Zootaxa 4059 (3): 555–568 ISSN 1175-5326 (print edition) www.mapress.com/zootaxa/ Article ZOOTAXA Copyright © 2015 Magnolia Press ISSN 1175-5334 (online edition) http://dx.doi.org/10.11646/zootaxa.4059.3.7 http://zoobank.org/urn:lsid:zoobank.org:pub:0D4F3447-2E65-4748-8DEF-92AB7876B431 Redescription of Nemachilichthys rueppelli, a senior synonym of N. shimogensis (Teleostei: )

ASHWINI KESKAR1,2,8, PRADEEP KUMKAR3,8, UNMESH KATWATE4, ANVAR ALI5, RAJEEV RAGHAVAN6,7 & NEELESH DAHANUKAR1,7,9 1Indian Institute of Science Education and Research, G1 Block, Dr. Homi Bhabha Road, Pashan, Pune 411008, India 2Department of Biodiversity, M.E.S. Abasaheb Garware College, Karve Road, Pune, Maharashtra 411004, India 3Department of Zoology, Modern College of Arts, Science and Commerce, Ganeshkhind, Pune 411053, India 4Bombay Natural History Society (BNHS), Hornbill House, Opp. Lion Gate, Shaheed Bhagat Singh Road, Mumbai, Maharashtra 400001, India 5Center for Aquatic Germplasm Conservation of Inland Aquatic Organisms, Kerala University of Fisheries and Ocean Studies (KUFOS), Kochi 682506, India 6Center for of Aquatic , Kerala University of Fisheries and Ocean Studies (KUFOS), Kochi 682506, India 7Systematics, Ecology and Conservation Laboratory, Zoo Outreach Organization, 96 Kumudham Nagar, Vilankurichi Road, Coim- batore 641035, India 8joint first authors 9Corresponding author. E-mail: [email protected]

Abstract

The hill-stream Nemachilichthys, an endemic of the Western Ghats of India, comprises two nominal species, N. rueppelli and N. shimogensis. The validity of the latter has been questioned by several authors. Here we show that there is only a marginal raw mitochondrial genetic distance (0.5% in cytochrome oxidase subunit I and 1.2% in cytochrome b) between topotypic specimens of the two nominal species. Further, although population-level morphometric variations ap- pear in a multivariate morphometric analysis, the two nominal species are morphologically similar, with apparently no significant characters separating them. We therefore consider N. shimogensis to be a junior synonym of N. rueppelli and redescribe the latter, providing further details on population variation and distribution.

Key words: Molecular distance, Discriminant Analysis, Hill stream loach, Western Ghats

Introduction

The Mongoose loach, Cobitis rupelli, was described by Sykes (1839) from the Beema river [=Bhima River] at Taimbournee [=Tembhurni] and Mota Mola [= Mula-Mutha River] near Poona [=Pune]. There are no known types for the species; however, a detailed description and illustration appear in Sykes (1841). Day (1878) established a new genus, Nemachilichthys, with N. rueppelli as its only member. The species was subsequently allocated to other genera, including Noemacheilus and (Bănărescu & Nalbant 1968; Menon 1987; Talwar & Jhingran 1991), though currently assigned to Nemachilichthys (Bănărescu & Nalbant 1995; Kottelat 2012). The correct spelling of the species-name is N. rueppelli (see Discussion, below). Narayan Rao (1920) described Nemachilichthys shimogensis from the Thunga River, Shimoga Town (Mysore), South India, a tributary of the Krishna River system, from which N. rueppelli too, was described. Narayan Rao (1920), however, did not distinguish N. shimogensis from N. rueppelli. While several authors considered N. shimogensis to be valid (Bănărescu & Nalbant 1995; Eschmeyer & Fricke 2015), others treated it as a junior synonym of N. rueppelli (Menon 1987; Talwar & Jhingran 1991). For his part, Kottelat (2012) noted: “Nemachilichthys shimogensis has long been listed as a synonym of N. ruppelli. It is treated as distinct by Bănărescu & Nalbant (1995: 448) but no proper comparison has been published. The figures in the respective original descriptions show some differences, but are not really conclusive since several figures in Sykes (1841) do not seem totally accurate. This can be solved only by examination of topotypes of both nominal species”.

Accepted by R. Pethiyagoda: 12 Nov. 2015; published: 23 Dec. 2015 555 In order to shed light on this problem, we herein examine topotypes of both nominal species using molecular and morphological analyses.

Methodology

Study site and sampling. Specimens of Nemachilichthys were collected from six localities, all in the upper reaches of the Krishna River system (Fig. 1). Topotypes of N. ruppelli were collected from Mula-Mutha River at Yerawada (18.540°N, 73.880°E, 556m a.s.l.), Pune, Maharashtra, and those of N. shimogensis from Tunga River at Shimoga (13.901°N, 75.564°E, 567m a.s.l.), Karnataka.

FIGURE 1. Sampling localities for Nemachilichthys rueppelli in the northern region of the Western Ghats. Locations: 1, Mula- Mutha River at Yerawada; 2, Nira River at Bhor; 3, Krishna River at Wai; 4, Koyna River at Patan; 5, Hiranyakeshi River at Ajara; and 6, Tunga River at Shimoga.

Voucher specimens and museum abbreviations. Voucher specimens examined in this paper are deposited in the collections of the Bombay Natural History Society (BNHS), Mumbai; the Wildlife Information Liaison Development (WILD) Society, Coimbatore; and the Western Regional Center of the Zoological Survey of India (ZSI-WRC), Pune. Syntypes of N. shimogensis were studied from the Natural History Museum (BMNH), London. The specimen of N. rueppelli from the Australian Museum, Sydney (AMS), New South Wales, Australia, listed as a syntype in Eschmeyer & Fricke (2015), was studied from photographs. Morphology, morphometry and osteology. Measurements were taken point-to-point to the nearest 0.1mm using dial calipers (Fig. 2). Subunits of the body are presented as percent of standard length (SL), and subunits of the head as percent of head length (HL). Values in parentheses after a count represent the frequency of that count. One specimen (BNHS FWF 189) was cleared and stained following the procedure of Potthoff (1984). Molecular analysis. Gills were harvested from eight fresh specimens (WILD-14-PIS-118, 119, 120, 122, 123, 215, 216, 225) and preserved in absolute ethanol. DNA extraction, PCR amplification for the cytochrome b (cytb) and cytochrome oxidase subunit I (COI) gene sequences and sequencing protocols follow Katwate et al. (2013) and Ali et al. (2013), respectively. Sequences were checked using the BLAST tool (Altschul et al. 1990) to investigate the affinity of the species to other known sequences in the NCBI database. Sequences generated as part of the study are deposited in GenBank under accession numbers KT380866–73 for COI and KT380874–81 for cytb. Gene sequences were aligned using MUSCLE (Edgar 2004). Raw (p) distances between pairs of sequences were calculated in MEGA 6 (Tamura et al. 2013). Statistical analysis of morphometric data. Statistical analysis of the morphometric data was performed on size-adjusted measurements of subunits of the body as proportions of standard length and subunits of head as proportions of head length. The null hypothesis that the data are multivariate-normal was checked using Doornik &

556 · Zootaxa 4059 (3) © 2015 Magnolia Press KESKAR ET AL. Hansen (2008) omnibus. Multivariate Analysis of Variance (MANOVA) was performed to understand whether the populations of the species formed significantly different clusters (Huberty & Olejnik 2006) using Pillay’s trace statistic (Harris 2001). Mahalanobis distances (Harris 2001) between pairs of individuals were calculated and were used for computing Fisher’s distances (distance between the centroids of the clusters, divided by the sum of their standard deviations) between two clusters to check if the clusters were significantly different from each other. Statistical analysis was performed in PAST 3.0 (Hammer et al. 2001).

