A taxonomic revision of Limnobaris Bedel in the strict sense (Coleoptera, Curculionidae, Baridinae), with particular emphasis on the species found in China

Abstract The genus name Limnobaris Bedel is applied in a restricted sense to baridine weevils with a covered pygidium and non-prominent, decussate mandibles which occur on sedges in the Palaearctic Region and immediately adjacent parts of tropical Southeast Asia. Calyptopygus Marshall and Pertorcus Voss are syn. n. of Limnobaris. Some species from Africa and the Americas are maintained provisionally in Limnobaris in the widest sense but will need to be transferred to other genera in future studies. A total of eleven species is recognized in Asia, two of which are widespread and occur also in the Western Palaearctic Region. Limnobaris martensi Korotyaev sp. n. is described from Nepal. Pertorcus tibialis basalis Voss is raised to species rank, as L. basalis (stat. prom.). New or reestablished synonyms are L. dolorosa (Goeze) (= L. jucunda Reitter, = L. koltzei Reitter), L. tibialis (Voss) (= Pertorcus tibialis pilifer Voss) and L. t-album (Linnaeus) (= L. bedeli Reitter, = Baridius crocopelmus Gyllenhal, = L. sahlbergi Reitter, = L. scutellaris Reitter, = Baris t-album sculpturata Faust). Calandra uniseriata Dufour is considered a junior synonym of Sitophilus oryzae (L.) (syn. n.). A key for identification and a distribution map are provided.


Introduction
Baridinae are hyperdiverse, oligophagous weevils with a worldwide distribution. Many are uniformly oblong-ovate and notoriously poor in taxonomically useful characters. A particular problem is that their higher classification is based largely on poorly known exotic species with aberrant shapes and bizarre character formations, while at the same time leaving the bulk of nondescript species in an unwieldy, phylogenetically meaningless disarray. The situation has escalated over the past century because regionally conducted studies increasingly failed to associate new collections and observations with previously published information. Even the relatively well-studied species of the temperate northern hemisphere are in need of much work Yoshihara 1996, Anderson 2002).
The present study is concerned with an ill-defined complex of small, slender, usually black species associated with Cyperaceae ( Fig. 1). Bedel (1885) was the first to place two Western Palaearctic species in a separate genus named Limnobaris Bedel, which differs from other European baridines by having the pygidium completely covered by the elytral apices. This simple but useful character was accepted readily by contemporary entomologists: Reitter (1888) honored Bedel for his keen observation with the patronym Limnobaris bedeli; Casey (1892) adopted the concept for Nearctic species, Hartmann (1904) for an African species and Faust (1896), Champion (1908Champion ( , 1909 and Hustache (1932) for Neotropical species. However, this worldwide concept no longer functioned when Casey (1920Casey ( , 1922 applied new generic names to North American and Brazilian species in his private collection but ignored many others not immediately available to him. In addition to these still existing but obsolete generic placements, further problems have evolved around certain East Asian species, which currently are placed in three genera. Most share a distinct ventromedian tooth on the male protibia and all lack the dense, lateral vestiture of the type species, Limnobaris t-album (Linnaeus). Marshall (1948) described the genus Calyptopygus for a species with the above characters from the Myanmar-China border region but did not recognize its similarity to Limnobaris. Likewise, Voss (1953) described Pertorcus for a species from China's Fujian Province, erroneously stating it has only six funicular segments. Korotyaev (1982) described Limnobaris kabakovi from Vietnam and was the first to make a connection between Limnobaris and Pertorcus, but was misled by Voss' faulty description of the antenna. Yoshihara and Morimoto (1994) did not recognize the latent problem so Calyptopygus and Pertorcus were disregarded in their revision of East Asian Limnobaris species. Subsequent regional studies used the name Calyptopygus without providing diagnostic criteria (Yoshihara and Morimoto 1997) or their criteria made Calyptopygus and Limnobaris paraphyletic to each other Yoshihara 1996, Zherikhin 1997). Kojima and Morimoto (2004) and Prena (2011) merely cataloged the status quo. The primary objective of this study is to resolve the taxonomic conflict in the current usage of the names Limnobaris, Calyptopygus and Pertorcus, and to provide a means for the identification of the species likely to be found in China. As this covers almost all known species of the genus in the restricted sense adopted herein, our results may help future students of the Limnobaris complex in other biogeographical regions.

