Re-description of Hypselobarbus lithopidos ( Teleostei : Cyprinidae ) , based on its rediscovery from the Western Ghats , India , with notes on H .

1 Flat L, Sri Balaji Apartments, 7th Main Road, Dhandeeswaram, Velachery, Chennai, Tamil Nadu 600042, India 2 Department of Fisheries Microbiology, College of Fisheries, Yekkur, Mangalore, Karnataka 575002, India 3 Department of Applied Zoology, Mangalore University, Mangalagangothri, Manglore, Karnataka 574199, India 1 jdmarcusknight@yahoo.co.in (corresponding author), 2 winrai@yahoo.com, 3 kevinroni@yahoo.com


INTRODUCTION
The genus Hypselobarbus Bleeker, 1860 has always interested ichthyologists and though there has been substantial work carried out on this genus (Mukerji 1931;Raj 1941;Jayaram 1997;Arunachalam et al. 2012;Pethiyagoda et al. 2012), the identities of individual species remain ambiguous beginning with the identity of H. mussullah (Sykes, 1839), which is the type species of Hypselobarbus.Currently, the genus includes at least 11 other species namely, Hypselobarbus curmuca (Hamilton, 1807), H. dobsoni (Day, 1876), H. dubius (Day, 1867), H. micropogon (Valenciennes, 1842), H. jerdoni (Day, 1870), H. kolus (Sykes, 1839), H. kurali (Menon & Rema Devi, 1995), H. lithopidos (Day, 1874), H. periyarensis (Raj, 1941), H. pulchellus (Day, 1870), and H. thomassi (Day, 1874) all endemic to the freshwater systems of peninsular India.The identity of certain other species such as H. mysorensis (Jerdon, 1849) is ambiguous with some considering it a valid species (Mukerji 1931) while others consider it a synonym of H. micropogon (Menon, 1999).Similarly, a few other species such as Gobio canarensis Jerdon, 1849, Barbus gracilis Jerdon, 1849, Barbus conirostris Günther, 1868 and Barbus guentheri Day, 1869 are either buried in synonymy or lost in time.In spite of the ambiguity, recent authors have even highlighted the possibility of undescribed species being concealed within this genus (Arunachalam et al. 2012).Of the known species, Hypselobarbus lithopidos is listed as Data Deficient in the IUCN Red List of Threatened Species (Raghavan & Ali 2011), H. pulchellus is listed as Critically Endangered (possibly extinct) (Devi & Ali 2011a) and H. thomassi as Critically Endangered (Devi & Ali 2011b).It is important to fill in knowledge gaps on the identity and the current status of the already known species before additional species are described under Hypselobarbus.
As there was an urgent need to do a complete taxonomic reassessment of this species, we carried out fresh surveys in the type locality (South Canara = Dakshina Kannada).During these surveys specimens of H. lithopidos which fit the original description by Day (1874) were collected.This highlighted the fact that this enigmatic fish was not extinct as thought previously (Ali & Raghavan 2011;Molur et al. 2011;Arunachalam et al. 2012).Incidentally, H. thomassi a very similar congener (type locality Canara) was also collected during these surveys.There is a certain amount of taxonomic ambiguity surrounding the identity of this species (Devi & Ali 2011b) with the population found below the Palghat/Palakkad gap being speculated as a different taxon.
In this paper, we confirm the presence of H. lithopidos in its type locality, and redescribe it thereby clearing any ambiguity that may surround its identity.The identity of H. thomassi which is the closest resembling congener is also discussed in this paper.