FIGURE 2. Illustration showing the lengths used for morphometric analysis. (A) lateral view and (B) dorsal view. Abbreviations: TL, Total length; SL, Standard length; HL, Lateral head length; PDL, Predorsal length; DCL, Dorsal-fin origin to caudal fin base; PPL, Pre-pectoral length; PVL, Pre-pelvic length; PAL, Pre-anus length; PAnL, Pre-anal length; BD(D), Body depth (at dorsal-fin origin); BD(A), Body depth (at anus); BW(D), Body width (at dorsal-fin origin); BW(A), Body width (at anus); HDF, Height of dorsal fin; BDF, Length of dorsal-fin base; PFL, Length of pectoral fin; PVL, Length of pelvic fin; AFL, Depth of anal fin; AFB, Length of anal fin base; CFL, Length of caudal fin; CPD, Depth of caudal peduncle; CPL, Length of caudal peduncle; DHL, Dorsal head length; HD(E) Head depth (at eye); HD(N), Head depth (at nape); HW(N), Head width (at nares); HW, Maximum head width; SNL, Snout length; ED, Eye diameter; and IOD, Interorbital width.

Geographical separation and population variation. All the populations studied are from the Krishna River system. We determined geographical separation using two methods. Approximate 'geographical distance' (in km) between populations was calculated by measuring the distance between the populations through current river channels connecting them. This creates an interesting observation that although the populations of Nemachilichthys in the Krishna River is just south of Nira River (Fig. 1), the geographical distribution between the two is large as the two rivers converge several kilometers downstream. Since fish phylogeographic patterns may reflect ancient, rather than modern river drainage connections (Burridge et al. 2006; Waters et al. 2001), it is possible that the original river connections may have been different from those seen today. Assuming that the populations that are

REDESCRIPTION OF NEMACHILICHTHYS RUEPPELLI Zootaxa 4059 (3) © 2015 Magnolia Press · 557 geographically closer were connected together in the past, we used approximate 'linear distances' (in km) between populations as another measure for geographical separation. We determined the relationship between the geographical separation and distance between population based on genetic (raw distances in COI and cytb gene sequences) and morphometric (Mahalanobis distances between centroids) analyses using Pearson's correlation coefficient.

Results

Identification of the two nominal species. Eschmeyer & Fricke (2015) mention AMS B.7528 as a syntype of Nemachilichthys rueppelli. Kottelat (2012), while mentioning that there is apparently no known type material, allowed that AMS B.7528 may be a doubtful syntype. Given, however, that the specimen AMS B.7528 (Fig. 3K) was collected from Poonah (= Pune) by Francis Day only in 1865, it cannot be a type specimen of N. rueppelli. We consider there to be no known types of Nemachilichthys rueppelli: the species is known only from the description and illustration provided by Sykes (1839, 1841). Loach species with a pointed snout and the local name 'Mooreh' as mentioned by Sykes (1841), collected from the type locality of N. rueppelli, closely match the original description. The only character that does not match is the presence of small scales on the body, whereas Sykes (1841) noted 'body naked'. These small scales were evidently overlooked by Sykes (1841), for he described Cobitis mooreh (= mooreh) and C. maya (=Lepidocephalichthys guntea) too, as lacking scales, although both possess minute scales throughout the body (Menon 1987; Talwar & Jhingran 1991). No other species of loach with a pointed snout occurring in the type locality (surveyed by the last author over the past 17 years) has a banded pattern that matches that illustrated in Sykes’s (1841) figure (see Fig. 3A). We observed, however, that the banded pattern varies with age: some specimens have triangular bands merging into one another near the lateral line (see Fig. 3E and F). We thus treat specimens collected from the Mula-Mutha River as topotypes of N. rueppelli. Specimens collected from the Tunga River at Shimoga match the description of N. shimogensis and are clearly conspecific with N. rueppelli, based on the syntypes BMNH 1919.11.19.13 and BMNH 1919.11.19.15. We thus treat the specimens collected from the Tunga River as topotypes of N. shimogensis. Molecular analysis. The pairwise raw genetic distances between topotypic N. rueppelli and N. shimogensis were 0.5% in COI and 1.2% in cytb (Table 1). Raw genetic distance increased as the geographical separation between the populations increased, when geographical separation was calculated as a linear distance between the populations (Fig. 4B and D).

TABLE 1. Pairwise percent raw genetic distances among six populations of Nemachilichthys rueppelli for the COI (above the diagonal) and cytb (below diagonal) gene partial sequences. Voucher number River (1) (2) (3) (4) (5) (6) (7) (8) WILD-15-PIS-225 Mula-Mutha (1) 0.0 0.2 0.2 0.3 0.3 0.3 0.5 WILD-14-PIS-122 Nira (2) 0.0 0.2 0.2 0.3 0.3 0.3 0.5 WILD-14-PIS-123 Nira (3) 0.0 0.0 0.0 0.2 0.2 0.3 0.3 WILD-15-PIS-215 Krishna river (4) 0.0 0.0 0.0 0.2 0.2 0.3 0.3 WILD-15-PIS-216 Krishna river (5) 0.1 0.1 0.1 0.1 0.0 0.5 0.5 WILD-14-PIS-120 Koyna (6) 0.3 0.3 0.3 0.3 0.1 0.5 0.5 WILD-14-PIS-119 Hiranyakeshi (7) 1.1 1.1 1.1 1.1 1.2 1.3 0.7 WILD-14-PIS-118 Tunga (8) 1.2 1.2 1.2 1.2 1.3 1.5 2.3

Morphological analysis. Morphometric data for the populations are provided in Table 2. All morphometric characters overlapped among the various populations. Morphometric characters of topotypic N. rueppelli from Mula-Mutha River and topotypic N. shimogensis from Tunga River overlapped strongly, indicating that the two nominal species cannot be distinguished based on univariate morphometric analysis.