Material and methods
This study deals with the European and Asian species currently placed in the genera Limnobaris, Calyptopygus and Pertorcus, but ignores one nominal Limnobaris species from Africa and 26 from the Americas. We have examined most of them and concluded that the taxonomic problems are too complex to be treated here. Rather than adding to the confusion with incomplete and preliminary new generic placements, we decided to leave them provisionally in Limnobaris sensu lato.
Several species treated herein have been documented in great detail by Dieckmann (1991) and Morimoto (1994, 1997), and we have used their results and observations directly without always checking the types or even representative material. However, we studied the types of most remaining species and refer in the text to their and other specimens' current repositories with the following acronyms: AKMB, Because of our emphasis on Chinese fauna, we list under "Material examined" only specimens collected not further than 1000 km away from the Chinese mainland. Collecting trips were made by us to Fujian, Jilin and Yunnan in 2013 to obtain biological data and fresh specimens for morphological and future molecular comparison of regional populations. For the map in Fig. 13, we combined our records with collecting data published by Zherikhin (1972), Egorov (1976Egorov ( , 1977, Yoshihara and Mo-rimoto (1994), Hong et al. (2000Hong et al. ( , 2011, Legalov (2009) and in original descriptions. Geographic distributions were explored with GoogleEarth directly from the graphical user interface of JP's database via GoogleEarth API. The map presented in this paper was generated with PanMap (www.pangaea.de). Illustrations were prepared from images taken with a Micropublisher 5.0 RTV digital CCD camera mounted on a Zeiss SteREO Discovery V12 or a Canon EOS 650D on a Leica DM2500 compound microscope. Aldus Freehand was used for vector graphics and Adobe Photoshop for pixel-based artwork. Total length was measured from the anterior margin of eye to the abdominal apex in dorsal view with an ocular micrometer in a dissecting microscope.

