DNA Barcoding and Systematic Review of Minervaryan Frogs (Dicroglossidae:Minerva

Sonali GARG and S.D.BIJU

Systematics Lab,Department of Environmental Studies,University of Delhi,Delhi 110007,India

Abstract The genus Minervarya is among the most widely distributed,commonly occurring,and taxonomically confusing groups of dicroglossid frogs in India.Recent studies have provided evidence that this genus contains complexes of morphologically conserved but genetically divergent taxa—some widely distributed across South and Southeast Asia,and many particularly restricted to the Western Ghats region of the Indian Peninsula—posing several challenges in resolving long-standing taxonomic confusions.Here,we present a systematic review of minervaryan species found in Peninsular India,based on extensive DNA barcoding with nearly 400 samples from the entire range of the genus,including 277 new samples and topotypic material for most available names from the study area,combined with detailed morphological studies.As a result,we recognise 18 species in Peninsular India,including a new species described herein as Minervarya pentali sp.nov.Due to the comprehensive nature of the study,including comparisons with all available types,certain long-standing taxonomic uncertainties on the status of ten previously known taxa are resolved. Rana (Tomopterna) parambikulamana Rao,1937 (=Minervarya parambikulamana),Rana (Hylorana)sauriceps Rao,1937 (=Minervarya sauriceps), and Fejervarya kudremukhensis Kuramoto,Joshy,Kurabayashi,and Sumida,2008“2007”(=Minervarya kudremukhensis),are considered as junior subjective synonyms of Rana (Rana) limnocharis mysorensis Rao,1922 (=Minervarya mysorensis);Nyctibatrachus sanctipalustris var.modestus Rao,1920 (=Minervarya modesta)is proposed to be a synonym of Rana limnocharis syhadrensis Annandale,1919 (=Minervarya syhadrensis);while Rana murthii Pillai,1979 (=Minervarya murthii) and Fejervarya mudduraja Kuramoto,Joshy,Kurabayashi,and Sumida,2008 “2007”(=Minervarya mudduraja) are considered as junior subjective synonyms of Rana nilagirica Jerdon,1853 (=Minervarya nilagirica).At the same time,Rana brevipalmata Peters,1871 (=Minervarya brevipalmata),previously known only from its original description and the type specimen,is recognised as a distinct species referable to live populations in the Western Ghats.The study results in taxonomic stability of all the currently recognised members of the genus in Peninsular India.Significant geographical range extensions of species previously known from single localities are also provided based on morphologically and genetically confirmed records.Additionally,we classify all the recognised species into eight species-groups,with the aim of facilitating a better working taxonomy and future systematic studies on minervaryan frogs across their entire known range in Asia.

Keywords amphibia,distribution,Fejervarya,integrative taxonomy,Minervarya pentali sp.nov.,morphology,species groups,Western Ghats Biodiversity Hotspot

1.Introduction

The systematics of the genus

Minervarya

Dubois,Ohler,and Biju,2001 (sensu lato) has been a subject of considerable discussion in the past few years,primarily owing to the conserved and confusing morphology of its members,as well as the largely unresolved genus-level relationships with the closely related

Fejervarya

Bolkay,1915 and

Sphaerotheca

Günther,1859 (e.g.,Dinesh

et al

.,2015;Garg and Biju,2017;Sanchez

et al

.,2018).As currently recognised (sensu Sanchez

et al

.,2018),the genus constitutes a relatively large radiation of small to largesized species (male SVL 16–50 mm;female SVL 19–65 mm),characterised by the presence of vomerine ridge;presence of pineal eye;presence of rictal glands,weakly to well developed(especially in males),and a light greyish-brown glandular stripe extending from the lower half of the tympanum up to the level of the shoulders on either side;presence of fejervaryan lines on either side of the belly;absence of webbing between fingers;presence of webbing between toes,ranging from reduced to extensive but never complete;absence of finger and toe discs;presence of skin folds,either continuous or discontinuous,with glandular warts on the dorsum;and smooth ventral skin(Dubois

et al

.,2001;Garg and Biju,2017).The recognised thirty-five species of the genus are predominantly distributed in South Asia (India including the Andaman and Nicobar Archipelago,Sri Lanka,Pakistan,Nepal,Bhutan,and Bangladesh),with only two members found in the Southeast Asian countries of Myanmar andThailand,and neighbouring regions of southern China (Frost,2021;Khatiwada

et al

.,2021).Most of these (31 species) are known to occur in India,with the highest diversity (23 species)and endemism (21 species) in the Western Ghats Biodiversity Hotspot,which encompasses the southwest coast of the Indian Peninsula (Dinesh

et al

.,2015;Frost,2021;Garg and Biju,2017).However,several

species of the region,such as

Minervarya brevipalmata

,

M.modesta

,

M.murthii

,

M.mysorensis

,

M.nilagirica

,

M.parambikulamana

,and

M.sauriceps

,are known only from their original descriptions or type specimens,in the absence of any new collections or reliable reports ever since their descriptions (Garg and Biju,2017).Although new species of this genus have been described in the recent years,of which 12 species alone are from regions in Peninsular India (e.g.,Kuramoto

et al

.,2008“2007”;Dinesh

et al

.,2015;Garg and Biju,2017;Raj

et al

.,2018;Phuge

et al

.,2019),attempts to either address or resolve the taxonomic status of the poorly known taxa have been limited.More recently,Garg and Biju (2017) assigned all the Western Ghats species into four morphologically diagnosed species groups.Phuge

et al

.(2020) clarified the taxonomic status of

M.syhadrensis

(Annandale,1919),also leading to the synonymy of

M.caperata

Kuramoto,Joshy,Kurabayashi,and Sumida,2008.Khatiwada

et al

.(2021) recently clarified also the taxonomic status and generic placement of

Minervarya

species found in Nepal based new collections along with molecular data,with implications for species found in India.However,the taxonomic status of a large number of named taxa still remains doubtful,thereby deterring proper identification and range delineation of minervaryan frogs,particularly within the Western Ghats.At the same time,integrated morphological and molecular studies based on comprehensive as well as wide taxon sampling for all known species of the genus,such as those undertaken on other groups of frogs in the region (e.g.,Biju and Bossuyt,2009;Biju

et al

.,2011,2014a,2014b;Dahanukar

et al

.,2016,2017;Garg and Biju,2016;Garg

et al

.,2018,2019,2021;Vijayakumar

et al

.,2014),remain unattempted on minervaryan frogs.

We undertook a study on minervaryan frogs,based on sampling of all known species of the genus in Peninsular India over a period of nearly two decades.Here,for the first time,we report specimens for all the recognised species of the region identified based on detailed morphological studies,along with molecular data from 277 newly sampled individuals combined into a DNA barcoding study using mitochondrial 16S rRNA gene sequences.The primary aim of our study was to resolve the taxonomic identities of species,particularly of members known only from their original descriptions and possibly confused with other widespread species.In turn,our wide sampling along with new topotypic material also enabled us to investigate the distribution ranges of species,the presence of potentially undescribed diversity,as well as understand systematic affinities for defining species-groups in this large and taxonomically challenging group of frogs.

2.Materials and Methods

2.2.Molecular study

Genomic DNA was extracted from 277 new samples using Qiagen DNeasy blood and tissue kit (Qiagen,Valencia,CA,USA),following manufacturer’s protocol.A fragment of the mitochondrial 16S rRNA gene was PCR-amplified using previously published primer set 16Sar and 16Sbr (Simon

et al

.,1994) and sequenced on both forward and reverse strands with BigDye Terminator v3.1 Cycle Sequencing kit using an ABI 3730 automated DNA sequencer (Applied Biosystems).The raw sequence data was assembled and checked in ChromasPro v1.4 (Technelysium Pty Ltd.),and deposited in the GenBank under accession numbers MZ156076–MZ156352.In addition to our newly sequenced data,we retrieved 120 homologous sequences for all known

Minervarya

species from the GenBank.

Limnonectes khasianus

was used as an outgroup taxon for the phylogenetic analyses.A dataset of total 398 sequences was assembled in MEGA 7.0 (Kumar

et al

.,2016).Sequences were aligned using the MUSCLE tool (Edgar,2004)in MEGA 7.0,and the alignment was compared using MAFFT(Madeira

et al

.,2019).The resultant dataset of 544 characters was analysed using the Maximum Likelihood (ML) and Bayesian approaches.The appropriate model of sequence evolution was determined by implementing Akaike Information Criteria in ModelTest 3.4 (Posada and Crandall,1998).Bayesian analyses were executed in MrBayes (Ronquist and Huelsenbeck,2003) for five million generations using uniform priors,four Metropolis-Coupled Markov Chain Monte Carlo (MCMCMC)chains,and the best-fit General Time Reversible (GTR) model with a proportion of invariant sites (+I) and gamma-distributed rate variation among sites (+G).Trees were sampled at every 1000th generation and the Bayesian posterior probabilities(BPP) for clades were summarised after discarding the first 25 percent trees as burn-in (Huelsenbeck

et al

.,2001).Clade support was also assessed in the ML framework through 10,000 ultrafast bootstrap replicates (UBS) executed using the GTR+I+G model in IQ-TREE (Minh

et al

.,2013),as implemented on the IQ-TREE webserver (Trifinopoulos

et al

.,2016).Intra-and interspecific uncorrected pairwise genetic distances among species were computed using MEGA 7.0 (Kumar

et al

.,2016).

