Revision of European species of the genus Rhabdomastix ( Diptera : Limoniidae ) . Part 2 : Subgenus Rhabdomastix s . str .

The second and final part of a revision of the European species of the genus Rhabdomastix Skuse, 1890 is presented. The subgenus Rhabdomastix s. str. is revised. Seven species are redescribed, Rhabdomastix (Rhabdomastix) japonica Alexander, 1924, R. (R.) laeta (Loew, 1873), R. (R.) borealis Alexander, 1924, R. (R.) edwardsi Tjeder, 1967, R. (R.) subparva Starý, 1971, R. (R.) hirticornis (Lackschewitz, 1940) and R. (R.) beckeri (Lackschewitz, 1935). Three new synonyms are proposed. Lectotypes of four pertinent nominal species are designated. Descriptions are provided of six species, viz. R. (R.) laetoidea sp. n. (Czech Republic, Slovakia, Bulgaria, Ukraine), R. (R.) crassa sp. n. (France, Czech Republic, Slovakia), R. (R.) corax sp. n. (Bulgaria, Greece), R. (R.) eugeni sp. n. (France, Switzerland, Germany, Czech Republic, Slovakia, Italy, Romania, Bulgaria, Greece, Ukraine, Armenia), R. (R.) filata sp. n. (Bulgaria, Greece, European Russia, Turkey, Georgia, Armenia) and R. (R.) georgica sp. n. (Georgia). Male and female terminalia are illustrated for all the species, and a key to species is appended.


INTRODUCTION
In the first part of this revision published recently (Starý, 2003), the taxonomic history of the genus Rhabdomastix Skuse, 1890 was reviewed, its classification outlined and re-assessed, and a new subgenus, Lurdia Starý, 2003, was established.Nine European Lurdia species were treated, and seven of these were described as new.
The second instalment, now presented, deals with the nominotypical subgenus Rhabdomastix s. str., as defined in the first part, i.e. in a broad sense, comprising the majority of species of the former subgenera Rhabdomastix s. str., Palaeogonomyia Meunier, 1899 and Sacandaga Alexander, 1911.Generally, the same morphological terminology is used as in the first part of this revision (Starý, 2003).The following should be added or repeated: Rs length (at half of Rs length or beyond it in European species).Sc2 lacking or faintly apparent at tip of Sc1 or some distance before it .R3 very short, from one-fifth to one-eighth of R4 length, vertical or virtually so, forming a 90 o with R4.Discal cell generally hexagonal, with proximal section of M3+4 (forming lower side of discal cell) distinctly angled near mid-length, at attachment of m-cu.Distal sections of M1+2 and M3 (beyond discal cell) considerably arched.
Abdomen.Male terminalia (cf.Fig. 17 and other relevant figures): Segment 9 generally parallel-sided, simple, with at most small lobe dorsally at posterior margin on each side of median interruption (except in beckeri, Fig. 44).Gonostyli generally shorter than those in Lurdia; outer one terminating in curved apical spine, or broadly rounded and without apparent spine; inner one fleshy, generally conical, variously swollen, generally a little shorter than outer gonostylus.Interbase membranous, pale, mostly dilated before apex into variously shaped apical blade, not connected membranously to its counterpart at about one third of its length.Female terminalia (cf.Fig. 19 and other relevant figures) with cercus and hypogynial valve of moderate length, the former at most slightly exceeding length of tergite 10.Spermathecae  two or three in number, spherical, oval, or reniform, subequal in size to each other, or, if three, one spermatheca sometimes tending to be smaller than other two.
Discussion.The two subgenera, Lurdia and Rhabdomastix s. str., were compared in the first part of this revision (Starý, 2003).It should be noted that the venational pattern in Rhabdomastix s. str., with its vertical (subvertical) R3, combined with the hexagonal discal cell and distinctly arched M1+2 and M3, is quite unique within the Limoniidae.The venation may vary in details even within a species, especially in aspects such as the verticality of R3 or the length ratio of R3 to R4.
Rhabdomastix s. str.species, as those of Lurdia, are mainly distinguished by details in the structure of the male and female terminalia, in the latter case predominantly by the number and size of the spermathecae.Other internal structures of the female terminalia, such as the infra-anal (supravaginal) plate, sternum 9 and genital fork (vaginal apodeme) are not sufficiently differentiated, and, of these, only the vaginal apodeme may sometimes provide some species-specific peculiarities (cf.Figs 18,25,28).If from the same region, some species may well be separated by the body colouration.However, this only applies to dry-mounted specimens. Distribution.Worldwide.

REVISION OF EUROPEAN SPECIES OF THE SUBGENUS RHABDOMASTIX S. STR.
Compared to the subgenus Lurdia, representatives of the European Rhabdomastix s. str.species are more diverse structurally, representing several species clusters, or evolutionary trends.However, within a cluster, the taxonomic situation parallels that in Lurdia: species may be very similar to each other in both external and genital characters that, in addition, vary in a certain degree.Moreover, some Rhabdomastix s. str.species have, or are presumed to have, wide ranges of distribution, and some show infraspecific variation in body colouration, a trait not observed in Lurdia.Consequently, to properly recognising species limits, large series of specimens have been examined from Central Europe where, fortunately, many species occur sympatrically, even syntopically.This was combined with the examination of additional specimens from more remote areas.Without this extensive and geographically diverse material, totalling over 2500 specimens, a re-assessment of the European Rhabdomastix s. str.species would not have been possible This revision of the European Rhabdomastix s. str.has been in progress, with interruptions, for an exceedingly long time.Many specimens from various institutions were examined as early as in the late 1970's.In the course of the study, some species concepts were modified.Therefore, as many as possible, relevant specimens, especially types, were re-examined quite recently (2002)(2003).