MATERIALS AND METHODS
The materials used in the present study are based mostly on collections from the Phalguni River, Mangalore and Kempu Hole River, a tributary of Netravathi in southern Karnataka.The specimens used in this study are registered in the Collections of the Zoological Survey of India, Southern Regional Centre, Chennai (ZSI/SRS) and the private collections of J.D. Marcus Knight (MKC).Measurements were taken using a digital caliper to the nearest 0.1mm.The standard length (SL) was measured using a foot ruler to the nearest 1.0mm.Quantification of characters follows Devi et al. (2010).Subunits of the head are also expressed in proportions of head length (HL).Numbers in parenthesis after a count denote the frequency of that count.Photographs of the syntypes of both H. lithopidos and H. thomassi from the Natural History Museum, London (BMNH), the Australian Museum, Sydney (AMS) and the Museum of Comparative Zoology, Harvard, (MCZ) were used to compare the general body shape and the lateral line scale count.All specimens used in this study were collected from the type locality (Dakshina Kannada) and are putative topotypes.
Morphometric and meristic data are given in Table 1.

Coloration
Formalin-fixed and alcohol-preserved specimens are dark grey at the back progressively becoming lighter at the abdomen with the fin edges becoming black.The outer edges of each scale has scattered pigments.In life, juveniles below the size of 50mm SL are grey with each scale having a dark outer edge.Pelvic fins are bright red in juveniles, which slowly lose colour as the fish ages (Images 2A,B & C). Adult specimens silvery grey with all fins grey, the principal rays of the caudal fin are white in colour.
Though Hypselobarbus lithopidos has been assessed as Data Deficient (Raghavan & Ali 2011) stating that there are no confirmed records of this species from its range since 1941, there have been sporadic reports of this species from the Western Ghats (David 1956;Indra & Devi 1990;Yazdani et al. 2001;Cherian et al. 2001;Vijaykumar et al. 2008;Vijaylaxmi & Vijaykumar 2011;Ahmad et al. 2011).Indra & Devi (1990) report H. lithopidos from Thekkady.Though their voucher specimens (ZSI/SRC F 1873 collected in 1975) were untraceable, the photograph of the fish provided in the paper (Fig. 2 in Indra & Devi 1990) portrayed a very deep bodied fish unlike the elongated streamlined fish that H. lithopidos is, raising doubts on the identity of the specimens they had examined.The other reports of H. lithopidos (David 1956;Vijaykumar et al. 2008;Vijaylaxmi & Vijaykumar 2011) were as a part of pollution studies or ichthyofaunal surveys and lacked the description or the voucher specimen of the fish identified as H. lithopidos thereby providing no clarity on the identity of this enigmatic barb.The recent report of this species by Ahmad et al. (2011), also needs verification as the record of this species fails to find mention in a subsequent phylogeny study of the genus by the same authors (Arunachalam et al. 2012) and has been stated to be different (M.Arunachalam pers. comm.: in Raghavan & Ali 2011).The other reports (Cherian et al. 2001;Yazdani et al. 2001) were merely based on reports prior to 1941.
Other than these unverified sporadic reports there have not been any confirmed records of this species from Kerala, Karnataka and Tamil Nadu parts of Western Ghats, although this area has been comprehensively explored (Easa & Basha 1995;Chandrashekhariah et al. 2000;Gopi 2000;Kurup et al. 2004) leading to the species being presumed extinct (Raghavan & Ali 2011) During our surveys, specimens of H. lithopidos which fit the original description of Day (1874) were collected from Phalguni River in Dakshina Kannada, highlighting the fact that this enigmatic fish is not extinct as previously thought.
Hypselobarbus lithopidos can also be distinguished from H. micropogon, H. periyarensis and H. dubius by having its last simple dorsal ray weak and articulated vs. strong osseous (Jayaram 1991).It can be distinguished from H. curmuca and H. kurali by the presence of a thin keratinized covering on the inside of the lower jaw vs. absence of the thin keratinized covering in the other two species.It can also be distinguished from the latter by the white principal rays of the caudal fin which is absent in H. curmuca and H. kurali (Talwar & Jhingran 1991).Moreover it can be further distinguished from H. curmuca by the presence of two pairs of barbels vs. one pair of barbel in the latter.H. lithopidos can be further distinguished from H. kolus by the presence of two pairs of barbels and 38-39 lateral line scales vs. one pair of barbel and 40-43 lateral line scales.H. lithopidos can also be distinguished from H. jerdoni, H. dobsoni and H. pulchellus by a higher lateral line scale count of 38-39 scales vs. 27-32 scales in the other three species (Jayaram 1991).