2.3.Morphological study

Morphological comparisons were made with the available type specimens,original descriptions,and new topotypic specimens for all the currently recognised

Minervarya

species known to occur in Peninsular India.The types of following species were examined:

M.agricola

and

M.murthii

(ZSI/SRS—Zoological Survey of India,South Regional Station,Chennai);

M.andamanensis

,

M.modesta

,and

M.syhadrensis

(ZSIC—Zoological Survey of India,Kolkata);

M.brevipalmata

(ZMB—Zoological Museum,Berlin);

M.cepfi

,

M.gomantaki

,

M.kadar

,

M.manoharani

,and

M.neilcoxi

(ZSI/WGRC—Zoological Survey of India,Western Ghats Regional Centre,Kozhikode);

M.goemchi

,

M.kalinga

,

M.krishnan

,and

M.marathi

(ZSI/WRC—Zoological Survey of India,Western Regional Centre,Pune);

M.keralensis

and

M.mysorensis

(NHM—Natural History Museum,London);

M.kudremukhensis

and

M.mudduraja

(BNHS—Bombay Natural History Society,Mumbai);

M.nilagirica

,

M.rufescens

,and

M.sahyadris

(MNHNP—Muséum National d’Histoire Naturelle,Paris).For two species

M.parambikulamana

and

M.sauriceps

,for which type specimens are lost,comparisons were made with the original descriptions and new topotypic specimens.Morphological characters for other congeners with their geographical ranges restricted outside the study region were largely derived from their original descriptions or subsequent literature,available types or other museum specimens,published photographs,or new collections.Shared morphological characters were also identified for grouping of species.We used only adult (mature) individuals for the morphometric studies.Sex and maturity were determined by the presence of secondary sexual characters (such as nuptial pads and vocal sacs in males) or examination of gonads.Measurements and associated terminologies follow Biju and Bossuyt (2009).The following measurements were taken to the nearest 0.1 mm using digital slide-calipers or a binocular microscope with a micrometer ocular:snout-vent length (SVL),head width (HW,at the angle of the jaws),head length (HL,from rear of mandible to tip of snout),MN (distance from the rear of the mandible to the nostril),MFE (distance from the rear of the mandible to the anterior orbital border),MBE(distance from the rear of the mandible to the posterior orbital border),snout length (SL,from tip of snout to anterior orbital border),eye length (EL,horizontal distance between bony orbital borders),inter upper eyelid width (IUE,the shortest distance between the upper eyelids),maximum upper eyelid width (UEW),internarial distance (IN),internal front of the eyes (IFE,shortest distance between the anterior orbital borders),internal back of the eyes (IBE,shortest distance between the posterior orbital borders),NS (distance from the nostril to the tip of the snout),EN (distance from the front of the eye to the nostril),TYD (greatest tympanum diameter),TYE (distance from the tympanum to the back of the eye),forearm length(FAL,from flexed elbow to base of outer palmar tubercle),hand length (HAL,from base of outer palmar tubercle to tip of third finger),FL(finger length),thigh length (TL,from vent to knee),shank length (SHL,from knee to heel),foot length(FOL,from base of inner metatarsal tubercle to tip of fourth toe),total foot length (TFOL,from heel to tip of fourth toe),FD(maximum disc width of finger),width of finger (FW,measured at the base of the disc),TD (maximum disc width of toe),width of toe (TW,measured at the base of the disc).Digit number is represented by roman numerals I–V in subscript.Measurements and photographs were largely taken of the right side of the specimen,and sometimes on the left side when a character was damaged.All measurements provided in the taxonomy section are in millimetres.For the convenience of discussion,species of the genus

Minervarya

are categorised based on their body size,as modified from Garg and Biju (2017):small (male SVL 16.0–25.0 mm),medium (male SVL 25.1–45.0 mm) and large (male SVL 45.1–65.0 mm).

2.4.ZooBank registration

This published work and the nomenclatural acts it contains have been registered in ZooBank,the online registration system for the International Commission on Zoological Nomenclature (ICZN).The ZooBank LSID(Life Science Identifier) for this publication is urn:lsid:zoobank.org:pub:3D20A723-AA21-456E-9C95-F3DF0F56A2FC and for the new species described herein LSID is urn:lsid:zoobank.org:act:1CBAF3D4-17A1-4A1F-B1A1-8E58A17C8FF9.The LSIDs and associated information can be resolved through any standard web browser by appending the LSID to the prefix http://zoobank.org.

3.Results

3.1.DNA barcoding and phylogenetic relationships

The results of our BI and ML phylogenetic analyses were highly congruent with respect to the monophyly of the major wellsupported clades.We recovered 26 species-level lineages(representing recognised or potential species) within the genus

Minervarya

(Figure 1).Our 277 new samples nested with 12 previously known species identified from topotypic material.Of the total 24 previously recognised species in the genus represented on the tree,16 were found in the Western Ghats or the Indian Peninsula.Of these,13 species were restricted to the region,while three species had their distributions extending towards central,eastern,and northern Indian regions (Table S1).In addition,our analyses recovered two hitherto unrecognised lineages—one identified as belonging to

Minervarya brevipalmata

(see section ‘3.2 Taxonomic status of some nominal taxa’),and the other representing a new species (formally described as

Minervarya pentali

sp.nov.

under the section ‘3.3 Description of a new species’).The species-level relationships were largely in agreement with previous studies (Dinesh

et al

.,2015;Garg and Biju,2017;Köhler

et al

.,2019;Phuge

et al

.,2019;Raj

et al

.,2018;Sanchez

et al

.,2018),and differed primarily with respect to either the level of support received for various clades or their taxonomic identities (as clarified in the subsequent section ‘3.2 Taxonomic status of some nominal taxa’).Twenty species recovered as highly supported (BPP ≥ 95,UBS ≥ 90)monophyletic clusters both in the BI and ML analyses (IQTREE) included

M.sahyadris

,

M.krishnan

,

M.agricola

,

M.asmati

,

M.teraiensis

(based on conspecific sequence identity with the typical sequences of

M.dhaka

,as per sequences available from Howlader

et al

.,2016;Khatiwada

et al

.,2021),

M.chiangmaiensis

,

M.syhadrensis

,

M.pentali

sp.nov.,

M.brevipalmata

,

M.goemchi

,

M.mysorensis

,

M.kirtisinghei

,

M.manoharani

,

M.cepfi

,

M.neilcoxi

,

M.marathi

,

M.muangkanensis

,

M.keralensis

,

M.kalinga

,and

M.nilagirica

.The only exceptions were:(1)

M.gomantaki

,for which the typical populations from Goa were recovered as a wellsupported cluster (BPP 100,UBS 99),however,the relationship of other genetically and geographically close populations from Maharashtra,which were also morphologically referable to this species,remained unresolved in both the analyses;and (2)

M.rufescens

,for which the monophyly of all the assigned populations was highly supported in our BI (BPP 100) but moderately supported in the ML analysis (UBS 80).Furthermore,the clade support could not be assessed for four species currently represented by a single GenBank sequence each (

M.nepalensis

,

M.greenii

,

M.kadar

,and

M.andamanensis

).All the

Minervarya

species included in our phylogenetic analyses were recovered in eight distinct,largely well supported,and some moderately supported (BPP ≥ 90,UBS ≥ 70) clades representing species-groups—(1)

Minervarya sahyadris

group:

M.gomantaki

,

M.krishnan

,and

M.sahyadris

;(2)

Minervarya syhadrensis

group:

M.syhadrensis

,

M.nepalensis

(sensu Khatiwada

et al

.,2021),and

M.pentali

sp.nov.;

(3)

Minervarya agricola

group:

M.agricola

,

M.asmati

,

M.chiangmaiensis

,and

M.teraiensis

;(4)

Minervarya mysorensis

group:

M.brevipalmata

,

M.goemchi

,and

M.mysorensis

;(5)

Minervarya rufescens

group:

M.cepfi

,

M.kadar

,

M.manoharani

,

M.neilcoxi

,and

M.rufescens

;(6)

Minervarya andamanensis

group:

M.andamanensis

and

M.muangkanensis

;(7)

Minervarya greenii

group:

M.greenii

and

M.kirtisinghei

;(8)

Minervarya nilagirica

group:

M.kalinga

,

M.keralensis

,and

M.nilagirica

.Based on further morphological studies,these clades are herein defined as species-groups,primarily adapted and expanded from our previous work (Garg and Biju,2017).A single species,

M.marathi

was provisionally assigned to the

Minervarya andamanensis

group based on morphology,although it showed a deeply divergent yet highly to moderately supported sister-group relationship with members of the morphologically disparate

Minervarya rufescens

group.

Figure 1 Bayesian consensus phylogram depicting phylogenetic relationships among 397 Minervarya samples from regions across the known range of the genus in Asia based on mitochondrial 16S rRNA sequences.The focal genus Minervarya is represented by 25 previously known and one new species,comprising eight species-groups discussed in the study.Coloured vertical bars alongside the terminal nodes indicate species;bold letters indicate the new species;and coloured circles at the internal nodes indicate species-groups.Asterisk symbol (＊) above the branch indicates Bayesian Posterior Probabilities (BPP) ≥ 95% and Ultrafast Bootstrap Support (UBS) values ≥ 90%;hash symbol (#) indicates BPP ≥ 90% and UBS values ≥ 70%;dash (-) indicates branches with BPP <90%or UBS <70% in cases where BPP ≥ 90% or UBS ≥ 70% in one of the analyses.The tree is rooted with Limnonectes khasianus as the outgroup taxon.

Figure 2 Geographical distribution of twenty-nine species in the genus Minervarya based on genetic samples used in the study and the known type localities.(A) Eleven species of the Minervarya andamanensis group (M.andamanensis,M.marathi,M.muangkanensis,and M.nicobariensis),Minervarya sahyadris group (M.chilapata,M.gomantaki,M.krishnan,and M.sahyadris),and Minervarya syhadrensis groups (M.pentali sp.nov.,M.nepalensis,and M.syhadrensis);(B) Eight species of the Minervarya agricola group (M.agricola,M.asmati,M.chiangmaiensis,M.sengupti,and M.teraiensis) and Minervarya nilagirica group (M.kalinga,M.keralensis,and M.nilagirica);(C) Ten species of the Minervarya greenii group (M.greenii and M.kirtisinghei),Minervarya mysorensis group (M.brevipalmata,M.goemchi,and M.mysorensis),and Minervarya rufescens group (M.cepfi,M.kadar,M.manoharani,M.neilcoxi,and M.rufescens).