In all, thirteen Rhabdomastix s. str.species are treated here.Seven species are redescribed, viz.R. (R.) japonica Alexander, 1924, R. (R.) laeta (Loew, 1873), R. (R.) borealis Alexander, 1924, R. (R.) edwardsi Tjeder, 1967, R. (R.) subparva Starý, 1971, R. (R.) hirticornis (Lackschewitz, 1940) and R. (R.) beckeri (Lackschewitz, 1935).Three new synonyms are proposed: R. hilaris Edwards, 1938 andR. cunctans Tjeder, 1955 are treated as junior synonyms of R. (R.) japonica, and R. lapponica Tjeder, 1936 as a junior synonym of R. (R.) borealis.The latter synonymy had been tentatively suggested by Savchenko et al. (1992).Lectotypes of hilaris, laeta, hirticornis and beckeri are designated.Descriptions are provided for six new species, viz.R. (R.) laetoidea sp.n., R. (R.) crassa sp.n., R. (R.) corax sp.n., R. (R.) eugeni sp.n., R. (R.) filata sp.n. and R. (R.) georgica sp.n., the latter being extra-European.Some European Rhabdomastix s. str.species have often been covered in the literature.The many species records listed below in the references sections under species headings are, however, largely suspect because of subsequently newly described species.Practically all the small, darkly coloured species had been identified as R. schistacea (Schummel, 1829), one of the species most commonly treated in the literature, until Tjeder (1967) described R. edwardsi.The taxonomic situation in Rhabdomastix at that time may best be illustrated by the fact that P. Lackschewitz, a distinguished student of Limoniidae who contributed considerably to a better knowledge of the group in Europe, had identified a series of specimens deposited in NHMW as belonging to R. schistacea (cf.Lackschewitz, 1940).Within this series, four species (edwardsi, crassa, filata, subparva) have now been differentiated (see respective Material examined sections).Schummel's description of Limnobia schistacea, based on a single female from Wroc aw (Breslau), Poland, although clearly representing a Rhabdomastix s. str.(cf.Schummel, 1829, Tab. 2, Fig. 2), deals with a small species ("2½" = ~ 5.5 mm) having the head and thorax of slate colour ("schiefergrau"), a yellowish grey abdomen with segments seamed with yellowish, R3 more than its own length beyond the tip of R1, and A2 long, ending beyond the origin of Rs.This combination of characters is not known within the European Rhabdomastix s. str.Therefore, Limnobia schistacea is considered a nomen dubium (cf.also Starý & Rozkošný, 1970).Since literature references are as complete as possible for each species treated, the same is provided for L. schistacea at the end of this paper.
Considering the material below, collected by me in the Czech Republic and Slovakia, many Rhabdomastix s. str.species may seem to be common.The rich material available from these territories, including that of new species (laetoidea, crassa, eugeni) is, however, a result of many years' collecting activity at specific habitats to which these species are strictly confined, namely sandy or gravelly banks of streams.Actually, there is a single common and largely eurytopic species in Central Europe, R. (R.) subparva.Other regions may show different relations.
The European Rhabdomastix species of the former subgenus Sacandaga were subdivided into three species groups by Savchenko (1982), the lurida, laeta and edwardsi groups.As the lurida group represents now the subgenus Lurdia (cf.Starý, 2003), and species such as R. hirticornis (formerly in the subgenus Palaeogonomyia) are treated here within Rhabdomastix s. str., the above concept had to be modified.Anyway, considering the world fauna, differences between species groups in Rhabdomastix s. str.should be considerably greater than are those found between the laeta and edwardsi groups of Savchenko (1982).Based on various characters, three clusters, or species complexes, may be distinguished preliminarily within the European Rhabdomastix s. str., namely those centred around R. (R.) laeta, R. (R.) edwardsi and R. (R.) hirticornis.In addition, R. (R.) beckeri is a distinctive species considerably different from all the others, and its affinities remain in question.

R. (R.) beckeri
Antenna short, with pubescence on flagellomeres distinct (Fig. 10); palpus short; R3 more than its own length beyond tip of R1 (Fig. 3); A2 ending opposite to origin of Rs (Fig. 3); differs considerably from the species above by having milky wings, with narrow darker seams along veins, and many details in structure of male terminalia.It most probably belongs to a different species cluster.
As already noted, the classification above is preliminary, since the characters used have unequal value.The laeta and edwardsi complexes are well defined, based on several independent features, whereas the so-called hirticornis complex, although distinguished at once by the long male antennae, can hardly be assigned the same weight.The conspicuousness of the long antennae is responsible for their being generally overvalued as a taxonomic character.The length of the antennae, however, vary extensively within genera and subgenera of the chioneine Limoniidae, unlike most other characters.
Diagnosis.General colouration yellow to pale yellow, sometimes conspicuously patterned with deep dark brown on thorax, including three stripes on prescutum.Antenna short, with very short pubescence on flagellomeres.Wing broad.A2 ending beyond origin of Rs.Legs yellow throughout.Male terminalia with apical blade of interbase very broad and aedeagus long and slender.Female terminalia with three spherical medium-sized spermathecae.
Colour.General colouration yellow to pale yellow, subshiny, with darker markings on thorax.Antenna dark brown, scape mostly yellow.Prescutum pale yellow laterally, with three broad darker stripes.Scutum and mediotergite (postscutellum) similarly darker, restrictedly patterned with yellow.Scutellum pale yellow.Pleuron mostly yellow, patterned with sulphur yellow in upper part, darker below, especially on lower portion of katepisternum and meron.Colouration of thorax practically identical in distribution of darker markings with that of two following species (laeta, laetoidea); in contrast to these, however, considerably variable in actual hue of pattern, varying from little-distinct, yellowish brown to sharply pronounced, deep dark brown.Wing tinged with yellowish.Halter pale yellow.Legs yellowish brown to yellow throughout, femora not darkened distally.Abdomen yellowish brown to brown.
Head.Antenna (Fig. 5) comparatively short, not reaching to base of wing.Proximal three flagellomeres nearly spherical, following ones gradually narrowed and lengthened towards apex of antenna.Longest verticils on flagellomeres slightly exceeding length of their respective segments.Pubescence very short, suberect, subequal in length to half breadth of respective segments, or even shorter, distinct only on proximal four or five flagellomeres.Palpus short.
Thorax.Wing (Fig. 1) rather broad, about three times as long as broad, with comparatively short stalk.Sc1 ending before fork of Rs, at about three quarters of Rs length.Sc2 not apparent, or slightly so some distance before tip of Sc1.R3 about its own length, or less, beyond tip of R1.R4 varying in number of macrotrichia, but generally with only few.A2 considerably sinuous, ending distinctly beyond origin of Rs.Halter moderately long, reaching to about posterior margin of abdominal tergite 2.
Abdomen.Male terminalia (Fig. 17).Segment 9 broader than long.Gonocoxite comparatively long and slender.Outer gonostylus short, less than half length of gonocoxite, gently and evenly arched, somewhat broadened before apex, with small apical spine.Inner gonostylus generally conical.Aedeagal complex as in Fig. 17.Interbase moderate in length, reaching to about half length of gonocoxite, abruptly expanded distally to form broad triangular, sometimes nearly quadrangular apical blade, microscopically serrate at distal margin.Aedeagus very long and slender, nearly twice length of vesica, the 661 latter comparatively small and narrow.Apodeme of vesica spine-like in dorsal aspect, subequal in length to vesica.Female terminalia 46).Cercus moderately broad, slightly exceeding length of tergite 10, gently upturned.Vaginal apodeme moderately broad.Spermathecae three, spherical, intermediate in size between those of R. (R.) laeta and R. (R.) laetoidea sp.n., with sclerotised parts of ducts thin, slightly shorter than spermathecal diameter.
