Gegeneophis tejaswini, Kotharambath, Ramachandran, Wilkinson, Mark, Oommen, Oommen V. & Gower, David J., 2015

Kotharambath, Ramachandran, Wilkinson, Mark, Oommen, Oommen V. & Gower, David J., 2015, A new species of Indian caecilian highlights challenges for species delimitation within Gegeneophis Peters, 1879 (Amphibia: Gymnophiona: Indotyphlidae), Zootaxa 3948 (1) : -

Gegeneophis tejaswini sp. nov.

( Figs. 1–2 ; Tables 1–2)

Gegeneophis sp. “sample 40”: Gower et al. (2011: 702, 705, fig. 2, tables 1, 2, 5)

Holotype. BNHS 5420 ( Fig. 1 ), adult female, collected from the village of Bedoor, near Kakkadav Bridge, near Cheemeni, Hosdurg Taluk, Kasaragod District, Kerala, India (12˚ 16' 06'' N, 75˚ 17' 14'' E, c. 50 m a.s.l.) by R. Kotharambath, Hareesh K., C. B. Binu and M. Wilkinson on 0 7 July 2010.

Paratopotypes (n = 7). ZSI 2532 (male) collected by R. Kotharambath, O.V. Oommen, Hareesh K. and Narayan P. on 10 April 2008; ZSI 2533 (male) collected by R. Kotharambath, Hareesh K. and Krishnan K. on 10 August 2008; BNHS 5421 (female) collected by R. Kotharambath, Hareesh K. and Sreerag K. on 0 4 September 2009; ZSI 2534 (male), BNHS 5422 (female), BNHS 5423 (male), and ZSI 2535 (male) collected by R. Kotharambath, Hareesh K., C. B. Binu and M. Wilkinson on 0 7 July 2010; all collected at the same locality as the holotype.

Diagnosis. Indotyphlids are the only teresomatan (non-rhinatrematid, non-ichthyophiid) caecilians in peninsular India (Wilkinson et al. 2011). The new species is clearly a teresomatan by virtue of having distinct primary and secondary AGs and a tentacle lying between eye and naris, and it is identified as a Gegeneophis on the basis that this is the only genus of indotyphlid in which the eye is under bone (Wilkinson et al. 2011). Gegeneophis tejaswini sp. nov. is an attenuate Gegeneophis with many primary annuli and relatively few SAGs. Differs from G. seshachari Ravichandran, Gower & Wilkinson, 2003 , G. pareshi Giri, Gower, Gaikwad & Wilkinson, 2011 and G. primus Kotharambath, Gower, Oommen & Wilkinson, 2012 in having SAGs. Differs from G. carnosus , G. ramaswamii Taylor, 1964 , G. madhavai Bhatta & Srinivasa, 2004 , and G. orientalis Agarwal, Wilkinson, Mohapatra, Dutta, Giri & Gower, 2013 in having more than 120 (versus fewer than 116) primary annuli. Differs from G. danieli Giri, Wilkinson & Gower, 2003 and G. goaensis Bhatta, Dinesh, Prashanth & Kulkarni, 2007 in having many fewer SAGs (<30 versus> 60). Differs from G. mhadeiensis in having more primary annuli (125–131 versus 117–122) and more primary annuli anterior to the first SAG (98–111 versus 88–93); eye much less visible (generally invisible versus clearly visible); more pink (less brown) body colour in larger specimens. Differs from G. krishni in usually having more SAGs (18-28 versus 11–18, see remarks) and in having more SAGs that are ventrally complete (5-7 versus 0–3); smaller ratios of head length to head widths (LH/WH <1.5 versus> 1.5, see remarks) and of total length to head width (usually <48 versus> 48, see remarks).