Delineation and classification of Limnobaris
The present ambiguities in the usage of the name Limnobaris relate to two different issues. The first is associated with its application to species in the New World and tropical Africa, the second with the placement of very similar East Asian species in three different genera.
The taxonomy of the American and African species presently placed in Limnobaris is not the objective of this study. Their descriptions were based on Casey (1892), who noted that Limnobaris (in the widest sense) differs from most New World baridines with a covered pygidium by non-prominent, decussate mandibles. When Casey (1920) substituted Limnobaris in his earlier sense with new genera, species not represented in his own collection remained in Limnobaris and only a few of them have been transferred to other genera by Buchanan (1932), Kuschel (1983) and Prena (2013). The remaining American Limnobaris species belong to eight poorly defined, primarily Neotropical genera and need revision. Limnobaris lineigera Hartmann, from Tanzania, is excluded from this paper for the same reason; it is misplaced in Limnobaris and needs to be compared with African material. We maintain the respective New World and African species in Limnobaris sensu lato and refer the issue to the next revising authors.
The chief problem in the current usage of the name Limnobaris in the Palaearctic Region is the paraphyly to Calyptopygus and Pertorcus. Voss (1953) stated incorrectly that Pertorcus tibialis has only six funicular segments. However, this species is part of a morphologically homogeneous complex which includes also L. albosparsa Reitter and the type species of Calyptopygus, C. ellipticus Marshall. This makes Pertorcus at least a junior subjective synonym of Calyptopygus, while the distinctness of Calyptopygus and Limnobaris needs to be addressed. Morimoto and Yoshihara (1996) distinguished between Limnobaris and Calyptopygus because they observed differences in (1) the length and shape of the rostrum, (2) the length proportions of the funicular segments and (3) how the mesosternum connects to the metasternum. These criteria seem to have been taken from Calyptopygus kumei Yoshihara & Morimoto, 1997 rather than from the type species, C. ellipticus. At least (2) does not hold for C. ellipticus, (1) only for females and (3) does not seem to work at all. Zherikhin (1997) did not say why he transferred L. albosparsa to Calyptopygus but he obviously took into account the male protibial tooth (Figs 3-5) and perhaps vestiture. These two criteria separate L. t-album, L. dolorosa and L. japonica (all with edentate male protibia and dense ventrolateral vestiture) from the remaining species, which have dentate male protibiae (except C. kumei) and sparse ventrolateral vestiture. All included species are associated with sedges as far as is known (see taxonomic treatment further below for data and references). Limnobaris and Calyptopygus each have species that develop either in Carex or Scirpus, or both. Our preliminary molecular data of five species indicate that L. t-album and L. dolorosa nest inside the Calyptopygus/Pertorcus clade and are derived from ancestors with sparse vestiture and a long rostrum. Until more species are sequenced and our understanding of the relationships between the relevant Old and New World groups has improved, we suggest treating the presently known eleven Eurasian species in a single genus and consider Calyptopygus Marshall and Pertorcus Voss as subjective junior synonyms of Limnobaris Bedel (syn. n.).
The unresolved relationships of the Limnobaris complex in the wide sense affect approximately 25 genera worldwide, 160 currently valid species-group names and at least two family-group names. Most of these genera occur in the Americas and are placed in Zygobaridina Pierce, 1907, currently with Limnobaridina Casey, 1922 andTorcina Bondar, 1943 included as junior synonyms (Prena 2012). The usage of Zygobaridina derives from obsolete views expressed by Jekel (1865), Lacordaire (1865) and Casey (1892). The name had been ignored in weevil systematics until Alonso-Zarazaga and Lyal (1999) used it to replace Centrinina Jekel, 1865, a junior homonym preoccupied in the sharks. However, Zygobaridina becomes paraphyletic to the more senior Leptoschoinina Lacordaire, 1865 when the exposure of the pygidium is disregarded (Prena 2008). The matter is complicated further by species in Anacentrinus Buchanan, 1932(currently synonymized with Apinocis Lea, 1927, which develop in grasses (Barber 1927, Böving 1927, Buchanan 1932, Wille 1934, Bryson 1941) but also in sedges (J. Prena, unpubl. data) and connect the Limnobaris complex morphologically and ecologically to Madopterina Lacordaire, 1865, still another family-group name with nomenclatural priority. Many of the American taxa relevant for a phylogenetic reconstruction are fraught with taxonomic problems and bristle with unrecognized synonyms and misinterpreted names. As long as these issues remain unresolved and nominal types for family-group names are either excluded or misidentified in phylogenetic studies, the prevailing confusion will continue to escalate.

Misplaced Palaearctic species
Calandra uniseriata Dufour, 1843 was described from the French Pyrenees. The author characterized briefly the genus and placed the species between Sphenophorus Schönherr and Sitophilus Schönherr in the modern sense. However, Heyden et al. (1891) trans-ferred C. uniseriata to Limnobaris, where it remained until L. Zerche (in Dieckmann 1991) questioned this placement. On our request, H. Perrin kindly searched for the specimen in the Dufour Collection at MNHN and confirmed its absence therein (in litt., 2.viii.2013). Likewise, we found no vouchers in the Heyden material in SDEI and SFFM. Because there is no clear evidence for a misplacement of C. uniseriata in Dryophthorinae and the type seems to be lost, we tentatively transfer the species to Sitophilus and consider it a synonym of S. oryzae (L.) (new placement, syn. n.). This placement is made with the purpose to protect the generally accepted name for the cosmopolitan storage pest, S. zeamais (Motschulsky), at least until all relevant senior nominal taxa have been reviewed.