Among the recognised species-groups,several notable intra-and inter-group relationships were observed.Within the

Minervarya sahyadris

group,

M.sahyadris

comprised of two well-supported divergent sub-lineages (uncorrected

p

-distances ranging between 1.3%–1.8%) from localities in Karnataka and Kerala,respectively (Figures 1 and 2).This species showed a well-supported sister-group relationship with the clade containing two weakly resolved but divergent sub-populations of

M.gomantaki

from Goa and Maharashtra (uncorrected

p

-distance ranging between 1.1%–1.7%) and

M.krishnan

.A sister-group relationship between

Minervarya sahyadris

group and the

Minervarya agricola

group was recovered in the BI analysis with low support (BPP 82),whereas the ML analysis suggested a possible sister relationship between the

Minervarya sahyadris

group and

Minervarya syhadrensis

group,albeit with much lower confidence (UBS 54,not shown).However,the close relationship of these three groups received moderate to weak support in our analyses (BPP 93,UBS 56).Within the

Minervarya agricola

group,a sister-group relationship between

M.asmati

and

M.teraiensis

was recovered in BI with low support (BPP 61),but together the two species showed a wellsupported sister-group relationship with

M.chiangmaiensis

.These three species were more distantly related to

M.agricola

,which also showed high levels of genetic differentiation(uncorrected

p

-distances ranging between 0.6%–4.0%) among various poorly resolved population sub-clusters,largely with either south-ranging (Tamil Nadu,Kerala,and Karnataka)or north-ranging (Maharashtra,Andhra Pradesh,Gujarat,Rajasthan,Punjab,Himachal Pradesh,Haryana,Delhi,Uttar Pradesh,Uttarakhand,Madhya Pradesh,Chhattisgarh,Odisha,Bihar,up to West Bengal) distribution pattern.The findings suggest that

M.agricola

could possibly also represent a complex of multiple species requiring further detailed population studies.Within the

Minervarya syhadrensis

group,the relationships between three recognised species received low support,but were suggestive of a sister relationship between

M.syhadrensis

and

M.nepalensis

(BPP 74,UBS <50),which were together likely to form the sister-group to the new species

M.pentali

sp.nov.At the same time,intraspecific divergence among the various widely distributed

M.syhadrensis

populations (from Kerala,Karnataka,Goa,Maharashtra,Gujarat,Andhra Pradesh,Madhya Pradesh,up to Odisha) reached up to 1.2%.The otherwise well-supported and distinct

Minervarya brevipalmata

group was recovered as having a weakly supported and unresolved relationship with three groups (

Minervarya sahyadris

group,

Minervarya syhadrensis

group,and

Minervarya agricola

group).Within the

Minervarya brevipalmata

group,

M.mysorensis

and

M.goemchi

showed a well-supported sister relationship,and together the two species formed a sister-group relationship with

M.brevipalmata

with high support (the taxonomic identities of these taxa are discussed in the section ‘3.2 Taxonomic status of some nominal taxa’).The intraspecific divergence among the various populations of

M.mysorensis

ranged up to 1.1%between three major population sub-clusters:(1) Anaimalais up to Nelliyampathy hills encompassing regions south of Palghat gap in Kerala and Tamil Nadu;(2) Wayanad plateau in northern Kerala up to Coorg plateau in southern Karnataka;and (3) Chikamagalur hills up to Sharavathi basin.For

M.goemchi

,the observed intraspecific divergences were up to 0.4%among populations distributed from northern Karnataka up to Maharashtra.On the other hand,the distribution of

M.brevipalmata

was restricted to regions south of the Palghat gap,from Anamalai and Cardamom hills down to Meghamalais.The

Minervarya greenii

group,comprising two species endemic to Sri Lanka,was recovered as having a sister relationship with the clade containing four groups (

Minervarya sahyadris

group,

Minervarya syhadrensis

group,

Minervarya agricola

group,and

Minervarya brevipalmata

group).The

Minervarya rufescens

group showed a well-supported sister-group relationship with

M.marathi

.Together,these members were recovered as having a close relationship with the

Minervarya andamanensis

group,albeit with weak support.The

Minervarya nilagirica

group formed the most distinct clade in the genus with a basal relationship to all the remaining known

Minervarya

species and species groups.Within the

Minervarya nilagirica

group,the sister relationship between

M.nilagirica

(predominantly restricted to the Western Ghats regions of Tamil Nadu,Kerala,and Karnataka) and

M.kalinga

(currently known from the northern Western Ghats states of Goa and Maharashtra,and also from the Eastern Ghats regions in Andhra Pradesh and Odisha,with intraspecific divergences of up to 1.1%) was well supported.The third member of the group,

M.keralensis

(restricted to south of the Palghat gap in Kerala and Tamil Nadu),formed a sistergroup to

M.nilagirica

and

M.kalinga

with high support (Figures 1 and 2).Although no other previous comparable study on minervaryan frogs of Peninsular India is available in terms of taxon sampling,the relationships recovered in our phylogenetic analyses among the major clades were largely congruent with the recent multi-gene and mitochondrial 16S rRNA phylogenies(e.g.,Dinesh

et al

.,2015;Garg and Biju,2017;Köhler

et al

.,2019;Phuge

et al

.,2019;Raj

et al

.,2018;Sanchez

et al

.,2018).

3.2.Taxonomic status of ten nominal taxa

(Peters,1871):taxonomic identity and nomenclatural stability.

While describing this species as

Rana brevipalmata

,Peters (1871) provided a detailed description of the holotype (Figure 3),which he stated to have been purchased from Pegü (Burma,Myanmar).Boulenger (1905“1904”) doubted the type locality of this taxon,apparently due to the availability of several comparable specimens“from the Nilgherry and Travancore hills”in southern India,which he identified as belonging to

Rana brevipalmata

Peters,1871.Subsequently,this taxon was subjected to frequent taxonomic changes.Boulenger(1890) synonymised it under

Rana limnocharis

Gravenhorst,1829 and referred to it as“var.

brevipalmata

”based on examination of some specimens from“Pegu and S.India”.Subsequent authors(e.g.,Annandale,1917;Boulenger,1920;Roux,1928;Bourret,1942;Gorham,1974;Daniel,1975) followed this by treating it either as a subspecies or a junior subjective synonym of

Rana limnocharis

.Pillai (1980) specifically addressed the taxonomic status of this taxon.Based on multiple new collections from Kerala and Tamil Nadu,he discussed morphological characters as well as behavioural patterns distinguishing

Rana brevipalmata

as a distinct species from

Rana limnocharis nilagirica

,with which he stated it to have been previously confused (Boulenger,1920).However,Dubois (1984) considered

Rana brevipalmata

to be a junior subjective synonym of

Rana nilagirica

Jerdon,1853,but subsequently treated it as

incertae sedis

in the genus

Fejervarya

(Dubois,1987“1986”).Dubois and Ohler (2000) later concluded that this name does not refer to any biological species of frog.Yet,

Rana brevipalmata

Peters,1871 continued to be treated as a valid species,in various genera such as

Limnonectes

(Dubois,1987;Dutta and Singh,1996),

Fejervarya

(Iskandar,1998;Fei

et al

.,2002;Dinesh

et al

.,2015),

Zakerana

(Howlader,2011),and the most recent transfer to

Minervarya

based on the separation of this predominantly South Asian genus from the largely Southeast Asian

Fejervarya

(Sanchez

et al

.,2018).Most available literature includes this species in the list of Western Ghats frogs,where it is also believed to be restricted (e.g.,Daniels,2005;Dinesh

et al

.,2009;Dutta,1997;Frost,2021).However,none of the recent studies have revisited the identity of this taxon and provided new collections or insights on its taxonomic status.Also surprisingly,even though the type of

Rana brevipalmata

was originally stated to have originated from Myanmar,it appears to have been included in the region’s amphibian inventories only rarely (Frost,2021),including its absence from a recent comprehensive study on

Fejervarya

and

Minervarya

frogs of that country (Köhler

et al

.,2019).

Figure 3 Type specimens of eight Minervarya taxa described from Peninsular India showing (left to right) dorsal view,ventral view of hand,and ventral view of foot.(A–C) holotype (ZMB 6130) of Rana brevipalmata;(D–F) holotype (BNHS 4653) of Fejervarya kudremukhensis;(G–I) holotype (BMNH 1947.2.1.80) of Rana (Rana) limnocharis mysorensis;(J–L) neotype (BNHS 6113) of Rana(Tomopterna) parambikulamana;(M–O) neotype (BNHS 6114) of Rana (Hylorana) sauriceps;(P–R) neotype (MNHNP 1984.2340) of Rana nilagirica;(S–U) holotype (BNHS 4645) of Fejervarya mudduraja;(V–Y) holotype (ZSI-SRS VA/773) of Rana murthii,(W) ventral view of throat and chest showing the absence of“pearl-like papillae”.Bottom row:Topotypes of four synonymised nomen in life.

We studied the type specimen of

Rana brevipalmata

Peters,1871 (ZMB 6130,an adult male) in order to compare it with various populations of minervaryan frogs from the Western Ghats.While describing this taxon,Peters (1871) provided a brief comparison with“

Rana limnocharis

Boie

(

Rana gracilis

Wiegm.)”and“the South African

Rana grayi

and

Rana fasciata

”,primarily distinguishing it based on very elongated limbs and reduced webbing between toes from the former,and well developed vomerine teeth from the latter species.Altogether,Peters (1871)described a number of morphological characters for

Rana brevipalmata

including head dimensions,position of nostril,tympanum diameter,presence of vomerine teeth,tongue shape,long limbs,relative length of fingers,extent of webbing on foot,skin texture (including dorsal skin with some short longitudinal folds and small rounded tubercles),body colouration and markings (including a bright mid dorsal line),and some body measurements,which were largely found to be matching with the available type.Boulenger (1890) discussed two characters distinguishing var.

brevipalmata

from

Rana limnocharis

:tibiotarsal articulation“reaching considerably beyond the end of the snout”and foot measuring“two thirds the distance between the end of the snout and the vent.”Boulenger (1920) also stated the first finger to be longer than the second in

Rana brevipalmata

.Later,Pillai (1980) studied populations from“Muthanga”,and“Chedleth”of Wayanad,Kerala state and“Anamalai”and“Valparai”of Anaimalai Hills,Tamil Nadu state,and provided further comparison between

Rana brevipalmata

from

Rana limnocharis

mainly stating differences in reduced foot webbing,absence of dermal fringe along toe V,tympanum less than half the diameter of the eye,and the tympanum-eye distance equal to the tympanum diameter.We found most of the characters discussed by Boulenger (1890,1920) and Pillai (1980),as well as several characters discussed in the original description provided by Peters (1871),to be comparable with our multiple new collections from Idukki district of Peninsular India (Table S1),which were also recovered as a distinct and previously unnamed clade within the genus

Minervarya

(Figure 1).Particularly notable among these are characters such as dorsal skin with relatively long longitudinal folds and small rounded tubercles,long hind limbs,reduced webbing between toes,and well developed inner metatarsal tubercle.Hence,building on the judgements made by previous researchers and the additional evidence gathered in the present study,we provisionally consider this taxon to have originated from the Western Ghats and identify live populations referable to this species (Table S1),thereby clarifying its taxonomic status.

(Rao,1937):rediscovery,neotypification,and synonymy.

