Rediscovery of Pseudophilautus hypomelas ( Günther , 1876 ) ( Amphibia : Anura : Rhacophoridae ) from the Peak Wilderness , Sri Lanka , a species thought to be extinct !

Author Contribution: LJMW was involved in designing the study, conducting the field survey, conducting museum reference work, gathering, and analyzing the data, and compiling the MS. DRVP was involved in conducting the field survey, conducting NMSL museum reference work, gathering, analyzing the data, and writing the MS. MDGR was involved in conducting the field survey, conducting NMSL museum reference work, gathering, analyzing the data, and writing the manuscript. NW was involved in field research work and writing of the manuscript.

Pseudophilautus hypomelas was only known from the syntype series deposited in the Natural History Museum, London (BMHH). Despite extensive field studies carried out over the past decade  in Sri Lanka this species was reported to be extinct in the scientific literature , IUCN & MENR 2007Bain et al. 2008;Chanson et al. 2008;Stuart et al. 2008;Manamendra-Arachchi & Meegaskumbura 2012). The species was first described by Günther as Ixalus hypomelas, later Kirtisinghe (1957) synonymized the species with P. leucorhinus, but considering its well distinguishing characters Dutta & Manamendra-Arachchi (1996), resurrected the species as Pseudophilautus hypomelas.  designated lectotype and para lectotypes from the syntypes of 14 specimens deposited at BMNH, most of which were in poor condition without having much information on external characters. Owing to the paucity of information, we conducted a survey in the Peak Wilderness Sanctuary of the Central Province, Sri Lanka and from our findings report the rediscovery of Pseudophilautus hypomelas, which was thought to be extinct.
The Peak Wilderness Sanctuary is one of the few remaining areas in Sri Lanka with a continuous natural forest with a cover of altitudinal graded forest types, ranging from lowland mixed dipterocarp forests to montane cloud forests and is an area of great biological diversity (Singhakumara 1995;;Fernando & Ranasinghe 1997). The Peak Wilderness Sanctuary, in particular, harbors a majority of the endemic and threatened bird species of Sri Lanka (Ranawana & Bambaradeniya 1998;Wickramasinghe et al. 2007).

MATERIALS AND METHODS
The field survey was carried out in the Peak Wilderness, to assess the diversity of amphibians in an elevation gradient for over two years. The field survey commenced at the end of 2009, and phases I and II were completed in December 2011. The sampling sites were selected in a random manner considering accessibility and to cover representative habitats in the Nature Reserve, through an initial reconnaissance survey. Fifteen field visits were made with each sampling session spanning eight continuous days. Sampling was nocturnal, and photographs of most species were taken in the wild to avoid any confusion of change in colour during captivity. Specimens collected in the field were first fixed in 90% ethanol for two hours and stored in 70% ethanol.
Sex and maturity were determined by examining secondary sexual characters, or when absent, by examining the gonads through a small lateral incision in to the specimen. The material referred to is deposited in the Natural History Museum London (BMNH), and the National Museum, Sri Lanka (NMSL). Three specimens were collected for the current work and are deposited in the NMSL and the Girithale National Wildlife Research and Training Center (Department of Wildlife Conservation (DWC)).
The rediscovered species was compared with all types from Sri Lanka deposited in the NMSL, and type specimens deposited in the BMNH. The specimens formerly belonged to the Wildlife Heritage Trust (WHT) bearing WHT numbers and are currently deposited in the NMSL, catalogued under the same numbers.
Forty-four external measurements of specimens were taken with a Mitutoyo digital vernier calliper to the nearest 0.1mm. Terminology of external morphology abbreviated in the text and external measurements for the description section follows Wickramasinghe et al. (2013a).
Snout angle (Manamendra-Arachchi & Pethiyagoda 2005) is not considered here due to an influence of preservation artifacts. Geographical coordinates were determined from GPS readings (Gamin eTrex Gista) with WGS84 (World Geodetic System) datum at the locality.

