Review of the Asian Thaumastodinae (Coleoptera, Byrrhoidea, Limnichidae), with a phylogeny of the genera

Abstract. The Asian species of the subfamily Thaumastodinae Champion, 1924 are reviewed. Seven new species are described: Acontosceles borneensis sp. nov., Pseudeucinetus papuanus sp. nov., Mexico ogasawaraensis sp. nov., M. baliensis sp. nov., M. papuanus sp. nov., M. palauensis sp. nov. and M. borneensis sp. nov. The genus Babalimnichus Satô, 1994 is treated as a junior synonym of the genus Mexico Spilman, 1972, and three known species of the genus Babalimnichus are transferred to Mexico, viz. M. taiwanus (Satô, 1994) comb. nov., M. masamii (Satô, 1994) comb. nov. and M. splendens (Hernando & Ribera, 2003) comb. nov. Additional specimen data are shown, and new distributional records are as follows: Acontosceles chujoi Yoshitomi & Satô, 2005 from Vietnam; A. zetteli Pütz, 2008 from Laos; Pseudeucinetus javanicus Yoshitomi & Putra, 2010 from Lombok Island; Mexico taiwanus (Satô, 1994) comb. nov. from Lutao, Lanhsu and the Yonaguni-jima Islands; and M. masamii (Satô, 1994) comb. nov. from Kume-jima. A species list of the subfamily Thaumastodinae is given, with ZooBank LSIDs. The phylogenetic relationships of the thaumastodine genera are discussed.


Introduction
Limnichidae Erichson, 1846 (minute marsh-loving beetles), a 'shore beetle' family (Jäch 1998), includes both riparian and littoral species, and has about 400 known species in 37 genera (Hernando & Ribera 2005a). This family is subdivided into four subfamilies (Limnichinae Erichson, 1846;Hyphalinae Britton, 1971;Cephalobyrrhinae Champion, 1925;and Thaumastodinae Champion, 1924) and has not been well studied taxonomically. Furthermore, the phylogenetic position of the family and its subfamilies is still not understood (Hernando & Ribera 2005a). The phylogenetic placement of Limnichidae and its possible non-monophyly has been considered previously by many authors, both informally (e.g., Hinton 1939;Crowson 1978) and with cladistic analyses (e.g., Lawrence 1988;Beutel 1995). Costa et al. (1999) mentioned the phylogenetic position of the family and its subfamilies, but their relationships remained uncertain. The monophyly of the family Limnichidae was not supported, but the families Limnichidae, Dryopidae and Heteroceridae constitute a monophyletic clade. Kundrata et al. (2017) analysed the molecular phylogeny of Byrrhoidea and Buprestoidea, and three subfamilies (all except

R e s e a r c h a r t i c l e
urn:lsid:zoobank.org:pub:4F834ADD-57E7-4CE3-B16A-127F1BE4AE14

Etymology
This species is named after its type locality.

Male
BODY. Oblong, subparallel-sided, slightly convex dorsally, weakly shiny, closely covered with suberect silver setae. Coloration of body blackish brown, but mouthparts and legs yellowish brown; elytra with obscure silver spots consisting of erect setae.
HEAD. Moderate in size; frons and clypeus closely covered with erect, long silver setae; clypeus with straight anterior margin. Eyes large, prominent dorso-laterally.
ANTENNAE. Reaching about proximal half of pronotum.

Redescription
For a detailed description of the external features, see the original description (Pütz 2008

Female
Unknown.

Remarks
This genus is easily distinguishable from other genera of the subfamily Thaumastodinae by the unique shape of the head: frons distinctly projecting anteriorly and narrow between the eyes.

Biological notes
The habitat of this genus ( Fig. 4A) is freshwater environments, i.e., rivers and streams. The adults are found in the splash zone of wet rocks, and frequently actively fl ying. In Java, Indonesia, Pseudeucinetus

Etymology
This species is named after its type locality.

Description
Male BODY. Oval, slightly convex dorsally, shiny, closely covered with suberect silver setae. Coloration of body black; antennae and legs blackish brown.
HEAD. Moderate in size; clypeus with straight anterior margin. Eyes large; distance between eyes about 3 eye facets.
ANTENNAE. Short, reaching about anterior third of pronotum.

Female
Unknown.

Remarks
This species is related to Pseudeucinetus novabritannica from the Bismarck Archipelago and P. solomonicus from the Solomon Islands, and differs from them in the following characteristics: 1) subapical inner projection of lateral lobe obvious (large and distinct in P. solomonicus); 2) subbasal inner projection of lateral lobe absent (present in P. novabritannica); and 3) basal piece long, about 0.7 times as long as lateral lobe (about 0.5 in P. novabritannica, about 0.8 in P. solomonicus).

Remarks
This is the fi rst record of this species from Lombok Island.

Remarks
The single female specimen labelled as having been collected in Brazil is questionable, because this genus is otherwise known only from the Oriental Region.

Remarks
As already mentioned by , Babalimnichus Satô, 1994 cannot be separated from Mexico, and it should be treated as a junior synonym of the latter.

YOSHITOMI H., Asian Thaumastodinae
The male genital structures are very peculiar, i.e., with very short lateral lobes with apical setae, longer median lobe and the presence of ventral plates (Fig. 1A).

Remarks
This species appears to be closely related to M. baliensis sp. nov. based on the shape of the aedeagus, but differs from it in the following characteristics: sternite IX broad (slender in M. baliensis sp. nov.), median lobe pointed at apex (somewhat obtuse in M. baliensis sp. nov.) and apical setae on lateral lobes long (short in M. baliensis sp. nov.).

