Channa nachi, Praveenraj & Moulitharan & Pavan-Kumar & Naveen & Thackeray & Yumnam & Gurumayum, 2025

Channa nachi , new species Praveenraj & Moulitharan

( Figs 1 View FIGURE 1 & 2 View FIGURE 2 )

Holotype. APRC /ZSI//P-2298, 96.4 mm SL; India: Meghalaya, South Garo hills District, a small stream draining into the Simsang River at Chokpot village , 25.34058°N; 90.45856°E, Coll. R. S. Naveen, 12 September, 2024. GoogleMaps

Paratypes. APRC /ZSI/P-2299, 12 ex., 63.3–118.6 mm SL (74.1, 88.7 & 118.6 mm SL; used for genetic analysis); Coll. Farham Sangma, 11 October 2020 ; CIARI /FF-102, 1 ex., 66.3 mm SL (cleared and stained for osteology), Coll. Farham Sangma, 11 October 2020 ; CIARI /FF-120, 3 ex., 73.3–122.0 mm SL, Coll . R.S. Naveen , 12 September, 2024 ; other information same as holotype.

Diagnosis. A member of the Gachua group distinguished from all other species in this group except C. gachua , C. lipor , and C. aurantipectoralis in having a unique colouration in life: body devoid of any spots or blotches, sides creamish-yellow to pale-brown, and bluish shades on posterior half (vs. different colour pattern and presence of numerous spots), and presence of 10–12 obliquely-arranged, saddle-like, pale-brown bars (vs. 5 to 8 distinct bars or absence of bars). It differs from C. gachua in having a shallower body (body depth 10.7–15.2 vs. 18.0–18.9% of SL) and more scales below lateral line at anal-fin origin (7.5–8.5 vs. 5); from C. lipor in having 10–12 pale-brown bars on sides and bluish shades below lateral line (vs. overall body orange to bronze-brown with six distinct brown, oblique bands), more anal-fin rays (23–24 vs. 20), more lateral-line scales (44–48 vs. 35–40), and vertebrae (43–45 vs. 40); and from C. aurantipectoralis in having a translucent orange pectoral fin with 3–4 faint orange or brown bands in life (vs. orange pectoral fin without any bands), fewer lateral-line scales (44–48 vs. 51–64), a slenderer head (head depth 37.4–45.0 vs. 45.5–52.8 % HL), and a shallower body (body depth 10.7–13.9 vs. 14.1–17.2 % SL).

Description. See Figs. 1 View FIGURE 1 & 2 View FIGURE 2 for general appearance and Table 1 View TABLE 1 for morphometric information. Body elongate, round in cross section anteriorly, laterally compressed at caudal peduncle, body depth 7.4–9.3 times in SL. Dorsal profile rising evenly from tip of snout to dorsal-fin origin, then almost straight to base of caudal fin. Ventral contour descending gently until below opercle, almost straight to base of caudal fin. Head large, length 3.1–3.6 times in SL. Head widest halfway between posterior margin of eye and opercle. Mouth large, posterior end of maxilla and premaxilla extending beyond posterior margin of eye. Lower jaw slightly protruding beyond upper jaw, lips fleshy. Premaxilla with multiple rows of small, sharp, pointed teeth. Dentary with two rows of teeth, inner row with 25 large canines; outer row with small, villiform teeth terminating in three rows at jaw symphysis ( Fig. 3A View FIGURE 3 ). Vomer with an inner series of six medium- to large teeth and outer series with smaller teeth. Palatine with external series of smaller teeth and an internal row of up to nine much larger canines ( Fig. 3B View FIGURE 3 ). Fifth ceratobranchial with 10 large teeth with anteriorly directed tips along posterior and medial margin, and smaller teeth in middle and along anterior margin ( Fig. 3C View FIGURE 3 ). Parasphenoid tooth patch with three, medium-sized teeth.

TABLE 1. (Continued)

Range (includes holotype and Morphometric characters Holotype paratypes) Mean±SD

(N=17) Post-orbital length 21.2 17.7–21.2 19.7±0.9 Inter-orbital width 11.5 8.9–11.5 10.1±0.7 Upper-jaw length 12.2 9.4–12.2 10.8±0.7

In percent of head length

Snout length 24.6 20.2–27.3 23.6±1.7 Eye diameter 17.3 14.2–18.2 15.9±1.2 Head depth 36.9 37.4–45.0 39.9±2.4 Head width 64.4 55.3–64.4 59.8±2.7 Pre-orbital depth 26.9 24.1–30.6 26.1±1.7 Post-orbital depth 36.9 33.1–37.6 35.4±1.2 Post-orbital length 66.8 60.8–69.5 66.0±2.0 Inter-orbital width 36.1 30.4–36.6 33.8±1.7

Upper-jaw length 38.4 31.6–40.5 36.3±2.4

Dorsal-fin rays 34(2), 35(5), 36(7), 37*(3). Anal-fin rays 22(3), 23*(5), 24(8), 25(1). Caudal fin with 6*(12), 7(4), or 8(1) + 6(6) or 7*(11) principal rays. Pectoral-fin rays 12(1), 13(3), 14*(11), 15(2). Predorsal scales 14*(13) or 15(4). Lateral line with 44(2), 45(4), 46*(7), 47 (1), 48(3) scales extending from shoulder girdle in horizontal line, but dropping one scale row at scale 11*(3), 12(4), 13(9), 14(1) then continuing horizontally to end of caudal peduncle. Transverse scale rows in dorsoventral direction 4.5*(14) or 5.5 (3) /1/ 7.5(10) or 8.5*(7). Postorbital scales 7*(17). Circumpeduncular scales 24(4), 26*(13). Prepelvic scales 10* (4), 12*(13). Total vertebrae 43 (3), 44 (3), 45 (7).

