An aberrant species of Nipponocercyon from Sichuan, China (Coleoptera, Hydrophilidae, Sphaeridiinae)

Abstract A detailed examination of specimens of Cryptopleurum sichuanicum Ryndevich, 2005 from high altitudes of Sichuan Province, China, revealed that the species belongs in the genus Nipponocercyon Satô, 1963 previously endemic to Japan. The species is here transferred in Nipponocercyon, and Nipponocercyon sichuanicus (Ryndevich, 2005), comb. n. is redescribed and compared with Nipponocercyon shibatai Satô, 1963. The male genitalia of Nipponocercyon sichuanicus is described for the first time. An adapted diagnosis of Nipponocercyon is provided, and reasons for the inclusion of Nipponocercyon sichuanicus into Nipponocercyon and the general distribution of the genus are discussed.


Introduction
Cryptopleurum sichuanicum Ryndevich, 2005 was described from a few female specimens collected in the mountains of Sichuan Province, China (Ryndevich 2005). Although considered "rather peculiar and isolated with some characters differing from The current study is largely based on newly collected material of N. sichuanicus (18 specimens) which were compared with one paratype of Cryptopleurum sichuanicum, and on the specimens of Nipponocercyon shibatai and of other megasternine genera deposited in the collection of NMPC.
Selected specimens were dissected, with genitalia embeded in a drop of water--soluble dimethyl hydantoin resin on a piece of transparent plastic pinned below the specimen, or of alcohol-soluble Euparal resin on a small piece of glass attached below the respective specimen. The external morphol ogy was examined using the Hitachi S-3700N environmental electron microscope at the Department of Entomology, National Museum in Prague. Habitus photographs were taken using Canon D-550 digital camera with attached Canon MP-E65mm f/2.8 1-5× macro lens, and subsequently adapted in Adobe Photo shop CS2. Figures of genitalia were prepared with the help of Photoshop CS4. The morphological terminology largely follows Komarek (2004) and Fikáček (2010), the higher-level taxo nomic nomenclature follows Hansen (1999) and Short and Fikáček (2011). taxonomy Nipponocercyon Satô, 1963 http://species-id.net/wiki/Nipponocercyon Adapted differential diagnosis. The inclusion of Cryptopleurum sichuanicum into Nipponocercyon (see below) requires a modification of the differential diagnosis of the genus as follows: Head without transverse interantennal ridge; eyes small, separated by 9× of one eye; mentum weakly bisinuate on anterior margin; antennae with 9 antennomeres; maxilla with or without sucking disc in males; maxillary palpomere 2 strongly widened distally; posterior tentorial pits minute; pronotum evenly convex, lateral margin not deflexed or slightly deflexed; transverse row of larger punctures along posterior margin of pronotum absent (large areas without microsculpture in N. sichuanicus may actually resemble enlarged punctures on the first view, but the punctures are as large as those in disc when examined in detail, see Fig. 8); median portion of prosternum weakly to distinctly separated from lateral portions, bearing coarse setiferous sculpture; median portion of prosternum carinate medially (carina distinct in N. shibatai, partly obliterated by the sculpture but still apparent in N. sichuanicus, compare Figs 12 and 15); prosternal process wide, deeply excised; antennal grooves moderately large to large, not reaching lateral margin of hypomeron (Figs 11,14); anteroventral margin of prothorax with a small denticle on the contact of prosternum and hypomeron; profemur with elongate ventral depression along anterior margin; elytron with 10 punctural series; elytral intervals flat or highly convex; lateral margins of elytra not denticulate nor serrate; mesoventral cavities for reception of procoxae large, reaching mesocoxae; preepisternal elevation subpentagonal, widely contacting metaventral process, median portion of metaventrite slightly to very distinctly elevated; postcoxal ridge lying parallel to posterior margin of mesocoxal cavity, not overlapping to lateral margin of metaventrite; lateral portions of metaventrite with coarse punctation (smaller punctures may be intermixed or absent); metanepisternum narrow, but distinct throughout; abdominal ventrite 1 carinate medially, with coarser punctation than ventrites 2-5; phallobase asymmetrical, much shorter than parameres; gonopore situated in basal half of median lobe; male sternite 9 with median tongue-like projection; male sternite 8 without median projection.
A few characters listed as diagnostic for Nipponocercyon by Hoshina and Fikáček (2010) have to be excluded as they are only present in N. shibatai but absent from N. sichuanicus: antennomeres 7-8 with groups of peg-like sensilla ventrally; metaventrite with two short mesal ridges anteriorly (but remnants of ridges seem to be retained in some specimens of N. sichuanicus; see the structure indicated as mtr in Fig. 20).
