Systematic status of Systomus rubrotinctus Jerdon ( Teleostei : Cyprinidae ) with notes on the Puntius arulius group of fishes

Systomus rubrotinctus Jerdon has long been considered a synonym of Puntius arulius. Examination of ‘P. arulius’ collected from various parts of the Western Ghats revealed the presence of two distinct species: and the fish considered by Day as ‘P. arulius’ from Wayanad is in fact P. rubrotinctus. Based on the colour pattern, morphometrics and meristics, P. rubrotinctus is retrieved from the synonymy of P. arulius and redescribed. Puntius rubrotinctus can be distinguished from P. arulius by the unique colour pattern consisting of three well-defined black blotches on the body, two scales high and three scales wide, as distinct from the diffused blotches in P. arulius. It further differs from P. arulius in having a shorter dorsal-to-hypural distance (49.8–54.6 % SL, vs. 57.3–57.6 % SL), a smaller interorbital width (28.4–35.6 % HL, vs. 39.1–39.7 % HL) and a higher number of gill rakers (10–11 in the first gill arch, vs. 8). Puntius rubrotinctus is wide distributed in the Cauvery River and its tributaries in the Western Ghats region across the states of Kerala, Karnataka and Tamil Nadu. The results revalidate the identity of the ‘Puntius filamentosus group’ of previous authors drawing attention to the presence of intermediate forms among this group, in which natural hybrids appear to occur, of which P. exclamatio may be one.


IntroductIon
Jerdon (1849) described two barbs, Systomus arulius and S. rubrotinctus, now be referred to Puntius.These fishes possess three black blotches on the body, the former with large diffused blotches, and the latter with smaller blotches.Jerdon described P. arulius as having "a large diffused black spot on side beneath the commencement of the dorsal fin, another over the anal, and another at base of caudal" and in the same work he described P. rubrotinctus a species with "3 small black spots on sides, one under dorsal, the 2 nd over the anal and the 3 rd near base of caudal".Though Jerdon (1849) does not give the exact type locality of P. arulius and P. rubritinctus, he mentions that P. arulius is called 'aruli' at Seringapatam [=Shrirangapattana] in Karnataka and that he procured P. rubrotinctus from the Manantoddy River (= Mananthavadi River) in Kerala.Subsequently, Day (1878) placed P. rubrotinctus in the synonymy of P. arulius.During an examination of the specimens deposited in the Southern Regional Centre, Zoological Survey of India and recent collections from the Western Ghats, the presence of two distinct species fitting the description of P. arulius and P. rubrotinctus given by Jerdon (1849) were observed.In this paper we retrieve P. rubrotinctus from its synonymy with P. arulius and distinguish it from its closely related congeners P. arulius, P. tambraparniei and P. Author Details: J.D. Marcus Knight is a naturalist based in Chennai.Amongst others, his interest is in exploring the freshwater habitats and is currently studying the Puntius filamentosus group of fishes.K. reMa Devi is a senior scientist in the Southern Regional Centre of the Zoological Survey of India and an ichthyologist who has published over hundred papers including descriptions of several new species.viDyaDhar atKore is a PhD student at ATREE working on conservation biology of native freshwater fishes in the Western Ghats, India.His interest lies in the community ecology of fishes and river restoration.
Author Contributions: JDMK recognized the validity of P. rubrotinctus and initiated the study.He speculated natural hybridization and all putative hybrids were collected by him.All the photographs used in the paper were provided by him.KRD carried out the morphometric study to distinguish P. rubrotinctus as a valid species.She provided help in writing the paper.VA also provided comparative material and helped in the morphometric study.
srilankensis, which we here refer to as the P. arulius group of fishes owing to their shared gross adult colour pattern.

MaterIals and Methods
The material for the present study is based mostly on the specimens from earlier surveys in the collections of the Southern Regional Centre of the Zoological Survey of India and recent collections from Bhadra River, Karnataka and Tambraparni River, Tamil Nadu (Fig. 1).The other specimens used in this study are registered in the Reserve Collections of the Zoological Survey of India, Southern Regional Centre, Chennai (ZSI/SRS) and Zoological Survey of India, Western Ghats Regional Centre, Kozhikode (ZSI/WGRC).Measurements were taken using a dial caliper to the nearest 0.1mm.Quantification of characters follows Devi et al. (2010).Subunits of the head are expressed in proportions of head length (HL).Numbers in parenthesis after a count denote the frequency of that count.Specimens of P. srilankensis were not examined and the data in Pethiyagoda & Kottelat (2005)