This taxon was originally described as

Rana (Tomopterna) parambikulamana

Rao,1937 based on a single specimen from“Parambikulam forests”,with an accompanying illustration.This type specimen is considered lost(Dubois,1984) owing to which the species is currently known only from its original description.The generic placement of this species has been confusing due to its vague description with assignment to the subgenus

Tomopterna

(Rao,1937),where it continued to be placed by several subsequent authors (e.g.,Dutta,1997;Chanda,2002;Matsui

et al

.,2008).It was also proposed as

incertae sedis

within

Fejervarya

(Dubois,1987“1986”) or stated to be“invalid”(Purkayastha and Matsui,2012),both without discussion.These confusions probably stemmed from Rao’s observations (1937):“I have compared this specimen with

R.rufescens,R.breviceps

and

R.dobsoni

,

from which it differs almost in every character,and generally resembles

R.tigrina

in external form,though differing in details both from this species and

R.limnocharis

(Forma typica) through which this new species is derivable.”Nonetheless,even though the species appears to have been confused with members belonging to five different genera,its assignment to

Minervarya

,as presently understood,can be considered more conclusive based on characters such as head longer than broad,snout pointed,projecting beyond the lower jaw;distinct tympanum;fourth toe considerably longer than the thigh or tibia;and presence of inverted“U”or“V”shaped skin fold,all of which are clearly comparable with the illustration of the type provided by Rao (1937).Within the genus

Minervarya

,this taxon cannot be a member of

Minervarya rufescens

group due to its large snout-vent size,presence of“light vertebral band”,“Hind limb long,the tibio-tarsal articulation reaching far beyond the tip of the snout”,“inner metatarsal tubercle large smaller than the first toe”,head“distinctly longer than broad”,and“snout longer than eye”.Instead,the species can be assigned to the

Minervarya mysorensis

group (see group definition in the section ‘3.4 Grouping of species in the genus

Minervarya

’) due to its adult size“39.00”mm,“head longer than broad”,“Hind limb long,the tibio-tarsal articulation reaching far beyond the tip of the snout”,“fourth toe considerably longer than the thigh or tibia”,toes“¼ webbed”,“inner metatarsal tubercle large”,and dorsum with“a light vertebral band”and“cutaneous folds edged black”.Furthermore,our field studies at Parambikulam have resulted in the collection of a species of the genus

Minervarya

that is comparable to many of the characters mentioned and illustrated in the original description of

M.parambikulamana

(Rao,1937).We provide a description of this new specimen and further find it necessary to designate the same as a neotype of

Rana

(

Tomopterna

)

parambikulamana

Rao,1937 in accordance with Article 75 of the ICZN (International Commission on Zoological Nomenclature 1999,hereafter,The Code),owing to the following reasons:(1) The unavailability of its name bearing type (Dubois,1984;Biju,2001;present study).(2) In the absence of a type,direct evidence on the holotype of

Rana

(

Tomopterna

)

parambikulamana

currently exists only from the original description and accompanying figure (Rao,1937).However,the description by Rao (1937) is imprecise and not particularly informative,except for a handful of characters such as presence of inverted“U”or“V”shaped skin fold on the dorsum and“inner metatarsal tubercle large smaller than the first toe”.These characters also largely overlap with other recognised and morphologically cryptic congeners,especially

M.brevipalmata

(Peters,1871),

M.mysorensis

(Rao,1920),

M.sauriceps

(Rao,1937),and

M.kudremukhensis

(Kuramoto

et al.

,2008“2007”),some of which in addition have overlapping or close distribution ranges.(3) Further,while describing this species Rao (1937) stated its overall resemblance with two taxa that are now known to be morphologically unrelated,

Rana tigrina

(=

Hoplobatrachus tigerinus

) and

R.limnocharis

(=

Fejervarya limnocharis

).This adds to the confusion in diagnosing

Rana

(

Tomopterna

)

parambikulamana

solely based on its original description.(4)Moreover,the identity and the specific taxonomic status of

Rana

(

Tomopterna

)

parambikulamana

itself have been previously doubted,with authors proposing it to be either

incertae sedis

(Dubois,1987“1986”) or invalid (Purkayastha and Matsui,2012)probably due to the difficulty in assigning it to any known living population since its single original record.Hence,based on new evidence gathered in the present study,designation of a neotype is considered essential for defining this taxon objectively,in order to identify it properly from other close congeners,as well as to stabilise the taxonomic status of the nomen.We hereby designate BNHS 6113 from the original type locality“Parambikulam”as the neotype of

Rana (Tomopterna)parambikulamana

Rao,1937 (=

Minervarya parambikulamama

).The neotype description provided below is largely consistent with what is known of the former name-bearing type.

Synonymisationof Rao,1937.

Due to the comprehensive nature of our study,we also compared our new collections of

Rana (Tomopterna)parambikulamana

Rao,1937 (=

Minervarya parambikulamama

)with all other known

Minervarya

species sampled from regions across Peninsular India.Based on a detailed morphological comparison of the original description and new topotypic material,including the designated neotype,we find this taxon to be extremely similar to

M.mysorensis

Rao,1922 with overlapping morphological traits,which do not support the two taxa to be reliably distinguished from each other.These include characters such as the adult size,snout sub-elliptical to pointed in dorsal and ventral view;short discontinued dorsal skin folds without a constant pattern;presence of an inverted“V”-shaped skin fold at the centre of the dorsum;foot including toes relatively long;and reduced webbing between toes.The skin texture disparity between Rao’s description of

M.parambikulamana

and

M.mysorensis

is probably a misinterpretation due to the dehydrated condition of the described specimen,as evident from the illustration (Rao,1937,Figure 1).In addition,we also sequenced molecular data from the neotype specimen of

M.parambikulamana

and included it in our phylogenetic analyses to further clarify the genealogical affinities of this taxon.Our results find

M.parambikulamana

to be nested together with various populations of

M.mysorensis

,including

M.kudremukhensis

and

M.sauriceps

(both subsequently synonymised with

M.mysorensis

),with shallow genetic distances of 0.8%–1.1%,which not only fall within the range of intraspecific distances observed for the various

Minervarya

species in our study (Table 1),but also do not correspond to any reliable morphological characters for interspecific comparison and may therefore be considered as intraspecific variations.Hence,in accordance with the Article 23 (Principle of Priority)of The Code we consider

Rana (Tomopterna) parambikulamana

Rao,1937 (=

Minervarya parambikulamana

) as a junior subjective synonym of

Rana (Rana) limnocharis mysorensis

Rao,1922 (=

Minervarya mysorensis

).

(Rao,1937):rediscovery,neotypification,and synonymy.

This taxon was originally described as

Rana (H ylorana) sauriceps

Rao,1937 based on“A number of specimens of all ages”from“Wattekole,Coorg,S.India”.However,the original description included measurements only for a single specimen with SVL measuring“30.00 mm”,along with illustrations (Rao,1937).Ever since,there have been no new reports or collections of this species,either from the type locality or other regions.Further,the original name-bearing type is considered lost (Dubois,1984),hence this nominal taxon was known only from its original description.Our surveys within the region of the type locality“Wattakole”(=Watekolli) resulted in collection of a ‘chocolate red’ coloured specimen belonging to a species of the genus

Minervarya

(Figure 3) that is found to be comparable with the original description and accompanying illustrations of

Rana(Hylorana) sauriceps

Rao,1937 with respect to several characters such as“hind limbs rather long,slender,tibio-tarsal articulation reaching the nostril”;“toes pointed;web not extending to the tip of the first phalangeal bone,rather stopping at the base”;“inner metatarsal about ½ the diameter of the eye”;“upper surface of the skin slightly granulate,with short interrupted longitudinal folds with a few tubercles”;a ∩-shaped mark found on the back behind the shoulders;dorsum chocolate red.We assign this new collection from the type locality as the neotype of

Rana (Hylorana) sauriceps

Rao,1937,an action we consider necessary in accordance with Article 75 of The Code,owing to the following reasons:(1) The unavailability of its name bearing type (Dubois,1984;Biju,2001;present study) or any new referable material since the original description.(2) In the absence of a type,direct evidence on the holotype of

Rana(Hylorana) sauriceps

currently exists only from the original description and accompanying figure (Rao,1937).However,the description by Rao (1937) is imprecise and not particularly informative.Several of the stated characters,particularly“Upper surface of the skin slightly granulate,with short interrupted longitudinal folds with a few tubercles”and“a ∩-shaped mark found on the back behind the shoulders”,also overlap with other recognised and morphologically cryptic congeners,especially

M.brevipalmata

(Peters,1871),

M.mysorensis

(Rao,1920),and

M.kudremukhensis

(Kuramoto

et al.

,2008“2007”),some of which in addition have overlapping or close distribution ranges.(3) Moreover,the specific taxonomic status of

Rana (Hylorana)sauriceps

itself has been previously doubted with a proposal to consider it as

incertae sedis

in the genus

Fejervarya

(Dubois,1987“1986”) that added to the confusion in assigning this nomen to any known living population ever since its original description.Hence,based on new evidence gathered in the present study,designation of a neotype is considered essential for defining this taxon objectively,in order to identify it properly from other close congeners,as well as to stabilise the taxonomic status of the nomen.We hereby designate BNHS 6114 from the original type locality“Wattekole”as the neotype of

Rana (Hylorana)sauriceps

Rao,1937 (=

Minervarya sauriceps

).The neotype description provided below is largely consistent with what is known of the former name-bearing type.

Description of neotypeof Rao,1937.

By present designation,Neotype:BNHS 6114 (Figure 3),from“Wattakole”(Watekolli),Karnataka state,India,collected by S.D.Biju on 17 August 2013.A medium-sized adult male(SVL 40.5);head longer (HL 14.5) than wide (HW 13.2);snout sub-elliptical to pointed in dorsal and ventral view,obtuse in lateral view,its length (SL 6.4) longer than horizontal diameter of eye (EL 3.9);loreal region flared and concave with indistinct canthus rostralis;interorbital space narrower (IUE 2.3) than upper eyelid width (UEW 2.8) and internarial distance (IN 3.4);tympanum oval,its horizontal diameter (TYD 2.1) nearly half (53.8%) of the eye diameter (EL 3.9);forearm (FAL 6.3)shorter than hand length (HAL 8.7);outer,inner,and middle metacarpal tubercles present on hand,well developed;thigh (TL 18.7) shorter than shank (SHL 20.0) and foot (FOL 23.4);foot webbing small:I1– 2II1–2III1–3IV3–1V;inner metatarsal tubercle prominent (IMT 2.0),oval,raised;outer metatarsal tubercle small,rounded.Dorsal skin shagreened to sparsely granular with short longitudinal skin folds;a faint inverted“V”-shaped ridge on anterior half of the dorsum,at the level of forearms;lateral surfaces of head shagreened;flanks shagreened to sparsely granular;fore-and hind limbs shagreened with a few scattered tubercles.Ventral surfaces shagreened,and posterior surfaces of thighs granular.

Colour in preservation

.Dorsum light reddish-brown without any prominent markings;upper half of tympanum and inner margins of supratympanic fold light brown;flanks with faint blackish-brown patches;fore-and hind limbs without cross-bands;dorsal surfaces of posterior half of thigh light brown with buff-coloured reticulations.Ventral surfaces greyish-white with a yellow tinge.S

ynonymisationof Rao,1937.

