Records of the Indian Sand Snake Psammophis condanarus (Merrem, 1820) (Reptilia: Lamprophiidae) in southern India

Acknowledgements: We thank our respective organisations for supporting our research activities. SRG thanks the Executive Chairman, all the Trustees and the Director of Chennai Snake Park for their encouragement. MBG and SRG thank the Andhra Pradesh Forest Department especially Principal Chief Conservator of Forests and Chief Wildlife Warden, Chief Conservator of forests, Wildlife Management Circle, Tirupati and District Forest Officer of Wildlife Management Division, DFO, Tirumala Tirupati Devasthanam (TTD), Tirupati for granting study permission. VRKS and SRG thank the District Forest Officer and Forest Range Officer, Bellari, Karnataka Forest Department for permission to study the snake and Mr. Adithya Vattam, Mr. Aslam Sayyed and Mr. Pompayya Malemath for their information, kind support and logistics. Records of the Indian Sand Snake Psammophis condanarus (Merrem, 1820) (Reptilia: Lamprophiidae) in southern India

The Indian Sand Snake Psammophis condanarus (Merrem, 1820) is distributed in eastern, northern and central India including parts of the Himalayan foothills, Bengal, Indo-gangetic plains, northwestern arid desert zones including Pakistan and northern parts of the Deccan plateau (Stoliczka 1872;Murray 1886;Wall 1908;Minton 1966;Whitaker & Captain 2004;Chandra & Gajbe 2005), making it the most widespread species of the genus in the Indian subcontinent. In fact it is the only congener in most of the central and eastern Abstract: We present new records of the Indian Sand Snake Psammophis condanarus from southern India, where its existence has remained doubtful till date. Our records are based on both live and preserved voucher specimens that are illustrated and described here. We furnish distribution records of this species from two sites belonging to two different ecoregions in southern India-Tirupati in the Eastern Ghats and Hospete in the Deccan plateau. Our work highlights the obscurity of certain, large-growing, diurnal land snakes that have as yet managed to evade the attention of field biologists largely due to a lack of field surveys in certain ecoregions.
The current day distribution of this species as such encompasses the type localities of these synonymised nomina in as far as they are known (Whitaker & Captain 2004). Ali (1943) reported this species from Bandipur (11.44 N & 76.50 E) near the Western Ghats, abutting the Mysore plateau, on the basis of a single specimen recovered from the stomach of a Short-toed Eagle Circaetus gallicus. Prasad (1992) highlighted the same and opined it to be a valid but overlooked record. He did not present any new material from southern India. We hereby confirm and elaborate on the distribution of this species in southern India.

Material and Methods
This study is based on the examination of both live and preserved specimens, one each from the Eastern Ghats and the Deccan plateau. Morphological examination terminology and protocols follow Whitaker & Captain (2004). Ventral scale counting follows Dowling (1951) and hemipenal description follows Dowling & Savage (1960). Body length was measured using a standard measuring tape (L.C 1mm) while other smaller measurements were taken using vernier callipers (L.C 0.1mm). Scale counts were done using a magnifying hand lens (5 X optical zoom). Scalation and distribution data were compared with literature. Photographs were taken using digital cameras. Description (also see Table 1): A thick-set, smooth and glossy-bodied snake with fairly large head, sharply protruding snout, concave loreal, distinct neck, robust body and tapering tail. Measurements of preserved specimen (in mm): head length: 23.50, head width: 11.50, head depth: 7.50, midbody width: 12.00, eye diameter: 3.30, eye-lip distance: 5.50, inter-narial distance: 3.50, frontal scale length: 6.00, frontal-rostral distance: 4.00, frontal width at midline: 2.20, frontal width at anterior end: 3.00. Measurements of live specimen (in mm): head length: 18.5; head width: 7.5; head depth: 7.0; body width: 10.5; eye diameter: 2.5. Scalation: Rostral visible from above, protruding, with a distinct cleft underenath; nasal scale only partially divided, sutured below the nostril, reaching