Pholidae, Gill, 1893

Koeda, Keita & Muto, Nozomu, 2019, An unexpected distribution record of the cold water fish Pholis fangi (Pholidae) from southern Taiwan, Zootaxa 4702 (1), pp. 87-93 : 88-91

Family Pholidae

Pholis Scopoli 1777

Pholis fangi (Wang & Wang 1935)

Figs. 1‒2

Enedrias fangi Wang & Wang 1935: 215 , fig. 39 (type locality: Chefoo, China); Yatsu 1981: 177.

Pholis fangi: Yatsu 1985: 281 ; Radchenko et al. 2010: 736; Sweetser & Hilton 2016: 1687.

Specimen examined. NMMB-P 28512, 131.5 mm SL, off Dong-gang (ca. 22°39ʹN, 120°24ʹE), Pingtung, South China Sea , Taiwan, 8 February 2018, obtained at fish-landing ground at Dong-gang.

Description of NMMB-P28512. Dorsal-fin rays 81; anal-fin rays II, 44; pectoral-fin rays 15; pelvic-fin rays I, 1; caudal-fin rays 8 (upper) +7 (lower) with 6 (upper) +5 (lower) procurrent rays; vertebrae 86 (36+50). Measurements: pre-dorsal-fin length 13.5% SL; pre-anus length 51.2%; body depth 13.0%; head length 14.0%. Snout length 18.0% HL; eye diameter 18.9%; upper jaw length 28.3%; interorbital width 14.9%; pectoral-fin length 61.3%; caudal-fin length 57.1%. Cephalic sensory canal openings: nasal series 2; interorbital pore 0; infraorbital series 6; anterior postorbital series + posterior postorbital series 3+3; occipital series 1-1-1; mandibular + preopercle series 4+5.

Body elongated, strongly compressed laterally, bar-like. Dorsal and ventral outline of body almost straight. Body and head covered with tiny scales. Lateral line absent. Head small, snout rounded. Mouth strongly oblique, forming an angle of ca. 60˚ to horizontal axis, lower jaw projecting slightly; maxilla extending posterior to vertical through anterior edge of orbit. Lips thick. Jaws with minute conical teeth, those on premaxilla and mandible in a single row with a patch anteriorly; teeth on maxilla few, widely spaced from each other; vomer and palatines toothless. Branchiostegal membranes on both sides joined.

Dorsal and anal fins low, bases very long; posterior end of dorsal and anal fins connected with caudal fin by membrane. Origin of dorsal fin above posterior tip of operculum; all dorsal fin rays spinous, similar length. Origin of anal fin at middle of body; first and second spines of anal fin shorter than soft rays. Pectoral fin rounded; base just below posterior tip of operculum. Pelvic fin tiny, its base anterior to pectoral-fin base. Caudal fin smaller than head, truncated with rounded uppermost and lowermost corners.

Fresh coloration. Body and head uniformly yellow, abdomen whitish; faint H-shaped dusky marking on dorsal and mid-lateral aspects; occipital, snout, lips, and ventral surface of lower jaw faintly dusky; fins generally pale yellow, with 13 equally-spaced short vertical white bands, margined with dark brown, on dorsal fin.

Distribution. Pholis fangi is currently known from the Bohai and Yellow seas ( Yatsu 1981; Yamada et al. 2007), newly collected from off southern Taiwan, northern South China Sea, but presently not known from the East China Sea.

Remarks. The morphological characters of the present specimen agree well with the key characters of P. fangi given by Yatsu (1981; as Enedrias fangi ), viz. interorbital pore absent, pectoral-fin length 61.3% of HL (vs. 58–64 % in the latter), and 13 equally-spaced short vertical white bands, margined with dark brown, on dorsal fin. All other count and measurement characters were also consistent with those given in previous studies ( Yatsu 1981; Yamada et al. 2007).