designation of neotype
As a thorough search of Jerdon's collections maintained in the Senckenberg Natural History Museum, Frankfurt and British Museum of Natural History, London revealed no specimen of Puntius rubrotinctus (R. Pethiyagoda, pers. comm.), a neotype is designated to stabilize the identity of Puntius rubrotinctus to differentiate it from P. arulius.As the type locality of S. rubrotinctus is the Manantoddy River (= Mananthavadi River) which falls in the Kabini River drainage from where the putative topotypes (ZSI/SRS F.8373) were collected, we designate the more intact smaller topotype (71.8mmSL, ZSI/SRS F.8373) as neotype as it was collected as nearly as practicable from the original type locality.The presence of three small black blotches on sides, one under dorsal, the 2 nd over the anal and the 3 rd near base of caudal, consistent with the original description by Jerdon, clearly distinguishes it from P. arulius, which has three large diffused black blotches on the side (Jerdon 1849).The photograph (Image 1) and the morphometric data of the neotype is provided in Table 1. to ensure recognition of the specimen designated.The neotype belongs to the registered collections of the Southern Regional Centre, Zoological Survey of India which is a recognized scientific institution that maintains a research collection, with proper facilities for preserving name-bearing types, and makes them accessible for study.1.General body shape and appearance as in Image 1 and Image 2A.Body elongate, laterally compressed; dorsal contour ascending, indented at nape, profile of back convex anterior to dorsal-fin origin, tapering gradually thereafter; ventral profile equally convex anterior to pelvic-fin origin, curving gently up to analfin origin; caudal peduncle deep, its depth a little less than its length.Head length 28.3-32.8% SL; eye large, its diameter 29.3-35.8% HL, positioned nearer to snout than to opercular margin; interorbital wide, almost equal to eye diameter; snout length less than eye diameter; males with a few minute tubercles on snout, extending up to infra-orbital region.Mouth terminal; lips thick, maxilla extending almost to anterior border of eye.A pair of short maxillary barbels present, 7.3-13.4% HL.Dorsal fin inserted nearer to tip of snout than to caudal-fin base, with three simple and 8½ branched rays, its distal margin slightly concave.Anal fin with three simple and 5½ branched rays.Pelvic fin with one simple and eight branched rays, its origin slightly posterior to dorsal-fin origin.Pectoral fin with one simple and 13(7) or 14(5) branched rays.Pectoral and pelvic fins short, not reaching pelvic and analfin origins, respectively.Caudal fin with 1+9+8+1 principal rays, deeply forked, with pointed lobes.7(4)-8(8) predorsal scales.Lateral line complete, with 20 (1), 21(10) or 22(1) scales on body including one scale on caudal-fin base.Transverse scales from dorsal-fin origin to pelvic-fin origin ½4+1+2.An axillary pelvic scale present.Eighteen circumferential scales (counted as number of scales around greatest depth of body beginning from first scale anterior to dorsal-fin origin), 12 circumpeduncular scales.Well-defied gill rakers, 2(6), 3(6) + 7(2), 8(10) on the first gill arch.

coloration
Formalin-fixed and alcohol-preserved specimens light brown above with a cream underside with three black, well-defined mid-body blotches, the first below the dorsal fin, the second above the anal fin, and the third on the caudal-fin base.Preserved specimens over time lose pigments and the blotch below the dorsal fin resembles the letters W or M. All fins hyaline.

distribution
The type locality of P. rubrotinctus is Manantoddy River (= Mananthavadi River) which falls in the Kabini River drainage in the Wyanad District of Kerala.Also specimens from Bhavali River drainage and Moyar River drainage in Karnataka and Tamil Nadu were examined showing that P. rubrotinctus is widespread in the Cauvery and its tributaries across the States of Karnataka, Kerala and Tamil Nadu.

