Barbatula gibba, Cao & Causse, 2012

Cao, Liang & Causse, Romain, 2012, Revision of the loach species Barbatula nuda (Bleeker 1865) (Pisces: Balitoridae) from North China, with a description of a new species from Inner Mongolia, Zootaxa 3586, pp. 236-248 : 244-246

Barbatula gibba sp. nov.

( Fig.1c)

Holotype. IHB 76 X2566, 70.7 mm SL; Dali-Nur Lake in Hexigten Banner , Inner Mongolia.

Paratypes. IHB 76 X2552–4, 76X2558–9, 76X2562, 76X 2564, 76X2567–8, 76X2574, 76X2576, 76X2584–5, 13 specimens, 46.7–75.7 mm SL, other data same as holotype .

Diagnosis. Barbatula gibba is distinguished from all other congeners from northern Asia by having a nearly columnar (vs. slightly compressed) anterior body, and a greatly convex (vs. slightly convex or straight) predorsal profile of the body. It is similar to B. nuda in having a scaleless predorsal body, by which both can be distinguished all other congeners from northern Asia, with sparsely, or fully scaled predorsal bodies. Barbatula gibba further differs from B. nuda in having widely separated (vs. closely-set) nostrils, and an upper lip with a slight and shallow (vs. marked and deep) median indentation.

Description. Body elongate, anteriorly nearly columnar with its width nearly equal to depth and posteriorly compressed laterally, scaled posterior to vertical through dorsal-fin origin; small scales sparsely scattered, and embedded beneath skin. Dorsal profile of head straight; predorsal profile of body greatly arched upward and declining rapidly along dorsal-fin base; and postdorsal body straight and slightly parallel with ventral. Ventral profiles of head and body from pectoral-fin insertion to anal-fin origin straight; anal-fin base and postanal profile slightly concave. Greatest body depth at vertical through midpoint between pectoral- and pelvic-fin insertion, and least depth of caudal peduncle at vertical through posterior end of anal-fin rays. Caudal peduncle laterally compressed, shorter than HL, and length of it 1.9 to 2.4 times its depth; width 1.5 to 2.3 times in its depth.

Head depressed and triangular in dorsal view, wider than deep. Snout obtuse, shorter than postorbital length of head and head height. Eyes small, close to dorsal profile of head. Nostrils widely separated by a gap twice distance between posterior nostril and anterior edge of eye.

Mouth inferior and arched. Lips thick; upper lip slightly papillated with an indistinct median incision and lower lip with small lateral expansions. Upper jaw unexposed or fully covered by upper lip, without processus dentiformis; lower jaw spoon-like with its anterior exposed, and laterally covered by lower lip ( Fig. 2c). Three pairs of barbels, one maxillary, and two rostral; inner rostral barbels extending to corner of mouth, and outer rostral ones reaching to vertical of anterior edge of posterior nostril; maxillary ones extending to vertical through post half of eye.

Lateral line complete, running slightly below middle of flank anterior to posterior end of anal-fin base, and then along middle of caudal peduncle. Supraorbital canal uninterrupted, not confluent with infraorbital canal; occipital canal continuous, confluent with infraorbital canal. Count of pores in cephalic sensory canal system: 7–8 in supraorbital canal, 12–13 in infraorbital, 3 in occipital, 8–10 in preoperculomandibular, and 60–69 in lateral line.

Fins flexible; D. iii, 7; P. i, 10–11; V. i, 6–7; A. iii, 5; C. 7–8+8 = 15–16 branched rays. Dorsal fin with a slightly concave distal margin, origin nearer to caudal-fin base than to tip of snout, last unbranched ray thickened near base. Pectoral fin inserted immediately anterior to vertical through posteriormost point of operculum, adpressed fin reaching nearly halfway to pelvic-fin insertion. Pelvic fin positioned opposite to dorsal-fin origin, adpressed fin not reaching anus. Anal fin with a straight distal margin. Caudal fin emarginated, its upper lobe as long as lower one. Vertebral count 40–42

Intestine forming a zigzag loop anteriorly, not touching U-shaped stomach. Gas bladder bipartite; anterior chamber fully enclosed in dumbbell-like capsule and posterior chamber strongly reduced.

