Taxonomic reassessment of two Indian shieldtail snakes in the Uropeltis ceylanicus species group (Reptilia: Uropeltidae)

 

S.R. Ganesh 1, R. Aengals 2 & Eric Ramanujam3

 

1Chennai Snake Park, Rajbhavan Post, Chennai, Tamil Nadu 600022, India

2 Zoological Surveyof India, Southern Regional Centre, Santhome High Road, Chennai, Tamil Nadu 600028, India

3 Pitchandikulam Bioresource Centre / Pitchandikulam Forest Consultants, Auroville, Tamil Nadu 605101, India

1 snakeranglerr@gmail.com (corresponding author), 2 aengalszsiramasamy@yahoo.com, 3 ericramanujamowl@yahoo.com

 

 

Abstract: Uropeltis is the most speciose of all shieldtail snake (uropeltid) genera, particularly in India, and has been bedeviled by a complex and intricate taxonomic history, with several weakly established synonyms and widely disjunct geographic ranges.  Our present work on two Indian Uropeltisspecies revealed greater species diversity than what is currently recognised.  We elevate Uropeltis arcticeps madurensis to species level, and revive Silybura shorttii (in the combination Uropeltis shorttii) from the subjective synonymy of U. ceylanicus.  We provide differential diagnoses, descriptions of examined material and comparisons with similar species based on an examination of voucher specimens as well as fresh, uncollected topotypes documented in the field.

 

Keywords: India, Silybura madurensis, shieldtail, S. shorttii, subspecies, synonym, taxonomy.

 

Abbreviations: BMNH—The Natural History Museum, London, United Kingdom; CSPT—Chennai Snake Park Trust, Chennai, India; CES—Centre for Ecological Sciences, Indian Institute of Sciences, Bangalore, India; MAD—Madras Government Museum, Chennai, India; MHNP—Museum of Natural History Paris, France; ZSI/SRC—Zoological Survey of India, Southern Regional Centre, Chennai, India.

 

 

 

doi: http://dx.doi.org/10.11609/JoTT.o3636.5305-14  | ZooBank: urn:lsid:zoobank.org:pub:6EBBAE45-9D0B-4E34-B1FE-16F45DF1433A

 

Editor: Gernot Vogel, Heidelberg, Germany. Date of publication: 26 January 2014 (online & print)

 

Manuscript details: Ms # o3636 | Received 25 May 2013 | Final received 16 January 2014 | Finally accepted 17 January 2014

 

Citation: Ganesh, S.R., R. Aengals & E. Ramanujam (2014).Taxonomic reassessment of two Indian shieldtail snakes in the Uropeltis ceylanicus species group (Reptilia: Uropeltidae). Journal of Threatened Taxa 6(1): 5305–5314; http://dx.doi.org/10.11609/JoTT.o3636.5305-14

 

Copyright: © Ganesh et al. 2014. Creative Commons Attribution 3.0 Unported License. JoTT allows unrestricted use of this article in any medium, reproduction and distribution by providing adequate credit to the authors and the source of publication.

 

Funding: None.

 

Competing Interest: The authors declare no competing interests.

 

Author Details: S.R. Ganesh is a researcher working on herpetofauna of southern India, based at the Chennai Snake Park. R. Aengals is a herpetologist, working as Scientist-C at the Zoological Survey of India, Southern Regional Centre, Chennai. Has worked extensively on the reptiles of southern India. Mario Eric Ramanujam is a biologist with the Pitchandikulam Bioresource Centre / Forest Consultants, Auroville. Has previously conducted surveys of the Kaliveli region under the European Commission and UBS projects. Has been researching imagery as a tool to enhance environmental sensitization.

 

Author Contribution: SRG studied both the species dealt with; RA and MER provided data on one snake species. SRG led the writing, in consultation with RA and MER who perused and approved the final text.

 

Acknowledgements: We thank our respective institutions for supporting our research activities. SRG thanks the Principal Commissioner and Secretary of Museums, Madras Govt. Museum, the Director, Zoological Survey of India and the Officer In-Charge, Southern Regional Centre, Chennai and the Karthik Shanker Lab, Centre for Ecological Sciences, Indian Inst. of Science for access to museum specimens; A.V.C College for organizing, Tamilnadu Forest Dept. for permitting and Wildlife Association of Rajapalayam (WAR) for funding his field work in High Wavys; Ruchira Somaweera and Sameera Karunarathna for sharing crucial information and photos of Srilankan shieldtail snakes; finally, David Gower (UK), Gernot Vogel (Germany) and Patrick David (France) for their valuable inputs and review comments.