558 · Zootaxa 4059 (3) © 2015 Magnolia Press KESKAR ET AL. 9 1 .9 .7 .0 .6 .0 .3 .5 .1 .6 .0 .9 .6 .7 2.2 21.6 19.5 15.6 49.7 54.7 37.2 –89.2 .4–110.3 , WILD-14-PIS-120–121 (2), WILD-15-PIS- (2), , WILD-14-PIS-120–121 -15-PIS-222–224 (3), ZSI-WRC P/4494 (4); ZSI-WRC (3), -15-PIS-222–224 (1); Tunga River: CRG-SAC.2014.03.190.1–3 CRG-SAC.2014.03.190.1–3 River: Tunga (1); . Character abbreviations as in Figure 2. Specimens used for morphometry - Mula-Mutha River: BNHS FWF FWF BNHS River: - Mula-Mutha morphometry for used Specimens 2. Figure in as abbreviations . Character ver, Bhor: BNHS FWF 175–177 (3), WILD-14-PIS-122–123 (2), WILD (2), WILD-14-PIS-122–123 (3), 175–177 FWF BNHS Bhor: ver, Nemachilichthys rueppelli Nemachilichthys Mula-Mutha River (n = 10) Nira River (n = 12) Krishna River (n = 5) Koyna River (n = 13) Hiranyakeshi River (n = 4) Tunga River (n = 4) All populations (n = 48) Mean (sd) Range Mean (sd) Range Mean (sd) Range Mean (sd) Range Mean (sd) Range Mean (sd) Range Mean (sd) Range Morphometric data for six populations of of populations six for data Morphometric Character TL, mmSL, mmHL, mm 93.1 (2.9)%SL 74.8 (2.6)HL 18.3 (0.6) 87.4–98.0PDLDCL 69.5–79.0 76.4 (9.2) 17.0–19.1PPL 61.4 24.5 (7.5) (0.5)PVL 61.8–88.5 50.4 (0.9) 16.1 (1.6)PAL 50.7 49.8–70.5 (1.0) 70.6 (9.0)PAnL 13.3–18.4 25.7 24.0–25.6 (0.9) 56.1 49.2–52.2BD(D) (6.6) 55.4 (1.6) 60.9–84.8 14.4 49.0–52.3 (1.6) 26.3BD(A) (0.8) 66.1 (1.6) 50.0–66.9 50.5 (1.5) 80.2 (1.3)BW(D) 81.4 24.5–27.4 (7.6) 13.0–16.8 50.4 16.1 25.2–27.9 (0.9) (1.1) 52.2–57.6BW(A) 66.3 48.0–53.0 (6.5) 13.5 26.4 63.5–93.5 (0.9) (1.4) 62.7–68.2 16.8HDF 49.3–51.4 (1.7) 25.7 12.3 (0.7) (1.1) 54.0 78.0–82.2 (1.3) 51.0–76.7 50.4 (1.0)BDF 14.8–18.2 90.7 23.7–29.0 10.7 (4.1) (1.0) 67.0 13.2–19.8 51.6 (1.2) 25.0–26.6 (1.0)PFL 79.9 52.3–56.1 12.5–15.6 73.3 (1.1) 49.5–51.9 (3.7) 15.9 27.2 (1.2) 10.4–14.6VFL (0.6) 21.7 64.3–69.0 18.2 50.3–52.7 (0.9) (0.9) 25.4 86.2–96.1 (0.9) 78.1–81.0 54.2 12.6 (1.2) 50.2 (0.9)AFL (1.1) 20.7 (0.8) 13.7–17.6 9.4–12.7 26.2–27.7 12.5 69.1–77.8 66.7 (1.2) 50.2 (1.4)AFB 73.0 24.4–27.5 (0.9) (25.6) 78.3 19.0 52.5–55.8 11.5–14.7 17.2–19.2 (1.5) (0.5) 48.3–52.0 20.3–23.1 14.1 10.4CFL 59.1 25.7 (0.6) (1.7) 16.0 (20.8) (1.2) 65.3–68.2 (0.9) 48.5–52.1 24.8 9.7–13.9 54.4–110.3 (0.2) 19.4–21.7 76.6–80.6 53.2CPD 12.1 (1.5) 49.9 (1.1) 16.4 15.3 (0.6) (0.6) 21.9 (4.6) (1.5) 13.3–14.9 24.1–28.7 65.2CPL 6.0–11.7 18.2–20.1 43.6–89.2 81.6 51.2 (1.9) 10.6 (12.1) (0.2) (0.6) 19.2 80.0 6.5 50.3–55.6 (1.2) 10.5–13.4 (1.1) (0.6) 15.1–17.5 24.6–25.1%HL 20.0–25.8 16.7 25.1 24.3 11.2–21.6 (1.8) (0.9) (0.5) 49.1–50.4 61.6–68.4 20.1 54 (0.9) 15.5–17.4 65.8DHL 9.2 17.3–21.1 (9.9) 53.7 78.1–81.8 (0.7) 13.5 9.6–11.1 (0.9) (0.9) 50.9–51.3 17.2 9.4 22.1 (0.9) (0.3) 26.2 (0.9) 13.4–19.5 (1.6)HD(E) 66.7 16.7 14.4 18.9–21.9 50.6 (2.0) (2.1) (0.6) 16.9 (3.4) 24.0–25.0 19.4 (1.0) 80.1 13.6 (1.1) 11.9–14.9 (1.7) 23.6–26.3 43.6 HD(N) (1.9) 5.5–7.2 8.5–10.2 15.6–18.5 51.1 21.2–23.1 17.4 (1.5) 52.6–54.4 (0.7) 24.3–27.8 20.6HW(N) (0.8) 94.8 15.8–19.1 (3.5) 11.2– 18.2–20.6 47.0–54.9 14.0 25.7 64.1–68.9 43.5 24.9 (0.5) (1.3) 11.3 (2.6) 17.7 (0.9) (1.9) 20.6 9.8–16.8 (0.5) 13.1–15.2HW (2.3) 7.4 8.8–9.9 77.7–81.4 (1.0) 49.4–52.7 51.4 48.6 19.8–21.7 (2.6) (3.0) 25.5 17.2 16.3–17.9 (1.0) 19.6 (1.2)SNL (1.2) 23.6–26.2 50.3 31.6 63.5 17.0–18.3 (1.4) (2.2) 14.2 23.9–26.7 16.9–25.1 (2.5) 14.5 (1.0) 7.1–13.0 (1.2) 88.4–99.9 13.6–14.5 79.4ED 9.2 19.3 (2.4) 50.7 40.5–48.3 (0.3) (0.7) (1.0) 15.9–18.9 47.8–53.4 6.3–9.2 16.8–21.8 15.5 24.0–27. 27.3 (1.8) 16.4 (0.7)IOD 20.1 (0.9) 43.6–53.1 12.4–16.0 47.0–54 (1.9) 12.1 91.9 60.1–65.5 25.9 48.9 (0.5) (1.9) (1.3) (2.7) 13.3 18.2–20.5 12.9–15.1 (1.2) 42.4 15.9 28.5–35.2 76.5–81.8 (1.8) 20.4 (0.7) 48.5–52 51.2 53.9 (1.0) (2.6) 26.9–28.6 (1.8) 7.8 8.6–9.6 14.8–17.6 13.6–17.4 (1.5) 47.9 13.7 (2.5) (0.6) 88.7–95.6 66.0 19.0 23.7–29 17.7–21.7 (1.9) 39.1–45.1 (0.9) 32.6 14.6–17.0 11.8 11.4–12.5 12.2–14.8 13.3 (2.8) (1.3) 79.8 (2.1) 23.9 (1.4) 44.8–53.4 16.4 (1.3) 47.8–57 19.4 43.3–51.9 (0.7) 12.7–14.3 19.0–21.3 (1.7) 16.1 9.0 95.4 6.6–10.4 (1.5) (0.4) (2.9) 48.2–56.6 40.3 15.7 27.2–35.6 (1.8) 60.1–69 (0.4) 13.1 21.2–26.0 18.0–19.9 10.4–13.6 51.1 (1.0) 9.9 76.5–8 22 (3.4) 42.6 14.0 (1.0) (1.8) (2.2) (1.0) 91.5–98.4 21.4 10.2–16.9 (1.3) 48.8 15.5–17.0 13.3– 6.8 38.6–43.1 (2.3) 31.9 (0.8) (2.0) 8.7–9.5 16.8–22.4 23.9 44.9–58.6 (1.2) 21.8 15.3–16.2 12.6 10.5– 39.5–45.0 12.1–15.7 (1.7) (1.7) 92.9 20.5–22.8 17.2 (3.2) 45.5–52.0 (1.4) 44.2 20.6–23.3 7.2–11.4 18.9 30.2–34.7 (2.9) (1.2) 19.7 (1.9) 49.3 9.3 5.1–8.2 (3.0) 49.3 (0.4) 14.1 17.2 (2.8) (0.4) 88.3–98.2 19.3–23.3 (0.7) 22.6–25.2 15.1–19.8 49.3 21.4 (1.7) 38.8–49.7 (1.7) 9.6–16.8 33.0 (2.4) 19.9 10.7 16.4–22.9 17.6–20.5 (1.3) (1.6) 45.4–52.3 43.7–53.2 6.8 90.4 16.4 (0.8) (1.4) 47.3–51.6 (3.5) 16.2–17.7 24.9 19.7 8.6–9.9 42.8 13.8–14.7 (1.2) (1.5) 28.9–37.2 (1.4) 17.8 16.9–25.8 (1.3) 52.6 16.8–23 16.9 (3.0) 51.2 (1.2) (1.5) 6.0–13.0 12.3–20.6 52.0 (2.1) 31.9 16.5 89.4–92.4 14.6 (1.7) 23.2–26.6 (0.9) 9.5 16.4–19.8 (0.9) (0.2) 48.4–57.8 41.7–44.6 18–23.3 6.2–7.9 14.8–20 17.9 49.4–55.2 (2.3) 49.2–52.4 22.7 95.5 (2.4) 14.6–19 24.8 51.8 (3.7) 13.9–15.9 (1.4) 29.9–34.0 (0.9) 41.5 (2.4) 13.9–20.9 51.4 (2.4) 7.9 51.2 9.3–9.8 (1.0) 18.2–26.7 (3.5) 91.4–99.5 14.2 21.2–28 32.2 (0.9) (2.8) 15.1 39.2–44.7 50.6–52.6 (2.8) 21.7 (2.7) 47.9–53.0 46.9–54.7 9.2 (0.4) 93.3 6.7–8.9 (3.2) 12.1–16 28.6–35.4 52.4 42.9 (4.9) (2.6) 11.9–18.4 54.4 48.4 (2.3) 19.6–25.5 (3.4) 88.3–99 32.3 (2.4) 7.1 47.9–59.1 38.6– 8.8–9.6 (1.0) 20.1 (7.4) 51.2–56.1 18.0 39.5– (1.9) 27.2– 51.0 (3.3) 9.3 5.1–10.4 (0.4) 9.8–26.2 50.9 (2.7) 15.5–19.9 44.8–59. 16.6 45.5–56 (3.4) 20.3 (2.6) 8.6–9.9 15.5–26 9.8–26.2 TABLE 2. 178–180 (3), WILD-15-PIS-225–227 (3), ZSI-WRC P/4495 (4); Nira Ri Nira (4); P/4495 ZSI-WRC (3), WILD-15-PIS-225–227 (3), 178–180 217–220 (4), ZSI-WRC P/4492 (3); Hiranyakeshi River: BNHS FWF 174 (1), WILD-14-PIS-119 (1), WILD-15-PIS-221 (1), ZSI-WRC P/4493 ZSI-WRC (1), WILD-15-PIS-221 (1), WILD-14-PIS-119 (1), 174 FWF BNHS River: Hiranyakeshi (3); P/4492 ZSI-WRC (4), 217–220 Krishna River, Wai: BNHS FWF 168–169 (2), WILD-15-PIS-215–216 (2), ZSI-WRC P/4491 (1); Koyna River, Patan: BNHS FWF 170–173 (4) 170–173 FWF BNHS Patan: River, Koyna (1); P/4491 ZSI-WRC (2), WILD-15-PIS-215–216 (2), 168–169 FWF BNHS Wai: River, Krishna (1). (3), WILD-14-PIS-118