Description of holotype. Some meristic and morphometric data are given in Table 2 . Condition good; c. 7 mm ventral incision into coelom c. 50 mm anterior to vent; mouth preserved slightly open causing a transverse crease dorsally in front of NG1; stratum corneum missing in a few small patches; some scale pockets opened on dorsum posteriorly. Overall shape fairly cylindrical, slightly dorsoventrally compressed and uniform throughout. In dorsal view head neither notably U- or V-shaped, its sides straight and converging substantially from back of head to distinct bulges of TPs, converging more substantially in front of TAs to blunt snout tip. In ventral view lower jaw and upper lip a little more rounded than snout, upper jaws visible anterior to CMs. In lateral view upper lip slightly concave.

Eyes faintly visible as tiny dark spots (c. half size of nares, TAs). TAs below imaginary lines between nares and CMs; approximately halfway between lips and imaginary lines between eyes and nares. In lateral view, CMs very slightly closer to bottom than to top of head, nares approximately equidistant from top and front of snout, slightly further from bottom. In dorsal view nares barely visible, very slightly inset, not visible in ventral view. TAs approximately same size as nares, marginally visible in ventral but not in dorsal view; TPs visible in dorsal and ventral views. Midventral crease from NG1 to close to tip of lower jaw.

No diastema between vomerine and palatine teeth. OMs monocusped; PMs smaller, more numerous, probably monocusped; elements of inner tooth rows small, number of cusps not ascertained; palate moderately concave; tongue unattached anteriorly, with two paramedian grooves posteriorly, narial plugs weakly developed; choanae subcircular, interchoanal distance a little more than twice width of each choana.

C2 but not C1 slightly more massive than head and anterior body. C1 approximately same length as first PA, much shorter than (approximately half length of) C2. NG1 faint except laterally, widely incomplete ventrally, narrowly incomplete dorsally. NG2 conspicuous, complete ventrally, perhaps narrowly incomplete dorsally. NG3 fainter than NG2, clearer than NG1, very faint ventrally where possibly widely incomplete. All NGs orthoplicate, with no obvious curvature. Single TGs very faintly indicated on both collars, that on C2 longer.

AGs well marked laterally, fainter but mostly complete or nearly so middorsally and midventrally, more conspicuous posteriorly; no substantial regional variation in lengths of PAs. Each AG with single row of pale enlarged granular glands posterior to narrow dark band. First SAG dorsolateral on 108th PA, none on 109th PA, dorsally complete on 110th PA, ventrally complete on 119th to 125th PAs. Posteriorly two to three rows of scales dorsally in pockets about 0.75 times as deep as length of PA. Small terminal cap, approximately same length as one and a half adjacent PAs. Posteriormost AG marginally behind level of centre of vent. No AGs unambiguously posterior to vent. Terminus bluntly rounded, slightly more so than head; no terminal keel. Disc fairly well circumscribed, subcircular, slightly wider than long, with 10 (five posterior, five anterior) slightly irregular denticulations, those posterior longer than those anterior; vent more or less circular.

Body grey in preservative, paler anteriorly, darker posteriorly, slightly paler ventrally. Some irregular pale markings. Faintly indicated, slightly darker longitudinal middorsal stripe apparent under microscope. Head whitish, paler than adjacent body, mostly lacking pigment anterior to bulges corresponding to the Mm. depressor mandibulares, some very small flecks of pigment middorsally, laterally to level of TAs, and under back of lower jaw; whitish lip lines and broad paler spots surrounding tentacles and (narrower around) nares, tip of snout separately pale. AGs mostly inconspicuous, slightly darker than body except posteriorly where relatively paler (whitish). Small pale patch at TT. Disc pale, with small darker streaks along midline of some denticulations. Immediately anterior to disc, row of granular glands along AG not complete across midline such that pale area of disc appears to extend anteriorly about the length of half the closest PA.

Variation and additional information from paratypes. See Table 2 for meristic and morphometric data. Generally good to fair condition, some (ZSI 2532, 2533) somewhat brown, dehydrated. External morphology of paratypes mostly very similar to holotype. Perhaps unsurprisingly, given the small sample size, multiple regressions of head length and of head width at CM on total length reveal no significant differences between the sexes (t-test, p = 0.852 and 0.112 respectively). None of type series has papillae (= anal glands of Taylor, 1968: 18) on the disc.