R. (R.) japonica, R. (R.) laeta and R. (R.) laetoidea sp.n. represent a group of closely related and exceedingly similar species.They are identical in distribution of darker markings on the thorax.In the female holotype of R. (R.) japonica (and two other specimens from Japan, see Material examined), these markings are dark brown, similar to the condition as described for R. cunctans from Europe.Such specimens with distinct dull or shiny markings on the thorax are sporadically represented in the material examined, from Sweden (holotype of cunctans), Switzerland, Germany, Italy and Algeria, and they show identity in structural characters with other specimens, paler in the pattern and more commonly collected, sometimes even at the same localities.This suggests that the variation is individual, independent of geographical distribution or ecological factors.This variation has not been observed in R. (R.) laeta and R. (R.) laetoidea sp.n., although this cannot be excluded (see note under laetoidea).A certain type of geographical variation may also be involved in R. (R.) japonica.British specimens (identified as hilaris) usually have a medium-dark pattern, with the markings somewhat diffuse, suffused with grey pruinosity, whereas members of Central European populations are mostly pale, with the pattern only slightly indicated.More material would be necessary from the eastern Palaearctic to decide whether specimens with the distinct pattern on the thorax are more frequent there than they are in Europe.In any case, a male from the Kuriles (see Material examined) has the pattern considerably paler than the Japanese specimens, as much as in Central European material.Wings are described as being broad in R. (R.) japonica (as they are in laeta); this, however, is also subject to a certain variation.First, females always have somewhat narrower wings than males, and, second, the smaller a specimen is, the narrower wings it has relative to its own wing length.
R. (R.) japonica is on average the largest of the species treated here.In general appearance, if not distinguished by a dark pattern on the thorax, specimens of R. (R.) japonica are exceedingly similar to R. (R.) laeta and R. (R.) laetoidea sp.n.The latter species is distinctive by a comparatively short aedeagus and broad vesica in males and large spermathecae in females, supported by markedly narrower wings in both sexes.The differences between R. (R.) japonica and R. (R.) laeta are best noticeable in the structure of the antennae, especially those of males (cf.Figs 5 and 6).In R. (R.) japonica, these are distinctly shorter, with the proximal flagellomeres more spherical, and the pubescence very short, subequal in length to at most half the breadth of the respective segments, distinct only on the proximal four or five flagellomeres.In R. (R.) laeta, the antennae are longer, with the proximal flagellomeres rather oval, and the pubescence is long, subequal in length to the entire breadth of the respective segments, distinct on almost all flagellomeres.This character, together with a slightly different shape of the interbases, had been indicated by Edwards (1938: 114, Text-figs 22e, f) as distinguishing his R. hilaris from R. laeta.Females of all the three species (japonica, laeta, laetoidea) differ in the size of the spermathecae (cf. Figs 19,21,23,(46)(47)(48).
Distribution.The species was reported from Japan (Hokkaido, Honshu, Shikoku, Kyushu), North Korea and the Russian Far East (Savchenko et al., 1992).There are probably no authentic literature records for Kyushu and North Korea.The synonymy with hilaris and cunctans, proposed here, extends its distribution into Europe (Great Britain, Sweden, cf.Savchenko et al., 1992) where it may also be among some records of R. (R.) laeta.Based on the material examined, the species is now recorded from Great Britain, Switzerland, Germany, Czech Republic, Slovakia, Austria, Italy, Macedonia, Albania, Bulgaria, Greece, Algeria, Azerbaijan, Russian Far East and Japan.
Colour.General colouration yellow to pale yellow, subshiny, with faintly indicated markings on thorax.Antenna dark brown, scape yellow.Prescutum pale yellow laterally, with indications of three broad darker stripes.Scutum and mediotergite (postscutellum) similarly darker, restrictedly patterned with yellow.Scutellum pale yellow.Pleuron mostly yellow, patterned with sulphur yellow in upper part, darker below, especially on lower portion of katepisternum and meron.Pattern on thorax generally pale, little-distinct, not varying to dark condition as in R. japonica.Wing tinged with yellowish.Halter pale yellow.Legs yellow throughout.Abdomen yellow to yellowish brown.
Head.Antenna (Fig. 6) moderately long, reaching to base of wing.Flagellomeres oval proximally, gradually narrowed and lengthened towards apex of antenna.Longest verticils on flagellomeres slightly exceeding length of their respective segments.Pubescence long, suberect, subequal in length to breadth of respective segments, distinct on almost all flagellomeres.Palpus short.
Thorax.Wing rather broad, about three times as long as broad, with comparatively short stalk.Sc1 ending before fork of Rs, at about three quarters of length of the latter.Sc2 not apparent or slightly so some distance before tip of Sc1.R3 about its own length or less beyond tip of R1.R4 with variable number of macrotrichia, but mostly with only few.A2 considerably sinuous, ending distinctly beyond origin of Rs.Halter moderately long, reaching to about posterior margin of abdominal tergite 2.
Abdomen.Male terminalia (Fig. 20).Segment 9 broader than long.Gonocoxite comparatively long and slender.Outer gonostylus short, less than half length of gonocoxite, gently and evenly arched, somewhat broadened before apex, with small, sometimes barely distinct apical spine.Inner gonostylus generally conical.Aedeagal complex as in Fig. 20.Interbase longer and much more slender than in R.(R.) japonica, slightly extending beyond half length of gonocoxite, moderately expanded distally to form roughly triangular apical blade with a few microscopic teeth at distal margin.Aedeagus very long and slender, yet not as long as in R. (R.) japonica.Vesica comparatively small and narrow.Apodeme of vesica spine-like in dorsal aspect, subequal in length to vesica.Female terminalia (Figs 21,47).Cercus moderately broad, slightly exceeding length of tergite 10, gently upturned.Vaginal apodeme moderately broad (much as in japonica).Spermathecae three, spherical, small, with sclerotised parts of ducts thin, subequal in length to spermathecal diameter.
Distribution.As one of those most commonly treated in the literature, the species was recorded from many European countries, also from West Siberia (Altai) and Mongolia (Savchenko et al., 1992).Although it most probably does occur throughout Europe, the actual records are unreliable because it could have been confused with R. (R.) japonica or R. (R.) laetoidea sp.n.Based on the material examined, the species is confirmed in Europe for Sweden, Finland, Netherlands, Switzerland, Germany, Czech Republic, Slovakia, Austria, Slovenia, Bulgaria and Ukraine, and newly recorded for Andorra and Italy.[The record from Italy by Mannheims (1964), originally accepted by Savchenko et al. (1992), refers to a Swiss locality (cf.Starý & Oosterbroek, 1996)  Diagnosis.General colouration yellow to pale yellow, with faintly indicated markings on thorax, including three stripes on prescutum.Antenna moderately long, with pubescence on flagellomeres subequal in length to breadth of respective segments.Wing narrow.A2 ending shortly beyond origin of Rs.Legs yellow throughout.Male terminalia with apical blade of interbase lanceolate and aedeagus short and broad.Female terminalia with three spherical, large spermathecae.