REDESCRIPTION OF NEMACHILICHTHYS RUEPPELLI Zootaxa 4059 (3) © 2015 Magnolia Press · 559 FIGURE 3. Nemachilichthys rueppelli. (A) Illustration of Cobitis rupelli from Sykes (1841); (B) illustration of Nemachilichthys shimogensis from Narayan Rao (1920); (C) WILD-15-PIS-226 (Mula-Mutha River, 94.1 mm SL); (D) WILD- 15-PIS-224 (Nira River, 55.9 mm SL); (E) WILD-15-PIS-215 (Krishna River, 60.9 mm SL); (F) WILD-15-PIS-217 (Koyna River, 61.3 mm SL); (G) WILD-15-PIS-221 (Hirnyakeshi River, 77.8 mm SL); (H) CRG-SAC.2014.03.190.2 (Tunga River, 89.2 mm SL); (I) BMNH 1919.11.19.13 (Tunga River, syntype of N. shimogensis); (J) BMNH 1919.11.19.15 (Tunga River, syntype of N. shimogensis) and (K) ASM B.7528 (Poonah). Illustration in Fig. 3(A) obtained from the Biodiversity Heritage Library (www.biodiversitylibrary.org), digitized by Natural History Museum Library, London (Sykes, W.H., 1841, On the fishes of the Dukhun, Transactions of the Zoological Society of London, 2, 349–378, Pls. 60–67.). Illustration in Fig. 3 (B) reprinted by permission of the publisher (C. R. Narayan Rao, 1920, II.—Some new species of Cyprinoid fish from Mysore, Annals and Magazine of Natural History, 6, 45–64, publisher Taylor & Francis Ltd, www.tandfonline.com).