Biological notes
The type series was collected from the splash zone of wet rocks on a rocky seashore (Fig. 4C).

Remarks
This species is similar to M. borneensis sp. nov. in the shape of the aedeagus, but differs from it in the following characteristics: median lobe obtuse at apex (pointed in M. borneensis sp. nov.); and apical setae on lateral lobes relatively short (long in M. borneensis sp. nov.).

Remarks
This species is similar to M. borneensis sp. nov. in the shape of the aedeagus, but differs from it in the following characteristics: ventral plates separated from median lobe (attached in M. borneensis sp. nov.) and apical setae on lateral lobes short (long in M. borneensis sp. nov.).

Etymology
This species is named after its type locality.

Redescription
Male For a detailed description of the external features, see the original description (Satô 1994 Sternite VII (Fig. 17A) with two pairs of long, stout extra setae, bearing about 20 long setae, bifi d in postero-lateral corners, gently projecting triangularly in median part of caudal margin. Sternite VIII (Fig. 17B) small, slightly sclerotized, Y-shaped. Sternite IX (Fig. 17C) slender, lightly curved in basal part; apical plates long and slender, pointed at apices. Aedeagus (Fig. 17D-F) relatively short, asymmetrical; basal piece oval; lateral lobes short, with 2-4 apical setae; median lobe a little wide, sparsely punctuate in apical part, slightly expanded in apical quarter, pointed at apex; ventral plates relatively slender, separated from median lobe, left one with thumb-like projection on inner margin of apex; ML / BL 2.42; ML / LL 2.42.

Remarks
This is the fi rst record of this species from Lutao and Lanhsu Islands, off Taiwan, and Yonaguni-jima. (Satô, 1994)

Redescription
Male For a detailed description of the external features, see the original description (Satô 1994 Sternite VII (Fig. 18A) bearing about 30 long setae, lacking extra setae; postero-lateral corners short and pointed; median part of caudal margin gently projecting triangularly. Sternite VIII (Fig. 18B) small, slightly sclerotized, Y-shaped. Sternite IX (Fig. 18C) well sclerotized, curved in basal part, shallowly concave at apex; apical plates rather stout, pointed at apices. Aedeagus (Fig. 18D-F) relatively short, asymmetrical; basal piece oval; lateral lobes short, with 3-4 apical setae; median lobe wide, sparsely punctuate in apical part, expanded in apical quarter, gently pointed at apex; ventral plates wide, separated from median lobe, left one long and curved interiorly in apical part, right one short and straight; ML / BL 2.24; ML / LL 2.43.

Remarks
I re-examined the specimens from Yonaguni-jima (Yoshitomi & Arai 2004), and it is clear that they do not represent this species but the preceding one. This is the fi rst record of this species from Kume-jima.
This species has the northernmost distribution (Sado Island) in the subfamily Thaumastodinae (Kamezawa 2017).

Distribution
Tonga Islands.

Remarks
This species was described based on the female holotype. No additional specimens have been found.

Discussion
In the present paper, I have reviewed the Asian species and genera of the subfamily Thaumastodinae, and recognized three genera and 25 species (adding seven new species) from this area.
Acontosceles is an Oriental genus, living in freshwater habitats. Some unnamed species were known (Pütz 2008) because the distributional area of each species is limited and some populations were represented by female specimens only. Some new species will be described in the near future.
Pseudeucinetus is also an Oriental genus, living in freshwater habitats. The adults of this genus are frequently attracted to light, and the distributional area of P. zygops is wide, from India to Indonesia.
Mexico is known as a Neotropical genus, and it is newly recorded from Asia in this study. The habitat of this genus is rocky seashores, and the distributional area is mainly the Oriental and the Neotropical, extending to the Palearctic (coastal area of Japan) and the Australian (Tonga Islands) Regions. The pattern is probably not a disjunct distribution, but rather a wide distribution in the Pacifi c. Undescribed species are expected to be found from unrecorded areas and countries, e.g., Indochina, the Philippines, many islands of Indonesia, Australia, New Caledonia and the Pacifi c Islands.
Martinius is a Neotropical genus, represented by three species. This genus shows a sister group relationship with the genus Mexico (Fig. 20).
Taking into consideration the factors mentioned above, Thaumastodinae does not seem to be an example of disjunct distribution. It is probable that the main distribution area of the subfamily is the Oriental Region, and it has spread to parts of the Pacifi c, the Palearctic, the Australian and the Neotropical areas.
In a preliminary phylogenetic analysis using morphological characters of nine genera in four subfamilies, the monophyly of the family Limnichidae was not supported (unpublished data). In addition, the monophyly of the subfamily Limnichinae, the most diverse subfamily within the family, was also doubtful as already mentioned by Hernando & Ribera (2005a). This subfamily is subdivided into three tribes, Bothriophorini, Limnichini and Wooldridgeni, and one genus group (Mandersia group, sensu Hernando & Ribera 2005b) in Limnichini. Most of the genera in the subfamily Limnichinae are classifi ed in the tribe Limnichini without a tribal defi nition. It is suggested that the family Limnichidae is a paraphyletic group, and the defi nition of the family must be reconstructed, including related families such as Heteroceridae (Crowson, 1978), Chelonariidae (Kundrata et al. 2017), or Dryopidae (Costa et al. 1999).