 

 

 

For images-- click here

 

 

Introduction

 

The shieldtail snakes of the family Uropeltidae Müller, 1832, are small to medium-sized, fossorial, primarily wet hill forest-dwelling, ovoviviparous snakes endemic to peninsular India and the adjacent island of Sri Lanka (Smith 1943; Murthy 1981, 1982; Daniel 2002; Das 2002). Currently, the following eight genera—Rhinophis Hemprich, 1820, UropeltisCuvier, 1829, Pseudotyphlops Schlegel, 1839, Plectrurus Duméril, 1851, Melanophidium Günther, 1864, Platyplectrurus Günther, 1868,Teretrurus Beddome, 1868 and Brachyophidium Wall, 1921 are considered valid (Smith 1943 [part]; McDiarmid et al. 1999).  Among these, Pseudotyphlops is endemic to Sri Lanka; Rhinophis and Uropeltis occur both in Western Ghats and Sri Lanka while all the other genera are endemic to the Western Ghats of India (Smith 1943; Rajendran 1985; McDiarmid et al. 1999; Whitaker & Captain 2004).  Many uropeltid species were described in India during its colonial rule in the 19thcentury and the presence of several weakly established subjective synonyms further complicates the systematics of this group as a whole (Gower et al. 2008).

The genus Uropeltis Cuvier, 1829 currently comprises 26 valid species and is reported to have a tricky and complicated taxonomy, badly needing a revision, which is exemplified by its type species U. ceylanicus Cuvier, 1829 (see Gower et al. 2008).  We refer to Smith’s (1943) “group II” (after Gower et al. 2008) containing U. ceylanicus, U. bicatenatus (Günther, 1864), U. arcticeps arcticeps (Günther, 1875), U. arcticeps madurensis (Beddome, 1878), U. macrolepis macrolepis (Peters, 1861), U. macrolepis mahabaleshwarensis Chari, 1955, U. rubromaculatus (Beddome, 1867), U. rubrolineatus (Günther, 1875), U. broughami (Beddome, 1878), U. phipsonii (Mason, 1888) and U. myhendrae (Beddome, 1886) as the Uropeltis ceylanicus species group.  In this paper, two species in this group, Silybura shorttii Beddome, 1863 and S. madurensis Beddome, 1878 that were subsequent to their original description as distinct species, considered either as a synonym or a subspecies of congeneric species, are taxonomically reassessed and elevated to species rank based on examination of preserved specimens and fresh, topotypic uncollected specimens sighted in the field that revealed diagnostic differences that were, in part, not considered by earlier workers.

 

Materials and Methods

 

This study is based on examination of five preserved specimens as well as four wild, live, uncollected individuals, all from the type locality of the taxa discussed.  Morphological details were noted using magnifying hand lenses. Measurements were taken using vernier calipers (least count 0.1mm) and in the case of snout-vent length, by a standard measuring tape (least count 1mm).  Where necessary, mean and standard deviation are given in brackets alongside ranges of numerical values.  Morphological character definitions and terminology follow Smith (1943), except for counting ventral scales for which we follow Gower & Ablett (2006), who included mental, postmental and preventrals in ventral counts.  Unequal symmetric scalation values are given in right, left order. Comparison is based on examination of 24 preserved specimens (see Appendix) and original description papers and subsequent taxonomic treatises (see literature cited).  Although the type specimens housed in European museums could not be accessed by us due to logistic constraints, our identifications of the mostly topotypic, non-types examined here were carefully cross-checked with the recently published type-redescriptions of U. ceylanicus group (except U. myhendrae) by Gower et al. (2008). Specific epithets used here follow Smith’s (1943) justified emendations, consequent upon his generic transfer to Uropeltis.

 

 

Taxonomy

 

Uropeltis shorttii (Beddome, 1863) comb. nov.

 

Silybura shorttii Beddome, 1863a (correct original spelling)

Silybura shortii – Beddome, 1863b (incorrect subsequent spelling)

Silybura shorttii – Günther, 1864: 191; Theobald, 1868: 43

Silybura shortii – Theobald, 1876: 134

Silybura nilgherriensis shortii – Beddome, 1886: 15

Silybura brevis – Boulenger, 1890: 269 in part; Boulenger, 1893: 158 in part

Uropeltis ceylanicus – Smith, 1943: 80 in part

U[ropeltis]. ceylanicus shortii – Murthy, 1990: 15

 