Diagnosis.
Small size (2.7-3.2 mm) and scattered, squamiform, appressed setae on the elytron make L. albosparsa a distinctive species around the Eastern Sea (Sea of Japan). The aedeagus is identical to that of the more southern L. tibialis except for its somewhat wider base ( Fig. 10 vs. Yoshihara and Morimoto 1994: 451), and the two species usually have a sharply pointed tooth on the ventral edge of the male protibia ( Fig. 5). However, they are allopatric ( Fig. 13) and L. tibialis has erect rather than appressed setae on the elytron. Other small species, like L. basalis, L. elliptica and L. martensi, have a blunt male protibial tooth and lack scattered squamiform setae on the elytron.
Biology. Nojima (in Yoshihara and Morimoto 1994) found specimens on Carex dickinsii Franch. and Savat. in Yoshikawa, Japan. Most known collecting dates fall between the middle of May and the end of June. Three specimens were collected in August and September (ZIN, data below) and indicate that the new generation emerges in the same year and overwinters in the adult stage. Material
Diagnosis. The Japanese L. babai is very similar to L. elliptica from Myanmar and South China. These two species can be distinguished from all others by the blunt ventromedian process on the male protibia and almost glabrous elytra. Limnobaris babai is on average larger than L. elliptica (3.5-4.5 mm vs. 3.2-3.8 mm), has shorter setae on the profemur, an apically rounded penis and the female protibia has a moderate ventromedian projection which is more subtle in L. elliptica. However, we have not compared specimens of the same size and these differences may not always hold. Limnobaris basalis, another morphologically similar species from Fujian, is smaller (2.3-3.1 mm) and has a longer rostrum. Distribution. The species occurs in Honshu and Kyushu, Japan (Yoshihara and Morimoto 1994

Diagnosis.
Limnobaris basalis is a small species (2.3-3.1 mm) with white, squamiform setae on the elytral base. Males can be distinguished from the equally small L. albosparsa and L. tibialis by the shape and position of the ventral process on the protibia (Fig. 3). Abraded females of L. basalis and L. tibialis may be distinguished by body proportion (L. basalis is slightly stouter) and host association, L. basalis and L. albosparsa by allopatry. Other similar but allopatric species are L. babai (Japan) and L. elliptica (South China, Myanmar). They are larger, have smaller and fewer scales on the elytral base and a shorter rostrum. Distribution. The species is known from several sites in Fujian Province, China (Fig. 13).
Biology. The overwintered weevil appears on Scirpus in early April and feeds on the leaves. In late April, when the plant flowers in Fujian, we observed numerous specimens on the inflorescence. The larva develops in the culm, apparently most successfully in the basal, often submerged internode. Occupants of more distal internodes frequently were parasitized by an unidentified species of the Eupelmus (s. str.) urozonus Dalman group (Eupelmidae, Hymenoptera). Pupation starts in June and newly eclosed adults appear in July. In middle July 2013, we found mostly larvae, approximately one third pupae and one eclosed adult in the culms we dissected.