It is relevant to note that Jayaram (1991) has also reported H. lithopidos from Thekkady, Kerala (ZSI/SRC F 2088) with a lateral line scale count of 38-40.As we could not locate these specimens, their identity remains unclear.Evidently the range of these enigmatic barbs is quite large.Dams and hydro-electric projects with other anthropogenic factors such as over exploitation and use of destructive fishing practices could be a reason for the decline in the population of these barbs in its historic range.Wallacean shortfall also plays a part in species being presumed extinct (Knight 2010) which in this case is clear, with the record of H. lithopidos from its type locality from where it was presumed extinct.
Hypselobarbus thomassi has been reported from several drainages north and south of Palghat Gap in the Western Ghats.However, only in Netravathi and Kabini rivers (part of the Cauvery catchment in Karnataka and Kerala) north of the Palghat Gap, are the reports of this species confirmed while the southern Western Ghats populations are considered a different taxon (Devi & Ali 2011b).Though recent surveys in the two areas have only reported one specimen from Netravathi (Devi & Ali 2011b) we recorded this species in the Kempu Hole river, a tributary of the Nethravathi-Kumaradhara river systems.Hypselobarbus thomassi is consumed locally as a prized food fish and was observed to grow to more than 600mm in length and weighing more than 4kg.The adults are deep red in colour with all fins becoming red.The scales are also red with the outer edges becoming dark; the coloring is true to the name given by Day (1874) as the 'red mahseer' of Canara (local name is Kempu Pervaul = Red Mahseer) (Image 2E).
It is relevant at this point to note that Arunachalam et al. ( 2012) have illustrated a Hypselobarbus sp. from Rosemala, Kerala [=Rosemalai (also spelt as 'Rusewalai' in pg.70 and 71 of Arunachalam et al. 2012 )], which they identify as H. lithopidos in figure 4 of pg.71.However, the image of this species provided by the authors in pg.66 (fig.3A of Arunachalam et al. 2012) clearly shows a species with approximately 34 lateral line scales on the body.This species which they speculate to be H. lithopidos could very well be H. thomassi.Interestingly, there is no mention of H. thomassi in that paper.
Incidentally, we examined one specimen of H. thomassi collected below Athirapally Falls, Chalakudy River, Kerala (ZSI/SRC F 8665) (Image 2D).We could not observe any valid differences between this specimen and the H. thomassi collected in Dakshina Kannada to warrant this specimen as a different taxon (morphometrics and meristics provided in Table 1).Therefore, the contention that the southern Western Ghats population below the Palghat gap is a different taxon (Devi & Ali 2011b) needs further validation and substantiation.
Interestingly, Hypselobarbus pulchellus, another species which is listed as 'Critically Endangered' and possibly extinct (Devi & Ali 2011c), was also collected during our surveys.We collected specimens with a lateral line scale count ranging from 32-35 + 1-2 from Sita River in Dakshina Kannada (Images 3A & B).Interestingly, there is a very recent report of a 'Puntius pulchellus' in a newspaper (Shrivana 2013).However, the common name given in the report being 'Hullugende' or 'Haragi' is used for either the cultured grass carp (Ctenopharyngodon idella) or Hypselobarbus dobsoni and not for H. pulchellus, which is called 'Katladi' by the local people.Moreover, the photograph given in the report clearly shows a fish with the dorsal fin tipped with black, which is a characteristic of H. dobsoni and not H. pulchellus.As the report does not mention any voucher specimens, the identity of the fish reported as H. pulchellus needs verification.Such reports not only hamper organized taxonomic work but also add to the ambiguity that surrounds such little known fish (Raghavan et al. 2013).We propose to elucidate the identity of H. pulchellus in a subsequent paper.