Taxonomic history

Cuvier (1829) first described Uropeltis ceylanicus from “Ceylan” (now Sri Lanka) but the type locality was subsequently considered erroneous, and the species is now considered endemic to India (Smith 1943).  This species has several very weakly characterised subjective synonyms originating from places far and wide in peninsular India (see Beddome 1886; Boulenger 1890, 1893; Smith 1943; Gower et al. 2008). Günther (1862) described Silybura brevis from Anamallays, Western Ghats.  Soon, this was followed by Beddome’s (1863) series of new species descriptions including S. shorttii from Shevaroy Hills in the Eastern Ghats, S. nilgherriensisfrom Nilgiris in the Western Ghats and S. nilgherriensis var. annulatafrom Wynaad, also in the Western Ghats. Lastly, Günther, (1864) described S. bicatenata from Deccan.  Gower et al. (2008) recently revived U. bicatenata (sic) from the synonymy of U. ceylanicus, thus demonstrating the potential existence of several other valid species hidden as synonyms within U. ceylanicus. Silybura shorttii was described by Beddome (1863) as a distinct species from the Shevaroy Hills in the Eastern Ghats based on syntypes BMNH 1946.1.15.91-94 (formerly 74.4.29.737-739) and MHNP 95.100 (McDiarmid et al. 1999). Silybura shorttii was considered as a subspecies of Günther’s S. brevis by Beddome (1886) who also relegated several other uropeltid species to subspecific status.  Boulenger (1890; 1893) further relegated the status of S. shorttii by listing it as a synonym of S. brevis.  This view was followed by Smith (1943), except that he rightly recognized the priority of Cuvier’s (1829) U. ceylanicus instead of erroneously perpetuating usage of Günther’s (1862) S. brevis. Murthy (1990) first allocated Silybura shorttii, as ‘shortti’ (sic) to Uropeltis, as a subspecies of U. ceylanicus (see chresonymy above).

 

Etymology

The specific epithet shorttii is a patronym, honouring its collector Dr. John Shortt, a medical physician with the then Madras Army (Playne et al. 1915), who donated the type specimens to Col. R.H. Beddome (see Beddome 1863a).

 

On the spelling of Silybura shorttii

The article Beddome (1863a) titled “Further notes upon the snakes of the Madras Presidency; with descriptions of new species” was published on 1 January 1863 in Madras Quarterly Journal of Medical Science, containing the original description of Silybura shorttii wherein the specimen collector was mentioned as Dr. Shortt.  The article Beddome (1863b) titled “Descriptions of new species of the family Uropeltidae from southern India, with notes on other little-known species” was published on 9 June 1863 in Proceedings of the Zoological Society of London. This latter paper contains an account on Silybura shortii wherein the specimen collector was miswritten as Dr. Short.  Silybura shorttiiBeddome, 1863a dated 1 January is obviously a precursor to Silybura shortiias mentioned in Beddome (1863b) dated 9 June, in keeping with Article 23 of theCode (ICZN 1999)—the Principle of Priority. But the spelling shortiihas been perpetuated in the literature (see chresonymy above).  Thus, Silybura shorttii Beddome, 1863 is the correct original spelling sensu Article 32 of the Code (ICZN 1999) and Silybura shortti used by Beddome, 1863b is an incorrect subsequent spelling.

 

Material examined (n=4, adults) (Image 1)

CSPT/S-80 (n=2, both adult males); ZSI/SRC/VRS 258 (n=2, subadults, a male and a female); coll. from Yercaud, Shevaroy Hills - a part of the Eastern Ghats, in Salem District of Tamil Nadu State, India.

 

Diagnosis

Uropeltis shorttii is diagnosed by the following combination of characters: tail shield with clearly defined, thickened, circumscribed disc; part of rostral visible from above not distinctly longer than its distance from frontal; rostral not fully separating nasals, shorter than rostral scale; dorsum dark blackish-brown with distinct yellow crossbars or annuli all over the body; ventrals 141–156 (148.5±10.06); venter with alternate rhomboidal large yellow and black spots or blotches, the two colours of equal intensities.

 

Description and variation (measurements in mm)

Snout-vent length 218.0–340.0 (279.0±86.2); tail length 13.0–17.0 (15.0±2.8); head length 7.9–12.0 (9.9±2.8); head width 4.9-8.5 (6.7±2.5); head depth 4.6–8.0 (6.3±2.0); body width 6.5–10.5 (8.5±2.8); eye-diameter 1.2–1.7 (1.4±0.2); eye-lip distance 0.9–1.1 (1.0±0.1); eye-nostril distance 2.0–2.9 (2.4±0.6); eye-rostrum distance 3.6–4.5 (4.0±0.6); interocular distance 3.3–4.3 (3.8±0.7); internarial distance 1.9–3.0 (2.4±0.7); snout-parietal distance 8.4–10.7 (9.5±1.6); posterior end of rostral to posterior end of parietal distance 6.9–8.9 (7.9±1.4); tail shield length 12.2–18.4 (15.3±4.3); tail shield width 5.7–11.4 (8.5±4.0); tail shield depth 6.2–9.3 (7.7±2.1); parietal scale length 2.9–4.7 (3.8±1.2); parietal scale width 2.1–3.8 (2.9±1.2); frontal scale length 2.7–3.4 (3.0±0.4); frontal scale width 2.8–3.1 (2.9±0.2); ocular scale length 1.9–3.0 (2.4±0.7); prefrontal scale length 1.5–2.7 (2.1±0.8); midbody ventral scale width 5.0–5.4 (5.2±0.2); midbody basal coastal scale width 2.3–2.7 (2.5±0.2).