Teeth examined in greater detail in ZSI 2532. Teeth in upper jaw strongly recurved, IMs and anterior OMs moderately recurved, posterior OMs slightly more recurved. Anteriormost PMs longest, OMs shorter but thicker with anterolateral elements stoutest. IMs closely spaced at midline, approximately half size of posteriormost OMs. VPs more uniform, anteriormost two or three slightly larger. Tips of crowns of VPs visible in lateral view.

Among paratypes, ZSI 2534 and possibly BNHS 5423 lack TGs on C1. Where clearly detectable, NG3 widely incomplete midventrally with possible exception of ZSI 2532 (where it bends notably posteromedially) and BNHS 5421 (offset). As in holotype, AGs mostly clearly marked throughout; PAGs faint on venter but clear laterally and dorsally, except in ZSI 2535 where faintly marked except close to terminus. Posteriormost AG approximately level with centre of vent.

All specimens more or less grey throughout, paler on head and slightly darker at tail end. Head whitish with little pigmentation in all paratypes, more notable but still diffuse (and paler than collars) pigmentation in slightly dehydrated ZSI 2532 and ZSI 2533. In ZSI 2535 body darker than other paratypes, with uneven dark and pale patches distributed throughout body. Eyes barely visible in ZSI 2532, 2533, BNHS 5422, 5423, not visible in BNHS 5421, ZSI 2534, 7472. Denticulations around vent unpigmented with possible exception of BNHS 5422.

Observations were made of BNHS 5421 in anaesthesia (MS222) before fixation. Head unpigmented, whitish at anterior, becoming pale pinkish towards collars. Body pink anteriorly, darkening posteriorly to purple, slightly paler grey on terminus. Body slightly darker above than below, more so posteriorly (including terminus) with weak, darker, broad middorsal stripe. Body surface spotted with glands throughout, AGs bordered with whitish colour, more conspicuous towards tail end. Eye faintly visible upon close observation. Disc around vent whitish. In smaller specimens anterior half of body pale pink, reducing intensity posteriorly, posteriormost quarter almost greyish. A photograph of the species in life is shown in Fig. 2 .

TABLE 1. Details of fieldwork conducted within 25 km radius from the holotype locality (Bedoor) of Gegeneophis tejaswini sp. nov. All localities are in Kasaragod district of Kerala state.

Locality Coordinates Elevation Distance to Date of General habitat Caecilian taxa found Person hours (m) Bedoor (km) fieldwork (number of specimens) digging/no. of persons

Bedoor N12.275069, c.50 0 10/04/2008 Mixed (coconut, Gegeneophis tejaswini (1) 8/3

E75.292987 arecanut and banana)

10/08/2008 plantation and home Gegeneophis tejaswini (2) 3/1 04/09/2009 gardens Gegeneophis tejaswini (3) 7/2 Taxonomic remarks. The new species is differentiated readily (using annulation characters) from all other described species of Gegeneophis except G. kr i s h n i and G. mhadeiensis , for which differences in annulation are generally not (or barely) absolute, in part because of outliers. For example, whereas the new species usually has more SAGs than G. krishni , this is not strictly true for one specimen of the latter species (BNHS 4176) which has a tiny isolated SAG on the 104th PA but otherwise has its anteriormost SAG on 110th PA. Similarly, whereas total length/LH = 48.4–60.3 (mean 53.93) in G. kr i s h n i versus 33.3–47.8 (mean 42.3) in most of the new species, this ignores one seemingly abnormal outlier, a specimen of the new species (ZSI 2535) for which this ratio is 56. LH/ WH in the new species is 1.18–1.44 (mean 1.34) versus 1.57–1.69 (mean 1.63) in G. krishni .