Within the amplified region of the COI gene sequence obtained from the present specimen, 569 bp were successfully aligned with those of 38 reference specimens (from INSDC or BOLD), including 14 specimens of P. fangi (Appendix 1). Five sequences from the latter 14 reference specimens were identical with that of the present specimen, suggesting their conspecificity. Furthermore, all of those 14 reference specimens and the present specimen together formed a monophyletic clade with a high bootstrap value (99) on a NJ tree ( Fig. 3 ).

The entire topology of the NJ tree was also largely consistent with the identification of the specimens. Each species formed a monophyletic clade with a high bootstrap value (99), except for P. picta (Kner 1868) and P. crassispina (Temminck & Schlegel 1845) , the former being paraphyletic with respect to the latter, with only a low bootstrap value (68) supporting the monophyly of the latter. Pholis fangi and P. nebulosa (Temminck & Schlegel 1845) together constituted a monophyletic clade (supported by a bootstrap value of 99), implying their status as closest relatives. This was matched by the proximity of their distributional ranges (the latter from the northwest Pacific coast southward from Hokkaido, Japan to the Bohai and Yellow Seas) as well as their morphological similarity (see below). Radcheko et al. (2012) also suggested that the two species may be the most closely related Pholis congeners, based on a phylogenetic analysis of mitochondrial Cytochrome b sequences (approximately 1,100 bp).

Pholis fangi is most similar to P. nebulosa but can be clearly distinguished in having a pectoral-fin length greater than 58% of HL (vs. less than 56% in P. nebulosa ), and a series of short vertical white bands, margined with dark brown, on the dorsal fin (vs. a series of triangular black markings) ( Yatsu 1981).

Based on a large-scale trawl survey of the Yellow Sea (including the Bohai Sea) and East China Sea, Yamada et al. (2007) reported that P. fangi was restricted to the former, and had at no time been collected in the East China Sea. In the Bohai and Yellow seas, specimens of P. fangi were generally collected in water temperatures below 13°C (mostly <10°C), except for two examples (collected in 16°C and 18°C). The World Ocean Atlas 2013 v.2 ( Boyer et al. 2013) noted that the water temperature in February (100 m depth) at 23° 5′N, 119°5′E (near the collection site), averaged over 8 years from 2005–2012, was 20.5, well above the known temperature range of the species. This indicates that the species may be able to tolerate higher-temperature waters than previously thought.

The previously-recognized southernmost record of P. fangi being from the southern Yellow Sea (33° N)( Yamada et al. 2007), the present specimen, collected off southwestern Taiwan, represents the first record of P. fangi and the family Pholidae from coastal Taiwanese waters, an extension of the southernmost record of the former by more than 1,000 km. It is also the first record of the typically temperate to subarctic family from subtropical waters ( Yatsu 1981).

It is uncertain whether the specimen in question represented a local population or was an example of rare dispersal from a more northern source (Bohai/Yellow Seas), not leading to population establishment. The lack of morphological and genetic differentiation between the former and those from the Bohai/Yellow Seas seems supportive of the latter explanation, although genetic divergence between Taiwanese and the Bohai/Yellow Sea populations may be inhibited by substantial gene flow or migration (seemingly unlikely given the absence of records from the East China Sea, despite intensive fisheries and ichthyofaunal surveys). In any case, the present report is a rare example of range extension of a cold water marine organism towards warmer latitudes, contrary to increasing recent observations of movements in the opposite direction, supposedly in relation to water currents (e.g., Koeda & Hibino 2017; Koeda et al. 2017; Koeda & Ho 2018a, b).

Figures

FIGURE 1. Lateral, dorsal, and ventral views (top to bottom) of a fresh specimen of Pholis fangi collected from southwestern Taiwan. NMMB-P28512, 131.5 mm SL.

FIGURE 2. Lateral and ventral views (top and bottom) of head of Pholis fangi collected from southwestern Taiwan. NMMB- P28512, 131.5 mm SL.

FIGURE 3. Neighbor joining tree based on partial (569 bp) sequences of the COI gene of Pholis fangi (NMMB-P28152) and other reference specimens. Node support indicated by bootstrap values (> 90 %) based on 1,000 replications. Sequence numbers in parentheses beside each species’ name. Scale bar represents the number of base differences per site based on the uncorrected p-distance.