 

Scalation

Rostral visible from above, smaller than nasal, not completely separating nasals; nasals in contact with one another posteriorly, prefrontals not in contact with rostral, subequal in size to nasal and ocular scales; nasals pierced by nostril, divided by rostral anteriorly but in contact with each other posteriorly; prefrontals somewhat larger than nasals and oculars, subequal to frontal; frontal longer than broad, distinctly smaller than parietal; parietals large, largest of all head scales; supralabials 4,4 (left, right), 1st and 2nd ones small, 3rdbelow eye, 4th the largest; infralabials 3,3 (left, right), elongate; mental scale small, subequal to 1st infralabial, but as wide as long; body scales imbricate, cycloid; dorsally around body in 19 (one head length after neck): 17 (at midbody): 17–15 (one head length before vent) rows; ventrals 141–156 (148.5±10.6), angulate laterally; anals 2, left overlapping right, each larger than a body scale; subcaudals 10–12 pairs +1 terminal scale; tail shield distinctly truncate above, mildly concave, circumscribed and ridged; covered with 30–31 (30.5±0.7), bi- and tri-carinate thickened scales; 10 scales across the length and 4–5 (4.5±0.6) across the width of the tail shield.

 

Colouration in life

Dorsum dark coffee brown with distinct bright yellow cross bars formed by series of yellow blotches across consecutive scales in dorsal scale rows; 34–47 (41±7.0) such cross bars present on body and tail; venter largely yellowish with dark brown spots and blotches, the dark spots restricted mostly to either side where ventral scales contact the outermost coastal scale rows; a pair of thick yellow stripes anteriorly along the neck and forebody on scale rows 3–5, the stripes extending to the level of the 36th–48th (42±8.4) ventral scale; the stripe passing through lower half of supralabials, below the ocular scale and upper half of infralabials; eye pale whitish-grey; inside of mouth pale pink; tongue of same colour, its tips lighter.

 

Colouration in preservative

Similar to in-life colouration, except that the bright yellow is faded to pale cream colour.

 

Comparisons and differential diagnosis

Uropeltis shorttii is herein compared with all the 26 currently recognized congeners from both India and Sri Lanka.  By having a thickened, circumscribed, mildly concave tail shield U. shorttiidiffers from the following 16 species: U. ellioti, U. nitidus, U. ocellatus, U. dindigalensis, U. beddomii, U. macrorhynchus, U. woodmasoni, U. broughami, U. maculatus, U. petersi, U. liura, U. pulneyensis, U. smithi from Western Ghats and U. melanogaster, U. phillipsi, U. ruhunae from Srilanka.  Further, U. shorttii also differs from the remaining congeners (after Gower et al. 2008) with a thickened, circumscribed, caudal shield categorized under Smith’s (1943) Group II A & B as follows (only opposing suite of character states of the congeners listed):U. macrolepis macrolepis found in northern Western Ghats: 15 midbody scale rows; ventral scales 127–140 (133.5±8.9); dorsum blackish-brown with yellow broken spots forming zig-zag crossbars or annuli; U. macrolepis mahabaleshwarensis found in northern Western Ghats: 15 midbody scale rows; ventral scales 120–130 (125±4.8); a pair of distinct, thick, yellowish-orange paravertebral stripes extending across most of the body except near neck, where there are two large orange spots; U. ceylanicus s. auct. here restricted to Western Ghats: dorsal body colouration uniform without distinct yellow crossbars; ventral scales 119–146 (132.5±19.0); U. arcticeps s. str. here restricted to Tirunelveli hills (see below): dorsally unpatterned, without distinct yellow crossbars; ventral scales 127–128; U. madurensis here restricted to High Wavys, Varushanad and Periyar hills (see below): dorsum uniform, without distinct yellow annuli; scales with distinct yellowish-rim over the body; venter with large orange and black blotched pattern; U. bicatenatusfound in northern Western Ghats: no yellowish scalloping chain-like pattern across both sides of the body; U. phipsonii found in northern Western Ghats: a pair of yellowish lateral streaks, one each, along both sides of the body; part of rostral visible from above distinctly longer than its distance from the frontal; U. myhendrae found in Tirunelveli, Ashambu and Travancore hills: part of rostral visible from above distinctly longer than its distance from frontal; U. broughami found in Nilgiri, Palni and Sirumalai hills: 19 midbody scale rows; rostral scale strongly developed and ridged with a dorsal keel; ventral scales 181–230 (205.2±34.6).