Colour.General colouration yellow to pale yellow, subshiny, with less distinct markings on thorax, compared to R. japonica and R. laeta, however, practically identical to the latter species in distribution of pattern.Antenna dark brown, scape yellow.Wing tinged with yellowish.Halter pale yellow.Legs yellow throughout.Abdomen yellow.
Head.Antenna moderately long, reaching to base of wing.Flagellomeres oval proximally, gradually narrowed and lengthened towards apex of antenna.Longest verticils on flagellomeres slightly exceeding length of their respective segments.Pubescence long, suberect, subequal in length to breadth of respective segments, distinct on almost all flagellomeres.Palpus short.
Thorax.Wing (Fig. 2) narrow, compared to both R. japonica and R. laeta, about four times as long as broad, with stalk longer than in two latter species.Sc1 ending just beyond mid-length of Rs.Sc2 slightly apparent, shortly before tip of Sc1.R3 about its own length beyond tip of R1.R4 with numerous macrotrichia both dorsally and ventrally.A2 slightly sinuous, ending shortly beyond origin of Rs.Halter moderately long, reaching to about posterior margin of abdominal tergite 2.
Abdomen.Male terminalia (Fig. 22).Segment 9 broader than long.Gonocoxite not as long as in R. (R.) japonica and R. (R.) laeta.Outer gonostylus short, less than half length of gonocoxite, gently and evenly arched, somewhat broadened before apex, with small, sometimes barely distinct apical spine.Inner gonostylus generally conical.Aedeagal complex as in Fig. 22. Interbase subequal to half length of gonocoxite, with apical blade more or less lanceolate, more slender than in R. (R.) laeta.Aedeagus comparatively short and broad, subequal in length to vesica, the latter rather broad.Apodeme of vesica spine-like in dorsal aspect, shorter than vesica.Female terminalia (Figs 23,48).Cercus moderately broad, slightly exceeding length of tergite 10, gently upturned.Vaginal apodeme narrow, compared to R. (R.) japonica and R. (R.) laeta.Spermathecae three, spherical, larger and paler than those of latter two species, with sclerotised parts of ducts thin, subequal in length to spermathecal diameter.
Etymology.The name of the new species, laetoidea, indicates its close relationship to R. (R.) laeta.An adjective in nominative singular.
Discussion.The new species differs from both R. (R.) japonica and R. (R.) laeta by its generally smaller size and distinctly narrower wings in both sexes.The latter character is correlated with the venation in that the veins run closer to each other and A2 is less sinuous, ending only shortly beyond the origin of Rs (Fig. 2).In R. (R.) laetoidea sp.n., the male terminalia are especially distinctive by having a comparatively short and broad aedeagus and a broad vesica (Fig. 22 Distribution.Czech Republic, Slovakia, Bulgaria, Ukraine.Note.This may be Rhabdomastix (Sacandaga) shardiana Alexander, 1957 described from a single male from Pakistan (Alexander, 1957: 292) [holotype %: North West Frontier Province, Shardi, 1.-10.viii.1953,altitude 6,130 feet, (F.Schmid leg.) (USNM)], a species with a rather distinct pattern on the thorax, generally conforming the condition in the laeta complex.Due to distortion of the male terminalia mounted on a slide, absence of the antennae in the holotype, and considering the disjunct occurrences of R. (R.) laetoidea sp.n. and R. shardiana, a clear decision on the conspecifity of the two forms could not be reached.Therefore, it is preferred to describe here R. (R.) laetoidea sp.n., a species fairly common in Central and South Europe, until more material, including females, is available from the area relevant to R. shardiana.Alexander, 1924 (Figs 24-26, 49) Rhabdomastix (Sacandaga) borealis Alexander, 1924a: 9 (description).Rhabdomastix (Sacandaga) borealis: Alexander, 1965 Diagnosis.General colouration yellowish brown, patterned with dark brown and pale yellow on thorax.Antenna moderately long, with pubescence on flagellomeres very short.Wing moderately broad.A2 ending beyond origin of Rs.Legs with femora considerably darkened distally.Male terminalia with apical blade of interbase triangular provided with tooth at outer margin, and aedeagus long.Female terminalia with three spherical spermathecae, somewhat smaller than those of R. (R.) laetoidea sp.n.
Colour.General colouration yellowish brown, with slight greyish pruinosity, patterned with dark brown and pale yellow on thorax.Antenna dark brown, scape yellowish brown.Prescutum dark brown, yellowed laterally.Sometimes two yellow longitudinal lines apparent, demarcating three broad dark brown stripes; yellow patch near posterior margin of prescutum.Scutum dark brown, patterned with yellow medially.Scutellum mostly yellow.Mediotergite (postscutellum) yellow anteriorly, dark brown in posterior half.Pleuron yellowish brown to brown, with slight greyish pruinosity, patterned with pale yellow in upper part, darkened on lower portions of katepisternum and meron.Wing tinged with brownish.Coxae yellowish brown.Trochanters and bases of femora yellow, the latter considerably darkened distally.Rest of legs yellowish brown.Halter pale yellow.Abdomen greyish brown.
Head.Antenna moderate in length, reaching to about base of wing.Proximal three or four flagellomeres shortoval to nearly spherical, following ones gradually narrowed and lengthened towards apex of antenna.Longest verticils on flagellomeres slightly exceeding length of their respective segments.Pubescence very short, suberect, distinct only on proximal four or five flagellomeres (much as in japonica).Palpus short.
Thorax.Wing moderately broad, more than three times as long as broad, with stalk comparatively short.Sc1 ending before fork of Rs, at about three quarters of length of the latter.Sc2, if apparent, considerably retracted from tip of Sc1, approximately opposite half length of Rs (much as in japonica).R3 less than its own length beyond tip of R1.R4 with a few macrotrichia both dorsally and ventrally.A2 considerably sinuous, ending distinctly beyond origin of Rs.Halter moderately long, reaching to about posterior margin of abdominal tergite 2.
Abdomen.Male terminalia (Fig. 24).Segment 9 broader than long.Gonocoxite comparatively long and slender.Outer gonostylus short, less than half length of gonocoxite, gently and evenly arched, parallel-sided, with small apical spine.Inner gonostylus generally conical, slender distally.Aedeagal complex as in Fig. 24.Interbase very long, considerably extending beyond half length of gonocoxite and expanded before apex to form roughly triangular apical blade terminating in large acute tooth at outer margin (at margin closer to long axis of hypopygium for crossing interbases).Aedeagus long (not as slender as in japonica and laeta), nearly twice length of vesica or less, the latter comparatively small and narrow.Apodeme of vesica spine-like in dorsal aspect, short, about half length of vesica.Female terminalia (Figs [25][26]49).Cercus moderately broad, slightly exceeding length of tergite 10, gently upturned.Vaginal apodeme abruptly expanded into very broad, transversally oblong or semicircular distal (caudal) portion.Spermathecae three, spherical, somewhat smaller than those of R. (R.) laetoidea sp.n., with sclerotised parts of ducts thin, shorter than spermathecal diameter.
