Careproctus pallidus (Vaillant 1888)

David L. Stein, 2005, Descriptions of four new species, redescription of Paraliparis membranaceus, and additional data on species of the fish family Liparidae (Pisces, Scorpaeniformes) from the west coast of South America and the Indian Ocean., Zootaxa 1019, pp. 1-25 : 13-17

Careproctus pallidus (Vaillant 1888)

(Fig. 5, Table 2)

Enantioliparis pallidus Vaillant 1888 : 22, Pl. 4, Figs. 3, 3a, 3b

Liparis pallidus Garman 1892: 70

Careproctus pallidus Norman1937: 130

Careproctus crassus de Buen 1961: 37

Material Examined. Enantioliparis pallidus (Vaillant 1888) .

Holotype. MNHN 1884-841, ripe? male, ~42 mm SL, 55° 31' S, 68° 03' W, Orange Bay, Tierra del Fuego, Chile, 28 m. Paratype. MNHN 1884-842, ripe? male, 44 mm SL, collected with holotype.

Other material. USNM 347660, ripe male, 41 mm TL, 34 mm SL, USARP Sta. 69- 24, 53° 51' 32" S, 70° 25' 52" W, kelp beds on SSE side of Harriss Bay, east coast of Dawson Is., Chile, 18 May 1969, 1.6 - 3 m. Careproctus crassus de Buen 1961 . Neotype, CAS 60515, female, 77 mm TL, 68 mm SL, in Macrocystis , Punta Toro, Isla Navarino, Tierra del Fuego, Chile, 15 Jan. 1980, 6-8 m.

Diagnosis. Vertebrae 36-37 (10+26-27), D 29-31, P 24?-29, C 12, radials 4 (3+1), round, unnotched. Interneural of first dorsal between vertebrae 5-6 or 6-7. Pectoral fin short, rounded, unnotched, middle rays somewhat more widely spaced but not shortened. Gill opening short, above pectoral fin base. Head length ca 24-30% SL, preanal fin l. 46.8-55.4, disk 38.2-45.9 % HL. Color in life orange, in alcohol, yellowish.

Description. Because Andriashev (1997) described and discussed this species, only additions and modifications to his account are provided herein.

Nostril diameter about equal to its length (vs. “its diameter 1.5 times in its length”). Opercular flap small, clearly lobed; a tissue fold extends ventrally as far as disk, appearing as though gill opening extends down entire side of body, but it does not. When opercular flap is closed, dorsal end of opening forms a membranous tube that remains open. Gill opening slightly larger than eye diameter. Cephalic pores small or tiny, easily overlooked, none tubular. Pore pattern 2-6-7-(+ 3-4 temporal pores). First temporal pore (t1) is sixth and last of infraorbital series; t3 and t4 are suprabranchial. Pores t1 and t3-4 apparently always present, presence of t2 variable. Chin pore pair spacing about equal to that of other pores (vs. “notably converged”). Orobranchial cavity and stomach pale.

Remarks. Specimen USNM 347660 was perfect before being cut open for internal examination. There was no damage to the skin or appendages whatsoever, and consequently presence or absence of the cephalic pores could be determined definitively. Their small size made it impossible to inject either liquid or air to clarify the canals, but there is no question regarding their presence and apparently irregular occurrence. Because dissection was not necessary for identification, to minimize damage pyloric caeca were not counted, nor was a fin removed for clearing and staining.

Careproctus pallidus is highly distinctive among Southern Hemisphere species because it is “Liparis-like” in general appearance, having a stout body with welldeveloped pectoral and caudal fins, short dorsal and anal fins, and living in shallow waters. In addition, its transverse genital fold and protruding papilla are unique. However, as Andriashev (1997) pointed out, its appearance is deceptive because unlike Liparis species, “Pectoral radialia in the specimen CAS 60515 are rounded (without notches), interradial fenestrae and pleural ribs are reduced, and hypural plates are completely fused and have no traces of a cleft.” Andriashev’s (1997) figure includes some differences from USNM 347660 and some inaccuracies, particularly of pores. It shows many, but not all, of the cephalic pores, and portrays some pores not present on the specimen. Some of those shown appear more clearly when the figure is enlarged. For instance, there appears to be a pore immediately posterior to the lower margin of the orbit, and another further back about half an eye diameter, but neither of these is on the specimen. The drawing omits io4 and t2. Furthermore, the opercular flap and the area surrounding it are not as structurally simple as shown.

The temporal pore patterns of the two recently collected specimens differ from each other (Fig. 5) (examination of the types for these characters was inconclusive owing to their condition), but the significance of these differences is unclear. In USNM 347660, t1-4 are present on the right side, but t2 is apparently absent on the left. In CAS 60515, t1 is present on both sides, but t2 could not be found on either side (Fig. 5). Differences between the specimens could easily be the result of size. USNM 347660 is considerably smaller than CAS 60515, and significantly smaller than either of the types (Table 1). Sokolovskii and Sokolovskaya (2003) point out that “all liparid larvae are characterized by the same order of fin formation” this may also be true of other characters (such as pores), but the size at which these characters develop differs among species (Grigor’ev, 2002). Variability in presence of these pores could be related to ontogeny of pore development, but little information is available about such development. Kido and Kitagawa (1986), Matarese et al. (1989), Grigor’ev (2002), showed or briefly described pores in liparid larvae or juveniles, but variability of appearance (and disappearance with growth) is unknown. In any case, t1 is clearly present on both sides in both specimens. In view of the small number of specimens known (four), it is possible that at small sizes (e.g., less than 40 mm SL), pores exist that later close and disappear. It is also possible that CAS 60515, which is also much larger than the other three specimens, represents a different species. In addition to the four specimens discussed above, Lloris and Rucabado (1991) described five recently hatched individuals 6.3-6.7 mm total length, and Moreno and Jara (1984) listed two more of 4.5-5.0 cm total length. I have been unable to examine any of these latter specimens to verify their identities or the cephalic pore pattern.

If the additional temporal pores normally occur in C. pallidus , I believe it would justify recognition of Enantioliparis as a valid genus. In that case, the diagnosis of the genus could be: “temporal pores three or four: an anterior (supraorbital) above and just behind orbit, with or without an intermediate about midway between supraorbital and first suprabranchial pores, suprabranchial pores two; coronal and postcoronal pores absent. Genital opening beneath a transverse fold of tissue, a posteriorly-directed genital papilla emerging from beneath it. Pectoral radials rounded, unnotched.”

Possession of the additional temporal pores suggests that C. pallidus may provide a shallow water link to the development of the hadal and abyssal liparid genera Pseudoliparis Andriashev 1955 , Notoliparis Andriashev 1975 , and Pseudonotoliparis Pitruk 1991 . All of these have numerous temporal, coronal, and postcoronal pores, considered by Andriashev (1998b) to be plesiomorphic characters similar to the arrangement in other cottoids.

Liparis antarctica Putnam 1874 is similar to C. crassus in many regards, including unnotched pectoral fin of about 30 rays, pale body and peritoneum color, low number of dorsal and anal fin rays, gill opening above pectoral fin, shape of teeth, and other characters. It differs most notably in having two nostrils, not mentioned by Putnam (1874), but clearly illustrated and described by Garman (1892 Pl. VI, Figs. 8, 10), and in apparently lacking the transverse anal fold (Garman 1892 Pl. VI, Fig. 9) described and shown by Andriashev (1997) and clearly evident in all the specimens listed above.