Uca (Xeruca) formosensis Rathbun, 1921

(Figs. 2–8)

Uca formosensis Rathbun, 1921: 155; Maki & Tsuchiya 1923: 205, pl. 23(4); Barnwell 1982: 79; Shih 1988: 105, fig. 9, pls. 11, 12; 1997: 68, figs. 5, 17, 41; 2000: 74, 4 unnumbered figs; Fukui et al. 1989: 227, fig. 3; Huang et al. 1989: 199, fig. 8, pl. 4D–F; Wang & Liu 1996a: 50, 1 unnumbered fig.; 1996b: 73, figs. 66–70; 1998: 73, figs. 66–70; 2003: 73, figs. 66–70; Ho & Hung 1997: 54; Shih et al. 1999: 164, figs. 1–4; Hung 2000: 140, fig. 440; Lee & Tung 2000: 34, 38, 7 unnumbered figs; Chen 2001: 202, fig. 637a, b; Lee 2001: 102, 3 unnumbered figs; Ng et al. 2001: 37; Rosenberg 2001: 848, 851, 860, 866; Shen & Jeng 2005: 164, 2 unnumbered figs; Beinlich & von Hagen 2006: 25; Wang 2009: 107 –109, 2 unnumbered figs; Liu & Wang 2010: 37, 5 unnumbered figs; Lee et al. 2013: 68, 2 unnumbered figs.

Gelasimus formosensis — Sakai 1939: 620, text-fig. 94b, pl. 105(1); Horikawa 1940: 28; Lin 1949: 26.

Uca (Thalassuca) formosensis — Crane 1975: 83, fig. 63C, pl. 14A–D; Sakai 1976: 604, text-fig. 331; Su & Lue 1984: 64, fig. 5; Dai et al. 1986: 425, fig. 235, pl. 59(1); Dai & Yang 1991: 465, fig. 235, pl. 59(1); Shih 1994: 82, figs. 15, 16, 56–59; Ho 1996: 9, figs. 1–7.

Uca (Tubuca) formosensis —Ng et al. 2008: 242.

Material examined: holotype ♂ (28.8 mm), paratype ♂ (27.6 mm), USNM 54472, Rokko (= Lugang), Changhua, Taiwan, coll. Taihoku Normal School, Aug. 1919. Others: see Appendix 1. Comparative material: see Appendix 2.

DNA analyses and discussion. A 578 bp segment of the 16S, 658 bp segment of COI and 633 bp segment of 28S from 26 species of fiddler crabs were amplified and aligned (Table 1). The phylogenetic tree of the combined markers was reconstructed from Bayesian inference analysis, with the support values from the maximum likelihood analysis (Fig. 9). It is clear that most subgenera (including Xeruca subgen. nov.) are highly supported by both methods. Xeruca is sister to Tubuca and Australuca. While Australuca is monophyletic, Tubuca becomes parapheletic. The “ Tubuca + Australuca ” clade, however, has high support. Gelasimus is sister to Cranuca, at least highly supported by Bayesian inference method. The clade of Austruca is monophyletic, has a high support value by Bayesian inference, with medium support by maximum likelihood. The American subgenus Uca is closely related to Afruca from eastern Atlantic.

After Crane (1975), the systematics of fiddler crabs has been revised by Rosenberg (2001) and Beinlich & von Hagen (2006), with the IWP broad-fronted groups studied by Shih et al. (2009a, 2013b) and Naderloo et al. (2010). Uca formosensis was established by Rathbun (1921), but only less than a dozen museum specimens could be examined until the 1980s (Crane 1975: 83; Barnwell 1982) and Crane mentioned it as an “elusive species.” Shih et al. (1999) revised and updated the taxonomy and morphology of the species, as well as adding ecological information after finding its preferred habitat, and examined more than 400 specimens. They suggested this species is closer to Uca (Tubuca), instead of the original designation under Uca (Thalassuca) Crane, 1975 (= Uca (Gelasimus) Bott, 1973), but notes the need for more evidence to confirm the status. Rosenberg’s (2001) detailed morphological analyses on fiddler crabs also found that U. formosensis does not clearly belong to any of the known subgenera, although closer to Tubuca and Australuca.