We also compared the original description and our new collections of this taxon with all other known

Minervarya

species found in the Western Ghats.Evidence gathered from new morphological and molecular data finds

Rana (Hylorana)sauriceps

Rao,1937 (=

Minervarya sauriceps

) to be conspecific with

M.mysorensis

.Most morphological characters (including adult size) were observed to be extremely similar or overlapping between these two currently recognised taxa,barring some minor intraspecific variations.Furthermore,mitochondrial 16S rRNA sequence data from the neotype specimen of

M.sauriceps

nested together with various populations of

M.mysorensis

(Figure 1) with maximum genetic distances of up to 1.0%,which fall within the range of intraspecific distances observed for the various

Minervarya

species in our study (Table 1).Hence,in accordance with the Article 23 (Principle of Priority) of The Code we consider

Rana (Hylorana) sauriceps

Rao,1937 (=

Minervarya sauriceps

) as a junior subjective synonym of

Rana(Rana) limnocharis mysorensis

Rao,1922 (=

Minervarya mysorensis

).

(Kuramoto,Joshy,Kurabayashi,and Sumida,2008“2007”):re-examination of taxonomic status and synonymy.

This taxon was originally described as

Fejervarya kudremukhensis

from“Kudremukh”based on a male holotype (Figure 3) and three paratypes,including a female specimen (Kuramoto

et al

.,2008“2007”).While describing this new species the authors compared and distinguished it mainly from four other known ‘large-sized’members of the genus:

M.brevipalmata

,

M.keralensis

,

M.nilagirica

,and

M.mudduraja

.Of these,the last was described as a new species in the same study as

M.kudremukhensis

(Kuramoto

et al

.,2008“2007”),while for the former three members the authors clearly stated that they did not collect any specimens and solely relied on museum specimens for largely morphometric comparisons.However,the authors did not specifically compare their new taxon with

M.mysorensis

,

M.parambikulamana

,and

M.sauriceps

,assuming them to be smaller-sized species.Our extensive collections of minervaryan frogs from the Western Ghats show that various populations of

M.

‘

kudremukhensis

’(including the examined type series) share overlapping size ranges and several other similar morphological characters with these three species.For example,adult size range,dorsal skin with short longitudinal skin folds and an inverted“V”-shaped ridge,snout subelliptical to pointed,tympanum about half of the eye diameter,hind limbs relatively long,relatively reduced foot webbing not extending beyond the second subarticular tubercle on either side of toe IV,mottling on posterior surface of thighs,and most specimens showing the presence of a middorsal vertebral line.Furthermore,since our study reports new collections of

M.mysorensis

,

M.parambikulamana

,and

M.sauriceps

,based on which their taxonomic identities have been clarified,there remains no doubt that

M.kudremukhensis

is conspecific with all of these taxa,and should hereafter be considered a junior subjective synonym of

M.mysorensis

,based not only on morphological but also conclusive phylogenetic evidence (Figure 1).Hence,in accordance with the Article 23(Principle of Priority) of The Code we consider

Fejervarya kudremukhensis

Kuramoto,Joshy,Kurabayashi,and Sumida,2008“2007”(=

Minervarya kudremukhensis

) as a junior subjective synonym of

Rana (Rana) limnocharis mysorensis

(Rao,1922)

(=

Minervarya mysorensis

).

(Jerdon,1853):taxonomic clarity and range extension.

This taxon was originally described as

Rana nilagirica

based on specimens from“marshes in the Wynaad and Neelgherries”,with a brief description (Jerdon,1853).However,the original name-bearing type of this nominal species was subsequently reported to be lost (Jerdon,1870).Most subsequent authors considered this taxon either as a subspecies or junior synonym of

Fejervarya limnocharis

(e.g.,Annandale,1917;Boulenger,1882,1920).However,Dubois (1984) regarded it as a distinct species under his subgenus

Fejervarya

and designated a neotype (MNHNP 1984.2340) from“Governor Shola”,Nilgiris,Tamil Nadu,India (Figure 3).Thereafter,the species was treated as a member of the genus

Fejervarya

sensu lato(e.g.,Iskandar,1998;Fei

et al

.,2002;Dinesh

et al

.,2015),until the recognition of the generic name

Minervarya

(sensu lato) for the predominantly South Asian radiation of what were formerly considered as fejervaryan frogs (Sanchez

et al

.,2018).However,despite its largely undoubted status as a distinct species as well as availability of a name-bearing type,

Minervarya nilagirica

is known only from the region of its original type locality.Our extensive collections of minervaryan frogs from the Western Ghats results in the following major findings:(1) significant extension of the geographical range of

Minervarya nilagirica

based on morphologically and genetically identified records from the Anaimalai hills (south of Palghat gap) to the hill ranges of Nilgiris,Siruvani,Wayanad,Coorg,and Chikmagalur,including Kempholey ghat,Shiradi ghat,and Kudremukh regions (north of Palghat gap),altogether encompassing the states of Tamil Nadu,Kerala,and Karnataka (Table S1;Figure 2);and (2) clarity on the identity and taxonomic status of two other nominal taxa from the Western Ghats—

Rana murthii

Pillai,1979

(=

Minervarya murthii

) and

Fejervarya mudduraja

Kuramoto,Joshy,Kurabayashi,and Sumida 2008“2007”(=

Minervarya mudduraja

)—both of which are shown to be junior subjective synonyms of

Minervarya nilagirica

(Jerdon,1853),with formal actions of synonymy undertaken below in the subsequent sections.

(Pillai,1979):re-examination of taxonomic status and synonymy.

This taxon was originally described as

Rana murthii

Pillai,1979 based on three male (including holotype) and two female specimens from“Naduvattom”,in the Nilgiris district of Tamil Nadu state.Pillai (1979) had discussed a single character to compare this species with

Fejervarya limnocharis

(referring to

Minervarya nilagirica

),

M.

brevipalmata

,

and

M.greenii

,i.e.,“anterior part of lower jaw and two triangular patches on breast beset with small pearl-like papillae”.However,an examination of the male holotype of

M.murthii

does not show“pearl-like papillae”on the breast or the lower jaw (Figure 3).Given that the holotype of

M.murthii

and the neotype of

M.nilagirica

cannot be distinguished based on this character,the distinct status of

M.murthii

becomes doubtful.We instead observed one of the male paratypes of

M.murthii

as having granulations on the chest and lower jaw,which were absent in the holotype and the other male paratype.Hence,the presence of ‘pearl-like papillae’ appears to be a variable character even in reproductively active males.In this context,Kuramoto

et al

.’s (2008“2007”) observation of“one exceptional individual”of

M.kudremukhensis

syn.nov.(=

M.mysorensis

) having“six faint longitudinal series of small dots on the throat and breast”,suggests that this character could be found variably in other members of the genus

Minervarya

.In our study,we also observed discolouration on the chest with sparse granulation in a few breeding males of

M.nilagirica

.Ruling out this character,we do not find any other reliable morphological differences between these two taxa.Although the size of the respective type specimens vary (

M.murthii

,holotype,male:SVL 31.1 mm;

M.nilagirica

,neotype,male:SVL 43.4 mm),it is considered a sampling disparity as multiple collections from our study show considerable size variation and overlap among specimens from Naduvattam (male:SVL 32–42 mm,

N

=5 and female:SVL 34–49 mm,

N

=5) and Governor Shola (male:SVL 28–47 mm,

N

=4 and female:SVL 38–51 mm,

N

=5).In addition to the lack of morphological differentiation,the type localities of

M.murthii

“Naduvattam”and

M.nilagirica

“Governor Shola”lie in close proximity within the same geographical region (less than 20 km apart,with similar elevation and habitat),further suggesting these to be continuous populations of the same species.This is also congruent with molecular evidence presented in this study based on 16S mtDNA,which shows negligible genetic distance between these taxa.Hence,in accordance with the Article 23 (Principle of Priority) of The Code we consider

Rana

murthii

Pillai,1979(=

Minervarya murthii

) as a junior subjective synonym of

Rana nilagirica

Jerdon,1853 (=

Minervarya nilagirica

).

(Kuramoto,Joshy,Kurabayashi,and Sumida,2008“2007”):re-examination of taxonomic status and synonymy.

This taxon was originally described as

Fejervarya mudduraja

Kuramoto,Joshy,Kurabayashi,and Sumida,2008“2007”based on four female specimens,of which the holotype (Figure 3) is from“Talapu,Madikeri”and the paratypes from close vicinities in Kodagu district of southern Karnataka.While recognizing this species,the authors compared and distinguished it mainly from four other known ‘large-sized’ members of the genus:

M.brevipalmata

,

M.keralensis

,

M.nilagirica

,and

M.kudremukhensis

.As clarified under the discussion for

M.kudremukhensis

,the authors did not collect any specimens of the former three taxa and relied on museum specimens for morphometric comparisons,largely with respect to relative proportions of body parts.For example,

M.mudduraja

(females only) was distinguished from the closely related

M.nilagirica

by longer hind limbs,foot,and hand;from

M.keralensis

by relatively smaller head width,tympanum,and inner metatarsal tubercle;from

M.brevipalmata

also by relatively smaller inner metatarsal tubercle and relatively larger eyelid width compared to inter-orbital distance;and from

M.kudremukhensis

by longer hands,fingers,and toes.Among these species,our further detailed morphological and molecular study reveals

M.mudduraja

to be conspecific with

M.nilagirica

.Based on examination of seven males and four females from the type locality Madikeri and surrounding regions,along with the holotypes of

M.mudduraja

and

M.murthii

,and the neotype of

M.nilagirica

,we found that various quantitative and qualitative characters do not permit a reliable morphological distinction of the three species from one another.In addition,a genealogical comparison of our new 16S mtDNA data from topotypic collections of

M.nilagirica

and

M.murthii

from Governor shola and Naduvattam,respectively,with the previously available sequences of

M.mudduraja

originating from the type specimens,shows maximum genetic distances of up to 0.8%,which fall within the range of intraspecific distances observed among populations of the same species in the genus

Minervarya

(Table 1).Hence,in accordance with the Article 23 (Principle of Priority) of The Code we consider

Fejervarya mudduraja

Kuramoto,Joshy,Kurabayashi,and Sumida,2008“2007”(=

Minervarya mudduraja

) as a junior subjective synonym of

Rana nilagirica

Jerdon,1853 (=

Minervarya nilagirica

).

(Rao,1920):re-examination of taxonomic status and synonymy.