Colouration in preservative. See Fig. 1 View FIGURE 1 . Dorsum, sides of head and body brown to beige, becoming lighter posteriorly. Ventral surface of body uniformly beige. Pectoral fin with 3–4 broad black bands and broad greyish interspace. Dorsal, pelvic, anal and caudal fins dark grey with narrow white rim.

Colouration in life. See Fig. 2 View FIGURE 2 . Body with series of 10–12 obliquely-arranged, saddle-like, pale-brown bars running from dorsal-fin base anteroventrally to upper third of body. Sides of head, lips and gular area bluish, postorbital area with brown blotches. Body creamish-yellow to pale-brown, and bluish shades on posterior half. Area anterior and posterior of eye bright orange, bluish below. Dorsal fin with light green base, blue middle, narrow dark-grey subdistal band and bright orange distal margin. Individual dorsal fin-rays sometimes with white distalmost tip, specifically rays at anterior portion of fin. Anal fin bluish, with dark-grey subdistal band and white distal margin. Caudal fin with light grey fin rays, bluish interradial membrane, narrow submarginal dark band and distal orange rim. Pectoral fin with orange base, and 3–4 semicircular light broad brown bands alternating with narrower chromeyellow bands. In some specimens, pectoral fin appearing yellowish with indistinct bands ( Fig. 2D View FIGURE 2 ). Pelvic fins light grey, with white distal margin.

Distribution and habitat. Channa nachi was collected from a small stream which feeds into the Simsang River system, passing through Chokpot village in South Garo Hills, Meghalaya. It is a shallow, slow-flowing stream with moderately clear water. The substrate of the stream is comprised of sand, leaf litter, and pebbles ( Fig. 4 View FIGURE 4 ). Other species encountered at the type locality were Schistura reticulofasciata , Dario kajal , Pseudolaguvia sp. , and Barilius bendelisis .

Etymology. The specific epithet nachi is the local vernacular name for the species in Garo language spoken by the Garo tribes of Meghalaya. It is used as a noun in apposition.

Proposed common name. Farham’s snakehead (after Farham Sangma, a local Garo resident and fish hobbyist, who first collected this species and made them available for the present study).

Genetic analysis. The interspecific genetic distance values between C. nachi , and its externally most similar species, C. gachua , C. aurantipectoralis and C. lipor was observed to be 14.1–14.3, 10.1–10.5 and 9.7–10.1%, respectively ( Table 2 View TABLE 2 ). Further, Channa nachi differs from all the species in Gachua group by a genetic distance ranging from 12.1–20.0% ( Table 2 View TABLE 2 ). In the maximum likelihood tree, C. nachi was recovered as a distinct clade with C. aurantipectoralis as its sister species ( Fig. 5 View FIGURE 5 ).

The identity of Channa melanostigma , C.pomanensis ,and C.brahmacharyi . We discuss the recent descriptions of three species of Northeast Indian snakeheads and their taxonomic issues. These include Channa melanostigma , C. pomanensis and C. brahmacharyi , all of which were described from the Brahmaputra drainage system, and their identities shrouded in confusion. We examined the type series, and fresh topotypes of Channa melanostigma ( Fig. 6 View FIGURE 6 ), and correct some errors in the original description by Geetakumari & Vishwanath (2010). The vertebral count (50–51) reported by Geetakumari & Vishwanath (2010) for this species is erroneous, as our radiographs reveal a count of only 47 (3) or 48* (5) vertebrae. The predorsal scale count range of 13–14 reported by Geetakumari & Vishwanath (2010) is the scale row count from the prenasal scale to a scale prior to the dorsal-fin ray origin, and the count from the dorsal-fin ray origin to the basal scale is 7 (2) and 8* (6), the latter methodology is often followed by some workers (see Fig. 7 View FIGURE 7 for explanation). The original description also failed to provide the total lateral-line scale count. Our counts from the type material, and fresh topotypes are 11 (1), 14* (5), and 15 (2) (pre-drop) scales, dropping one scale, followed by 32* (2), 34 (5), and 37 (1) (post-drop) scales with a total of 46* (1), 48 (1), 49 (5), and 50 (1) scales. The original description of C. melanostigma was based on preserved specimens, and lacked a colour photograph. Because channids are often difficult to identify from preserved specimens, we complement the description here with a photograph of a live specimen ( Fig. 8 View FIGURE 8 ). We also report that C. melanostigma occurs in the Brahmaputra drainage basin in Assam and Nagaland, and provide COI gene sequences of the specimens collected from the type locality in Arunachal Pradesh, and from additional localities in Nagaland and Assam for reference (see Fig. 5 View FIGURE 5 ).