 

Distribution and field observations

Uropeltis shorttii in as far as is known, is restricted to Shevaroy Hills (11’N & 78’E; 350–1600 m), a part of southern Eastern Ghats located in Salem District of Tamil Nadu.  Shevaroy Hills that is largely cultivated with coffee and silver oak today, has some remnant patches of tropical evergreen cloud forests and has been reported to harbor wet-forest taxa (e.g., the endemic Yercaud Day Gecko Cnemaspis yercaudensis Das & Bauer, 2002), even though set amidst the drier, rocky, Eastern Ghats hill range. Although Smith (1943) reported the Shevaroys in the distribution of his U. ceylanicus, it was inclusive of U. shorttii, which as currently understood is allopatric with respect to the Western Ghats species U. ceylanicus s. auct. (also see Gower et al. 2008).  The live specimen was encountered on the way to Kiliyur Falls, a seasonal waterfalls surrounded by silver oak and coffee plantations and smaller patches of evergreen forests in Yercaud (1550m) at the summit of Shevaroys.  The only syntopic congener occurring with U. shorttii is U. ellioti, an apparently widespread species belonging to a different species group (see Smith 1943).  We doubt the identity of the species mentioned as ‘S. shortii’ by Beddome (1886) from Anamalais and based on our field observations including part of the data presented herein, we do not expect U. shorttii and U. ceylanicus to co-occur either in the Western or the Eastern Ghats.

 

Uropeltis madurensis (Beddome, 1878) comb. nov.

 

Silybura madurensis Beddome, 1878: 802

Silybura nilgherriensis arcticeps -–Beddome, 1886:16 in part

Silybura madurensis – Boulenger, 1890: 267, 1893: 156

Uropeltis arcticeps madurensis – Smith, 1943: 81 in part; Whitaker & Captain, 2004: 71; Chandramouli & Ganesh, 2010: 79

Uropeltis arcticeps (non Silybura arcticeps Günther, 1875) – Hutton & David, 2009: 310

 

Taxonomic history

Beddome (1878) described Silybura madurensis based on syntypes BMNH 1946.1.16.38-39 (formerly 82.1.12.11-12) (McDiarmid et al. 1999) collected from High Wavy Mountains, Madura[i] District, [Tamilnadu State] elevation 5500 feet, southern India.  Beddome (1886) relegated S. madurensisto the synonymy of another congener S. arcticeps Günther, 1875.  Again, S. arcticeps was considered as a ‘variety’ of S. nilgherriensis Beddome, 1863.  Beddome (1886) did this rather reluctantly, as evidenced by his reporting of two distinct, non-overlapping ventral scale count ranges “127-128” for arcticeps and “146-157” for madurensis. Some subsequent workers on Indian snakes (Boulenger 1890, 1893) continued to recognise madurensis as a valid species.  However, Smith (1943) again followed Beddome (1886) and listed S. madurensis as a synonym of S. arcticepsalong with another ‘variety’ S. nilgherriensis var. pictaBeddome, 1886 originating from Peermade, Travancore.  The status of pictawas not discussed by Smith (1943), for reasons not made clear, but it is worth mentioning here that Boulenger (1890, 1893) listed var. picta as a synonym of U. madurensis and not of U. arcticeps.  Murthy (1990) and Whitaker & Captain (2004) have regarded U. madurensis as a subspecies of U. arcticeps,and their nomenclatural usages have been–Uropeltis arcticeps arcticeps(Günther, 1875) and U. a. madurensis (Beddome, 1878).

 

Etymology

Although not mentioned in the original description, the specific epithet madurensis is a toponym, named after its “type locality” sensu lato ‘Madura’ (= Madurai District, Tamilnadu State), southern India.

 

Material examined (Image 2)

CSPT/S-6 an adult female from (the greater) Madurai District, Tamil Nadu, southern India; formalin-preserved; collector and date unknown.  Four more uncollected topotypic specimens documented in the field by the first author in High Wavy Mountains, Tamil Nadu, southern India.

 

Diagnosis

Uropeltis madurensis can be diagnosed by the following combination of characters: tail shield with clearly defined, thickened, circumscribed disc; part of rostral visible from above not distinctly longer than its distance from frontal; rostral not fully separating nasals; dorsum uniform brown, each scale with a well-defined lighter golden yellowish outline; ventrals 144–157; venter with alternate rhomboidal large brown and orange spots or blotches, the two colours of equal intensities.