Diagnosis.
Limnobaris dolorosa has characteristic ventrolateral vestiture of dense, squamiform setae, which occurs also in L. t-album and L. japonica. All three species can be distinguished from each other based on details of the male genitalia and vestiture. In addition, L. japonica is allopatric (see under this species for further details). The safest way to separate L. dolorosa from L. t-album is the short and wide penis (Figs 6, 7). Moreover, L. dolorosa has evenly dense vestiture on the entire flank, whereas L. t-album usually has less dense vestiture on the first ventrites and the metasternum. Notes. Specimens with a dolorosa-type of aedeagus show regional variation in size, body shape, vestiture and surface sculpture. Dieckmann (1991) mentioned disjunct, aberrant populations of unusually large specimens, one of which was found on Cladium mariscus (L.) Pohl rather than Carex. This scarce material was collected at sites with mild winters and cool summers (Swedish Baltic islands and Adige River valley in South Tyrol). Our own study showed that the East Asian L. jucunda is a population with a dolorosa-type of aedeagus that is morphologically indistinguishable from European Carex-associated specimens. However, small, densely squamose specimens with the same male genitalia occur in Transbaikal, Mongolia and adjacent Northeast China. Further studies with inclusion of molecular and ecological data of geographically representative material are needed to better understand the nature of this variation. We therefore follow Dieckmann (1991) and consider L. dolorosa (and, likewise, L. t-album) as a morphologically and ecologically polytypic species.
Distribution. The range of L. dolorosa extends from Western Europe (without Iberian Peninsula) to the Pacific coast (apparently without offshore islands, although Legalov (2010) reported the species from the Kuril Islands, possibly a misidentification of L. japonica). In China, the species has been found in Heilongjiang, Inner Mongolia and Jilin (Fig. 13).
Biology. Because of taxonomic confusion with L. t-album, published life history data are unreliable and need verification. Specimens of the typical size range occur on Carex rostrata Stokes ex With. in Scotland (Cawthra 1957; see Morris 2009 for weevil identification) and Northern Germany (J. Prena, unpubl. data). Palm (1957) found very large specimens (up to 7 mm) on Cladium mariscus in the Baltic islands Öland and Gotland. Hoffmann (1955) mentioned Scirpus sylvaticus L. as a host plant. In Jilin Province, we found L. dolorosa in mixed stands of Scirpus and tall Carex species (mostly C. rhynchophysa C. A. Meyer, some C. drymophila Turczaninow ex Steudel), occasionally in pure Carex stands. Feeding occurred on leaves of C. rhynchophysa and Scirpus. Life history data published by Kleine (1910), Hoffmann (1955) and Cawthra (1957) on L. t-album almost certainly apply to L. dolorosa. According to Kleine (1910), the larva develops in the stem of Cladium mariscus. Cawthra (1957) reported that eggs are laid in the basal part of the leaf of Carex rostrata and that the larva bores down to the rhizome, overwinters, and pupates inside the plant in late May. Hoffmann (1955)

Diagnosis.
Limnobaris elliptica is very similar to L. babai from Japan. These two species can be distinguished from all others by the short, curved rostrum, the blunt ventromedian process on the male protibia and almost glabrous elytra. Limnobaris elliptica is on average somewhat smaller than L. babai (3.2-3.8 mm vs. 3.5-4.5 mm), has an apically pointed penis (rounded in L. babai) and at least the males have longer setae on the underside of the profemur. Limnobaris basalis, another similar species, has a longer rostrum, larger scales on the elytral base and occurs in Fujian. The Nepalese L. martensi is slenderer and almost entirely glabrous. Distribution. The species is known from three sites west of the Gaoligong Mountains in the border region between Myanmar and Yunnan Province, China (Fig. 13).
Biology. We found larvae, pupae and freshly eclosed adults inside flowering culms of Scirpus wichurai Böckeler in late September. A few specimens already had exited the plant and one was observed on a tall Scleria species, which does not appear to be a larval host. Many larvae were killed by three to five specimens of an apparently undescribed Entedon Dalman species (Eulophidae, Hymenoptera), a large and widespread genus known to parasitize beetle larvae (C. Hansson, in litt.  -3.iv.1934-3.iv. (3x), 14.iv.1934.

Diagnosis.
Limnobaris japonica is the only species in Hokkaido, Honshu and the Kuril Islands that has dense lateroventral vestiture. Unlike L. dolorosa and L. t-album, the two other species with this trait, L. japonica has squamiform vestiture also on the prosternum. The penis is similarly elongate as in L. t-album but apically more gradually narrowed. Distribution. The species occurs in Hokkaido and Honshu (Yoshihara and Morimoto 1994) and Sakhalin Oblast (Egorov et al. 1996) (Fig. 13).
Biology. Adult weevils have been collected from Carex thunbergii Steud. and unidentified Carex species (Yoshihara and Morimoto 1994).

