Seleniolycus Anderson

Peter R. Møller & Andrew L. Stewart, 2006, Two new species of eelpouts (Teleostei, Zoarcidae) of the genus Seleniolycus from the Ross Dependency, Antarctica., Zootaxa 1376, pp. 53-67 : 61-65

[[ Genus Seleniolycus Anderson ]]

Key to the species of Seleniolycus

1a Body entirely naked; predorsal distance 20-23.8% SL; suborbital head pores 5; anal fin rays 53-58 ....................................................................... Seleniolycus laevifasciatus

1b Body scaled posteriorly; predorsal distance> 27% SL; suborbital pores 6 or 7; anal fin rays 64-73 .................................................................................................................... 2

2a Pectoral fin rounded with no exserted rays; distance from snout to anterior scales 48.9-55.4% SL; head length 16.1-18.8% SL; anal fin rays 69-73 ................................ ........................................................................................................ Seleniolycus robertsi

2b Pectoral fin wedge-shaped with the lower fin rays thickened and exserted; distance from snout to anterior scales 58.7-61.8% SL; head length 19.8-24.6% SL; anal fin rays 64-66 .................................................................................. Seleniolycus pectoralis

Discussion

Seleniolycus has been proposed as the sister taxon to the widespread mesopelagic genus Melanostigma Guenther, 1881 (Anderson, 1988, 1994), indicating a southern hemisphere origin of the lineage. Morphology of this clade was supported by seven homoplastic character states: squamation absent, parasphenoid wing low, frontal-parasphenoid articulation separated by pterosphenoid, interorbital pore absent, palatopterygoid series reduced, upper pharyngeals 2, number of vertebrae 72-105 (see Anderson, 1994). Although it is uncertain if all the osteological characters states of the two new species agree with the states given above, the lack of an interorbital pore, vertebrae number, and upper pharyngeals agree with Anderson’s (1994) observations. However, scales present posteriorly in both new species conflict with Anderson’s scale character, representing a new character state in this clade. This type of squamation, with scales on the posterior half of the body only, is found in several unrelated zoarcid lineages and is thought to be an intermediate, reduced, apomorphic state. In future studies, polarization of squamation should be treated as multi-state and include the half scaled condition. The presence of posterior squamation supports the interpretation that these two new species are the least reduced, basal-most branches in the Seleniolycus-Melanostigma linage (sensu Anderson, 1988, 1994), both having larger gill slits and more head pores than S. laevifasciatus and Melanostigma species.

As noted above, Seleniolycus is endemic to the Southern Ocean. The type species S. laevifasciatus is known from approximately 26 specimens collected from the South Shetland Islands (~70° W), the Falkland Rise, Scotia Arc to the Banzare Bank, and south of the Kerguelen Plateau in the eastern Antarctic (~80° E) (Anderson, 1994, 2006). Anderson (1994, p. 45) suggested that S. laevifasciatus was probably benthopelagic and circumantarctic in distribution, possibly extending north into the Subantarctic (Anderson, 2006, p. 13). In spite of circumantarctic currents being favourable to the distribution of S. laevifasciatus via larval transport into the western Antarctic, none have been taken by toothfish droplines in this sector. Several trawl-based expeditions by research vessels in the Ross Sea and adjacent regions (e.g. Eastman & Hubold, 1999; Mitchell & Clarke, 2004) have also failed so far to capture any specimens of S. laevifasciatus . Therefore, available data indicates that S. laevifasciatus has a distribution restricted to eastern Antarctic seas.

In spite of the two new species described herein being caught at greater depths than Seleniolycus laevifasciatus , the upper depth range of S. robertsi (1455 m) overlaps the lower depth range of S. laevifasciatus (1500 m). The capture method of Seleniolycus robertsi and S. pectoralis (bottom longline with baited hooks on traces ~ 1 m long) suggests these new zoarcids are closely associated with the benthos, probably benthic, rather than benthopelagic. In this respect, they differ from their possible sister group Melanostigma , which are pelagic, except during benthic spawning (Markle & Wenner, 1979; Silverberg et al., 1987).

Geography and isolation of the Pacific-Antarctic Ridge suggest these two new species of Seleniolycus may be endemic to the ridge. This hypothesis is supported be Springer (1982) who showed that the Pacific Plate margins were areas of high endemism. The Pacific-Antarctic Ridge forms the southernmost edge of the Pacific Plate and separates the Pacific and Antarctic plates. It rises up abruptly out of the abyssal plains of the Australian-Antarctic Basin to the west, the Amundsen Abyssal Plain to the south, and the Pacific Basin to the north. It is also a region of high geological activity; an active spreading centre since the Late Cretaceous, currently at ~54-74 mm/yr (Mayers et al., 1990; Géli et al., 1997), and crossed by numerous fracture zones (Marks & Stock, 1994). These complex features have combined to create deep-sea habitats high in morphological diversity (e.g. Ondréas et al., 2001), which probably support high biological diversity.

However, remoteness, prevailing bad sea conditions, substantial depth, and broken bottom topography have all contributed to a paucity of biological sampling and museum specimens from the ridge, and indeed from Antarctic bathyal and bathypelagic depths (1000 to 4000 m) in general (e.g. Eastman, 1993, p. 51; Møller et al., 2005, p. 31). Analyses of Antarctic marine ichthyofauna have focused on the origins, evolution and radiation within the Southern Ocean, or biogeography of the shallow inshore marine fauna (0-1400 m depth) (e.g. Anderson, 1990; Briggs, 2003; Clarke & Johnson, 1996; Eastman & McCune, 2000). Further collecting is required along the Pacific-Antarctic Ridge and adjacent areas to test the endemism hypothesis of this southernmost Pacific Plate boundary feature, and to better determine the extent of distribution of these two new species.