A Review of the Snailfishes (Liparidae, Scorpaeniformes) of New Zealand, Including Descriptions of a New Genus and Sixteen New Species Author Stein, David L. text Zootaxa 2012 3588 1 54 journal article 10.5281/zenodo.283120 62898615-ab09-4d0a-9341-9cbece286c1a 1175-5326 283120 Osteodiscus rhepostomias n. sp. Fig. 16 Holotype . NMNZ P.046408, sex unknown, 40mm SL, TL unknown, 46°38.80' S , 178°30.03' E , Bounty Trough, R/V Tangaroa , Stn. TAN 0 116, 10 October 2001 , 2786– 2821 m . NMNZ P.046408/1, cleared and stained right pectoral girdle. Paratypes . NMNZ P.046409 (two specimens), sex unknown, 29 mm , 30 mm SL, TL unknown, collected with holotype . Diagnosis. Mouth distinctly oblique, not horizontal. Dorsal fin rays 46–47, anal fin rays 41– ~44, caudal fin rays 10, pectoral fin rays 24. Gill opening above pectoral fin and extending ventrally in front of about nine rays. Radials apparently 2 (1+0+0+1). FIGURE 16. Osteodiscus rhepostomias n.sp. Holotype. NMNZ P.046408, 40mm SL. A. lateral view of preserved specimen; B. NMNZ P.046408/1, right pectoral girdle; C. ventral view. Description. Paratype data (where extant) in order of decreasing length. Counts: V 51 (55, 51), D 47 (46, 42?), A 41 (~44, 41), C na (10, 10), P 24, radials 2? Ratios: HL 27.3 (27.8, 28.2), HW 17.6 (na, 19.9), sn 10.2 (na, 11.5), E 5.4 (4.7, na), orbit 8.4, uj 10.9, UPL ~17.5 (na, 20.2), md 11.9, disk 6.9 (9.3, 8.7), da 3.2, bd 20.6, preD 29.0, preA 40.5, ma 18.6, aAf 17.9. In % HL: HW 64.5 (na, 70.4), sn 37.3 (na, 40.7), E 16.7 , orbit 30.9, uj 40.0, UPL ~63 (na, 71.6), md 43.6 (na, 38.3), disk 25.4 (33.3, 30.9), da 11.8, bd 75.4, preD 106.4, preA 147.3, ma 68.2, aAf 65.4. Head globular, dorsal profile more or less flat to snout. Snout blunt, angled anteroventrally. Nostrils single, horizontal with middle of pupil. Mouth terminal, large, oblique at an angle of about 30 degrees. Premaxilla reaching to below anterior margin of orbit, upper jaw to below mid orbit, lower jaw with a prominent symphyseal process. Teeth slender sharp canines, arranged in about 14 oblique rows of up to 6 teeth each. Posterior teeth uniserial in jaws; anteriorly, outer teeth smallest, gradually larger medially. A clear symphyseal gap present in lower jaw. Lower jaw teeth similar to those of upper jaw, symphyseal gap absent. Eye prominent, moderately large. Gill opening long, above pectoral fin and extending ventrally over about nine rays. Opercle a prominent long spine, directed almost straight down. Opercular flap large, triangular, its apex pointed ventrally and terminating in opercular spine, posterior margin almost vertical. Cephalic pores apparently large, oval, not prominent, without raised margin, very difficult to distinguish, pattern undeterminable. Symphyseal pore pair well separated, difficult to distinguish from skin folds. Pectoral fin upper ray on horizontal just above posterior of upper jaw. Pectoral fin rays 15+4+5. Upper lobe about 2/3 HL. Notch moderately deep, ray spacing distinctly wider than in upper and lower lobes, rudimentary rays absent. Lowest (anteriormost) ray base of lower lobe below anterior margin of orbit, below posterior of premaxilla. Pectoral girdle of holotype cleared but poorly stained probably owing to reduced calcification. Radials apparently 2 (1+0+0+1), the upper (larger) forming upper part of fin base and supporting entire upper lobe, lower radial much smaller, at ventral end of girdle. Trunk thin, distinctly shallower than head depth, tapering gradually to tail. Dorsal fin origin probably between vertebrae 4–5, anal fin origin between vertebrae 10–11. Disk length about 1/4 HL, its width slightly greater than its length, its anterior margin just below posterior end of oral cleft, posterior margin below mid orbit. Disk skeletal, structure clearly visible through skin of ventral surface. Anus close to posterior edge of disk and below rear of head. Hypural fused, supporting caudal of 10 (1+4/5) rays. Internal organs unknown, deteriorated from poor preservation. All caudal rays of holotype missing, hypural present; caudal rays of paratypes present. Skin thin, fragile, translucent. Prickles absent. Color in life unknown. In alcohol, body color brown, tail unpigmented. Orobranchial cavity brownish, peritoneum black. Distribution. Known only from the types , collected in the Bounty Trough southeast of New Zealand at 2786–2821 m . Etymology. From the Greek, rhepo , incline or slope, and stomias , mouth, to denote the upwardly angled mouth of the species. Comparisons. Although the type specimens are not in good condition, the new species is clearly distinct from previously known congeners in its angled mouth (vs horizontal). This is the third known species and the first record of Osteodiscus from outside the North Pacific, and establishes the genus as biantitropical. The two other known species are O . cascadiae Stein 1978 from off Oregon USA , and O . andriashevi Pitruk and Fedorov 1990 , from the Sea of Okhotsk. Both differ from O . rhepostomias in having horizontal (vs oblique) mouths and fewer caudal fin rays (10 vs 6–7 and 7–8 respectively). In addition, O . rhepostomias differs from O . cascadiae in mandible-anus distance (68 vs 73–109% HL) and disk to anus distance (12 vs 1–4% HL). It differs from O . andriashevi in number of vertebrae (51–55 vs 55–59), number of dorsal fin rays (46–47 vs 52–53), anal fin rays (41–44 vs 46–49) and in many proportions. Comments. Owing to the small size of the specimens and their condition, it was difficult to examine all three as thoroughly as desirable. However, the disks are in good condition with no evidence of damage. It is possible but unlikely that these specimens are a new species of Careproctus rather than of Osteodiscus ; however, the disks are skin covered but have no thickened margins or well-defined tissue segments such as are present in Careproctus species, and this characteristic is clearly evident in all three specimens. Pectoral girdle structure is unusual regarding the large size and odd shape of R1. Kido (1988: 148) described the pectoral girdle of O . cascadiae as having a similar structure of two radials (1+0+0+1), although he described R1 as small and round and about the same size as R4. However, O . andriashevi has 3 (2+0+1) or 4 (1+1+1+1) radials. More specimens are needed to determine whether R 1 in O . rhepostomias as described above is normal, the result of poor staining (lack of calcification), incomplete development, or an anomaly.