The data were multivariate normal (within-group Doornik and Hansen omnibus: Ep = 50.03, p = 0.699). The null hypothesis that the mean vectors of all the six populations are equal was rejected in MANOVA (Pillai's trace =

3.775, F95,140 = 2.091, p < 0.0001) indicating that at least some of the populations formed significantly different clusters. Discriminant analysis (DA) of the data is illustrated in Fig. 4A and factor loadings are shown in Table 3. DA extracted five axes (Fig. 5B) the first two of which explained 71% of the total variability in the data. Pairwise comparison (Fig. 5C) of the clusters suggests the population from Mula-Mutha and Tunga had the highest inter- cluster distance and the two clusters were significantly different from each other (F = 5.081, p = 0.001). There was

560 · Zootaxa 4059 (3) © 2015 Magnolia Press KESKAR ET AL. a significant positive relationship between the morphological distances and geographical separation among the populations when geographical separation was calculated as a linear distance between the populations (Fig. 4F).

TABLE 3. Factor loadings for discriminant analysis. Variable abbreviations as in Figure 2. Variable Axis 1 Axis 2 Axis 3 Axis 4 Axis 5 HL 0.09 -0.21 -0.15 0.13 -0.03 DHL -0.20 -0.30 0.46 -0.15 0.99 HD(E) 0.08 0.49 -0.03 0.12 0.49 HD(N) 0.45 0.52 0.49 0.48 -0.24 HW(N) 0.11 0.02 -0.18 0.02 0.20 HW 0.42 0.07 -0.17 -0.16 0.13 SNL 0.26 0.15 0.59 -0.30 0.65 ED 0.48 -0.43 -0.24 -0.10 0.50 IOD 0.25 0.59 -0.32 -0.31 0.29 PDL -0.02 -0.06 0.01 0.08 0.05 DCL -0.05 -0.10 0.10 -0.17 -0.18 PPL -0.09 -0.22 -0.15 0.04 0.24 PVL -0.29 0.17 -0.05 0.18 -0.12 PAL -0.18 0.03 -0.28 0.12 -0.61 PAnL 0.04 0.20 0.03 0.21 -0.01 BD(D) 0.12 0.33 0.00 0.07 -0.13 BD(A) 0.05 0.19 0.14 -0.04 -0.00 BW(D) 0.17 0.36 -0.08 -0.03 -0.13 BW(A) 0.09 0.30 -0.02 -0.04 -0.14 HDF -0.10 -0.23 0.04 0.24 0.06 BDF 0.01 0.03 0.35 0.04 -0.00 PFL 0.08 -0.35 0.05 0.05 0.01 VFL 0.08 -0.34 0.05 0.01 -0.10 AFL -0.04 -0.21 0.01 0.13 -0.03 AFB 0.03 -0.21 -0.03 -0.01 -0.11 CFL -0.20 -0.32 0.01 -0.14 -0.02 CPD 0.00 0.06 0.03 0.00 -0.03 CPL 0.02 -0.01 0.03 0.19 -0.05

Taxonomy

Nemachilichthys rueppelli (Sykes, 1839)

Cobitis rupelli Sykes, 1839: Sykes (1839, p. 162), Sykes (1841, p. 366, Pl. 64, fig. 1) Nemachilichthys ruppelli: Day (1878, p. 611, Pl. 155, fig. 7), Bănărescu & Nalbant (1995, fig. 19A, B) Kottelat (2012, p. 93, fig. 10.22.1) Nemacheilus ruppelli: Talwar and Jhingran (1991, p. 499) Noemacheilus ruppelli: Bănărescu & Nalbant (1968, p. 329) Noemacheilus (Noemacheilichthys) ruppelli: Menon (1987, p. 158, Pl. 12, fig. 1) Nemachilichthys rueppelli: Eschmeyer & Fricke (2015) Nemachilichthys shimogensis Narayan Rao, 1920: Narayan Rao (1920, p. 62, Pl. 2, fig.5, 5a, 5b)

REDESCRIPTION OF NEMACHILICHTHYS RUEPPELLI Zootaxa 4059 (3) © 2015 Magnolia Press · 561 FIGURE 4. Genetic and morphometric distances versus geographical and linear distance between the sampled populations of Nemachilichthys rueppelli. (A, C and E) raw percent genetic distance in COI, cytb and morphometric Mahalanobis distance between centroids versus the geographical distances between the populations and (B, D and F) raw percent genetic distance in COI, cytb and morphometric Mahalanobis distance between centroids versus the linear distances between the populations. Note that correlation coefficients are significant in B, D and F, even after Bonferroni correction, but not in A, C and E.

Material examined. Krishna River System: 3 ex., BNHS FWF 178–180, 74.2–79.0mm SL, Mula-Mutha River, Yerawada (18.540°N, 73.880°E, 553m ASL), coll. N. Dahanukar and M. Paingankar, 6 October 2012 (topotypes of N. rueppelli); 3 ex., WILD-15-PIS-225–227, 73.8–77.5mm SL, Mula-Mutha River, Yerawada (18.540°N, 73.880°E, 553m ASL), coll. N. Dahanukar and M. Paingankar, 6 October 2012 (topotypes of N. rueppelli); 4 ex., ZSI-WRC P/4495, 69.5–77.1mm SL, Mula-Mutha River, Yerawada (18.540°N, 73.880°E, 553m ASL), coll. N. Dahanukar, M. Paingankar, 6 October 2012 (topotypes of N. rueppelli); 1 ex., AMS B.7528, Poonah (18° 34' N, 73° 58' E), coll. F. Day, 1865; 2 ex., BNHS FWF 168–169, 53.8–66.9mm SL, Krishna River, Wai (17.955°N, 73.880°E, 707m ASL), coll. N. Dahanukar and M. Paingankar, 6 October 2012; 4 ex., BNHS FWF 170–173, 84.7- 93.5mm SL, Koyna River, Patan (17.368°N, 73.903°E, 571m ASL), coll. P. Kumkar and S. Gosavi, 11 October