Although clearly belonging to the species complex comprising the three species above (japonica, laeta, laetoidea), R. (R.) borealis is very distinctive in both general appearance and the structure of the male terminalia.Its general colouration is distinctly darker, yellowish brown, with greyish pruinosity, patterned with dark brown on the thorax, extensively so dorsally.In contrast to the species above, the femora are considerably darkened distally in R. (R.) borealis (yellow throughout in japonica, laeta and laetoidea).Within the species treated here, the male terminalia of R. (R.) borealis are unique in the shape of the interbases (Fig. 24).The female terminalia are similar to those of the related species, with the spermathecae somewhat smaller than in R. (R.) laetoidea sp.n., although they are well characterised by a very broad vaginal apodeme (Fig. 25).
Distribution.In the new concept presented here, the species appears to be Holarctic in distribution, and was recorded from the USA (Alaska) (Alexander, 1965), Norway, Sweden and the Russian Far East (Savchenko et al., 1992).Herewith confirmed for all the countries.Distribution in Canada is practically beyond any doubt, but not proved by the two wings on the slide tentatively listed here.The identification by Alexander, who must have had the rest of the specimen, is reliable insofar as made possible by a comparison of external characters.In any case, an authentic record for Canada is needed.A short diagnosis of Gonomyia schistacea from Finland by Lundström (1907: 21), added with a figure of the wing showing R3 less than its own length beyond the tip of R1 (cf.Lundström, 1907, Fig. 25), suggests that a species of the laeta complex is involved, most probably R. (R.) borealis.This, however, should also be confirmed.Tjeder, 1967 (Figs 27-29, 50) Rhabdomastix parva : Edwards, 1938: 113, 115  Diagnosis.General colouration dark greyish brown, with bluish pruinosity on pleuron.Antenna short.Wing narrow, infuscated.A2 ending before origin of Rs.Legs brown with coxae greyish brown.Male terminalia with outer gonostylus generally straight, with distinct apical spine, and apical blade of interbase spoon-shaped.Female terminalia with three spherical, small spermathecae.
Colour.General colouration dark greyish brown with bluish tinge, dull, without conspicuous markings on thorax, more brownish in middle of prescutum.Antenna dark brown throughout.Pleuron heavily suffused with dark bluish grey pruinosity, somewhat variable in extent and bluish hue.Wing infuscated.Halter whitish.Coxae generally dark, brown to greyish brown.Trochanters and bases of femora yellowish brown, the latter darkened distally.Rest of legs generally brown.Abdomen dark greyish brown.
Head.Antenna short, not reaching to base of wing.Flagellomeres short-oval.Longest verticils on flagellomeres slightly exceeding length of their respective segments.Pubescence indistinct.Palpus short.
Thorax.Wing rather narrow, about four times as long as broad, with stalk comparatively short.Sc1 ending before half length of Rs.Sc2 lacking.R3 more than its own length beyond tip of R1.R4 bare or with at most a few macrotrichia dorsally.A2 sinuous, ending before origin of Rs.Halter comparatively short, not reaching to posterior margin of abdominal tergite 2.
Abdomen.Male terminalia (Fig. 27).Segment 9 longer than broad.Gonocoxite sometimes rather stout, broad.Outer gonostylus comparatively short, about half length of gonocoxite, bent only at base, straight distally, generally parallel-sided, with distinct apical spine.Inner gonostylus generally conical.Aedeagal complex as in Fig. 27.Interbase moderate in length, reaching to about half length of gonocoxite, very slender near mid-length, at most very slightly bent distally to form spoon-shaped apical blade, mostly rounded at apex (pointed in some cases, cf.Fig. 27).Aedeagus slender, subequal in length to vesica, the latter broad, bulbous, with long apodeme, narrowly fan-shaped in dorsal aspect, about same length as vesica.Female terminalia 50).Cercus comparatively slender and rather long, longer than tergite 10, gently upturned.Spermathecae three, small, spherical, with sclerotised parts of ducts very short.One spermatheca sometimes smaller than other two.

Material examined. Holotype & (original designation):
Great Britain, England, South Devon, Sidmouth, 10.v.1936 (F.W. Edwards leg.) (BMNH), labelled "S.Devon: Sidmouth.10.V.1936.F.W. Edwards. B.M. 1936-366" (printed), "Holotype" (a red-margined circular label, printed), "Holotypus & Rhabdomastix edwardsi Tjed.Bo Tjeder 1966" ("Holotypus" printed, the rest in Tjeder's hand, red label).The specimen is micro-pinned on a celluloid slide, with only left fore and hind Discussion.There is a certain variation in the body size and colouration.Specimens from Great Britain and South Europe are, on the average, smaller, somewhat more robust, and darker, rather dark greyish brown, with the wings strongly infuscated and with the bluish pruinosity limited to the pleuron.Members of Central European populations are larger, more slender, generally somewhat paler, with a very distinct bluish suffusion that passes from the pleuron onto the prescutum and other dorsal parts of the thorax.
The species concept of R. (R.) edwardsi has been one of the controversies of this revision, since specimens from various regions differ in various aspects (see above).Although the species had been described and illustrated adequately (cf.Tjeder, 1967: 225, Figs 1-10), it does not seem to have been well recognised, and it only was accepted as occurring in Great Britain.It has never been reported outside that country, except for records from the Czech Republic and Slovakia (Starý, 1987), later doubted (see Distribution).From the beginning of my studies on Rhabdomastix, the sympatric material of two forms was available to me from the former Czechoslovakia, differing considerably in the body colouration from both R. (R.) subparva, the most common regional species, and from each other.The one form was distinctive by a bluish pruinosity on the pleuron (these specimens may occur in various collections labelled by me as "caesia"), whereas the other form was entirely black.The sympatric occurrence of these forms supported the view that they represent valid species.The solution reached after a thorough comparison, repeated many times, is now believed to be a correct one: form No. 1 is R. (R.) edwardsi and form No. 2 is a new species described below as R. (R.) crassa sp.n.
The bluish pruinosity, rather dark and intensive, differentiates R. (R.) edwardsi from all species treated here, except perhaps R. (R.) hirticornis, which, however, is sufficiently distinctive by its very long male antennae.In the structure of the male terminalia, R. (R.) edwardsi appears to be most closely related to R.(R.) crassa sp.n. and R. (R.) corax sp.n., both entirely black species.R. (R.) edwardsi differs from these species in having a generally straight outer gonostylus (gently and evenly arched in corax) and rather long, slender interbases (shorter and broader in crassa, more rounded at apex).Some other external and male genital features differentiating the three species are specified in the discussions of the two latter.The female terminalia of the three species are very similar to each other, having the spermathecae of approximately the same size.R. (R.) edwardsi has the cerci slightly longer and more slender than the other two.