Recognition. By the combination of median portion of the prosternum differentiated from lateral portions, subpentagonal preepisternal elevation of the mesothorax widely contacting the metaventrite, large mesothoracic cavities for reception procoxae (reaching to anterior margin of mesoxocal cavity) and metanepisternum well developed both anteriorly and posteriorly, Nipponocercyon is most similar to the genus Australocyon Hansen, 1990. It may be easily distinguished from the Australian and Neotropical species of Australocyon by the male sternite 9 with tongue-like median portion (Fig. 4c), and male sternite 8 without median projection; from Australocyon pilocnemoides group it may be distinguished by the undifferentiated surface of the subpentagonal mesoventral plate (with a semicircular median portion defined by a wide bead in Australocyon pilocnemoides group, see Fig. 6 in Hansen (2003)), unmodified antennal morphology (in contrast to long antennae with prolonged antennal club pointed at apex and antennomere 6 not cup-like in shape in Australocyon pilocnemoides group), and the excavate ventral surface of profemora (without any sculptured depression in Australocyon pilocnemoides group).
When the size of mesoventral cavities for reception of procoxae is not taken into consideration, Nipponocercyon may resemble other megasternine genera with small subpentagonal mesoventral plate, clearly defined prosternal plate and male sternite 9 tongue-like medially (characters distinguishing the respective genus from Nipponocercyon are listed in parentheses: Agna (prosternal plate without deeply excised prosternal process, antennal grooves very small and angular in shape, profemur without sculptured depression); Bolbonotum and Kahanga (elytral grooves deep and wide, reaching total base of elytra, prosternal plate projecting both anteriad and posteriad, profemur without ventral impression, mesoventral plate rhomboid when examined in detail, gonopore apical), Deltostethus (mesoventral plate with wide marginal bead, profemur without ventral depression, gonopore apical), and Pelocyon (metavetrite with complete femoral lines, prosternal plate longer than wide). Nipponocercyon sichuanicus may resemble some species of the genera Cryptopleurum, Pachysternum and Cyrtonion by its large antennal grooves, large grooves for reception of procoxae, reduced epipleura and strongly sculptured body. See below under that species for characters distinguishing it from the mentioned genera.
Composition and distribution. The genus now includes two species, one distributed in Kyushu, Shikoku and the southern part of the Honshu, the other occuring in high altitudes of the mountain ranges in the Chinese province of Sichuan (Fig. 23).
Head. Clypeus widely rounded, constricted above antennal bases, with very distinct marginal bead. Dorsal surface of clypeus and frons with strong scale-like microsculpture obscuring the punctation, sparsely arranged punctures not apparent among microsculpture, only evident according to long thin setae arising from punctures. Frontoclypeal suture apparent as a non-sculptured stripe directing mesad, strongly bent posteriad submesally. Mentum with sparsely arranged fine setiferous punctures medially and posteriorly, interstices with strong scale-like microsculpture. Antennal club without distinct groups of peg-like sensilla dorsally or ventrally. Maxilla of male without sucking disc ventrally.
Prothorax. Pronotum with sparsely arranged fine setiferous punctures, larger punctures along posterior margin absent. Whole dorsal surface with mesh-like microsculpture, microsculpture strong along anterior and posterior margins and on lateral portions of pronotum, obsolete on pronotal disc; pronotal disc with  Hoshina and Fikáček (2010) in combination with the photo sent to us by H. Hoshina, we did not examine any male specimen for this study]). Parts of male genitalia: a tegmen, b median lobe, c sternite 9. Abbreviations: aes3 metanepisternum, ang antennal groove, epi epipleuron, grp grooves for reception of procoxae, mtr anteromesal ridge of metaventrite, pfd profemoral depression. irregular longitudinal striae. Lateral portions of pronotum slightly deflexed (and hence seen in ventral view). Prosternum with well defined median plate 2.0× wider than long, bearing strong rugose sculpture, indistinctly carinate mesally. Anterolateral corners of prosternum (at contact with hypomeron) with small but distinct tooth. Antennal grooves large, but not quite reaching lateral margin of hypomeron. Profemur with a rather shallow sculptured depression on a large portion of ventral surface. Protibia angulate distally. Mesothorax. Scutellar shield with sparse fine punctation, without microsculpture. Elytra with 10 punctural series, all series deeply impressed, lateral striae deeper than median ones; serial punctures minute and rather inconspicuous; elytral intervals highly convex, bearing sparsely arranged fine setiferous punctation, interstices with strong microsculpture consisting of small bumps; lateral portions of elytra deflexed laterally (hence, visible in ventral view); epipleuron present only on elytral base, reduced to extremelly narrow stripe behind level of mesocoxae. Mesoventrite with pentagonal posteromedian elevation, the elevation 1.3× wider than long, with rugose setiferous sculpture.