Another poorly known species,

Minervarya modesta

,was originally described as

N yctibatrachus sanctipalustris

var.

modestus

Rao,1920 based on a holotype from“Jog,Shimoga,Mysore”.New specimens of this species have not been recorded ever since its original description.Biju

et al

.(2011) reviewed this taxon,discussed its taxonomic history,and formally transferred it to the genus

Fejervarya

(sensu lato) based on examination of the available holotype.In the present study,we collected several individuals of a

Minervarya

species from Mavinagundi (near Jog falls)in Uttara Kannada district and Shimoga in Shimoga district of Karnataka state.Although the original description of

Nyctibatrachus sanctipalustris

var.

modestus

Rao,1920 is brief,ambiguous,and mostly includes a comparison with the originally assigned species,in addition to the holotype (ZSIC 19179) being severally damaged,we found our new collections to largely match with the available name-bearning type and some of the described characters,chiefly with respect to the adult snout-vent size (SVL 23.1 mm) and dorsal skin with“long longitudinal folds on the body and limbs”.Furthermore,our newly generated 16S mtDNA sequence from a topotypic specimen (SDBDU 2017.3677) nests together with various studied populations of

M.sahyadrensis

with negligible to maximum 0.8% intraspecific divergence.Hence,we propose

N yctibatrachus sanctipalustris

var.

modestus

Rao,1920 (=

Minervarya modesta

) to be considered as a junior subjective synonym of

Minervarya sahyadrensis

(Annandale,1919).

(Bahuguna,2018):re-evaluation of generic placement.

Although this study focused on

Minervarya

frogs of the Western Ghats,we sampled some populations of species with geographical ranges extending outside the region.During our surveys,we encountered a population morphologically identical to

Fejervarya jhilmilensis

(=

Minervarya jhilmilensis

) (Sanchez

et al

.,2018) at Haridwar,in close proximity to its type locality.This species was genetically close to members of the genus

Fejervarya

rather than members of the genus

Minervarya

,with zero to maximum genetic distances of up to 2.5% from

F.

‘

orissaensis

’ for the mitochondrial 16S gene sequences.Therefore,the species is removed from the genus

Minervarya

and formally placed in the genus

Fejervarya

.However,we defer the evaluation of the specific taxonomic status of

Fejervarya jhilmilensis

for a future study of fejervaryan frogs.

3.3.Description of a new species

sp.nov.

Pental’s Minervaryan Frog

(Tables 1–2;Figures 1–2,4)

Etymology:

The species is named after Prof.Deepak Pental,a renowned Indian Plant Genetist and former Vice Chancellor of University of Delhi,in appreciation of his contributions to science.We also acknowledge his support and encouragement in setting-up of the Systematics Lab at Department of Environmental Studies,University of Delhi.The species epithet

pentali

is treated as a noun in the genitive case.

Holotype:

BNHS 6115,an adult male,from Nedumbassery,Cochin (10°09’42.5”N 76°23’22.4”E,11 m asl),Ernakulam district,Kerala state,India,collected by S.D.Biju and Sonali Garg on 29 July 2014.

Paratypes:

BNHS 6116–BNHS 6119,four adult males,from Nedumbassery,Cochin (10°09’42.5”N 76°23’22.4”E,11 m asl),Ernakulam district,collected along with the holotype.

Referred specimen:

SDBDU 2017.3628,an adult female from Kuriyode,Kollam district,Kerala state,collected by Sonali Gargand S.D.Biju on 1 August 2017.

Table 2 Morphometric measurements for the type series and referred specimen of Minervarya pentali sp.nov. (in mm).

Description of holotype:

A small-sized adult male (SVL 19.3 mm),body rather slender;head longer (HL 8.1 mm) than wide(HW 5.9 mm;MN 6.2 mm;MFE 5.0 mm;MBE 3.1 mm);snout shape sub-elliptical in dorsal view,rounded in lateral view,slightly protruding,snout length (SL 3.1 mm) longer than horizontal diameter of eye (EL 2.2 mm);loreal region acute,rounded canthus rostralis;interorbital space flat,nearly equal to(IUE 1.6 mm) upper eyelid width (UEW 1.5 mm) and sub-equal to internarial distance (IN 1.7 mm);nostril oval,as close to the tip of snout (NS 1.2 mm) as to the eye (EN 1.2 mm);tympanum(TYD 1.2 mm) 54.5% of eye diameter (EL 2.2 mm);tympanum to eye distance (TYE 0.7 mm) 58.3% of the tympanum diameter;pineal ocellus present;supratympanic fold well developed,extends from posterior corner of the eye to near the shoulder;vomerine ridge present,bearing small teeth,at an angle of 45°to the body axis,as close to choanae as to each other;tongue moderately long,emarginated (Figure 4).

Arms short,forearm length (FAL 3.4 mm) shorter than hand length (HAL 4.6 mm);fingers rather long,relative length of fingers II

Hind limbs short,thigh (TL 8.2 mm) shorter than shank(SHL 8.9 mm) and foot (FOL 9.0 mm),distance from the base of tarsus to the tip of toe IV (TFOL 13.9 mm);toes long,relative length of toes I

Skin of dorsum shagreened with scattered glandular projections and short longitudinal glandular folds;an interrupted inverse V-shaped ridge at the centre of dorsum;snout shagreened,upper eyelids sparsely tuberculate;upper and lower parts of flank shagreened with a few granular projections.Dorsal surfaces of forelimb,thigh,and shank shagreened with sparsely glandular projections.Ventral surfaces of throat,chest,belly,and limbs smooth;and posterior parts of thigh and region surrounding the vent sparsely granular;fejervaryan lines present on both sides of the belly;rictal glands present,just behind the labial commissure of the mouth (Figure 4).

Colour in life

.Dorsum greyish-brown with blackish-brown lining on skin folds;lateral surfaces of head lighter than dorsum(Figure 4);upper and lower lip with prominent crossbands,greyish-brown alternating with light grey;mid-dorsal line light yellowish-grey;rictal gland creamy white;anterior parts of flank light greyish-brown,forelimbs and hind limbs (including toes) lighter greyish-brown compared to the dorsum;limbs with dark greyish-brown transverse bands;groin greyishyellow;anterior parts of thigh brown with yellowish-grey reticulations;webbing light grey.Ventral surface of throat flesh colour with a black calling patch;belly white;forearm and foreleg light flesh white with dark brown mottling on the margins.

Colour in preservation

(alcohol)

.Dorsum brownishgrey,a thin creamy white mid-dorsal line extending from snout to the vent,a greyish-brown stripe between the eyes that extends over the upper eyelids;upper and lower jaw margins with faint light brown crossbands alternating with light grey;forelimbs and hind limbs lighter brown in colour than dorsum,with dark grey transverse bands;posterior parts of thigh light brown with dark brown reticulations.Ventral surface of throat with a dark grey to blackish calling patch;belly off-white;limbs light greyish-brown,with dark greyish-brown mottling on the margins;hand and foot light greyish-brown (Figure 4).

Figure 4 (A–G) Holotype of Minervarya pentali sp.nov. in preservation showing (A) dorsal view;(B) ventral view;(C) lateral view of head;(D) ventral view of hand;(E) ventral view of foot;(F) schematic illustration of foot webbing;(G) Size (SVL 19.3 mm) in comparison to the Indian five-rupee coin (23 mm diameter);(H–O) Minervarya pentali sp.nov. in life;(H–K) holotype showing (H) dorsolateral view;(I) dorsal view;(J) posterior view of thighs;(K) lateral view of groin;(L) paratype showing dorsolateral view;(M) paratype showing dorsolateral view;(N) paratype showing dorsolateral view;(O) paratype showing dorsolateral view;(P–S) male advertisement calls of Minervarya pentali sp.nov. showing (P) 10 second call segment with a single call type;(Q) 1 second call segment;(R) 0.2 second call segment;(S) spectrogram of the 0.2 second call segment shown in Figure 4R.

Morphological variation:

Morphometric data from five adult males and one adult female,including the holotype,is given in Table 2.

In preservation

:BNHS 6116,BNHS 6118,and BNHS 6119:absence of a mid-dorsal line,and dorsal skin with relatively more granulations and skin folds;BNHS 6117:limbs relatively more granular;and SDBDU 2017.3628 (female):presence of a faint and thin mid-dorsal line,prominent dorsal skin folds and granulation,and loreal region nearly obtuse.

Secondary sexual characters:

Male

(HT).Nuptial pad on finger I present,one single patch on prepollex and finger I up to half the length of the penultimate phalange.Vocal sacs present,marked by loose blackish skin on throat.

Female

(SDBDU 2017.3628).Gravid with mature ova,pigmented on poles,diameter 0.6–0.7 mm (

N

=30).

Morphological diagnosis:

Minervarya pentali

sp.nov.

can be distinguished from other congeners by the following combination of morphological characters:(1) small adult size(male SVL 19–22 mm,

N

=5;female SVL 24–26 mm,

N

=2);(2)slender body;(3) snout shape sub-elliptical in dorsal view and rounded in lateral view;(4) presence of vomerine teeth;(5) finger and toe tips rounded,not enlarged into discs,without grooves;(6) dorsal skin sparsely granular with short discontinuous longitudinal skin folds;(7) presence of fejervaryan lines;(8)presence of rictal glands at labial commissure of the mouth;(9)eye length shorter than snout length (male EL/SL ratio 0.64–0.75,

N

=5);(10) inter upper eyelid width narrower or nearly equal to upper eyelid width (male IUE/UEW ratio 0.88–1.06,

N

=5);and inter upper eyelid width shorter or equal to internarial distance (male IUE/IN ratio 0.83–1,

N

=5);(11) shank length nearly equal to foot length (male SHL/FOL ratio 0.98–1,

N

=5);(12) foot webbing small;(13) absence of a white band along the upper lip margin;(14) first finger length equal or sub-equal to the second finger length;(15) nostril nearly equidistant from the snout and eye (male EN/NS ratio 0.93–1.02,

N

=5);(16)shorter hind limbs in comparison to the body length with tibiotarsal articulation reaching up to the axilla when hind limb is stretched along the body.

Morphological comparison:

Minervarya pentali

sp.nov.

cannot be confused with other known members of the genus

Minervarya

due to its small adult male body size,SVL≤ 22 mm (

vs.

SVL >25 mm),except four members of the

Minervarya sahyadris

group (

M.chilapata

,

M.gomantaki

,

M.krishnan

,and

M.sahyadris

).It differs from all the members of

Minervarya sahyadris

group in having granular dorsal skin with discontinuous longitudinal skin folds (

vs.

shagreened to sparsely granular,without or with weakly developed longitudinal skin folds);absence of white band along the upper lip margin (

vs.

upper lip with white horizontal band in all the four species);and relatively more foot webbing that extends well above the first subarticular tubercle on the outer side of toes I and II,and above the second subarticular tubercle on inner side of toe V(

vs.

relatively less,up to or below).Within the

Minervarya syhadrensis

group,the new species resembles the morphologically and geographically close

M.syhadrensis

;however

Minervarya pentali

sp.nov.

differs from the former by its smaller adult size,male SVL 19–22 mm,

N

=5;female SVL 26 mm,

N

=1 (

vs.

larger,male SVL 25–32 mm,

N

=6;female SVL 34–41 mm,

N

=3);snout sub-elliptical in dorsal view (

vs.

nearly pointed);shank nearly equal to foot length,male SHL/FOL ratio 0.98–1,

N

=5,female SHL/FOL ratio 0.98,

N

=1 (

vs.

shank shorter than foot,male SHL/FOL ratio 0.89–0.92,

N

=6,female SHL/FOL ratio 0.96,

N

=1);small foot webbing,I1½–2II1½–2½III2–3IV3–1½V (

vs.

moderate,I1–2II1–2III1–3IV3–1V);and short discontinuous dorsal skin folds in fewer rows (

vs.

relatively more prominent and longer longitudinal skin folds in more rows).