Distribution.So far the species has been known from Great Britain only.Records from the Czech Republic and Slovakia (Starý, 1987), based on unpublished material, were later withdrawn (Starý, 1993(Starý, , 1996)).Records are presented here for Great Britain, Spain, France, Germany, Czech Republic, Slovakia, Austria, Italy, Slovenia, Bosnia and Hercegovina, Albania and Bulgaria.
Diagnosis.General colouration black throughout.Antenna short.Wing narrow, strongly infuscated.A2 ending far before origin of Rs.Legs dark brown, including coxae.Male terminalia with outer gonostylus generally straight, with distinct apical spine, and apical blade of interbase spoon-shaped.Female terminalia with three spherical, small spermathecae.
Description.Very small species, plump in general appearance, with all body appendages (antennae, palpi, wings, legs) slightly shorter, compared to other species.Body length 3-6 mm, wing length 3-5 mm.
Colour.General colouration black, dull (deep dark brown in faded dried specimens), without conspicuous markings on thorax.Antenna almost black throughout.Pleuron heavily suffused with dark greyish black pruinosity.Wing strongly tinged with blackish.Halter infuscated, especially on stem.Coxae deep dark greyish brown.Trochanters and bases of femora brown, the latter darkened distally, deep dark brown.Rest of legs dark brown.Abdomen almost black, somewhat shiny.
Thorax.Wing (Fig. 4) rather narrow, about four times as long as broad, with stalk very short.Sc1 ending at about half length of Rs or slightly before it.Sc2 littledistinct at tip of Sc1.R3 more than its own length beyond tip of R1.R4 bare or with at most a few macrotrichia dor-sally.A2 slightly sinuous, ending far before origin of Rs.Halter comparatively short, not reaching to posterior margin of abdominal tergite 2. Legs rather thick and short, compared to other species.
Etymology.The name of the new species, crassa (= thick, stout), refers to its somewhat plump general appearance, with all body appendages slightly shorter than in the other species.An adjective in nominative singular.
Discussion.R. (R.) crassa sp.n. is, on the average, the smallest of the species treated here, long known from numerous localities in the Czech Republic and Slovakia.It is distinctive by its generally plump appearance, short verticils on antennae and the black body colouration.Within European species, the latter character is only shared by R. (R.) corax sp.n., which differs from R. (R.) crassa sp.n. in having a light grey pruinosity on the pleuron and more macrotrichia on R4.In the structure of the male terminalia, R. (R.) crassa sp.n. is very similar to R. (R.) edwardsi and R. (R.) corax sp.n., differing from either or both in having the outer gonostylus generally straight (gently and evenly arched in corax), the inner gonostylus moderately broad (more slender in corax), the interbase short and comparatively broad (longer and more slender in both edwardsi and corax), and the apodeme of the vesica narrowly fan-shaped (more slender in corax, rod-like or spine-like).The female terminalia of the three species are not well distinguished having the spermathecae of approximately the same size, but the features listed above clearly validate R. (R.) crassa sp.n. as a separate species.
Colour.General colouration black, dull (deep dark brown in faded dried specimens), without conspicuous markings on thorax.Antenna almost black throughout.Pleuron slightly suffused with light grey pruinosity.Wing strongly infuscated, blackish.Halter infuscated throughout.Coxae deep dark greyish brown.Trochanters and bases of femora brown, the latter still darkened distally, deep dark brown.Rest of legs dark brown.Abdomen almost black.
Head.Antenna comparatively short, not reaching to base of wing.Flagellomeres short-oval.Longest verticils on flagellomeres slightly exceeding length of their respective segments.Pubescence indistinct.Palpus short.
Thorax.Wing rather narrow, about four times as long as broad, with stalk comparatively long, longer than that of R. (R.) crassa sp.n.Sc1 ending at about half length of Rs or slightly before it.Sc2 not apparent.R3 more than its own length beyond tip of R1.R4 with about 10 macrotrichia, mostly dorsally.A2 slightly sinuous, ending before origin of Rs.Halter moderate in length, reaching to posterior margin of abdominal tergite 2.
Abdomen.Male terminalia (Fig. 32).Segment 9 longer than broad.Gonocoxite moderate in length and breadth.Outer gonostylus comparatively short, about half length of gonocoxite, considerably bent at base, otherwise gently and evenly arched, generally parallel-sided, sometimes slightly tapered distally, with barely distinct apical spine.Inner gonostylus generally conical, more slender than in other related species.Aedeagal complex as in Fig. 32.Interbase moderate in length, reaching slightly beyond half length of gonocoxite, slender, bent near two-thirds of its length to form lanceolate apical blade, similar to that of R. (R.) subparva.Aedeagus short, subequal in length to vesica, the latter broad, bulbous, with long apodeme, subequal in length to vesica, rod-like or spine-like in dorsal aspect.Female terminalia (Fig. 36).Cercus moderate in length and breadth, subequal in length to tergite 10, gently upturned.Spermathecae three, small, spherical, subequal in size to both R. (R.) edwardsi and R. (R.) crassa sp.n., with sclerotised parts of ducts practically not apparent.One spermatheca sometimes smaller than other two.
Etymology.The new species is named corax (= the rook) for its notably black body.A noun in nominative singular standing in apposition to generic name.
Discussion.This new species is distinctive by its black body colouration, being thus similar to R. (R.) crassa sp.n., from which it differs by some external and genital characters, such as the overall slender appearance, the setosity of R4, the shape of both gonostyli (the outer gonostylus of corax somewhat suggesting that of subparva) and the structure of the aedeagal complex.For details, see the discussion of R. (R.) crassa sp.n.
Colour.General colouration dark greyish brown to light grey, dull, without conspicuous markings on thorax, more brownish in middle of prescutum and sometimes still more so on abdomen.Antenna brown throughout.Pleuron heavily suffused with grey, sometimes greyish brown pruinosity.Wing slightly infuscated.Halter dirty white to pale yellow.Coxae, trochanters and bases of femora mostly yellow to yellowish brown, otherwise legs somewhat darker, brown.Abdomen greyish brown, proximal segments sometimes paler, brown, more conspicuously so ventrally.Sometimes abdomen entirely brown.
Head.Antenna (Fig. 8) comparatively short, not reaching to base of wing.Flagellomeres short-oval to oval.Longest verticils on flagellomeres slightly exceeding length of their respective segments.Pubescence indistinct.Palpus short.
Thorax.Wing rather narrow, about four to five times as long as broad, with short stalk (Fig. 15).Sc1 ending at about half length of Rs.Sc2 not apparent.R3 more than its own length beyond tip of R1.Macrotrichia on R4 somewhat varying in number, about 10, mostly placed dorsally.A2 sinuous, ending before origin of Rs.Halter comparatively short, not reaching to posterior margin of abdominal tergite 2 (Fig. 15).