Metathorax. Anteromedian process with very weakly developed short longitudinal ridges, in many individuals completely obscured by microsculpture; median portion of metaventrite slightly elevated bearing densely arranged coarse setiferous punctures separated by 0.5-1.2× puncture diameter; lateral portions of metaventrite with extremelly large setiferous punctures; whole surface of metaventrite except its posterome-  dian portion with mesh-like microsculpture on interstices, microsculpture stronger laterally than medially. Hind wings well developed.
Abdomen. All abdominal ventrites with strong scale-like microsculpture, punctation of ventrite 1 consisting of extremelly large setiferous punctures similar to that on lateral portions of metaventrite; punctation of ventrites 2-5 sparse and very fine, nearly completely obscured by microsculpture.
Differential diagnosis. See the identification key above for characters distinguishing N. sichuanicus from N. shibatai. Nipponocercyon sichuanicus may be confused with some species of Cryptopleurum, Pachysternum or Cyrtonion (the latter not occurring in Asia, however) which are also characterized by large antennal grooves and strongly sculptured dorsal surface. Nipponocercyon shibatai may be easily distinguished from them by the combination of following characters: (1) metaventrite without femoral lines (femoral lines present in Cryptopleurum, Pachysternum and Cyrtonion, see e.g. Fig. 22, feml); (2) metanepisternum wide throughout (Fig. 20) (reduced anteriorly and widening posteriad in the above genera as well as in all other genera of the Megasternum group characterized by large antennal grooves, see e.g. Fig. 22, aes3); (3) gonopore situated in basal portion of the median lobe ( Fig. 4; this character distinguishes both species of Nipponocercyon from all other Megasternini); (4) mesoventral plate only slightly wider than long, without acute angles (Fig. 19) (mesoventral plate large and distinctly transverse in Cryptopleurum, see e.g. Fig. 22); (5) anterolateral corners of mentum not sharply angulate (Fig. 10) (sharply angulate in Cryptopleurum, as in Fig. 16).
Biology. No details on the biology are known. The terrestrial habits of N. shibatai (Hoshina & Fikáček 2010) as well as the vast majority of the megasternine taxa suggest that N. sichuanicus is a terrestrial species.
Distribution. The species occurs in the mountains of the Sichuan province in South China, at altitudes between 2500-3500 m a.s.l. (Fig. 24).

Discussion
The inclusion of Cryptopleurum sichuanicum into the genus Nipponocercyon may appear contradictory, as the species differs in its external morphology from N. shibatai in many characters, some of which were previously considered as diagnostic at the generic level. However, a detailed comparison reveals that despite many differences, both spe-cies are very similar in terms of the morphology of the ventral side and genitalia (compare Figs 3-6) which is crucial for generic assignment of the megasternine taxa. Especially important in this respect are characters which are present in both Nipponocercyon species but absent from all other Megasternini: the basal position of the gonopore on the median lobe (Figs 4b, 6b), and the presence of a depression on the ventral surface of profemur (Figs 3, 5, pfd). Another character infrequent in the Megasternini but shared by both Nipponocercyon species is the presence of large mesothoracic grooves for the reception of the procoxae (Fig. 20, grp). Except Nipponocercyon, large grooves are only present in Australocyon and the Megasternum group of genera, of which the representatives of the latter generic group and part of Australocyon clearly seem unrelated to Nipponocercyon based on other external characters (see the differential diagnosis of Nipponocercyon for details). Moreover, general morphology of the genitalia of both species is very similar, with principal differences found only in the shape of the apical portion of the median lobe. Eyes with extremelly small dorsal portion are also unusual within the Megasternini but shared by both species. All these characters indicate that both species are more closely related to each other than they are to other megasternine taxa, which justifies the tranfer of Cryptopleurum sichuanicum into Nipponocercyon.
Characters in which Nipponocercyon sichuanicus is contradicting the diagnosis of the genus Nipponocercyon used by previous authors (Satô 1963, Hansen 1991, Hoshina and Fikáček 2010) seem now to be autapomorphies of N. shibatai or N. sichuanicus. For example, the groups of peg-like sensilla on antennomeres 7-8 are autapomorphies of N. shibatai; highly sculptured dorsal and ventral surfaces, reduced epipleura (Fig. 20,epi), extremely large antennal grooves (Fig. 11), and the absence of male sucking discs on the maxilla are autapomorphies of N. sichuanicus. The reduction of sucking discs in N. sichuanicus may be possibly correlated with the strong microsculpture on dorsal surface of this species, which makes the suckers uneffective for male's holding on the female during the copulation. Two diagnostic characters are seemingly absent from N. sichuanicus, but a careful examination of multiple specimens revealed that they are present in both Nipponocercyon species: remnants of a pair of short anteromesal ridges on the metaventrite (present even in some specimens of N. sichuanicus; Fig.  20, mtr) and the median carina of prosternal plate.