Whereas overlap in ranges of morphometric variables is expected to increase with larger samples, we suspect that interspecific differences in the numbers of PAs and (relative and/or absolute) position of anteriormost SAGs, and ratios of head and body dimensions will be significantly different when larger samples are available. We are also impressed by the difference between G. tejaswini sp. nov. and G. k r i s h n i in the number of ventrally complete SAGs, although substantially greater, possibly overlapping, variation in this feature should not be unexpected also when larger samples are examined.

Molecular data (Gower et al., 2011) provide additional evidence that these three nominal species are distinct. This is particularly the case for the superficially similar G. mhadeiensis and G. tejaswini sp. nov. which are not each other’s closest relatives (Gower et al., 2011: fig. 2) and which differ by more than 9.5% in 883 aligned sites of mitochondrial 12S and 16S (Gower et al., 2011: table 5). The molecular difference between G. tejaswini sp. nov. and the other superficially most similar (in terms of annulation) species G. k r i s h n i are less extreme and they are much more closely related (Gower et al., 2011: fig. 2), but they are nonetheless separated by a large uncorrected pdistance of 6.1% (Gower et al. 2011). Molecular data have yet to be published for G. pareshi and G. orientalis but these two species are easily distinguished from G. tejaswini sp. nov. in that the latter has, for example, SAGs (absent in G. pareshi ) and more than 120 PAs (<110 in G. orientalis ). Gower et al. (2011: tables 1, 2) mistakenly reported their G. tejaswini sp. nov. voucher specimen as “ UK MW3421” – it should have been UK MW3417, which is now paratype ZSI 2532.

Etymology: The specific epithet is in reference to the type locality, which lies close to (less than 1 km north of) the Tejaswini (= Thejaswini) river. The origin of the river’s name is the Sanskrit word tejas, meaning spiritually splendorous radiance. The Tejaswini river is associated historically with agricultural communities rising up against feudalism and British imperialism. For nomenclatural purposes the specific epithet is considered to be a noun in apposition.

Suggested common name: Tejaswini Geg (English).

Distribution and Natural History: Gegeneophis tejaswini sp. nov. is known only from the type locality. Animals were found during four field visits conducted 2008–2010, between April and September, but none was found in June 2014 when the ground was dry before the monsoon. All the specimens were dug from soil in mixed (including coconut, arecanut, banana) home gardens bordering commercial rubber (to the north and west) and arecanut (to the south) plantations and the disturbed Kamballur Reserve Forest (to the east). Specimens were mostly dug from around the bases of trees, on sloping, partly terraced ground among housing ( Fig. 3 ). Below the slope and away from housing lies a flat area with more intensive agriculture and regular drainage channels though no specimens were found here in the small amount of digging carried out (only in June 2014).

Three poorly preserved specimens of G. tejaswini sp. nov. were neither included in the type series nor yet deposited in a permanent collection. These are smaller than any of the types, with total lengths of approximately 85–95 mm (fresh lengths approximately 95–105 mm). They were collected from the type locality on 10 August 2008 and 0 4 September 2009. One was found approximately 0.5 m from an adult female and is shown in Fig. 2 .

Conservation Status: Given that Gegeneophis tejaswini sp. nov. is known only from a small series of specimens from a single locality, and that very little is known of its general ecology and nothing of its reproductive biology, we expect that it is likely to qualify for ‘Data Deficient’ status under IUCN criteria. The species is able to tolerate some agricultural habitats close to human habitation, which are extensive in this region, and it has been recorded at the type locality on four separate occasions between April 2008 and July 2010, such that we are hopeful it might qualify for a Least Concern categorization if additional localities are discovered. Clearly it is not very abundant in the region given that 39 person hours of digging in localities less than 20 km from the type locality failed to yield additional specimens (Table 1).

