Gymnothorax pseudokidako, Huang & Smith & Loh & Liao, 2021

Huang, Wen-Chien, Smith, David G., Loh, Kar-Hoe & Liao, Te-Yu, 2021, Fig. 4 in Periclimenaeus apomonosi Park & De Grave 2021, sp. nov., Zoological Studies (Zool. Stud.) 60 (24), pp. 1-17 : 14-16

Gymnothorax pseudokidako

Gymnothorax pseudokidako might be confused with G. kidako , a morphologically similar species distributed in the North-West Pacific from Taiwan to Japan and sympatrically occurring with G. pseudokidako in northern Taiwan. The most important difference in the coloration pattern between the two species is that the white margin of anal fin is absent in G. pseudokidako but present in G. kidako . Moreover, the color of blotches on G. pseudokidako is creamy white, whereas the blotches are often bright-yellowish on G. kidako . The plainly colored lower jaw and throat is also a diagnostic character of G. pseudokidako , whereas G. kidako usually has a mottled lower jaw and throat ( Fig. 9 ). In morphometric and meristic characters, G. pseudokidako can be distinguished from G. kidako by having a relatively short tail (50.5–53.0% vs. 52.9– 56.4% of TL), more dentary teeth (17–26 vs. 16–20), and fewer total vertebrae (134–139 vs. 137–143) ( Table 3 and Fig. 12 ). In molecular analysis, the reciprocal monophyly of G. pseudokidako and G. kidako is not supported by the topology of the COI tree. A similar phenomenon can be observed in a number of sibling muraenid species, e.g., Gymnothorax griseus ( Lacepède, 1803) vs. Gymnothorax thyrsoideus ( Richardson, 1845) , and Gymnothorax margaritophorus Bleeker, 1864 vs. Gymnothorax pharaonis Smith, Bogorodsky, Mal and Alpermann, 2019 ( Smith et al. 2019). By contrast, the more conservative nuclear EGR3 gene shows reciprocal monophyly between G. pseudokidako and G. kidako ( Fig. 11B ). Despite the mito-nuclear discordance, the more conservative, but separable EGR3 gene highly supported G. pseudokidako as a separated species. Discordance between COI and other molecular markers is not a rare phenomenon in marine fishes. For instance, the eight tuna species of genus Thunnus can be well differentiated exclusively by the mitochondrial control region ( Viñas and Tudela 2009); damselfishes Abudefduf sexfasciatus and A. vaigiensis are indistinguishable in mitochondrial COI and cytochrome b genes but found to be distinct based on nuclear genes (COI sequences published on BoldSystems; Bertrand et al. 2017). The conflicting results between molecular markers may attribute to different evolutionary histories of genes.

Gymnothorax mucifer and G. niphostigmus are also sympatric species with G. pseudokidako in that they have similar coloration patterns and overlap in most of their morphometric measurements and meristic counts ( Table 3). However, G. pseudokidako can be easily distinguished from both species by its lack of the white margin of the anal fin, and having a brown saddle-like marking on top of head (vs. dense pale spots on top of head). Gymnothorax pseudokidako further differs from G. niphostigmus in the number of total vertebrae (134–139 vs. 140–146). ML trees of COI and EGR3 genes also support G. pseudokidako is a different species from G. mucifer and G. niphostigmus ( Fig. 11 ). Lastly, G. pseudokidako is apparently different from M. similis , a synonym of G. kidako , by the lack of white margin on the anal fin, a shorter tail (50.5–53.0% vs. 54.5% of TL), a longer head (12.6–14.3% vs. 12.0% of TL), and more dentary teeth (17–26 vs. 13–14), although the vertebrae count is not available from the holotype of M. similis ( Böhlke and Smith 2002) . Based on morphological and molecular evidence, G. pseudokidako is clearly a new species well separated from G. kidako and other congeners.

Sexual dimorphism in the dentition, e.g., females and immatures have an additional inner row of maxillary teeth but is lost in mature males, has been reported in several muraenids, including two pale-spotted species: Gymnothorax baranesi Smith, Brokovich and Einbinder, 2008 and G. mucifer ( Smith et al. 2008; Huang et al. 2019). Gymnothorax niphostigmus was also reported to have 1–2 inner maxillary teeth in smaller individuals ( Chen et al. 1996). However, the sexual dimorphism of dentition is not observed in G. pseudokidako or G. kidako . The inner row of maxillary teeth is absent in all mature males and females, except for the smallest paratype of G. pseudokidako (ASIZP0080929, a 608 mm female), which has two inner teeth on each side. Furthermore, the number of teeth is neither related to sex nor total length (data not shown, see Fig. 12B for example). Based on our observation, no dental change can be found in G. kidako ; for G. pseudokidako , all small individuals might have an inner row of maxillary teeth and would be lost when growing larger regardless of sex. Thus the dental change in G. pseudokidako is more likely to be ontogeny-dependent rather than sexdependent.

Figures

Fig. 9. Comparison of coloration patterns between (A, C, E, G) Gymnothorax pseudokidako sp. nov., and (B, D, F, H) G. kidako. (A–B) lateral view; (C–D) lateral view of head; (E–F) dorsal view of head; (G–H) lateral view of tail. Arrows point out the origin of dorsal fin. (A, C, G) ASIZP0080920, holotype, 801 mm TL; (E) ASIZP0080924, paratype, 913 mm TL; (B, D, H) DOS 06258, 631 mm TL; (F) DOS 08048-1, 840 mm TL.

Fig. 12. Relationship of selected characteristics of Gymnothorax pseudokidako sp. nov. and G. kidako. (A) tail length in % of TL to TL; (B) number of dentary teeth to TL; (C) frequency distribution of total vertebrae. Black for G. pseudokidako; white for G. kidako. Teeth counted from each side of the jaw, two counts for each individual.

Fig. 11. Maximum likelihood trees of Gymnothorax pseudokidako sp. nov. and closely-related species, with Uropterygius macrocephalus as the outgroup. (A) partial mitochondrial COI gene sequences (666 bp) based on the HKY + I model; (B) partial nuclear EGR3 gene sequences (767 bp) based on the GTR + I model. Numerals beside the internal branches are bootstrap values.