Abdomen.Male terminalia (Fig. 33).Segment 9 longer than broad.Gonocoxite moderate in length and breadth.Outer gonostylus, on average, slightly longer than that of related species (except that of filata), more than half length of gonocoxite, gently and evenly arched or nearly straight distally, generally parallel-sided, rounded at apex, with apical spine barely distinct or lacking.Inner gonostylus rather broad, generally conical.Aedeagal complex as in Fig. 33.Interbase comparatively short and slender, reaching to about half length of gonocoxite, bent shortly before apex to form short and narrow, roughly lanceolate apical blade.Aedeagus slender, subequal in length to vesica, the latter broad, bulbous, with apodeme comparatively short, fan-shaped or rounded in dorsal aspect, shorter than vesica.Female terminalia (Figs 37,51).
Etymology.The new species is named in honour of the late Dr. Yevgeniy (= Eugen) Nikolaevich Savchenko (Kiev, Ukraine), a distinguished specialist on the Limoniidae (s.lat.) and Tipulidae, who first collected this species and who was able, under conditions inconceivable to his colleagues from the then so-called free world, to do so much on the craneflies.A noun in genitive singular.
Discussion.Whereas members of Central European populations are dark greyish brown, most resembling R. (R.) subparva, the specimens examined from the Balkans are somewhat paler, bicoloured, having a light grey thorax and brown abdomen.Surprisingly, a few specimens examined from Calabria (southern Italy) are rather dark, much as the Central European individuals.
R. (R.) eugeni sp.n., if from Central Europe, most resembles R. (R.) subparva by its dark greyish brown body colouration and yellow coxae.It differs from the latter, as do other related species, in that it is not as slender in general appearance as R.(R.) subparva, having the wing stalk and halter shorter (cf.Figs 15 and 16).In R. (R.) eugeni sp.n., R4 bears about 10 macrotrichia (more numerous, about 20, in subparva, fewer or none in edwardsi, crassa and filata).Specimens of R. (R.) eugeni sp.n. from the Balkans are paler, bicoloured, thus somewhat resembling R. (R.) filata sp.n. and the Balkan specimens of R. (R.) hirticornis, both quite different in the structure of the antennae.R. (R.) eugeni sp.n. is probably most closely related to R. (R.) filata sp.n. in having the outer gonostylus rounded at apex, without an apparent apical spine, it differs, however, from the latter by a small apodeme of vesica in males (large, broadly fan-shaped in filata) and reniform spermathecae in females (spherical in filata).
Colour.General colouration grey to light grey, dull on thorax, without conspicuous markings.Antenna pale brown to brown throughout.Pleuron variably suffused with whitish grey pruinosity.Wing hyaline or slightly tinged with yellowish.Halter whitish yellow.Coxae, trochanters and bases of femora yellow, rest of legs somewhat darker.Abdomen light brown, subshiny.
Head.Antenna (Fig. 9) short, not reaching to base of wing, with pedicel large, compared to other species, and flagellum very thin, filiform.Flagellomeres mostly elongate-oval, slender, except more rounded first flagellomere.Longest verticils on flagellomeres exceeding length of their respective segments.Pubescence indistinct.Palpus short.
Thorax.Wing rather narrow, about four times as long as broad, with short stalk.Sc1 ending at about half length of Rs.Sc2 sometimes slightly apparent at tip of Sc1.R3 more than its own length beyond tip of R1.R4 with at most a few macrotrichia dorsally.A2 sinuous, ending opposite to origin of Rs or shortly before it.Halter short, not reaching to posterior margin of abdominal tergite 2.
Abdomen.Male terminalia (Fig. 34).Segment 9 longer than broad.Gonocoxite moderate in length and breadth.Outer gonostylus slightly more than half length of gonocoxite, bent at base, otherwise straight, parallel-sided, rounded at apex, with apical spine sometimes not apparent.Inner gonostylus generally conical, somewhat swollen.Aedeagal complex as in Fig. 34.Interbase reaching beyond half length of gonocoxite, bent shortly before apex to form narrow, roughly lanceolate blade.Aedeagus short, rather broad, subequal in length to vesica; the latter broad, bulbous.Apodeme of vesica large, broadly fanshaped in dorsal aspect, subequal in length to vesica.Female terminalia (Figs 38,52).Cercus moderately broad, slightly exceeding length of tergite 10, gently upturned.Spermathecae two, nearly spherical, mediumsized, with sclerotised parts of ducts comparatively thin, extending about one-third of spermathecal diameter.
Discussion.R. (R.) filata sp.n. is distinctive by the structure of the antennae, having the pedicel large, compared to other species treated here, and the flagellum thin, filiform (Fig. 9).Body colouration is generally pale, bicoloured, with the light grey thorax and pale brown abdomen, similar to populations of R. (R.) hirticornis and R. (R.) eugeni sp.n. from the Balkans.R. (R.) filata sp.n. is close to R. (R.) eugeni sp.n., sharing with it the overall shape of the outer gonostylus, with rounded apex, without a distinct apical spine.It differs, however, from the latter by other details in the structure of the male and female terminalia, particularly the large, broadly fan-shaped apodeme of vesica in males and the spherical spermathecae in females.
Colour.General colouration dark greyish brown, dull, without conspicuous markings on thorax.Antenna brown throughout.Pleuron heavily suffused with grey pruinosity.Wing slightly infuscated.Halter dirty white to pale yellow.Coxae, trochanters and bases of femora mostly yellow to yellowish brown, otherwise legs somewhat darker, brown.Abdomen greyish brown.
Head.Antenna of moderate length, reaching to base of wing.Flagellomeres oval.Longest verticils on flagellomeres exceeding length of their respective segments.Pubescence indistinct.Palpus short.
Thorax.Wing rather narrow, about four times as long as broad, with comparatively long stalk (Fig. 16).Sc1 ending at about half length of Rs.Sc2 slightly apparent at tip of Sc1.R3 more than its own length beyond tip of R1.R4 with numerous macrotrichia, about 20, both dorsally and ventrally.A2 sinuous, ending shortly before origin of Rs.Halter comparatively long (compared to other species), reaching to about posterior margin of abdominal tergite 2 (Fig. 16).
