The genus Nipponodrasterius Kishii (Coleoptera, Elateridae, Agrypninae), a junior synonym of the genus Gamepenthes Fleutiaux (Coleoptera, Elateridae, Elaterinae), with review of the Japanese Gamepenthes species

Abstract The genus Nipponodrasterius Kishii, 1966 was established as a member of the subfamily Conoderinae Fleutiaux, 1919 (now tribe Oophorini Gistel, 1848; subfamily Agrypninae Candèze, 1857) based on N. alpicola. The genus was suggested to be unlikely to belong to Agrypninae because it lacks diagnostic features of the Agrypninae. However, there are no taxonomic treatments for the genus or species. Here, we review the status of the genus and species by examining the holotype of N. alpicola. Consequently, the genus was found to be a junior synonym of the genus Gamepenthes Fleutiaux, 1928 and N. alpicola was found to be a junior synonym of G. pictipennis (Lewis, 1894). We review all species of Gamepenthes in Japan and provide a key to species.


introduction
The monotypic elaterid genus Nipponodrasterius Kishii, 1966 was established as a member of the subfamily Conoderinae Fleutiaux, 1919(currently tribe Oophorini Gistel, 1848 subfamily Agrypninae Candèze, 1857) based on Nipponodrasterius alpicola Kishii, 1966. This species is known only from the holotype collected on a mountain (altitude 2800 meters) in central Honshu, Japan. Its sex was not determined in the original description. The original description separates the genus from known members of tribe Oophorini using the following characteristics: rather flat body, antennomeres II and III small, antennomeres IV-X clearly serrate, elytral surface clearly granulated, and tarsomere IV shortly bilobed. Kishii (1987) redescribed this genus, modifying the original description slightly. He changed the description "tarsomere IV shortly bilobed" in Kishii (1966) to "tarsomere IV slightly enlarged apically". Moreover, he added an important character state to the generic description, claws without basal setae. Ôhira (1994) focused on the status of the claws and suggested that this genus is unlikely to belong to Agrypninae, because the presence of basal setae on the claws is a diagnostic feature of members of Agrypninae, absent only in genus Danosoma Thomson, 1859(Hayek 1973. Ôhira (1994) did not examine specimens of the genus directly and did not revise the status of the genus.
We observed that N. alpicola is similar to species of the genus Gamepenthes Fleutiaux, 1928 in the tribe Megapenthini Gurjeva, 1973, subfamily Elaterinae Leach, 1815 based on these diagnostic features and a drawing of the dorsal habitus in the original description. In considering the systematic position of N. alpicola, the direction of the mouth-parts and the structure of the head capsule, procoxal cavity, prosternal process, mesocoxal cavity and mesosternum are also important characters (Ôhira 1970a;Stibick 1979;Schimmel 2003), however, these features were not described in Kishii (1966Kishii ( , 1987. The holotype of the type species required re-examination. Five Gamepenthes species are distributed in Japan: G. ornatus (Lewis, 1894); G. pictipennis (Lewis, 1894); G. similis (Lewis, 1894); G. versipellis (Lewis, 1894); G. yoshidai Ôhira, 1995. Ôhira (1995a) reviewed the four Japanese species, except for G. yoshidai, and showed that they are clearly separated from each other by their aedeagi. Ôhira (1995a) provided a key to distinguish them, based mainly on body color, although members of the genus often show color variation. Subsequently, Ôhira (1995b) described a new species as G. yoshidai without showing its aedeagus. Hence, we examine the holotype of N. alpicola and review the Japanese Gamepenthes species including G. yoshidai in order to resolve the problem of Nipponodrasterius. Additionally, we provide a new key to these species based on external morphology.

Material and methods
We examined the holotype of N. alpicola and non-type specimens of the five Japanese Gamepenthes species, as well as the only paratype of G. yoshidai. The type specimens are in the collection of the Osaka Museum of Natural History (OMNH; Osaka, Japan).
The non-type specimens are in the personal collection of the authors (Osaka, Japan) and will be donated to the OMNH in the future. Unique identifier numbers of the non-type specimens are: GO001-GO014; GP001-GP012; GS001-GS010; GV001-GV017; GY001, GY002. We found a slide of the male genitalia of the holotype of N. alpicola in OMNH (Fig. 5L), although its sex and genitalia were not mentioned in the original description. On the slide, the aedeagus of the holotype, missing the phallobase, had been mounted in balsam and distorted due to pressure exerted by the cover slip (Fig. 5J).
The methods used for observing and dissecting specimens, taking photographs, creating line drawings, and depositing the dissected parts follow Arimoto and Suzuki (2020).