 

Description of preserved specimen (measurements in mm)

Snout-vent length 320.0; tail length 10.3; head length 10.0; head width 9.05; head depth 5.5; body width 10.5; eye-diameter 1.9; eye-lip distance 1.1; eye-nostril distance 2.9; eye-rostrum distance 4.0; interocular distance 4.7; internarial distance 3.5; snout-parietal distance 10.7; posterior end of rostral to posterior end of parietal distance 8.9; tail shield length 13.4; tail shield width 7.8; tail shield depth 8.7; parietal scale length 4.1; parietal scale width 3.8; frontal scale length 4.0; frontal scale width 3.1; ocular scale length 3.0; prefrontal scale length 3.0; midbody ventral scale width 5.4; midbody basal coastal scale width 2.7.

 

Scalation

Rostral visible from above, not fully separating nasals; portion of rostral visible from above less than distance from frontal; nasals pierced by nostril, divided by rostral anteriorly but in contact with each other posteriorly; prefrontals slightly larger than nasals / oculars, subequal to frontal; frontal longer than broad, distinctly smaller than parietal; parietals large, largest of all head scales; a small rhomboid accessory scale just behind parietal; supralabials 4,4, 1st and 2ndones small, 3rd below eye, 4th the largest; infralabials 3,3, elongate, 1st pair slightly curved anteriorly; mental scale small, subequal to 1st infralabial, but as wide as long; dorsal scales in 17 (one head length after neck): 17 (at midbody): 15 (one head length before vent) rows; ventrals 144, angulate laterally; anals 2, right overlapping left, each larger than a body scale; subcaudals 7 pairs + 1 terminal scale; tail shield distinctly truncate above, ridged and slightly concave; covered with 30, bi- and tri-carinate thickened scales; 7 scales across the length and 4–5 across the width of the disc.

 

Colouration in preservative

Overall dorsal body colour pale brownish-grey, with contrastingly coloured pale whitish scale border around all dorsal scales; neck with two parallel, longitudinal pale stripes one on each side, that converge towards the mental; each stripe extends dorsally to the supralabials and two to three outermost scalerows, posteriorly to the first lateral blotch present on the neck; eye pale white; inside of mouth pale rose to grey deeper inside; venter brownish-grey with 32, paired, off-white blotches, each blotch two to three scales large; blotches either alternate or rarely conjoin to form cross bars; subcaudals deep brown enclosed by two parallel stripes laterally, joined together anteriorly by a cross bar at the anal scale; tail shield dorsally with dark brown swatches on a pale whitish background.

 

Colouration in life based on topotypes sighted in the field (n=4)

Dorsum rich brown; each scale with a yellowish-orange outline; venter heavily blotched with alternate large spots of orange and dark brown, the two colours being more or less equal in proportions; a pair of thick, orange stripes extending from last supralabial scale to the anterior 1/3rdof the body, across the first blotch near the neck; anal and subcaudal scales orange with smaller blackish-brown spots.

 

Distribution and field observations

Uropeltis madurensis is now considered endemic to HighWavys-Varushanad-Periyar hill complex.  This region is located between the Palnis to the north and the Srivilliputhur Hills to the south (Hutton & David 2009).  The first author studied U. madurensis in cloud forest-plantation matrix in High Wavys between December 2007 and January 2008, in the post-monsoon season.  Four adults were sighted on a high elevation mountainous plateau ca. 1300–1600 m.  The first snake was sighted dormant under a small rock in a streamside rainforest tract near Cloud Land Estate.  The second one was actively moving about on forest floor near tea estates in Upper Manalar on a rainy day.  The third one was a road-kill sighted amidst coffee plantations in Manalar/Sand River Cottage.  The fourth one was found resting under a small cement slab near an estate bungalow in Eravangalar.  Important characters of these specimens include mildly concave, thickened, circumscribed tail shield, 17 midbody scalerows, dorsal body with contrasting yellow-orange outline and 148–157 ventrals.  Syntopic congeners include Uropeltis cf. dindigalensis (see Chandramouli & Ganesh 2010), U. liura (see Smith 1943) and U. ceylanicus, U. ellioti, U. pulneyensis, U. rubromaculatus and U. woodmasoni (after Hutton & David 2009).  All syntopic congeners except U. ceylanicus and U. rubromaculatusbelong to different species groups (Smith 1943).