Diagnosis.
Limnobaris kabakovi is a relatively large (4.1-4.6 mm) species with ventral projection on the male protibia and scattered, appressed, squamiform setae on the elytron. Similarly appressed elytral vestiture occurs also in L. albosparsa, but that species is smaller and occurs only around the Eastern Sea. At least the type series of L. kabakovi has brownish elytra. Distribution. The species is known only from the type locality, a mountain resort and national park in Northern Vietnam (Fig. 13). This is the most southern record for a species of Limnobaris.

Diagnosis.
Limnobaris kumei is the only known species that has neither dense lateroventral vestiture nor a male protibial projection. Moreover, the rostrum is longer and sexually more dimorphic than in other congeners. Females have the antenna inserted in the basal half of the rostrum, a condition otherwise noticed only in L. albosparsa and L. tibialis. Distribution. The species is known from the Ryūkyū Islands (Okinawa) and Taiwan (Fig. 13).

Limnobaris martensi Korotyaev, sp. n.
http://zoobank.org/A7C0F62E-6BD3-453B-BB1C-97342A7EBDCC Diagnosis. Limnobaris martensi is a shiny, almost entirely glabrous species that has just a few inconspicuous setae on the ventral side and the legs (Fig. 2). The slightly bulging apical section of the tenth interstria is nearly smooth in L. martensi but more or less serrated in the other glabrous species L. babai and L. elliptica. Very slender, abraded L. tibialis can be distinguished from L. martensi by the sharply pointed, more basally inserted male protibial projection and apically round penis (Figs 8, 10). Description. Rostrum as long as pronotum, weakly and evenly curved, cylindrical, parallel-sided in basal 2/3, slightly dilated to apex; depression before eyes very weak but distinct. Dorsal surface of rostrum evenly convex, with short striole at level of antennal insertion; shiny, with sparse minute round punctures. Sides of rostrum at base with denser and larger elongate punctures. Antennae inserted at 0.57 × length of rostrum from base, antennal scrobe shortly continued beyond base of antenna. Ventral margin of antennal scrobe merging with lateroventral edge of rostrum at half way to eye; dorsal margin of scrobe reaching eye. Scape of antenna slender, shortly widened at apex. First segment of funicle 1.5 × as long as wide, 2nd slightly longer than wide, 3rd weakly transverse, 4-7th moderately to strongly transverse. Base of club very broadly rounded, almost truncate, but clearly separated from broad 7th funicular segment, apex of club broadly rounded. Frons weakly convex, at anterior margin as broad as base of rostrum, slightly widened posteriad, shiny, with sparse small punctures. Vertex with reticulate microsculpture. Eyes large.
Pronotum 1.1 × as wide as long, parallel-sided in basal half, then weakly narrowed to shallow apical constriction. Base of pronotum feebly bisinuate. Disc weekly and evenly convex, sub-matt due to reticulate microsculpture, with rather sparse fine, somewhat angular, round or oblong punctures, separated usually by not less than own diameter, in some places by 2-3 × diameter. Median line without microreticulation and punctures. Scutellum shiny, nearly rectangular, feebly widened at base. Elytra 1.8 × as long as wide, with well-pronounced humeri, parallel-sided in basal half, rather narrowly rounded at apex; sutural angle slightly sinuate. Disc flattened, with fairly abrupt declivity; preapical prominences very distinct. Striae deep and narrow, intervals flat, about 4 × as wide as striae, shiny, with 1 row of small punctures and much finer microreticulation than on pronotum. Intervals in many places distinctly impressed around sparse and inconspicuous punctures in striae.