562 · Zootaxa 4059 (3) © 2015 Magnolia Press KESKAR ET AL. 2013; 1ex., BNHS FWF 174, 74.2 mm SL, Hiranyakeshi River, Ajara (16.130°N, 74.210°E, 647m ASL), coll. P. Kumkar, 22 December 2013; 3 ex., BNHS FWF 175–177, 67.9–70.5mm SL, Nira River, Bhor (18.154°N, 73.840°E, 598m ASL), coll. A. Keskar and P. Kumkar, 31 October 2013; 2ex., WILD-15-PIS-215–216, 50.0– 57.4mm SL, Krishna River, Wai (17.955°N, 73.880°E, 707m ASL), coll. N. Dahanukar and M. Paingankar, 6 October 2012; 6 ex., WILD-14-PIS-120–121 and WILD-15-PIS-217–220, 51.0–65.0mm SL, Koyna River, Patan (17.368°N, 73.903°E, 571m ASL), coll. P. Kumkar and S. Gosavi, 11 October 2013; 2 ex., WILD-14-PIS-119 and WILD-15-PIS-221, 72.1–77.8mm SL, Hiranyakeshi River, Ajara (16.130°N, 74.210°E, 647m ASL), coll. P. Kumkar, 22 December 2013; 5 ex., WILD-14-PIS-122–123 and WILD-15-PIS-222–224, 56.0–68.9mm SL, Nira River, Bhor (18.154°N, 73.840°E, 598m ASL), coll. A. Keskar and P. Kumkar, 31 October 2013; 1 ex., ZSI-WRC P/4491, 52.6mm SL, Krishna River, Wai (17.955°N, 73.880°E, 707m ASL), coll. N. Dahanukar and M. Paingankar, 6 October 2012; 3 ex., ZSI-WRC P/4492, 65.1–67.6mm SL, Koyna River, Patan (17.368°N, 73.903°E, 571m ASL), coll. P. Kumkar and S. Gosavi, 11 October 2013; 1ex., ZSI-WRC P/4493, 69.1 mm SL, Hiranyakeshi River, Ajara (16.130°N, 74.210°E, 647m ASL), coll. P. Kumkar, 22 December 2013; 4 ex., ZSI-WRC P/4494, 49.8–58.6mm SL, Nira River, Bhor (18.154°N, 73.840°E, 598m ASL), coll. A. Keskar, P. Kumkar, 31 October 2013; 3 ex., CRG-SAC.2014.03.190.1–3, 43.6–89.2mm SL, Tunga River, Shimoga (13.901°N, 75.564°E, 569m ASL), coll. A. Ali, R. Britz and N. Sood, 1 March 2014 (topotypes of N. shimogensis); 1 ex., WILD-14-PIS-118, 47.3mm SL, Tunga River, Shimoga (13.901°N, 75.564°E, 569m ASL), coll. A. Ali, R. Britz and N. Sood, 1 March 2014 (topotype of N. shimogensis); 2 ex., BMNH 1919.11.19.13 and BMNH 1919.11.19.15, Tunga River, Shimoga, coll. Narayan Rao (syntypes of N. shimogensis), 1919. Description. General appearance as in Fig. 6. Mouth structure as in Fig. 7. Variation in body form and coloration as in Fig. 3. Morphometric data, separately for the various populations and pooled, provided in Table 2. Body elongate, sub-cylindrical, slightly compressed laterally; dorsal profile convex; ventral profile straight to slightly convex. Body deepest at dorsal-fin origin. Head large, about a quarter of SL. Eyes large, positioned dorsolaterally, in posterior half of head, closer to opercular margin than to tip of snout. Interorbital width slightly greater than eye diameter. Nares positioned dorsolaterally, slightly closer to anterior border of eye than to tip of snout, with large nasal flap. Barbels in three pairs, long, inner rostral shorter than outer rostral, outer rostral not reaching anterior border of eye, maxillary shorter than or equal to outer rostral, extending to perpendicular from anterior border of eye. Lips fleshy, separated by a deep groove, joined at isthmus (Fig. 7). Upper lip with a median groove, with 2 to 4 large papillae on either side of groove. Lower lip with deep median groove, 2 to 4 large papillae on either side of groove. Body with minute scales throughout, except on head and ventral surface from belly to anal fin base. Lateral line complete. Caudal peduncle long, its length 1.3 to 1.8 times its depth. Dorsal fin origin almost midway between tip of snout and caudal-fin base, with 3 (38) or 4 (10) unbranched rays and 10 (48) branched rays. Pectoral fin with 1 (48) unbranched and 10 (2), 11 (17) or 12 (29) branched rays, not reaching pelvic-fin origin. Pelvic fin originating slightly posterior to the vertical from dorsal-fin origin, with 1 (48) unbranched and 7 (48) branched rays, extending beyond anus, not reaching anal fin. Anal fin with 3 (48) unbranched and 5 (48) branched rays. Anus closer to ventral-fin base than to anal-fin origin. Caudal fin forked, with 10 (48) dorsal and 9 (48) ventral principle rays. Vertebral column with 22 abdominal vertebrae (including 4 vertebrae in the Weberian apparatus); 12 caudal vertebrae (including compound centrum); total vertebrae 34 (c&s specimen BNHS FWF 189). Coloration. In life, body beige to pale yellow, with brown bars covering dorsal surface, extending ventrolaterally below the lateral line, not reaching ventral surface; brown bars irregular, 12–20 in number. A rosette of large, brown spots on dorsal surface of head. Cheeks uniform pale brown, turning pale yellow to beige ventrolaterally. Ventral surface uniform beige to pale yellow throughout. Dorsal fin with three or four rows of brown spots, present only on fin rays; dorsal fin membrane hyaline. Caudal fin with 4–6 posteriorly-directed 'V'- shaped brown bands; a distinct black spot on the middle of the caudal-fin base. Pectoral, pelvic and anal fins hyaline, often with two rows of brown bands, sometimes faint or absent. Barbels beige, turning orange to red during breeding season from June to September. Body suffused faint-orange during breeding season, along with pectoral and pelvic fins; anal and caudal fins yellow. Rarely, snout becoming red during breeding season. In alcohol preservation, coloration similar to that in life, but more faded. Banding pattern varying with age and size (see examples in Fig. 3). Distribution. Nemachilichthys rueppelli is currently known only from the upper reaches of the east-flowing Krishna River system in the Western Ghats. In particular, the species is reported from seven rivers, Indrayani

REDESCRIPTION OF NEMACHILICHTHYS RUEPPELLI Zootaxa 4059 (3) © 2015 Magnolia Press · 563 (Dahanukar et al. 2012), Mula-Mutha, Nira, Krishna, Koyna, Hiranyakeshi and Tunga (Fig. 1). Bhat (2004) reported the species from the west-flowing Aghanashini River in the central region of the Western Ghats; this record is not supported by voucher specimens and has been omitted from Fig.1. Similarly, reports of the species from three other west-flowing rivers, viz. Amba, Kundalika and Savitri (Katwate et al. 2012), are disregarded here as those were clearly based on misidentifications of species.

FIGURE 5. Discriminant analysis depicting the population variation in six populations of Nemachilichthys rueppelli. (a) scatter diagram of discriminant analysis, (b) scree plot and (c) Fisher's distances between clusters (blue cells) and associated p values (red cells). In (c), values of p in bold are significant after Bonferroni correction.

Habitat. Nemachilichthys rueppelli inhabits moderate to fast-flowing streams and rivers. It is often found associated with aquatic vegetation and the submerged roots of riparian vegetation. The species is usually found on the stream or river bed. Preferred substrates include mud, silt, pebbles, small rocks and boulders. Syntopic species include Cyprinidae: Garra mullya (Sykes), Devario aequipinnatus (McClelland), Rasbora daniconius (Hamilton) and Pethia sp.; Nemacheilidae: Annandale, Paracanthocobitis mooreh (Sykes); Cobitidae: Lepidocephalichthys thermalis (Valenciennes), Gobiidae: Glossogobius giuris (Hamilton) and Mastacembelidae: Mastacembelus armatus (Lacepède).