between 1 st and 2 nd supralabials, loreal oval, posterior genials slightly longer than anterior pair, dorsal scales imbricate, smooth but with distinct and deep apical pits, outermost coastal scale rows slightly larger than the rest, vertebral scale rows not larger than the rest, scales on dorsal tail larger, ventral scales very wide, extending on to ventrolateral parts, not angulate laterally. Colour in preservation of voucher specimen (formalin-darkened): slaty dark grey above with white and black spots on labia, chin and outermost scalerows; dorsum with three dark greyish-brown stripes -one vertebral stripe that is five scalerows wide (at midbody) narrowing posteriorly to three scalerows wide; this one flanked by two lateral stripes on either side that are one scale row wide; each ventral scale dotted with black on either extremities forming a ventro-lateral line. Colouration in life (based on both specimens): dorsum light rosy grey, with a broad, five scales-wide dark coffee brown, black and white edged vertebral stripe; laterally flanked by two thinner stripes three scaleswide, partly of fully black-bordered similar dark brown bilateral stripes on each side. Top of head dark brown being the origin of the dark broad vertebral stripe; sides of head covered by similar dark brown stripe across eye, separated above a thin lighter supraocular stripe; rest of the head (including labia), chin and underside of head pale white with brownish spots; a brown-bordered white ventrolateral stripe covering the confluence of Southern Indian records of Psammophis condanarus Ganesh et al. ventral and outermost coastal scalerows; ventral and subcaudals scales pale yellow. Hemipenis (n=1, of the preserved specimen): organ everted, with a single side exposed out; organ smooth, slender and without a broad lobe-head, not quite forked at tip; pedicel narrower, at the level of tubular part; pedicel head lacking spiny projections or other distinct architecture; sulcul lips not prominent, smooth; organ 18mm long and 4mm wide, extending upto 5 th subcaudal scale. Field observations: The snake from Tirupati was observed actively moving about in a grassy patch during early morning hours in Kapilatheertham, at the foot of the Tirumala Hills. The one from Hospet was captured from suburban outskirts of the city during daytime when it was taking shelter along the boundary wall of a building. The area was vegetated with thorny bushes and human settlements.

Discussion
We attribute the Eastern Ghats and Deccan populations to P. condanarus based on the following combination of characters which are compared with all Indian congeners: anal scale divided (vs. single in P. leithi), preocular not touching frontal (vs. touching in P. leithi and P. schokari), frontal anteriorly not twice as wide as at midline (vs. anteriorly twice as wide as at midline in P. leithi and P. schokari), nasal only partially divided (vs. completely divided in P. leithi and P.   schokari), frontal distinctly longer than its distance from snout-tip (vs. not longer than its distance from snouttip in P. longifrons), body with 3 or 5 longitudinal stripes with vertebral stripe darker (vs. 4 with vertebral stripe light in P. leithi, 4 or uniformly dark with vertebral stripe light in P. schokari, with brownish body and scales edged with black in P. longifrons) (see Smith 1943;Whitaker & Captain 2004). Smith (1943) described the Indo-Chinese population of P. condanarus as subspecies P. condanarus indochinensis based on the differences in ventral, subcaudal counts and dorsal stripe patterns. Later it was given full species status by Hughes (1999) in his review on primarily African species. Psammophis condanarus is distinct from P. indochinensis by having higher ventrals (165-179 vs. 156-173 in P. indochinensis), higher subcaudals (75-93 vs. 66-85 in P. indochinensis), dorsal pattern (vertebral stripes darker vs. vertebral stripe lighter/absent or variable), number of dark dorsal stripes (3 or 5 vs. 4 in P. indochinensis) and their different geographical distributions. It is noteworthy here that the ventral counts of the new material resemble P. indochinensis much more than P. condanarus. We provisionally consider them to represent P. condanarus, based on congruence of other morphological characters and distribution.