The autapomorphies of N. sichuanicus make the species similar in habitus to Cryptopleurum and related genera, which is the reason why the species was originally described within Cryptopleurum. However, a detailed examination shows that this similarity is due to the parallelism, as the body parts which are responsible for the Cryptopleurum-like appearance of N. sichuanicus differ between Cryptopleurum and N. sichuanicus in detailed morphology: (1) The antennal grooves of N. sichuanicus are large, but still do not reach the lateral margins of hypomeron, as they do in Cryptopleurum and related taxa (compare Figs 11 and 17).
(2) The meso-and metathorax of N. sichuanicus resemble Cryptopleurum and related genera by their extremely coarse punctation and large grooves for reception of procoxae, but the overall thoracic morphology is totally different from these genera due to its well developed metepisternum (not reduced anteriorly, Fig.  20: aes3) and the ventral side of mesothorax not steeply declined, possessing unmodified mesepimera. (3) The prosternal plate of N. sichuanicus is feebly carinate even though the carina is partly obscured by the strong microsculpture (ecarinate in Crypopleurum and related genera, see Fig. 18).
The above differences indicate that despite its superficial resemblance, Nipponocercyon sichuanicus is not related to the Cryptopleurum group of genera, but probably represents another example of highly sculptured sphaeridiine taxon derived from a non-sculptured ancestor. The shift from non-sculptured to sculptured phenotype was shown to lead to similar ‚sculptured' morphology in distantly related taxa in the Megasternini (Oosternum and Emmidolium, see Fikáček 2007), and similar examples also exist in some other groups of the Sphaeridiinae (e.g., the omicrine genera Noteropagus and Peratogonus resemble the megasternine genus Cryptopleurum by the same characters which are responsible for the Cryptopleurum-like appearance of Nipponocercyon sichuanicus).
The inclusion of Cryptopleurum sichuanicum into Nipponocercyon extends the range of the genus (previously endemic to Japan) to the mainland Asia. The isolated occurrence of the genus at high altitudes of the mountain ranges in Sichuan and in mountain areas of southern Japan (Fig. 23) suggests that the current distribution may represent relictual remnants of the former wider distribution of the genus. The situation hence resembles that of the myxophagan genus Satonius Endrödy-Younga, 1997 which was originally considered as Japanese endemic (Satô 1963;Endrödy-Younga 1997), but was later found to be distributed in various areas in central and southern China (Hájek and Fikáček 2008;Hájek et al. 2011). High altitude areas of other mountain ranges of China need to be sampled in order to test if the occurrence of Nipponocercyon on the continent is really restricted to Sichuan, or if the genus is more widely distributed in the mountains of the transitional zone between Palaearctic and Oriental regions. Hoshina and Fikáček (2008) examined 35 specimens of Japanese Nipponocercyon from the entire Japanese range of the genus, including the types of all three taxa described from Japan by Satô (1963) and Nakane (1968): N. shibarati Satô, 1963, N. shibarati omayanum Nakane, 1968and N. monticola Nakane, 1968. They found a wide variation in several characters among the specimens they examined (especially the shape of mesoventral plate and metaventral ridges, the presence of dorsal sculpture and the impression of elytral series) which was considered as intraspecific variation due to a weak correlation of the characters and extreme uniformity of male geniatalia. The above taxa were therefore synonymized with N. shibatai, although the authors expressed the need for more detailed studies based on more extensive material to clarify the taxonomy of Japanese Nipponocercyon. The inclusion of N. sichuanicus into the genus may bring new insights into the problem. We have examined specimens of N. sichuanicus from a rather wide area in Sichuan (Fig. 24), but all of them are very constant in the proportion of the mesoventral plate. This is in contrast to the high variability of the shape of the mesoventral plate of N. shibatai proposed by Hoshina and Fikáček (2008). Moreover, Nipponocercyon shibatai and N. sichuanicus are very similar in the morphology of the aedeagus (the width of the apex of the median lobe is the only relevant difference, in contrast to many differences in external morphology), indicating that genital morphology is very conservative in Nipponocercyon. This suggests that no significant differences in aedeagal morphology should be expected among closely related species, whereas the shape of mesoventral plate may be a good indicator of species limits. This further demostrates the need for additional studies of Japanese Nipponocercyon, as the existence of more than one species in Japan cannot be excluded.