Figures

FIGURE 1. Holotype (BNHS 5420) of Gegeneophis tejaswini sp. nov. Whole body shown in dorsal (A) and ventral (B) views; head, collars and anteriormost two primary annuli shown in dorsal (C), right lateral (D) and ventral (E) views; body terminus shown in dorsal (F), left lateral (G) and ventral (H) views. Scale bar for whole body views 10 mm, for anterior and posterior close-up views 5 mm.

FIGURE 2. Gegeneophis tejaswini sp. nov. in life, two topotypic specimens found approximately 0.5 m apart. The larger specimen (preserved TL = 196 mm) is adult female paratype BNHS 5421.

FIGURE 3. Habitat at Bedoor, type locality of Gegeneophis tejaswini sp. nov.

Tables

TABLE 2. Some morphometric (in mm) and meristic data for the type series of Gegeneophis tejaswini sp. nov. * indicates holotype; all other specimens are paratopotypes. † specimen used in molecular phylogenetic study (sample 40) of Gower et al. (2011). ** obscured by partial eversion of phallodeum.

Collection	BNHS	BNHS	BNHS	BNHS	ZSI	ZSI	ZSI	ZSI
Specimen Number Sex	5420* f	5421 f	5422 f	5423 m	2532† m	2533 m	2534 m	2535 m
Length PAs SAGs	193 127 20	196 126 18	135 128 20	140 127 23	180 126 28	200 125 25	207 131 22	224 131 20
First PA with dorsally complete SAG	110	112	111	109	105	106	115	115
PAs with ventrally complete SAGs	119–125	118–123	120–125	120–124	119–124	117–123	124–128	125–129
TGs on C1, C2 ST–NG1	1, 1 5.6	1,1 5.3	1,1 4.5	0,1 4.5	1,1 5.5	1,1 5.8	0,1 6.3	1,1 5.6
Length of CI (N1–N2) Length of C2 (N2–N3)	1.4 2.4	1.2 1.6	1.0 1.5	1.2 1.6	1.5 1.9	1.4 2.2	1.6 2.2	1.8 2.0
ST–CM	4.1	3.7	3.2	3.3	4.0	4.4	4.8	5.2
CM–N1 WH	1.6 3.9	1.6 4.1	1.3 3.2	1.3 3.8	1.9 3.9	1.7 4.2	1.5 4.5	2.1 4.6
Width of head at CM N–N	3.4 1.5	3.5 1.3	2.8 1.2	2.9 1.1	3.2 1.2	3.6 1.3	4.1 1.5	4.3 1.6
TA –TA N–TA	2.5 0.7	2.3 0.7	2.0 0.7	2.0 0.6	2.0 0.8	2.2 0.7	2.5 0.9	2.7 0.9
TA –L	0.5	0.3	0.4	0.4	0.4	0.3	0.5	0.5
N–L N–CM	1.0 3.4	0.9 3.4	0.8 2.5	0.8 2.7	0.8 4.2	0.9 3.7	1.1 4.0	1.2 4.3
TA –ST	1.7	1.6	1.4	1.5	1.5	1.6	1.9	2.1
TA –CM ST–AM	2.6 1.3	2.4 1.1	1.9 1.2	2.0 1.2	2.5 1.1	2.7 1.2	3.1 1.3	3.3 1.5
Width at midbody Circumference	4.5 15	4.1 14	3.7 13	4.2 15	4.5 15.5	4.6 15	4.6 15	3.9 14
Width at level of vent Vent –TT	2.9 0.8	2.9 0.9	2.3 0.5	2.6 0.8	3.0 1.0	3.1 1.1	3.3 1.0	2.6 1.0
Width of disc Vent denticulations PMs	1.3 10 21	1.2 11 23	1.6 11 24	1.6 ** 20	1.7 11 22	1.3 15 23	1.4 10 or 11 23	1.6 10 26
VPs OMs IMs	22 14 2	22 13 2	22 14 2	21 12 2	21 16 2	19 14 2	21 14 2	27 16 2