Coloration in preservative. Back and side of body brownish, ventral body grayish. Five or six transverse brown bars on predorsal region of body, and four or five on postdorsal region. Spots scattered over lateral body, and on pectoral and pelvic fins. Spots scattered over dorsal and caudal fins, forming three black lines.

Sexual dimorphism. Males with broadened and widened unbranched rays and 4–5 outer branched rays of pectoral fin. Dorsal surfaces of pectoral fins covered with breeding tubercles; breeding tubercles sparsely scattered over head and body in males.

Distribution. Currently known from Dali-Nur Lake in Hexigten Banner, Inner Mongolia Autonomous Region ( Fig. 4).

Etymology. The specific epithet is made from the Latin words ‘ gibbus ’ (humped), referring to the greatly convex predorsal profile of the body.

Remarks. We had no access to specimens of these species outside China: B. compressirostris (or B. golubtsovi ), B. dgebuadzei , and B. sawadai . The first one was resurrected by Kottelat (2006) as a valid species closely related to B. golubtsovi . According to Kottelat, B. compressirostris differs from B. golubtsovi only in the absence of projections on the skin of the body (vs. presence). For facilitating the description of this new species, we tentatively consider B. compressirostris and B. golubtsovi as the same species. The data used here for the aforementioned species and B. potaninorum are from Prokofiev (2004).

Barbatula potaninorum was originally described in Orthrias by Prokofiev (2007) from North China. Its exact type locality remains unclear. The original description stated that the type specimen was collected by the Russian explorer G. N. Potanin in 1887 from Gansu Province and nearby Inner Mongolia bordering with Mongolia, or the upper Xilingol River in Inner Mongolia. However, his figure 23 indicated that the type locality was Gaxun-Nur Lake in Inner Mongolia. Potanin made two expeditions to China: the first one from 1884–1886, and the second one from 1892–1893. The maps of these two expeditions were provided by Wang (1993) (see Luo 2005: pp. 202 and 205). In April 1886, Potanin traveled west to Qinghai Lake, turned to the north, and, through several ridges, got to the sources of the Zhoshuy River (today’s Hei-He). Then, he followed this river down to Gaxun-Nur Lake in Inner Mongolia. His expedition finally came to the Orkhon River, Kiakhta, Mongolia, in November 1886. If the type material of B. potaninorum was caught in 1887, it was not from Gaxun-Nur Lake, and Potanin was not the collector. The material was more likely captured from the upper Xilingol River basin in Inner Mongolia. According to Wang (1993), the Russian military officer E. Harnack made an expedition to Xing’an Mountains, Northeast China, in 1887 ( Luo 2003). His expedition started from Peking, turned northeast at Kalgan (today’s Zhangjiakou City), and arrived at Donlon (Inner Mongolia); from there, they headed north to Dali-Nur Lake, which is south of the upper Xilingol River, and eventually entered into Xing’an Mountains. The possibility that the type specimen of B. potaninorum was collected by E. Harnack from the upper Xilingol River nearby the Dali-Nur cannot be ruled out.

According to the original description, B. potaninorum possesses a scaled body, pelvic fins anterior to the vertical through the dorsal-fin origin, and widely separated nostrils. These characters are shared with Nemacheilus pechiliensis Fowler 1899 , a species tentatively considered to be a synonym of B. toni in this study. The type locality of this species is the Tan Lan Ho, a tributary of the Shu Lan Ho (in the Luan-He basin), around 30 miles northeast of Lama-Miau or Dolon-Nor), Pechili Province, northern China. It is not too far away from the upper Xilingol River basin. Whether B. potaninorum and Nemacheilus pechiliensis are the same species remain uncertain. This hypothesis needs testing based on examination on the type material of the former and additional specimens from the type locality of the latter. For the time being, B. potaninorum is regarded as valid.