elevations (1,350–1,650 m asl.), L. murphyi was abundant, and L. fuliginosa was also collected (Matsui et al., 2017a). Together with L. melanoleuca, L. fuliginosa occupies a unique position in Leptobrachella phylogeny (Matsui et al., 2017a; Chen et al., 2018), basal to Chinese and Vietnamese clades. Thus, Chen et al. (2018) placed it in their clade of Tenasserim Range species, but such a geographic division is not valid. Chen et al. (2021) also recorded L. minima on Doi Inthanon, indicating that there are four congeneric species on this mountain. Other than these Leptobrachella species, this high mountain harbors other unique anurans, such as Ansonia inthanon (Matsui et al., 1998), Gracixalus seesom (Matsui et al., 2015), and possibly G. carinensis (Matsui et al., 2017b). Future extensive surveys might reveal presence of more interesting species.
Huay Sa Kud, in Mae Hong Son Province, where L. sinorensis sp. nov. was obtained, is very close to the type locality of L. pelodytoides, Thao, Kayah State of Myanmar (Capocaccia, 1957). Although records of L. pelodytoides frequently occur in the surrounding countries, no reliable data have been obtained after the original collection by Leonardo Fea in the late 1880s. We first expected our Thai specimens to represent that species, but close morphological comparison with museum specimens of L. pelodytoides revealed slight morphological differences, as shown in our species comparisons. Molecular information on topotypic L. pelodytoides is still unavailable. A recent monograph of Myanmar amphibians and reptiles (Zug, 2022: 27) refers to the species, but only with general comments. It is thus desirable in the future to obtain detailed information of L. pelodytoides from Myanmar, to clarify its phylogenetic relationships within the genus Leptobrachella. Similarly, a male specimen of L. korifi sp. nov. and female specimens of L. sinorensis sp. nov. have not yet been obtained, and their calls are unknown. Urgent intensive field surveys in the region from northern Thailand to eastern Myanmar must be undertaken immediately, because of ongoing environmental degradation.
Fig.5.
Ventral view of foot of Leptobrachella korifi sp. nov. (A: KUHE 19134, female holotype), L. sinorensis sp. nov. (B: KUHE 19816, male holotype), L. murphyi (C: KUHE 19221, a male), and L. cf. minima (D: KUHE 19200, a male). Not to scale.
From the known phylogenetic relationships, the two new species described herein form a clade with species from Southern China and Vietnam to Indo-Burma and assigned to the former subgenus Lalos of the genus Leptolalax (Ohler et al., 2011). Such a classification is much more useful and readily understandable than using Leptobrachella. We here add some comments about this problem. The small Southeast Asian megophryid frogs, to which the species treated herein belong, were first known as Leptobrachium (Boulenger, 1893), but were later moved to Megophrys (Boulenger, 1908). Then they were again placed in Leptobrachium as its subgenus Leptolalax (Dubois, 1980). It was elevated to a distinct genus Leptolalax (Dubois, 1983), which was later split into two subgenera, Bornean Leptolalax and continental Lalos (Dubois et al., 2010).
A mitochondrial phylogenetic study, however, revealed that Leptolalax forms a clade with Leptobrachella (Smith, 1925), another genus of more tiny megophryids, mainly from Borneo (Matsui et al., 2017a). Monophyly of Leptolalax and Leptobrachella with respect to Leptobrachium was fully supported, and three monophyletic lineages (Bornean Leptolalax, Leptobrachella, and continental Leptolalax) were recognized in this clade, but their relationships were unresolved. Within the continental clade, the species from the Malay Peninsula formed a sister clade to the clade of the remaining regions, and the species from southern Thailand formed a sister clade to the species from more northern regions. Unfortunately, the study lacked samples from Cambodia and southern Vietnam, which seem to form a distinct clade (Poyarkov et al., 2015; Rowley et al., 2016), because sequences available from GenBank were not suitable for detailed comparisons.
Finally, employing larger samples, including Cambodian and southern Vietnamese ones, Chen et al. (2018) obtained a similar result that Leptobrachella is nested within Leptolalax and is the sister group of Bornean Leptolalax, rendering Leptolalax paraphyletic. Chen et al. (2018) recognized six clades (A: Southern China and Indo-Burma, B: Central Highlands, C: Mount Fansipan, D: Tenasserim Range, E: Borneo, and F: Malay Peninsula). For these clades, Chen et al. (2018) adopted a one-genus concept, placing all species into Leptobrachella, because they thought such an idea to be the most conservative. However, a four-genera concept, retaining Leptobrachella, assigning Leptolalax to the Bornean species (Clade E), and erecting new genera for the species from Southern China to the Tenasserim Range (Clades A–D) and for those from Malay Peninsula (Clade F), better reflects phylogenetic relationships.
The one-genus concept of Chen et al. (2018) is now prevalent, but it usually requires notation like “diagnostic characters for species occurring north of the Isthmus of Kra” (Table 4 of Chen et al., 2021). This is highly inconvenient from the view of taxonomic usefulness. Moreover, such an idea strongly contradicts the case in Megophrys (sensu lato, including species only slightly more divergent than (about 1.3 times) the species number of Leptobrachella), in which molecular clades each of which was originally recognized as a subgenus (Mahony et al., 2017), were later arbitrarily treated as distinct genera (e.g., Deuti et al., 2017; Lyu et al., 2021). In order to solve the generic problem of Leptobrachella (sensu lato), confirmation of morphological/ecological synapomorphies for each clade is required.
Acknowledgments
The National Research Council of Thailand and the Royal Forest Department of Thailand kindly permitted MM to conduct the project, and Chulalongkorn University kindly provided all the facilities for conducting the research. SP would like to express gratitude to Chulalongkorn University for providing some research funding. We thank the late J. Nabhitabhata, T. Hikida, H. Ota, and K. Araya for help in the field, and A. Hamidy for laboratory assistance. MM is grateful to C. McCarthy and B. T. Clarke (BM), R. Günther and U. Manthey (ZMB), G. Doria (MSNG), and the late R. F. Inger (FMNH) for allowing him to examine specimens under their care. We thank A. Savitzky and anonymous reviewers for improving an earlier version of the manuscript. This study was supported by grants under The Monbusho International Scientific Research Program (Field Research, Nos. 06041066, 08041144, and 10041166) to MM.