 

Comparisons and differential diagnosis

Uropeltis madurensis is herein compared with all the 26 currently recognized congeneric taxa from both India and Sri Lanka.  By having a thickened, circumscribed, concave tail shield U. madurensisinstantly differs from the following 16 species: U. ellioti, U. nitidus, U. ocellatus, U. dindigalensis, U. beddomii, U. macrorhynchus, U. woodmasoni, U. broughami, U. maculatus, U. petersi, U. liura, U. pulneyensis, U. smithi from Western Ghats andU. melanogaster, U. phillipsi, U. ruhunae from Sri Lanka.  Further, U. madurensis also differs from the remaining congeners (after Gower et al. 2008; Whitaker & Captain 2004) with a thickened, circumscribed, caudal shield categorized under Smith’s (1943) Group II A & B as follows (only opposing suite of character states listed): U. macrolepis macrolepis found in northern Western Ghats: 15 midbody scale rows; ventral scales 127–140 (133.5±8.9); dorsum blackish-brown with yellow broken spots forming zig-zag crossbars or annuli; U. macrolepis mahabaleshwarensis found in northern Western Ghats: 15 midbody scale rows; ventral scales 120–130 (125±4.8); a pair of distinct, thick, yellowish-orange paravertebral stripes extending across most of the body except near neck, where there are two large orange spots; U. ceylanicus s. auct. here restricted to Western Ghats: dorsal scales lacking a clearly defined yellowish border; ventral scales 119–146 (132.5±19.0); U. arcticeps s. str. here restricted to Tirunelveli hills: dorsal scales lacking a clearly defined yellow scale border; ventral scales 127–128; U. shorttii here restricted to Shevaroys, Eastern Ghats: dorsal body with distinct yellowish annuli or crossbars, each body scale without yellowish-orange border or rim; distribution Eastern Ghats; U. bicatenatus found in northern Western Ghats: no yellowish scalloping chain-like pattern across both sides of the body; U. phipsonii found in northern Western Ghats: a pair of yellowish lateral streaks along both sides of the body; part of rostral visible from above distinctly longer than its distance from the frontal; U. myhendrae found in Tirunelveli, Ashambu and Travancore hills: dorsum with brownish-black body, each scale with yellowish posterior border forming more or less complete band or annuli; part of rostral visible from above distinctly longer than its distance from frontal;U. broughami found in Nilgiris, Palni and Sirumalai hills: 19 midbody scale rows; rostral scale strongly developed and ridged with a dorsal keel; dorsum brown with distinct small, yellow-black-edged transverse ocelli; ventral scales 181–230 (205.5±34.5).

 

 

Discussion

 

Uropeltis ceylanicus, as treated here in a partially conservative approach, still contains three junior subjective synonyms—brevis Günther, 1862 from Anamallays; nilgherriensis Beddome, 1863 from Nilgiris and nilgherriensisvar. annulata Beddome, 1863 from Wynaad, all in the Western Ghats.  We repeat the statement of Gower et al. (2008) that full re-evaluation of the taxonomy of U. ceylanicus is beyond the scope of the present study, pending new range-wide fieldwork and collections.  But this in no way hinders our hypothesis that the Eastern Ghats endemic U. shorttii is specifically distinct from the Western Ghats endemic U. ceylanicus which is also reflected in their extensive morphological differences. Indeed U. ceylanicus is so far unknown from the Eastern Ghats except for the implied record of the type locality of U. shorttii, which, from now on, is no more U. ceylanicus.

With regards to U. arcticeps andU. madurensis, as stated above, another available name Silybura nilgherriensis picta Beddome, 1886 has been associated as a subjective junior synonym of these species (Boulenger 1890, 1893; Smith 1943).  Its characters mentioned in the original description (Beddome 1886) including dorsum and venter with orange-yellow and black blotches (vs. uniform brownish-black in arcticeps; brown with each scale outlined in yellow in madurensis), ventrals 150 that is consistent with U. madurensis (vs. 127–128 in arcticeps) and distribution: Peermade 1000–1300m (vs. Tirunevlei hills for arcticeps; High Wavys for madurensis) denote that owing to incongruence in both morphology and distribution, it must not be considered as a synonym of either species.  We also feel that its ventral count 150 might have prompted Boulenger (1890, 1893) to consider S.n. picta as a subjective junior synonym of U. madurensis, rather than U. arcticeps.  As was the case with some other nomina relegated to the synonymy of U. ceylanicus, lack of data precludes us from commenting any further on the taxonomic status of Silybura nilgherriensis var. picta.