Minervarya pentali

sp.nov.differs from another known member of the group from Nepal,

M.nepalensis

(based on the type series),by its smaller adult size,male SVL 19–22 mm,

N

=5;female SVL 26 mm,

N

=1 (

vs.

larger,male SVL 29–35 mm,

N

=8;female SVL 30–45 mm,

N

=12);and small foot webbing,below the second subarticular tubercle on either side of toe IV,I1½–2II1½–2½III2–3IV3–1½V(

vs.

moderate,up to third subarticular tubercle on either side of toe IV,I1–2II1–2III1–2IV2–1V).Although the new species is distinguishable from all other known members of the genus by its small size,in order to avoid confusion,specific comparison is provided with all other known members by their respective species groups.

Minervarya pentali

sp.nov.

differs from members of the

M.agricola

group by its smaller adult size,male SVL ≤ 22 mm (

vs.

male SVL 22–41 mm),and smaller foot webbing that does not extend beyond the second subarticular tubercle on either side of toe IV (

vs.

above);differs from members of the

M.andamanensis

group by its smaller adult male size,SVL ≤ 22 mm (

vs.

male SVL 33–47 mm),relatively more granular dorsal skin with longitudinal skin folds (

vs.

less granulations and weakly developed or absence of skin folds),and smaller foot webbing that does not extend beyond the second subarticular tubercle on either side of toe IV(

vs.

above);differs from members of the

M.greenii

group by its smaller adult male size,SVL ≤ 22 mm (

vs.

male SVL 25–41 mm),and smaller foot webbing that does not extend beyond the second subarticular tubercle on either side of toe IV (

vs.

above);differs from members of the

M.mysorensis

group by its smaller adult male size,SVL ≤ 22 mm (

vs.

male SVL 35–53 mm),and shorter hind limbs in comparison to the body length with tibiotarsal articulation reaching up to the axilla when hind limb is stretched along the body (

vs.

longer hind limbs with tibio-tarsal articulation reaching up to the snout);differs from members of the

M.nilagirica

group by its smaller adult male size,SVL ≤ 22 mm (

vs.

male SVL 32–46 mm),small foot webbing that does not extend beyond the second subarticular tubercle on either side of toe IV (

vs.

above),and thighs light brown with light yellowishgrey reticulations (

vs.

dark blackish-brown with prominent and contrasting golden yellow reticulations);and differs from members of the

M.rufescens

group by its smaller adult male size,SVL ≤ 22 mm (

vs.

male SVL 24–35 mm),slender body (

vs.

stout body),head longer than wide (

vs.

head wider than long or sub-equal),short and cylindrical inner metatarsal tubercles(

vs.

shovel-shaped),and dorsal skin with longitudinal skin folds(

vs.

without longitudinal skin folds).Further,

Minervarya pentali

sp.nov.

differs from

M.marathi

,provisionally considered a member of the

M.andamanensis

group,by its smaller adult size,male SVL 19–22 mm,

N

=5,female SVL 26 mm,

N

=1 (

vs.

larger,male SVL 34–44 mm,

N

=4,female SVL 41–49 mm,

N

=6).

Phylogenetic relationship:

Minervarya pentali

sp.nov.

is a well-supported distinct lineage and member of the

Minervarya syhadrensis

group (Figure 1).At the mitochondrial 16S locus,it is divergent from other members of the group by average uncorrected pairwise genetic distances of:4.9% (range 4.7%–4.9%) from

M.nepalensis

;and 4.4% (range 4.0%–5.1%) from

M.syhadrensis

.

Vocalisation:

The male calls of

Minervarya pentali

sp.nov.

were recorded at Nedumbassery on 29 July 2014,between 20:00–21:00 h,in ambient air temperature:25.0°C dry bulb,24.0°C wet bulb.The male produced a single type of call with pulsatile temporal structure.The calls were not delivered in groups and had uniform intervals.A typical call was 223.1 ms long in duration;showed a rise time of 35.2 ms and fall time of 22.1 ms;with the call envelope comprising 19 pulses delivered at a rate of 83.6 pulses/second.The call spectrum was characterised by two broad peaks with mean dominant frequency of 3.1 kHz(Figure 4).

Distribution and natural history:

Minervarya pentali

sp.nov.

is endemic to the southern Western Ghats,where it is currently known to be widely distributed in the low-lying regions,from sea level up to the elevations of 220 m asl,nearly across the entire state of Kerala and the adjoining areas of Tamil Nadu.Kerala state:Thiruvananthapuram district (Chathankod and Karamana),Kollam district (Kuriyode and Nilamel),Pathanamthitta district (Kulanada),Alappuhza district (Alleppy and Karthikappally),Ernakulam district (Angamally and Nedumbaserry),Thrissur district (Thrissur town),Malappuram district (Thenjipalam and Kadakkattupara),and Kozhikode district (Calicut city and suburban areas,and Ramanattukara);Tamil Nadu state:Kanyakumari district:Kiriparai (Figure 2).The species presumably has a wider distribution within its known range in Kerala and adjoining lowland regions of Tamil Nadu.In the present study,the species was predominantly observed in wayside vegetated areas close to permanent water bodies or temporary water collection sites,agricultural fields,and small-scale plantations within human settlements,but not inside forested areas.During the breeding season (July to September) males aggregate in large numbers on the ground,among grasses,or vegetation edges,near water bodies and temporary puddles.Calling activities have been recorded between 18:00–22:00 hours.

3.4.Grouping of species in the genus

Species belonging to the genus

Minervarya

can be assigned to eight distinguishable groups,primarily adapted from Garg and Biju (2017) and expanded herein based on further evidence.Although the revised scheme of grouping presented below is defined using morphological traits,most groups also find support in our preliminary phylogenetic analyses (Figures 1,5 and 6).In the future,robust multi-locus phylogenies and phenotypic traits such as acoustics and behavior can be used to provide additional evidence as well as diagnostic characters for these species-groups,which are proposed with the aim of facilitating a better working taxonomy for this group of morphologically confusing frogs.

group

Members included:

Minervar ya agricola

,

M.asmati,M.chiangmaiensis

,

M.teraiensis

,and

M.sengupti

.

Morphological definition:

This group can be distinguished from other minervaryan groups by the following suite of characters:small to medium adult size (male SVL 20–47 mm,female SVL 27–54 mm);elongate to robust body;head longer than wide;moderate foot webbing,beyond the second subarticular tubercle on either side of toe IV;long and cylindrical inner metatarsal tubercles;dorsal skin with short discontinuous skin folds,with or without scattered granular projections;dorsal chevron present;groin without prominent markings or reticulations;thigh with faint reticulations (Figures 5 and 6).

Phylogenetic relationship:

Based on the 16S rRNA locus,the

Minervarya agricola

group is closely related to the

Minervarya sahyadris

and

Minervarya syhadrensis

groups (Figure 1).

Distribution:

India,Sri Lanka,Bhutan,Nepal,Bangladesh,Myanmar,Thailand,up to southern China (Figure 2).

group

Members included:

Minervar ya andamanensis

and

M.muangkanensis

,and provisionally

M.marathi

and

M.

nicobariensis

.

Morphological definition:

This group can be distinguished from other minervaryan groups by the following suite of characters:medium adult size (male SVL 33–47 mm,female SVL 40–54 mm);elongate to robust body;head longer than wide or sub-equal;moderate foot webbing,up to or above the second subarticular tubercle on either side of toe IV,and below the second subarticular tubercle on the inside of toe III;long and cylindrical inner metatarsal tubercles;dorsum shagreened to sparsely granular,with weakly developed short discontinuous skin folds;groin without prominent reticulations;thighs with faint reticulations (Figures 5 and 6).

Phylogenetic relationship:

Based on the 16S rRNA locus,the

Minervarya andamanensis

group shows a poorly resolved sister-group relationship with members of the

Minervarya rufescens

group and

M.marathi

(Figure 1).

Distribution:

India (Andaman and Nicobar Archipelago and northern Western Ghats),Myanmar,and Thailand (Figure 2).

group

Members included:

Minervarya greenii

and

M.kirtisinghei

.

Morphological definition:

Medium adult size (male SVL 25–41 mm,female SVL 32–50 mm);elongate to robust body;head longer than wide;first finger longer then second finger;moderate foot webbing,above the second subarticular tubercle on either side of toe IV and below the second subarticular tubercle on the inside of toe III;long and cylindrical inner metatarsal tubercles;dorsum sparsely to prominently granular with long longitudinal skin folds;posterior surface of thighs with faint reticulations (Figures 5 and 6).

Phylogenetic relationship:

This group is provisionally proposed to accommodate two species currently known only from Sri Lanka,which are together recovered as a monophyletic group showing a distinct phylogenetic position within the genus based on the mitochondrial 16S locus,however with inconsistent group-level relationships (e.g.,Dinesh

et al

.,2015;Köhler

et al

.,2019;Kotaki

et al

.,2010;Phuge

et al

.,2019;Raj

et al

.,2018;Sanchez

et al

.,2018;Sumida

et al

.,2007;present study,Figure 1).

Distribution:

Endemic to Sri Lanka (Figure 2).

group

Figure 5 Morphological characters (clock-wise:dorsal skin texture;ventral surface of foot showing the extent of webbing,and inner and our metatarsal tubercles;and schematic illustration of foot webbing) for twenty-nine recognised Minervarya species,arranged in eight species groups identified in the study.Frog silhouettes for each species indicate the species group by colour and the proportionate adult snout-vent size (SVL) in millimeters.

Figure 6 Twenty-nine recognised Minervarya species depicted in life.Species are arranged in eight species groups identified in the study.

Members included:

Minervarya

brevipalmata,M.goemchi

,and

M.mysorensis

.