TABLE 1. Haplotypes of 16S rDNA, COI and 28S rDNA for the subgenus Xeruca subgen. nov. and other related subgenera used in this study. See Material and methods for abbreviations of institutions.

Madagascar: Sarodrano ZRC THH 04-17 AB535405 LC053377 LC053396 Cranuca U.inversa Kenya: Gazi MZUF 1024 AB813658 AB813674 AB813703 Ghana: Elmina 13654 LC053362 LC053380 LC053399 The new subgenus of Xeruca is established herein for Uca formosensis based on a suite of morphological characters, including carapace, major chela, handedness, G1 and gastric mill, which are supported by the phylogenetic analyses using mitochondrial and nuclear markers. Previous studies on the morphology showed that U. formosensis is closer to Tubuca and Australuca (Shih et al. 1999; Rosenberg 2001), which is also revealed by the molecular evidence (Fig. 9). Xeruca can be easily separated from Tubuca by the carapace shape (Fig. 4) and the fingers of the chela (Fig. 5). Although Xeruca is more similar to Australuca, the latter can be distinguished by its smaller adult size, relatively shorter dorsolateral margins, and the possession of a longer gape between the closed fingers of the adult major chela (Figs. 4, 5).

The molecular evidence (Fig. 9) shows that the subgenera Tubuca and Australuca form an unresolved clade in conflict with the present taxonomy based on morphological characters (Crane 1975; Rosenberg 2001), although the Australuca species belong to their own clade. It is likely that Australuca Crane, 1975, is a synonym of Tubuca Bott, 1973. Further studies that include more species into the analysis may clarify the problem.

The morphology of gastric mills is related to different functions of feeding and seems unsuitable for phylogenetic study. This may be true for the lateral gastric teeth, but not for the foregut ossicles (Brösing & Türkay 2011), which has been reflected by other studies: e.g. Yang (1986) on various crabs, especially Macrophthamidae; Sakai et al. (2006) on the species of the Helice and Chasmagnathus complex; Brösing (2010) on several brachyuran families; and Naderloo et al. (2010) on the Uca lactea species complex. Based on the systematic study of the morphology of the gastric mills of IWP Uca species (in preparation), some groups have unique characters and could be used for the systematics of fiddler crabs. As mentioned above, there are several consistent characters for each of the following subgenera: Gelasimus, Uca, and Afruca (Fig. 8).

Uca formosensis was published in 1921, however, its systematic position was uncertain, either in Uca (Gelasimus) (Crane 1975) or Uca (Tubuca) (Shih et al. 1999). The morphological analyses by Rosenberg (2001) have shown this species is substantially different from other species and should be placed in its own subgenus. According to the evidence of morphology and DNA sequences of this study, as well as the peculiar ecology and reproduction behavior (Shih et al. 2005), the establishment of a new subgenus Xeruca is appropriate to solve the uncertain position of this species.

The monotypic Xeruca becomes a subgenus with the narrowest distribution among the known subgenera of fiddler crabs. Its distribution restricted to Taiwan and the offshore Penghu Islands also implies the cladogenesis of this subgenus is related with the geological history of Taiwan Island during early Pliocene (reviewed by Shih et al. 2006, 2009b, 2011) and more studies are necessary to clarify it. Shih et al. (1999) have suggested this species to be included in the list of protected species of the Wildlife Conservation Law, R.O.C., because several habitats have been destroyed by industrial development, illegal fishing ponds, and mangrove overgrowth (Fig. 1; Shih et al. 1999). However, this species has not be considered as a protected species yet, although some populations of U. formosensis have been under different pressures, e. g. the Penghu population has become extinct in 2008 after the inappropriate replanting of introduced mangroves for about 15 years; and an unthoughtful ecological engineering on the habitat of Shengang, Changhua in 2006 nearly exterminated the local population (Shih 2008a, b). The conservation policy and management for U. formosensis have to be reconsidered by government of Taiwan, because it is not only an endemic species to Taiwan, but also an endemic subgenus of fiddler crabs.