Hexapinus Manning & Holthuis 1981

Polyphyly of Hexapinus Molecular phylogenetic analyses conducted in this study clearly showed the polyphyly of Hexapinus based on sequences of Hn. okinawa sp. nov., Hn. patuma and Hn. simplex (Fig. 5), with 14.4–14.7% K2P distance between Hn. okinawa sp. nov. + Hn. simplex clade and Hn. patuma singleton (Table 2). This generic distance is close to those between the Hn. okinawa sp. nov. + Hn. simplex clade and species of other genera: 12.6–15.1% from Rayapinus maenosonoi, 14.6–16.4% from Mariaplax chenae and 16.5–17.0% from M. narusei (Table 2). Our detailed morphological examinations of the above three species, as well as of Hn. latipes, the type species of Hexapinus, and Hn. latus, revealed that the species can be divided into two morphological groups; the Hn. latipes group composed of Hn. latipes, Hn. okinawa sp. nov., and Hn. simplex, and the Hn. patuma group including Hn. patuma, Hn. latus and probably Hn. ceres. In the present molecular phylogeny, the Hn. latipes group is supported by 97% bootstrapping value containing Hn. latipes, Hn. okinawa sp. nov. and Hn. simplex, although the Hn. patuma group was recovered as paraphyletic (Fig. 5). The former group can be distinguished from the latter by the following characters: epistome posterior margin laterally keeled and interrupted medially, without a median tooth in Hn. latipes group (Figs. 2B, 3D, E; Rahayu & Ng 2014: fig. 12C) (versus epistome posterior margin entirely keeled, with a median tooth in Hn. patuma group, Fig. 4C, E); carapace posterolateral margin laterally produced irregularly, forming subtrapezoidal protuberance in Hn. latipes group (Figs. 1A, 2A, 3A–C) (versus carapace posterolateral margin lacking such protuberance in Hn. patuma group, Fig. 4A, B); male pleon distinctively slender and elongated, G1 slender, simply curved anterolaterally in Hn. latipes group (Rahayu & Ng 2014: fig. 21A, B, E–G; Ng & Rahayu 2015: figs. 5A, B, 6A, B) (versus having stout male pleon and male G 1 in Hn. patuma group, Naruse 2020: fig. 3B–D); female sternopleonal cavity anteriorly closed at the border of thoracic sternite 3/ 4 in Hn. latipes group (Figs. 1D, 3F, G; Ng & Rahayu 2015: fig. 5H) (versus female sternopleonal cavity abruptly narrowed at anterior part of thoracic sternite 4, anteriorly with a distinct, short, narrow longitudinal groove reaching to thoracic sternite 3 in Hn. patuma group, Fig. 4D, F). Although no specimen of Hn. ceres was available for this study, the photographs by Rahayu & Ng (2014: fig. 14B, D) indicate a short longitudinal groove at the anterior part of thoracic sternum in female holotype, which is analogous to Hn. patuma group. Species belonging to the Hn. patuma group, show morphological similarities to some Mariaplax Rahayu & Ng, 2014 and Rayapinus Rahayu & Ng, 2014 in taxonomically important characters (e.g., male pleon, G1, and female sternopleonal cavity) (T. Sato, unpublished data). Our molecular phylogenetic analyses also indicated that Mariaplax itself is polyphyletic (Fig. 5). While M. chenae and M. narusei are in the same clade (99% bootstrapping value), M. ourabay shared a clade with R. maenosonoi in relatively low generic divergence (9.1%; Table 2). The phylogram calculated in this study, however, lacks high bootstrap values at deeper clades (Fig. 5). Additional taxa sampling and more gene sequences are needed to create a more reliable tree for a comprehensive revision of these genera, but such a large-scale revision is beyond the scope of this study. Our ongoing revision will provide a better hypothesis of the systematics of these genera in the future.