RESULTS AND DISCUSSION
There are 14 specimens of P. hypomelas deposited in the BMNH collection, of which one specimen had been designated as a lectotype, mature female 20.9mm SVL (BMNH 1947.2.27.8, collector-Beddome, locality-Ceylon) by . From the remaining 13 paralectotypes 11 were critically examined while the two remaining paralectotypes BMNH 1947.2.27.9, female specimens (19.7 SVL) and BMNH 1947.2.27.10 (20.7mm SVL) collected by W. Ferguson, from Ceylon were separately listed as 'others' (data not provided) in the same publication. Since all the 11 paralectotypes considered by Manamendra-Arachchi & Pethiyagoda (2005) have been completely dried out, and the two specimens collected by W. Furguson are in better condition, we have considered only the latter in the present study. The lectotype and the two paralectotypes we consider here were originally registered in 1876 as lot BMNH 1876.3.21.31-33 Ixalus hypomela. They were re-registered in 1947 as BMNH 1947.2.27.8-10.
During a nocturnal sampling on the 06 April 2010, about 40 frogs resembling P. hypomelas were noticed from the Peak Wilderness Sanctuary (06 0 48'28.02"N & 80 0 28'14.46"E, elevation 1300m). Initially these frogs were thought to be new to science, of which only three specimens were collected to ascertain their taxonomic identity.
Careful museum studies confirmed the identity of the above collected specimens as Pseudophilautus hypomelas. Since the original description by Günther (Appendix 2) is precise here we redescribe the species with a report of rediscovery after a span of 137 years for scientific clarity of the species.
Skin on dorsal and lateral snout and between eyes smooth; upper eyelid smooth; head smooth laterally, shagreened near the gape of mouth; dorsum smooth; thin median dermal ridge on mid dorsum from tip of snout to back of head; upper and lower part of flank smooth; supratympanic fold prominent; upper arm smooth; forearm and hand smooth; inner, outer and dorsal thigh smooth; leg dorsally smooth; small tubercles present at heel; tarsus and foot smooth.
Ventral side of body: Throat and chest smooth; belly granular; forearm and upper arm smooth; thigh weakly granular; leg and tarsus smooth.
Colour in life: Dorsum cream colour, a pair of broad bronze longitudinal dorsal bands extends from the back of the eye to the groin, bronze band between the eyes forms a prominent 'T' shaped patch centrally projecting towards vent; laterally bronze band on canthal edge, bronze patch below eye, dark brown band below supratympanic fold; limbs dorsally cream, forelimb, hind limb, fingers and toes with bronze bands; ventrally off white with dark brown blotches, throat darker, belly off white, hands, feet, and webbing dark brown (Image 1).
Colour in alcohol: Colour pattern remains with a little fading, bronze changes to brown and off white to a yellowish tinge (Image 2A,D); ventral side blotching preserved with a little fading.

Natural history
The species was found in elevations of 750-1400m in lower montane rain forests (Fig. 1). Commonly observed in bushes of less than 1m high, and preferred grassy habitats with a less canopy cover such as those in disturbed areas.
P. hypomelas was identified by its identical resemblance to the material found in the BMNH with its comparable sizes, smooth skin, similar ratios of morphometric measurements, above all by the marking on head (Image 4 A-C), and the blotching on ventral side (Image 4 D-F) which are still preserved in the type material confirming its identity. Apart from which, in some collected material a prominent cross marking was noticed on the ventral side (Image 5A) as mentioned in variations section; interestingly this marking was also seen in some type material specimens (Image 5B). Also the type description mentions "a fine white line runs along the middle of the back and of the abdomen, beginning from the snout….." (Image 3), and Günther also states that all or some of these lines may be absent, which is true for some specimens (mentioned under variations) which did not possess these lines. In the original type description Günther erroneously states that its "tympanum hidden", "Fingers not webbed; web of the hind foot rudimentary", but a distinct tympanum can be seen (Image 6) in all the specimens found in the BMNH, and in our specimens, also the webbing formula of fingers are I3-3II3-3 ½ III2 2/3-3 -IV (Fig.  3A); toes I2-2 -II2 + -3III2 -3 -IV3 --2V (Fig. 3B). Dutta & Manamendra-Arachchi (1996), states that the species possesses rudimentary webbing on toes but again in 2005, Manamendra-Arachchi & Pethiyagoda states that P. hypomelas, "Toes free (webbing absent)". Although the toes of the lectotype specimen is in poor condition on close observation the rudimentary webbing can still be determined (Image 7). A description of the species along with a distribution was last provided by Manamendra-Arachchi & Dutta in 1996, which states; "apart from the syntype specimens, the species has been recorded from 1800m Kandy", however the photograph of a live frog provided, ' Figure  165', was erroneous and not a specimen of P. hypomelas, but the ' Figure 164' of a syntype specimen clearly shows its dried out condition. Interestingly, one of the above authors in 2005 stated that P. hypomelas was extinct for the reason that they had never encountered the species from their 10 years of Island-wide survey. Although the type locality for the species at the BMNH is mentioned as "Ceylon" (Sri Lanka), the locality of the two specimens provided by W. Ferguson in his "Reptile Fauna of Ceylon, Letter on a collection sent to the Colombo Museum", on page 29 states under point 164, that they were collected from Le Vallon Estate, Nilambe District. Presently, this area comes under the Nuwara Eliya District. Also in this same reference (Appendix 3) Ferguson mentions that the description of Ixalus hypomelas by Dr. Günther was based on specimens sent by Col. Beddome and himself. Above all, the statements clearly have the mention about specific identity of the 'type locality', in spite of this and without making an effort to searching this species from 'type locality' the scientific community had concluded the species to be extinct.
The conservation status of this species redescribed here, should be considered as Critically Endangered under the IUCN Red List Categories and Criteria because of the extent of occurrence (EOO) is less than 100km 2 , the area of occupancy (AOO) is less than 10km 2 , and is recorded from a single location. The habitat is under severe anthropogenic activities such as over exploitation of natural resources for tea cultivation, forest fragmentation, use of agrochemicals, soil erosion, inadequately planned constructions and illegal constructions, mini-hydro power plants, forest die back, and discharge of pollutants to the environment. There seems to be an ambiguity with the type locality from the literature for P. hypomelas and our collections from a different locality could be considered as an undescribed species owing to the fact of high levels of point endemism within Sri Lanka, but we take the most parsimonious solution of fixing our collections as P. hypomelas due to its overall resemblance in morphology to stabilize the taxonomy for this species.