564 · Zootaxa 4059 (3) © 2015 Magnolia Press KESKAR ET AL. Ecology. Gut content revealed insect parts with higher frequency, followed by ostracods, with relatively low frequency of plant matter and diatoms (based on 12 unregistered unsexed specimens from Mula-Mutha River at Yerawada, Koyna River at Patan and Nira River at Bhor).

FIGURE 6. Nemachilichthys rueppelli WILD-15-PIS-217 (Koyna River, 61.3 mm SL) in (A) lateral, (B) dorsal and (C) ventral view.

Discussion

The spelling of the specific epithet of Nemachilichthys rueppelli needs to be clarified. In the original description (Sykes 1839), the species’ name was spelt Cobitis rupelli. However, Sykes (1839) specifically mentioned that the species was named for Rüppell [sic]. Therefore, the species-name needs to be corrected (International Commission on Zoological Nomenclature (ICZN, 1999), art. 32.5) to N. rueppelli where the umlaut is signified by the letter 'e' after the vowel. Kottelat (2012: p. 11) mentions "... omission of the umlaut is correct; if Sykes had written "Rüppell", ü should have been corrected into ue, but as Sykes used "Ruppell", u should not be corrected". This is in error, as in the original publication (Sykes 1839) Rüppell’s name is in fact spelt with an umlaut, which may not have been distinct in the copy of Sykes (1839) examined by Kottelat (2012). In his original description of Nemachilichthys shimogensis, Narayan Rao (1920) did not compare the species with N. rueppelli. He doubted, however, the conspecificity of the species identified as N. rueppelli by Day (1878, p. 612; pl. 155, fig. 7) and Sykes’s (1841, p. 366; pl. 64, fig. 1) description and illustration, suggesting that Sykes's Cobitis rupelli could be a local variety of Paracanthocobitis botia (Hamilton, 1822). It is known that several of Sykes's (1841) illustrations are not accurate (see Hora 1943; Kottelat 2012; Knight et al. 2013). Nevertheless, the identity of the species is immediately established from its pointed snout (see Fig. 3A), a character not shared by any other loach co-occurring in its type locality, the Mula-Mutha River of Pune. Topotypic N. rueppelli are similar to the illustration of N. shimogensis in Narayan Rao (1920; see Fig. 3B). Bănărescu & Nalbant (1995) considered both species valid but did not mention the distinction between them, while Kottelat (2012) considered N. shimogensis as a questionable synonym of N. rueppelli.

REDESCRIPTION OF NEMACHILICHTHYS RUEPPELLI Zootaxa 4059 (3) © 2015 Magnolia Press · 565 FIGURE 7. Nemachilichthys rueppelli mouth (illustrated based on WILD-15-PIS-221, Hirnyakeshi River, 77.8 mm SL). Scale bar is 5 mm.

Our genetic and morphometric analysis suggests that the two nominal species are in fact the same and the marginal differences seen in the genetic markers and multivariate morphometric analysis are the result of geographical separation of populations. Furthermore, the statistically significant linear relationship (p < 0.01) observed between geographical separation and genetic and morphometric analyses suggests that the genetic and morphometric variations are continuous. Thus, these variations cannot be used for determination of species boundaries. As a result, the genetic and morphometric differences observed between topotypic N. rueppelli and N. shimogensis is due to variation in two widely separated populations and is not indicative of two distinct species. Further, it has been suggested that a genetic distance of around 3% in COI is indicative of distinct species in fishes (Ward 2009), so the distance of 0.5% between topotypic N. rueppelli and N. shimogensis is not sufficient for separating the two nominal species. An interesting result revealed in our study was the presence of a positive significant correlation between genetic and morphometric distance and geographic separation measured as linear distance, rather than geographical distance based on current river drainage connections. This finding suggests that the ancient drainage connections were likely to have been different from those observed at present, indicating possible events of freshwater vicariance that could have isolated the populations studied here. While such inferences are available also for fishes of other regions (see Waters et al. 2001; Burridge et al. 2006), no such studies are available for fishes of the Western Ghats.

566 · Zootaxa 4059 (3) © 2015 Magnolia Press KESKAR ET AL. Acknowledgements

This work was funded by INSPIRE Research Grant [IFA12-LSBM-21] to ND and the Mohammed Bin Zayed Species Conservation Fund [No. 1225670] to RR. ND and AK are thankful to Director, Indian Institute of Science Education and Research, Pune; AK is also thankful to Principal; Head, Department of Biodiversity; and Anand Padhye, Abasaheb Garware College, Pune; and PK is thankful to Principal; Head, Department of Zoology, Modern College of Arts, Science and Commerce, Ganeshkhind, Pune, for providing facilities. We thank Deepak Apte, Director; and Rahul Khot, Curator of Natural History Collection department, for their help during the study of museum specimens and registration of specimens in Bombay Natural History Society (BNHS), Mumbai; Sanjay Molur, Executive Director; and Priyanka Iyer, curator of fish collection, for holding our specimen vouchers in the museum collection of Wildlife Information Liaison Development (WILD) Society, Coimbatore; P.S. Bhatnagar, Officer In-Charge and Shrikant Jadhav for registration of specimens in the Zoological Survey of India-Western Regional Center (ZSI-WRC). We are thankful to Ralf Britz for providing live photograph of the species from Tunga River, and to Siby Philip, Nikhil Sood and Sachin Gosavi for their assistance in the field. We are also thankful to Andrew Polaszek, Louise Allcock and Debbie East for helping us with obtaining permission for the use of illustration of N. shimogensis from Annals and Magazine of Natural History, Taylor & Francis Group. We are grateful to Mark McGrouther and Sally Reader, Australian Museum, Sydney (AMS), for providing collection details and photographs of the specimen of N. rueppelli (AMS B.7528) in their care. An anonymous reviewer and Rohan Pethiyagoda provided valuable comments on an earlier draft of the manuscript.