dIscussIon
Even though Kortmulder (1972) highlighted the similarities between the juveniles and adults of P. filamentosus and P. arulius, it was the revision of the P. filamentosus group by Pethiyagoda & Kottelat (2005) which brought P. filamentosus and P. arulius under the same group on the basis of their treating the juvenile colour pattern of these fishes as synapomorphic.The P. filamentosus group currently involves eight valid species, six of which are endemic to India: Puntius arulius (Jerdon), P. assimilis (Jerdon), P. exclamatio Pethiyagoda & Kottelat, P. filamentosus (Valenciennes), P. rohani Devi et al. and P. tambraparniei (Silas), and two to Sri Lanka: P. singhala (Duncker) and P. srilankensis (Senanayake).Of these Puntius arulius, P. rubrotinctus, P. tambraparniei and P. srilankensis can be termed as the 'P.arulius group of fishes', characterized by three mid-body blotches in adults, in comparison to the other members of the P. filamentosus group, which have either a single blotch other authors have shown a wider range of P. arulius extending up to Tenmalai reservoir in southern Kerala (Day 1878;Jenkins 1909;Mukerji 1932;Silas 1953;Menon 1999;Gopi 2000;Arunachalam & Murugan 2007) with Arunachalam et al. (2005) confirming the presence of a fish very closely ressembling P. arulius from Thenmalai, Kulathupuzha and Kottayam.
Incidentally Thenmalai is the type locality of P. exclamatio which gives rise to the question whether P. exclamatio is a hybrid between the P. arulius likefish and the widespread P. filamentosus as that (i.e., P. exclamatio) is the only fish in this group which has a mid-body 'W' shaped blotch like P. rubrotinctus, which belongs to the P. arulius group and an elongated caudal peduncle blotch which resembles the other fishes of the P. filamentosus group.Moreover P. exclamatio was described as a fish with sub-terminal mouth (Pethiyagoda & Kottelat 2005) but the specimen collected from Kallada River which we observed had terminal mouth which is the character of P. rubrotinctus.Pethiyagoda & Kottelat (2005) stated that the males of P. exclamatio lacked the filamentous prolongations of the dorsal-fin branched rays.In our earlier work (Devi et al. 2010) we attributed this to seasonal shedding of the filaments as the specimens we observed from Kallada possessed dorsal fin filaments; but the lack of the dorsal fin filaments in some populations of P. exclamatio could also be attributed to hybridization as P. rubrotinctus too, lacks dorsal filaments.Moreover a specimen of P. exclamatio we observed from Varkala, Kallada River drainage in Kerala had black caudal-fin tips, which are characteristic of P. filamentosus (Image 2J).
To add impetus to the question of possible hybridization within the P. filamentosus group of fishes, we found intermediate forms between P. filamentosus and P. tambraparniei (Image 2 F, G, H, I) at Cheramadevi in Tambraparni River drainage.These intermediate fishes were found where P. filamentosus and P. tambraparniei co-existed in the Tambraparni River at Cheremadevi.These intermediate fishes either resembled P. filamentosus with an extra midbody blotch, which is a character of P. tambraparniei or they looked like P. tambraparniei with a black bar on the tips of the caudal fin, which is a character of P. filamentosus absent in the typical P. tambraparniei.The possibility of these two species hybridizing is high as they are closely related and the exchange of genes among P. arulius and P. filamentosus has already been speculated (Arunachalam & Murugan 2007) and could be the reason for the high genetic variation among P. filamentosus in the Tambraparni drainage (Johnson et al. 2007).
Hybridization in cyprinidae is more common than in any other group of freshwater fish (Scribner et al. 2000).Hybridization in the genus Puntius was investigated by Kortmulder (1972) and he produced viable hybrids by crossing P. conchonius × P. stoliczkanus, P. cumingi × P. nigrofasciatus and P. stoliczkanus × P. nigrofasciatus.He even observed fertile eggs being produced in a cross between a P. conchonius female and P. filamentosus male and speculated that the reason for the larvae not surviving could also be due to certain specific conditions of the experiment.
Recent studies have shown that hybridization does lead to adaptation through the creation of new genes and morphologies (Seehausen 2004;Bell & Travis 2005;Pfennig et al. 2007;Schwenk et al. 2008;Hayden et al. 2010).Some of these hybrids are fertile (Wood & Jordan 1987;Wyatt et al. 2006;Hayden et al. 2010) and possess unique phenotypic characters which are intermediate between the parental species, which helps them exploit niches unavailable to the parental species, thereby out-competing them, particularly in novel habitats (Seehausen 2004;Nolte et al. 2005).Thus the presence of intermediate forms in the P. filamentosus group could also be attributed to hybridization as it plays a role in adaptive radiation and the evolution of new lineages.We hope to investigate this question further in the Puntius filamentosus group by future genetic studies.
was used for comparison.