Morphological definition:

This group can be distinguished from other minervaryan groups by the following suite of characters:medium to large adult size (male SVL 35–54 mm,female SVL 45–65 mm);elongate or robust body;head longer than wide;small foot webbing,below or up to the second subarticular tubercle on either side of toe IV,below or up to the first subarticular tubercle on the outer side of toe III,and below or up to the second subarticular tubercle on the inner side of toe V;long and cylindrical inner metatarsal tubercles;dorsal skin with prominent folds,either continuous or discontinuous,and scattered granular projections;dorsal chevron present;groin without prominent reticulations;posterior surfaces of thigh with reticulations (Figures 5 and 6).

Phylogenetic relationship:

Based on the 16S rRNA locus,the

Minervarya mysorensis

group forms a well-supported monophyletic clade,however with weak to moderate support for group level relationships (Figure 1).

Distribution:

Endemic to the Western Ghats (Figure 2).

group

Members included:

Minervarya kalinga,M.keralensis

,

and

M.nilagirica

.

Morphological definition:

This group can be distinguished from other minervaryan groups by the following suite of characters:medium to large adult size (male SVL 32–46 mm,female SVL 38–65 mm);robust body;head longer than wide or sub-equal;first finger longer than second finger(except in

M.nilagirica

);moderate to large foot webbing,up to or above the second subarticular tubercle on either side of toe IV in

M.nilagirica

,and up to or above the third subarticular tubercle on either side of toe IV in

M.kalinga

and

M.keralensis

;long and cylindrical inner metatarsal tubercles;dorsal skin with prominent discontinuous folds,and glandular warts or prominent granulations;dorsal chevron present;groin and thighs with prominent reticulations (Figures 5 and 6).

Phylogenetic relationship:

Based on the 16S rRNA locus,the

Minervarya nilagirica

group is the most distinct group showing a well-supported sister-group relationship with all the remaining

Minervarya

species studied in a phylogenetic framework so far(Figure 1).

Distribution:

Endemic to Peninsular India both in the Western and Eastern Ghats (Figure 2).

group

Members included:

Minervarya cepfi

,

M.kadar

,

M.manoharani

,

M.neilcoxi

,and

M.rufescens

.

Morphological definition:

This group can be distinguished from other minervaryan groups by the following suite of characters:small to medium adult size (male SVL 24–35 mm,female SVL 32–41 mm);stout body;head wider than long or sub-equal;presence of rictal glands at labial commissures of the mouth;small foot webbing,below the second subarticular tubercle on either side of toe IV;shovelshaped inner metatarsal tubercles;dorsal skin with glandular projections,spines or warts,but without longitudinal skin folds;dorsal chevron present;groin without prominent reticulations;thighs with faint reticulations (Figures 5 and 6).

Phylogenetic relationship:

Based on the 16S rRNA locus,the

Minervarya rufescens

group shows a well-supported sequential sister relationship with

M.marathi

and members of the

Minervarya andamanensis

group (present study,Figure 1;Phuge

et al

.,2019;Sanchez

et al

.,2018).

Distribution:

Endemic to the Western Ghats (Figure 2).

group

Members included:

Minervarya pentali

sp.nov.,

M.nepalensis

,and

M.syhadrensis

.

Morphological definition:

This group can be distinguished from other minervaryan groups by the following suite of characters:small to medium adult size (male SVL 19–35 mm,female SVL 23–45 mm);elongate body;head longer than wide;moderate foot webbing,up to the third subarticular tubercle on either side of toe IV;long and cylindrical inner metatarsal tubercles;dorsal skin with continuous or discontinuous skin folds,with or without scattered granular projections;dorsal chevron present;groin without reticulations;thighs with faint reticulations (Figures 5 and 6).

Phylogenetic relationship:

Based on the 16S rRNA locus,the

Minervarya syhadrensis

group is closely related to

Minervarya agricola

and

Minervarya sahyadris

groups(Figure 1).

Distribution:

Peninsular and Central India,and Nepal (Figure 2).

group

Members included:

Minervarya chilapata

,

M.gomantaki

,

M.krishnan

,

and

M.sahyadris

.

Morphological definition:

This group can be distinguished from other minervaryan groups by the following suite of characters:small adult size (male SVL 12–22 mm,female SVL 23–24 mm);slender body;head longer than wide;small foot webbing,up to or below the second subarticular tubercle on either side of toe IV;presence of rictal glands at labial commissures of the mouth and a white horizontal band along the upper lip;long and cylindrical inner metatarsal tubercles;dorsal skin with weakly developed longitudinal skin folds or granulations;dorsal chevron absent;groin without reticulations;thighs with faint reticulations(Figures 5 and 6).

Phylogenetic relationship:

Based on the 16S rRNA locus,the

Minervarya sahyadris

group is closely related to

Minervarya agricola

and

Minervarya syhadrensis

groups (Figure 1).

Distribution:

Currently a disjunct distribution in Peninsular India and Northeast India (Figure 2).

4.Discussion

Frogs of the genus

Minervarya

constitute a long-standing taxonomic enigma,especially because several species originally described from the Western Ghats had either not been reported ever since their original description or remained poorly known due to lack of new collections.In this study,we resolve numerous uncertainties regarding the identity as well as taxonomic status of ten nominal taxa,along with revisiting all of the 18 presently recognised members of the genus from Peninsular India,based on extensive sampling combined with comprehensive morphological and molecular studies.As a result,our study recognises 29

Minervarya

species as valid,in contrast to the previously 35 known species(Frost,2021;Khatiwada

et al

.,2021).The primary aim of our study was to resolve a taxonomic conundrum that had long deterred comprehensive and meaningful studies on this group of morphologically confusing frogs.In turn,our results also facilitate a better understanding of morphological characters useful for diagnosis of species and species-groups in the genus,systematic relationships among all known species,and extended geographical ranges,in addition to description of a new species.Our extensive sampling and DNA barcoding of

Minervarya

frogs from regions across the Western Ghats also provide clarifications and significant extensions to the geographical ranges previously known for all Peninsular Indian species.The insights gathered on the patterns of distribution and endemism,will help future researchers in proper identification and better range delineation,especially considering the morphologically conserved and cryptic nature of species in the genus.The range of the entire genus is also better understood after this study and can aid future phylogenetic and biogeographical studies,not only in relation to all of minervaryan frogs but also the closely related genera

Fejervarya

and

Sphaerotheca

.The genus

Fejervarya

,with which minervaryan species still continue to be confused—largely due to the overall conserved morphology,extended and overlapping geographical ranges outside of their major centres of diversity in South (

Minervarya

) and Southeast Asia (

Fejervarya

),and unresolved phylogenies (Dinesh

et al

.,2015;Sanchez

et al

.,2018)—is also confirmed to be distributed up to North India based on the transfer of

Minervarya jhilmilensis

to

Fejervarya

,and the recent reports of fejervaryan members from Nepal (Khatiwada

et al

.,2021).Nonetheless,future studies are still essential to delineate the boundaries of these two genera and the factors possibly contributing to their broadly restricted geographical distributions.Our study also indicates the presence of previously unknown cryptic diversity within the genus.Although the current evidence does not support recognition of putative new taxa delimited in the

Minervarya sahyadris

and

Minervarya agricola

groups (Figure 1),new insights are certainly likely to be gained from the observed population clusters.For example,members of the

Minervarya sahyadris

group are geographically restricted and appear to have narrower ranges.On the other hand,the

Minervarya agricola

group represents the most wide-ranging radiation within the genus,especially its namesake species

Minervarya agricola

that based on our sampling is known to occur right from the southern tip of the Indian Peninsula up to central,north,and east India.The high intraspecific distances observed among various populations of this taxon also suggest it to be a species complex that can prove to be a good model for understanding the patterns of genetic and possibly phenotypic differentiation,diversification,and distribution in minervaryan frogs.Such studies will also require proper evaluation of the taxonomic status and ranges of several other available names from Northeast India,Nepal,and the neighbouring Southeast Asia.Even within the Western Ghats,with the taxonomy of several species having been resolved,future surveys and detailed studies are likely to yield more undescribed taxa.Hence,in order to facilitate a better working taxonomy for minervaryan frogs,and to eventually enable a much more comprehensive understanding of the genus across its entire range in South,Southeast,and East Asia,we recognise and redefine eight species-groups to accommodate all of the currently known

Minervarya

species.In the future,additional morphological data and robust phylogenies based on extensive taxon sampling from regions outside of Peninsular India (such as Nepal,Andaman and Nicobar Archipelago of India,and Sri Lanka) will be required to conclusively address the taxonomic status and systematic affinities of some poorly known taxa that were outside of the present work’s focal study area.Acknowledgements We thank the various Indian State forest departments for study permissions and field support;several colleagues,friends,and local communities,particularly Robin Suyesh,Raju Vyas,Samadhan Phuge,A.K.Pradeep,K.Anilkumar,Gopinath Sricandane,K.V.Sreenivasan,K.Jayayam,M.Sathyamoorthy,Varad Giri,Romulus Whitaker,K.V.Gururaja,Aravid Madhyastha,K.P.Dinesh,Abhijit Das,Caesar Sengupta,Amit Rane,and past and present members of the Systematics Lab for logistic and/or field support during our various surveys;Chatmongkon Suwannapoom (

M.chiangmaiensis

and

M.muangkanensis

),Suranjan Karunarathna(

M.greenii

),and Nayana Wijayathilaka (

M.kirtisinghei

) for providing photographs;Mark Oliver Rödel and Frank Tillack(ZMB,Berlin),David Gower and Barry Clarke (NHM,London),Alain Dubois and Annemarie Ohler (MNHN,Paris),Kailash Chandra and Kaushik Deuti (ZSI,Kolkata),P.M.Sureshan(ZSI-WGRC,Calicut),K.Subramanian (ZSI-SRS,Chennai),K.P.Dinesh (ZSI-WGC,Pune),and Rahul Khot (BNHS,Mumbai)for providing museum support and access to specimens in their care;and Rohan Pethiyagoda and an anonymous reviewer for comments on an earlier version.This study was partially supported by the following grants to SDB:Critical Ecosystem partnership Fund,Conservation International,USA (Project 55918/2009);University of Delhi Research and Development Grants;DST Purse Grants Phase I and Phase II,Department of Science and Technology,Ministry of Science and Technology,Government of India.SG received research fellowships from University of Delhi (2010/56562) and the Council for Scientific and Industrial Research (CSIR No.9/45(1381)/2015-EMR-I and CSIR No.09/045(1694)/2019-EMR-I);and a small research grant(5409-0260) from Global Wildlife Conservation,USA.

Appendix

Table S1 Details of the mitochondrial 16S rRNA genetic samples included in the study.The list is arranged by the order of terminal nodes in the phylogenetic tree shown in Figure 1.

Continued Table S1

Continued Table S1

Continued Table S1

Continued Table S1

Continued Table S1

Continued Table S1

Continued Table S1