Smith (1943) in his accounts of U. ceylanicus and U. arcticeps had written “with yellowish spots transversely arranged (shorttii)” and “ventrals 146–157 (madurensis)”.  This implicit recognition of names with an express association with consistent features that are diagnosable, serving them to be identified from their nearest related congeners explains how these species got lumped in an over-circumscribed taxon-boundary.  This lumping is exemplified by the fact that Beddome (1886) considered var. shortii and arcticeps(including madurensis) as both belonging to a single species Silybura nilgherriensis Beddome, 1863. We are of the opinion that most historical workers apparently felt contended keeping these diagnosable and allopatric (see Image 3) species to a low profile, as ‘varieties’ or ‘subspecies’. In this case, U. shorttiifrom Shevaroy Hills in the Eastern Ghats having distinct yellow cross bars (vs. uniform in U. ceylanicus) and U. madurensis from the High Wavys having considerably high number of ventral scales (vs. lower no. of ventrals in U. arcticeps) testifies this.  Our present examination of preserved material and wild-caught topotypes revealed subtle differences that disclosed the true taxonomic status of some of the available names.  Further specimen examinations and field observations are necessary to reassess the status of other nomina and the true taxonomic diversity of this group.

 

 

REFERENCES

 

Beddome, R.H. (1863a). Further notes upon the snakes of the Madras Presidency; with descriptions of new species. Madras Quarterly Journal of medical Science 6: 41–48, 1 pl.

Beddome, R.H. (1863b). Descriptions of new species of the family Uropeltidae from southern India, with notes on other little-known species. Proceedings of the Zoological Society of London225–229, 3 pl.

Beddome, R.H. (1886). An account of the earth-snakes of the peninsula of India and Ceylon. Annals and Magazines of Natural History 17(5): 3–33.

Boulenger, G.A. (1890). Fauna of British India. Reptilia and Batrachia.Taylor and Francis, London, UK.

Boulenger, G.A. (1893). Catalogue of Snakes in the British Museum. Vol. I. Taylor and Francis, London, UK.

Chandramouli, S.R. & S.R. Ganesh (2010). Herpetofauna of southern Western Ghats, India - reinvestigated after decades. Taprobanica 2(2): 8–21, pls.4.

Daniel, J.C. (2002). The Book on Indian Reptiles and Amphibians. Oxford University Press, Bombay Natural History Society, India, 248pp.

Das, I. (2002). A Photographic Guide to Snakes and other Reptiles of India. New Holland Publishers, U.K., 144pp.

Gower, D.J. & J.D. Ablett (2006). Counting ventral scales in Asian anilioid snakes. The Herpetological Journal16: 259–263.

Gower, D.J., A. Captain & S.S. Thakur (2008). On the taxonomic status of Uropeltis bicatenatus (Günther) (Reptilia: Serpentes: Uropeltidae). Hamadryad33(1): 64–82.

Günther, A.C.L.G. (1864). Reptiles of British India. Taylor and Francis, London, UK.

Hutton, A.F. & P. David (2009). Note on a collection of snakes from south India, with emphasis on the snake fauna of the Meghamalai Hills (High Wavy Mountains). Journal of the Bombay Natural History Society 105 (3): 299–316.

ICZN (International Commission on Zoological Nomenclature). (1999). International Code of Zoological Nomenclature.Fourth edition, adopted by the International Union of Biological Sciences, International Trust of Zoological Nomenclature, London, xxix+306pp.

McDiarmid, R.W., J.A. Campbell & T.A. Toure (1999). Snake Species of the World: A Taxonomic and Geographic Reference. The Herpetologists’ League, Washington DC.

Murthy, T.S.N. (1981). Checklist and key to Uropeltid snakes of India. The Snake 13: 142–150.

Murthy, T.S.N. (1982). An illustrated field guide to the rough tailed snakes of India. The Snake 14: 119–135.

Murthy, T.S.N. (1990). Illustrated guide to the snakes of the Western Ghats, India. Records of the Zoological Survey of India, Occasional Paper 114: 1–69.

Playne, S., J.W. Bond & A. Wright (1915). Southern India: Its history, People, Commerce and Industrial Resources. Foreign and Colonial Compiling & Pub. Co., London, UK, 766 pp.

Rajendran, M.V. (1985). Studies in Uropeltid Snakes. Madurai Kamaraj University, Madurai, India.

Smith, M.A. (1943). Fauna of British India including Ceylon and Burma. Vol - III Serpentes. Taylor & Francis, London, UK, 583pp.

Theobald, W.E. (1868). Catalogue of reptiles in the Museum of the Asiatic Society of Bengal. Journal of the Asiatic Society of Bengal. Extra number: 88pp.

Theobald, W. (1876). Descriptive catalogue of the reptiles of British India. Thacker Spink and Co., India, 238pp.

Whitaker, R. & A. Captain (2004). Snakes of India - The Field Guide. Draco Books, Chengelpet, South India, 481.