Channa pomanensis Gurumayum & Tamang (2016) was described from the Poma River in Arunachal Pradesh. It possesses a similar colour pattern to that of C. melanostigma and the authors claimed both species to be sympatric. We note that the characters provided by Gurumayum & Tamang (2016) to differentiate C. pomanensis from C. melanostigma are inconsistent. For example, in the discussion section, the authors mention: “ C. pomanensis can be differentiated from C. melanostigma in having (7 vs. 12–14) transverse bands on body”. Our examination of the types of C. pomanensis and C. melanostigma reveals that there are 12–14 alternating black (6–7) and grey (6–7) bars in both species, thus Gurumayum & Tamang’s (2016) count of 7 bars in C. pomanensis seems almost certainly based solely on the count of the black bars. The predorsal scale count differences (7–8 vs. 13–14) used in Gurumayum & Tamang (2016) to distinguish C. pomanensis from C. melanostigma are based on different ways of counting these scales as mentioned above, as otherwise, the count overlap. Additional characters used were the differences in the transverse scale rows above the lateral line (4½–5½ vs. 3½–4½) which has now been found to overlap, and in the vertebral count (42–45 vs. 50–51), which is erroneous as explained above. Gurumayum & Tamang (2016) use the dorsal-fin length for comparison and mention that the “posterior tip of dorsal fin exceeding (vs. not exceeding) base of first procurrent caudal-fin ray (compare Fig. 1 View FIGURE 1 here and fig. 1 in Geetakumari & Vishwanath, 2010)”. This is however, not a valid character as the fin length varies between sexes. From the above discussion it is evident that Gurumayum & Tamang’s (2016) comparison of C. pomanensis with C. melanostigma was solely based on published literature, and the authors did not examine either the types or fresh specimens of C. melanostigma . Though there are several overlapping meristic and morphometric characters between C. melanostigma and C. pomanensis , our COI gene sequences generated for a C. pomanensis paratype and topotypes reveal genetic distances of 3.8–4% with C. melanostigma . This has also been supported by the recent analysis in Britz et al. (2024) (see Fig. 18). Based on the examination of a fresh specimen, we find that C. melanostigma possess black spots on each scale on the sides, while C. pomanensis possesses dumbbell-shaped orange to maroon blotches on the sides. We also note that C. melanostigma has a narrow snout and thick lips, compared to a broadly rounded snout with thin lips in C. pomanensis ( Fig. 6 View FIGURE 6 ). There is also a vertebral count difference of 42 (1), 45 (1), 46 (3) vs. 47 (3) 48* (5) in C. pomanensis and C. melanostigma , respectively. Based on these distinguishing characters, we accept the validity of both C. melanostigma and C. pomanensis , and provide photographs of live specimens and DNA sequences ( Figs. 5 View FIGURE 5 , 6 View FIGURE 6 , & 8). Channa melanostigma and C. pomanensis are known in the international aquarium trade as Channa sp. “emerald green” and Channa sp. “Garo green”, respectively.

Another species of the Gachua group that needs discussion is Channa brahmacharyi , described from the Simsang River, Meghalaya, by Chakraborty et al. (2020). In Chakraborty et al. (2020), the fin ray, scale, and vertebral counts provided for C. brahmacharyi overlap with those of C. melanostigma and C. pomanensis . Chakraborty et al. (2020) not fully aware of the taxonomic confusion in the Gachua group omitted comparing C. brahmacharyi with C. pomanensis . Erroneously, Chakraborty et al. (2020) use a different name, Channa viridimaculata (a nomen nudum) in their Table 1 View TABLE 1 , which provides morphometric information on C. brahmacharyi . Also, in Chakraborty et al. (2020), the holotype of C. brahmacharyi lacks a museum accession number thus not satisfying Article 72.10 of the Code, which suggest that when designating a holotype, syntype, lectotype, or neotype, researchers should clearly label these specimens to indicate their type status, and when publishing the description, they should include all information present on the specimen labels to ensure future identification of the type specimens. During the year 2020-2023, we attempted to locate the type specimens of C. brahmacharyi at the Zoological Survey of India, Kolkata, and Bharati Vidyapeeth Institute of Environment Education and Research, Pune, India, but were unsuccessful. Therefore, we collected new specimens from the type locality, Simsang River, East Garo Hills, and found that the meristic, morphometric, and genetic data overlap with those of C. pomanensis . The COI gene sequences of C. brahmacharyi differ from those of C. pomanensis by a genetic distance of only 1.8–2.0%, which is well within the interspecific variation encountered in species of the Gachua group. Given the similar colour pattern ( Fig. 8 View FIGURE 8 ), overlapping characters, minor genetic differences, and distribution of C. pomanensis and C. brahmacharyi in the same river system, we treat C. brahmacharyi as a synonym of C. pomanensis .