Measurements and indices
Measurements and indices were made following Arimoto and Suzuki (2020 Antennae serrate from antennomere IV, without median longitudinal carina; antennomere IV longer than II-III combined in many, shorter than II-III combined in a few. Pronotum with median basal furrow, without sublateral incision near hind angles; hind angles unicarinate. Pronotosternal sutures not grooved or very shallowly grooved in front. Posterior edge of hypomeron straight mesally and then broadly rounded. Procoxal cavity partly closed behind by mesal projection of hypomeron. Prosternal process concave between procoxae, with subapical tooth in lateral view. Side of scutellum parallel on anterior half. Mesocoxal cavity open to mesepimeron and mesepisternum. Mesosternum separated by suture from metasternum. Outer edge of metacoxal plates wide in most, but narrowed in a few. Elytral surface with rasp-like punctures; elytral apical edge serrate and with small spines, serration very slight in some and then almost rounded. Tarsi simple; claws simple, without basal setae. Ôhira (1970a, 1995a) stated that the apical elytral edge is more or less truncate; however, this is based on a misunderstanding.
Ecology. Adults of the genus are often observed visiting flowers during the daytime. In Japan, Gamepenthes has been recorded visiting the flowers of Sambucus sp.  (Ôhira, 1995a). This study adds flowers of Tilia japonica (Miq.) Simonk. (Malvaceae) to the records.
While most adult Gamepenthes individuals are not attracted to lights at night, G. similis has been collected by light traps (Ootsuka et al., 1981). We examined a few specimens of G. ornatus and G. pictipennis collected using simple light traps made with weak fluorescent lights.
Included species from Japan. Five species: G. ornatus (Lewis, 1894); G. pictipennis (Lewis, 1894); G. similis (Lewis, 1894); G. versipellis (Lewis, 1894); G. yoshidai Ôhira, 1995. Key to species of the genus Gamepenthes from Japan  . Prothorax black in male, brown in female. Each elytron with pale orange U-shaped marking basally. Antenno-mere II shorter than wide in male, almost as long as wide in female; III shorter than wide in male, longer than wide in female; IV robust (1.36-1.53 times longer than wide in male, 1.69-1.79 times longer than wide in female), longer than II-III combined (1.53-1.99 times as long as II-III combined in male, 1.02-1.10 times as long as in female). Prosternal process weakly inclined dorsad (24-38°). Metacoxal plates not narrowed outwards on lateral half; outer edge wide. Apical edge of elytra serrate. Apex of parameres beyond pre-apical expansions equilateral triangular (apex length 0.8-0.9 times width of parameres at expansion).
Measurements. Male (6 spec.  (Lewis, 1894) are often found sympatrically and are remarkably similar. They are distinguished by apical expansion of parameres beyond apical-lateral hooks (G. ornatus, apex length 0.8-0.9 times width of parameres at expansion; G. versipellis, 0.5-0.6 times width of parameres at expansion) (Fig. 7A, D). Ôhira (1995a) distinguished G. ornatus from G. versipellis in the key by the antennomere II (G. ornatus, longer than wide; G. versipellis, shorter than wide) and color of the basal outer edge of the elytra (G. ornatus, pale orange, Fig. 1A; G. versipellis, black, Fig. 1D). However, in both species, the males' antennomere II is shorter than wide (Fig. 3A), and in females, it is almost as long as or slightly longer than wide (Fig. 3B). Moreover, their elytral base is variable in color, and we found specimens [GO001, GO016] of G. versipellis with a pale orange basal outer edge of the elytra. The two species are difficult to distinguish using antennomere II and elytral color. In females, pronotum coloration is a good diagnostic character (G. ornatus, brown, Fig. 4B; G. versipellis, red, Fig. 4D) because no pronotum color variation has been found in either species. Considering the possibility that females with non-red pronota have been found, pronotum shape is the best diagnostic character for both sexes of the two species except for aedeagus (G. ornatus, PWI 150-156 in male and 161-174 in female; G. versipellis, PWI 172-184 in male and 176-191 in female) (Fig. 4A-D). Figures 1B, 2A-D  . Prothorax black; hind angles of pronotum yellow to orange in many, black in some. Each elytron with three separate yellow markings basally; basal markings connected in many, variably reduced in some. Antennomere II shorter than wide; antennomere III shorter than wide in male, slightly longer than wide in female; IV elongate (1.74-2.00 times longer than wide in male, 1.96-2.05 times longer than wide in female), longer than II-III combined (1.47-1.79 times as long as II-III combined in male, 1.21-1.40 times as long as II-III combined in female). Prosternal process weakly inclined dorsad (20-25°). Metacoxal plates narrowed outwards, with outer edge narrow. Apical edge of elytra rounded, without spines. Apex of parameres beyond pre-apical expansions widely triangular (apex length 0.1-0.2 times width of parameres at expansion).
We found a slide of the male genitalia labeled as belonging to the holotype of N. alpicola (matching specimen number and label information) in OMNH (Fig. 5K, L), although the sex and genitalia of the holotype were not mentioned in the original description. We found that N. alpicola should be a junior synonym of G. pictipennis by comparing the aedeagi of the Japanese Gamepenthes species (Figs 5J, 6, 7) and their aedeagus shown in Ôhira (1970b, 1995a). Gamepenthes pictipennis is also distinguished from its congeners by a combination of body size, antennae, prosternal pro- cess, metacoxal plates, and apical edge of the elytra (see above key and diagnoses). The holotype of N. alpicola shares these features (Fig. 5). We also found that G. pictipennis has variable pronotum and elytral coloration, ranging from a black base with clear yellow markings (Figs 1B, 2D) to having fewer yellow markings ( Fig. 2A) to entirely black (Fig. 2B). Even the black specimens show slight yellow markings, and the hind angles of the pronotum and anterior parts of that elytra are slightly yellow tinged (Fig. 2B), as in the holotype (Figs 2C, 5C). We conclude the holotype of N. alpicola is a melanic specimen of G. pictipennis.