References

Ali, A., Philip, S., Dahanukar, N., Renjithkumar, C.R., Bijukumar, A. & Raghavan, R. (2013) Distribution, threats and conservation status of Hypselobarbus thomassi (Day, 1874), a poorly known cyprinid fish of the Western Ghats freshwater ecoregion. Journal of Threatened Taxa, 5, 5202–5213. http://dx.doi.org/10.11609/JoTT.o3838.5202-13 Altschul, S.F., Gish, W., Miller, W., Myers, E.W. & Lipman, D.J. (1990) Basic local alignment search tool. Journal of Molecular Biology, 215, 403–410. http://dx.doi.org/10.1016/S0022-2836(05)80360-2 Bănărescu, P. & Nalbant, T.T. (1995) A generical classification of Nemacheilinae with description of two new genera (Teleostei: : Cobitidae). Travaux du Muséum National d'Histoire Naturelle "Grigore Antipa", 35, 429–496. Bănărescu, P. & Nalbant, T.T. (1968) Cobitidae (Pisces, Cypriniformes) collected by the German India Expedition. Mitteilungen aus dem Hamburgischen Zoologischen Museum und Institut, 65, 327–351. Bhat, A. (2004) Patterns in the distribution of freshwater fishes in rivers of Central Western Ghats, India and their associations with environmental gradients. Hydrobiology, 529, 83–97. http://dx.doi.org/10.1007/s10750-004-4949-1 Burridge, C.P., Craw, D. & Waters, J.M. (2006) River capture, range expansion, and cladogenesis: the genetic signature of freshwater vicariance. Evolution, 60 (5), 1038–1049. http://dx.doi.org/10.1111/j.0014-3820.2006.tb01181.x Dahanukar, N., Paingankar, M., Raut, R.N. & Kharat, S.S. (2012) Fish fauna of Indrayani River, northern Western Ghats, India. Journal of Threatened Taxa, 4 (1), 2310–2317. http://dx.doi.org/10.11609/JoTT.o2771.2310–7 Day, F. (1878) The fishes of India; being a natural history of the fishes known to inhabit the seas and fresh waters of India, Burma, and Ceylon. Part 4. Taylor and Francis, London, xx+ 227 pp., 57 pls. [pp. xx + 553–779 pp., pls. 139–195] Doornik, J.A. & Hansen, H. (2008) An omnibus test for univariate and multivariate normality. Oxford Bulletin of Economics and Statistics, 70, 927–939. http://dx.doi.org/10.1111/j.1468-0084.2008.00537.x Edgar, R.C. (2004) MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucleic Acids Research, 32, 1792–1797. http://dx.doi.org/10.1093/nar/gkh340 Eschmeyer, W.N. & Fricke, R. (Eds.) (2015) Catalog of Fishes: Genera, Species, References. Available from: http:// researcharchive.calacademy.org/research/ichthyology/catalog/fishcatmain.asp (accessed 8 November 2015) Hammer, Ø., Harper, D.A.T. & Ryan, P.D. (2001) PAST: paleontological statistics software package for education and data analysis. Palaeontologia Electronica, 4, 1–9. Harris, R.J. (2001) A primer for multivariate statistics. 3rd Edition. Lawrence Erlbaum Associates Publishers, London, 609 pp. Hora, S.L. (1943) The game fishes of India: XV. The mahseers or the large scaled barbels of India. 9. Further observations on

REDESCRIPTION OF NEMACHILICHTHYS RUEPPELLI Zootaxa 4059 (3) © 2015 Magnolia Press · 567 the large-scaled barbels of India. Journal of the Bombay Natural History Society, 44 (1), 1–8. Huberty, C.J. & Olejnik, S. (2006) Applied MANOVA and Discriminant Analysis. John Wiley & Sons, Hoboken, NJ, 488 pp. Katwate, U., Paingankar, M.S., Jadhav, S. & Dahanukar, N. (2013) Phylogenetic position and osteology of Pethia setnai (Chhapgar & Sane, 1992), an endemic barb (Teleostei: Cyprinidae) of the Western Ghats, India, with notes on its distribution and threats. Journal of Threatened Taxa, 5, 5214–5227. http://dx.doi.org/10.11609/JoTT.o3857.5214-27 Katwate, U., Raut, R. & Advani, S. (2012) An overview of fish fauna of Raigad District, northern Western Ghats, India. Journal of Threatened Taxa, 4, 2569–2577. http://dx.doi.org/10.11609/JoTT.o2760.2569–77 Knight, J.D.M., Rai, A. & d'Souza, R.K.P. (2013) On the identities of Barbus mussullah Sykes and Cyprinus curmuca Hamilton with notes on the status of Gobio canarensis Jerdon (Teleostei: Cyprinidae). Zootaxa, 3750 (3), 201–215. http://dx.doi.org/10.11646/zootaxa.3750.3.1 Kottelat, M. (2012) Conspectus Cobitidum: an inventory of the of the world (Teleostei: Cypriniformes: Cobitoidei). Raffles Bulletin of Zoology, Supplement 26, 1–199. Mafakheri, P., Eagderi, S., Farahmand, H. & Mosavii-Sabet, H. (2015) Descriptive osteology of kermanshahensis (Bănărescu and Nalbant, 1966) (Cypriniformes, Nemacheilidae). Croatian Journal of Fisheries, 73, 115–123. http://dx.doi.org/10.14798/73.3.840 Menon, A.G.K. (1987) The fauna of India and the adjacent countries. Pisces. Vol. IV. Teleostei - Cobitoidea. Part 1. Homalopteridae. Zoological Survey of India, Calcutta, x + 259 pp., 16 pls. Narayan Rao, C.R. (1920) II. — Some new species of cyprinoid fish from Mysore. Annals and Magazine of Natural History, 6, 45–64, Pls. 1–2. http://dx.doi.org/10.1080/00222932008632409 Potthoff, T. (1984) Clearing and staining techniques,. In: Moser, H.G., Richards, W.J., Cohen, D.M., Fahay, M.P., Kendall, A.W. Jr. & Richardson, S.L. (eds.). Ontogeny and Systematics of Fishes. No. 1. American Society for Ichthyology and Herpetology, Special Publication, pp. 35–37. Sykes, W.H. (1839) On the fishes of the Deccan. Proceedings of the Zoological Society of London, 1838 (6), 157–165. Sykes, W.H. (1841) On the fishes of the Dukhun. Transactions of the Zoological Society of London, 2, 349–378, pls. 60–67. Talwar, P.K. & Jhingran, A.G. (1991) Inland fishes of India and adjacent countries. Vol. 1–2. Oxford & IBH Publishing Co., New Delhi, Bombay, Calcutta, xvii + 1158 pp. Tamura, K., Stecher, G., Peterson, D., Filipski, A. & Kumar, S. (2013) MEGA6: Molecular Evolutionary Genetics Analysis version 6.0. Molecular Biology and Evolution, 30, 2725–2729. http://dx.doi.org/10.1093/molbev/mst197 Ward, R.D. (2009) DNA barcode divergence among species and genera of birds and fishes. Molecular Ecology Resources, 9 (4), 1077–1085. http://dx.doi.org/10.1111/j.1755-0998.2009.02541.x Waters, J.M., Craw, D., Youngson, J.H. & Wallis, G.P. (2001) Genes meet geology: fish phylogeographic pattern reflects ancient, rather than modern, drainage connections. Evolution, 55 (9), 1844–1851. http://dx.doi.org/10.1111/j.0014-3820.2001.tb00833.x

568 · Zootaxa 4059 (3) © 2015 Magnolia Press KESKAR ET AL.