Legs slender and fairly long. Fore tibia with large tooth slightly proximal of middle of inner surface, apex of tooth blunt (Fig. 4). 1st tarsite 1.5 × as long as wide, 2nd clearly transverse, rounded at sides, 3rd in fore tarsus 1.7 × as broad as 2nd. 5th tarsite slender, weakly widened toward apex, by one-half of its length projecting beyond lobes of 3rd tarsite. Length of claw 1.5 × width of claw-segment at apex. Penis as in Fig. 8, moderately bent ventrally at base and apex, basal apodemes shorter than in other species.
Body black; scape of antenna in basal 2/3 light brown, apical third of scape, 1st and base of 2nd segments of antennal funicle, and tarsi dark brown, humeral callus brownish. Upper side bare, legs with sparse short recumbent white setae, sides of abdomen with few inconspicuous setae.
Pertorcus tibialis Voss, 1953: 75  directed process on the ventral edge of the protibia, which is located more basally than the blunt tooth of other species (Fig. 5). Limnobaris albosparsa, the only other species with a sharply pointed process, has a slightly wider penis basally, appressed rather than erect setae on the elytron and is allopatric. The number and width of white setae on the elytral interstriae vary considerably even in series taken at the same locality. Females may be confused with sympatrically occurring L. basalis, but the latter species has white setae crowded at the elytral base and is slightly less elongate. Most female L. tibialis can be distinguished by more basally inserted antennae. Notes. Voss (1958) distinguished a pilose "form" with slightly different protibial process. Even though the size of setae can differ noticeably between individual specimens, transitional forms occur even in the same series. Pertocus tibialis pilifer is a new junior synonym of the nominal species.
Biology. In Northern Fujian, we found this species on several, ca. 20-30 cm tall Carex species. The larva develops in the basal, ca. 2 mm wide section of the culm. Pupation starts in July and newly eclosed adults appear in the same month.
Distribution. Limnobaris t-album is widespread in the Palaearctic Region and apparently extends further north than the other common species, L. dolorosa. In fact, L. t-album is the only baridine weevil worldwide that reaches either of the polar circles. The most eastern records are from Mongolia (ca. 95°E), Xinjiang Province, China (ca. 90°E) and the adjacent Russian territory (Legalov 2010).
Biology. Because of taxonomic confusion with L. dolorosa, published life history data are unreliable and need verification. The adult occurs on Carex acutiformis Ehrh. between middle May and early July in Northern Germany (J. Prena, unpubl. data). Carex atherodes Spreng., C. pseudocyperus L., C. rostrata and Scirpus sylvaticus grew occasionally nearby but received no attention by the weevil. Reports from C. rostrata, C. vesicaria L. and the tall sedges Cladium mariscus and Schoenoplectus lacustris (i.e., Heyden in Brisout 1870, Heyden 1877, Sahlberg 1892, Kleine 1910, Hoffmann 1955, Cawthra 1957, Scherf 1964  Mesosternum, metasternum and abdominal ventrites laterally with wide, imbricate setae completely covering integument; male with ventrites 1-2 strongly depressed; penis shorter and wider (Fig. 6); transpalaearctic ........... L. dolorosa -Mesepisternum, mesepimeron and metepisternum with imbricate scales forming T-shaped pattern, adjacent sclerites (flanks of metasternum, often also first two ventrites) less densely covered with wide setae, leaving integument in between visible; male with ventrites 1-2 slightly depressed; penis longer and narrower (Fig. 7) checked the language and, together with an anonymous reviewer and subject editor Miguel Alonso-Zarazaga, commented on the manuscript. The first author was funded by a 1-year grant for senior international scientists awarded by the Chinese Academy of Sciences (2012T1S0025). The second author was supported by the Russian Foundation for Basic Research (Grant No 13-04-01002). Fieldwork was supported by NSFC programs 31210103909, 31172130 and J1210002. All of these contributions were important and much appreciated.