Due to this morphological incongruence, it is essential here to deal with the synonyms of P. condanarus (Günther 1864; also see Uetz & Hosek 2017). Jerdon (1854) described Leptophis bellii from Jalna. Günther (1864) remarks that he was able to categorically identify Jerdon's Snake based on Walter Elliot's drawing named Leptophis bellii that depicted a snake identical to P. condanarus. Günther (1864) also associated another nomen Psammophis taeniata Günther, 1862 from Forts Pitt's museum specimens from India as a synonym of P. codanarus. Though P. taeniata was later not recognised as a synonym (Smith 1943) it was then again listed so (Wallach et al. 2014). It is noteworthy here that Psammophis taeniata mentions four dark dorsal stripes (Günther 1862). The next synonym Psammophis indicus Beddome, 1863 was erected on the basis of a holotype from the Eastern Ghats (Nallamalais, as Nullay Mullay hills of Kurnool district; Beddome 1863). It also specifies three dark stripes, although, Beddome (1863) did not mention any ventral count value, precluding us to associate or recognise our geographically discrete morphological variation.
Another synonym Phayrea isabellina Theobald, 1868 has had a rather unsettled past. In its original description, Theobald (1868) provided a very brief account and later authors often just repeated the same information. This short-lived nomen remained valid for just two years and  and "no [collector] history" is associated with the specimen. Later, Wall (1921) redescribed the type and opined it to be more related to an Amphiesma pointing out discrepancies in dorsal scale row counts and the consequent taxonomic interpretations that entail. Das et al. (1998) mention this nomen as a synonym of an unrelated Neotropical snake Liophis lineataus (Linnaeus, 1758). Without discussion, Wallach et al. (2014) again list Phayrea isabellina in the synonymy of P. condanarus. In essence, due to the complicated taxonomy of P. condanarus, a population systematics study is needed to resolve the problems. Our individuals are similar to those from Maharashtra plateau in having only three dark dorsal stripes where vertebral stripe covers 3 or 5 vertebral scale rows. In the past records it is also mentioned that Eastern Maharashtra's population bears only three dark stripes (Whitaker & Captain 2004). Smith (1943) mentions a juvenile from the same area having a vertebral stripe that occupies five scale rows. No other data or discussion is available except these small variations, perhaps indicative of the very few specimens examined from these parts. To get further support, one of us (VS) examined and confirmed the above mentioned characters in the population of Nashik of western Maharashtra where one specimen had a vertebral stripe covering five dorsal scales with 172 ventrals and 73 subcaudals while another one had the same covering three dorsal rows with 173 ventrals and 73 subcaudals (Image 2). To know the pattern in the nominotypical population, we examined the plate of Russell (1796) and found that the drawn specimen had only three stripes.
In southern India, this species was doubtfully known from Bandipur of Mysore plateau (Ali 1943) so far. This work confirms and extends the distribution of P. condanarus in southern India where it is likely to be the only representative congener in most of its range unlike other parts further north. It may be noted that recent surveys also revealed the presence of at least one other congener, the stout Sand Snake Psammophis longifrons Boulenger, 1890 in southern India (Shikaripur, Shimoga of Karnataka by Premkumar & Sharma 2017), where its existence had been doubted previously (Boulenger 1890;Smith 1943;Whitaker & Captain 2004). Large variation in ventral and subcaudal scale counts warrants taxonomic studies on this population. It is hoped that our current findings will instigate and encourage further work on the population systematics of Psammophis condanarus.
In peninsular India, studies on snakes had mainly targeted the evergreen forest-clad Western Ghats biodiversity hotspot in the past (e.g., Hutton & David 2009;Ganesh et al. 2014), while other ecoregions (such as the Eastern Ghats, Deccan plateau, etc.,) that are much drier, have been rather neglected. But recent works in these under-studied regions (e.g., Rao et al. 2005;Srinivasulu & Das 2008;Ganesh & Arumugam 2016;Pompayya & Aslam 2016) reveal higher herpetological diversity than earlier presumed. Both the historical reports of P. condanarus from Nallamala (Beddome 1863) and P. longifrons from Cudappah (Boulenger 1890) were not mentioned in the recent works on the Nallamala herpetofauna (Rao et al. 2005;Srinivasulu & Das 2008). The recent findings show that the biodiversity, especially of the herpetofauna, of the dry forests in several ecoregions of southern India, was grossly underestimated.