Abdomen.Male terminalia (Fig. 35).Segment 9 longer than broad.Gonocoxite moderate in length and breadth.Outer gonostylus comparatively short, about half length of gonocoxite, gently and evenly arched, tapered distally, with distinct apical spine.Inner gonostylus generally conical.Aedeagal complex as in Fig. 35.Interbase moderate in length, reaching to about half length of gonocoxite, generally slender, bent at two-thirds of its length to form long and slender, lanceolate apical blade.Aedeagus slender, longer than vesica; the latter narrow, with long apodeme, rod-like or spine-like in dorsal aspect, subequal in length to vesica.Female terminalia (Figs 39,53).Cercus rather broad and short, subequal in length to tergite 10, abruptly tapered and upturned before apex.Spermathecae two, exceedingly large, irregularly short-oval to reniform, with sclerotised parts of ducts short and curved.plex, thus actually representing the opposite of R. (R.) crassa sp.n.R. (R.) subparva has, e.g., the wing stalk and halter rather long, the latter reaching to about the posterior margin of the abdominal tergite 2 (Fig. 16) (shorter in all related species, cf.Fig. 15).The numerous macrotrichia on R4 likewise are unique within the complex.The dark greyish brown body colouration combined with yellow coxae is only present in the Central European specimens of R. (R.) eugeni sp.n.The male terminalia of R. (R.) subparva are characterised by an evenly arched outer gonostylus tapered distally and provided with a distinct apical spine, and by a generally slender aedeagal complex with a narrow vesica and a rod-like apodeme.The female terminalia are distinctive in having rather broad, abruptly upturned cerci and, especially, two shortoval, exceedingly large spermathecae.
Distribution.The species was described comparatively recently, at the time when more attention was paid to small dark Rhabdomastix s. str.species, previously lumped under "schistacea".Therefore, it may be assumed that most of the country records published by Savchenko et al. (1992) are correct.These are as follows: Switzerland, Germany, Poland, Czech Republic, Slovakia, Austria, Ukraine, Italy, the former Yugoslavia (Slovenia, Serbia), Albania (?), Romania and Bulgaria (modified according to the present political boundaries).Based on the material examined, the species is here confirmed for Switzerland, Germany, Czech Republic, Slovakia, Austria, Italy, Slovenia, Romania, Bulgaria and Ukraine, and newly recorded for European Russia (southeast).Not confirmed from Poland, Serbia or Albania.
Diagnosis.General colouration grey on thorax, more brownish on abdomen.Male antenna very long, subequal to entire body.Wing narrow, slightly infuscated.A2 ending before origin of Rs.Legs yellowish brown to brown, with yellow coxae.Male terminalia with outer gonostylus gently arched, club-shaped, with apical spine not apparent; apical blade of interbase spoon-shaped.Female terminalia with three reniform, medium-sized spermathecae.
Redescription.Rather small species, yet somewhat larger than many others described here as "small".Body length 4-6.5 mm, wing length 5-7 mm.
Colour.General colouration grey to light grey, dull, without conspicuous markings on thorax, sometimes more brownish in middle of prescutum and still more so on abdomen.Antenna dark brown throughout.Pleuron heavily suffused with bluish grey pruinosity.Wing slightly infuscated.Halter dirty white to pale yellow, with knob faintly infuscated.Coxae yellow to yellowish brown, fore coxa more greyish.Trochanters and proximal half of femora yellow, the latter darkened distally.Rest of legs yellowish brown.Abdomen greyish brown to pale brown.
Head.Male antenna (Fig. 11) very long, subequal to entire body.Flagellomeres very long, first one rather conical, following ones cylindrical, longest near midlength of antenna, terminal flagellomere minute.Verticils inconspicuous, about one-fourth to one-fifth length of their respective segments, distinct at base of eight proximal flagellomeres, largely getting lost among greatly developed erect pubescence subequal to one-third length of longest flagellomeres.Female antenna (Fig. 12) considerably shorter than that of male, yet distinctly longer than antenna of any other species treated here (except for male of georgica), extending beyond base of wing by about one-fourth of its length.Flagellomeres mostly elongate-oval, progressively narrowed towards apex of antenna.Verticils on flagellomeres slightly shorter than their respective segments.Pubescence indistinct.Palpus long, distinctly exceeding diameter of head; terminal palpomere nearly twice as long as penultimate (Fig. 13).
Thorax.Wing rather long and narrow, about four to five times as long as broad, with comparatively short stalk.Sc1 ending at about half length of Rs.Sc2 faintly apparent at tip of Sc1.R3 more than its own length beyond tip of R1.R4 with numerous macrotrichia, about 20, both dorsally and ventrally.A2 sinuous, ending far before origin of Rs.Halter appearing rather long, but not reaching to posterior margin of abdominal tergite 2.
Abdomen.Male terminalia (Fig. 40).Segment 9 longer than broad.Gonocoxite moderate in length and breadth.Outer gonostylus more than half length of gonocoxite, generally slender, club-shaped, gently and evenly arched, slightly broadened before apex, with apical spine mostly not apparent, concealed by expanded apical portion of gonostylus.Inner gonostylus somewhat swollen, with short obtuse point.Aedeagal complex as in Fig. 40.Interbase comparatively long, extending beyond half length of gonocoxite, expanded distally to form spoon-like apical blade, sometimes pointed at apex.Aedeagus slender, longer than comparatively narrow vesica.Apodeme of vesica short, mostly spine-like from dorsal aspect, shorter than vesica.Female terminalia (Figs 42,54).Cercus slender, rather long, longer than tergite 10, gently upturned.Spermathecae three, medium-sized, short-oval to reniform, with sclerotised parts of ducts short, somewhat curved.One spermatheca sometimes smaller than other two.
Discussion.Members of Central European populations are distinctly darker than specimens from the Balkans (including the type series).The latter are more pronouncedly bicoloured, having a light grey thorax and a pale brown abdomen.This may be the case for other southern regions of Europe, as indicated by several specimens from southern Ukraine and a dry-mounted female from southern Switzerland (Ticino).On the other hand, the examined specimens from Algeria are still darker than those from Central Europe.There is a certain variation in the shape of the outer gonostylus; the apical spine is sometimes well apparent (cf.Lackschewitz, 1940, Tab. 3, Figs 28a, b), which, however, may be an artefact caused by compression of the hypopygium between the celluloid slides.
From all the species treated (except georgica), males of R. (R.) hirticornis may at once be separated by the long antennae (Fig. 11).The problem of long antennae in Rhabdomastix was discussed in some detail in the first part of this revision (Starý, 2003) and it is again mentioned in the discussion of the species complexes above.Due to the bicoloured appearance, females of R. (R.) hirticornis from South Europe may be confused with R. (R.) filata sp.n. or South European specimens of R. (R.) eugeni sp.n.It should, however, be emphasised that females of R. (R.) hirticornis are also clearly separable by the structure of the antennae.These are considerably shorter than those of males, yet distinctly longer than the antennae of any other species treated (Fig. 12).The long palpi in both sexes of R. (R.) hirticornis, distinctly exceeding the diameter of the head, with the terminal palpomere nearly twice as long as the penultimate, represent another distinguishing character (Fig. 13) (in the other species treated, except georgica, the palpi are short, subequal to the diameter of the head, with the terminal palpomere only slightly longer than the penultimate, cf.Fig. 14).The considerable distinctness of the male antennae in R. (R.) hirticornis is not reflected in the structure of the male terminalia, which provide no essential differences as compared to the other species, and only differ by details, such as the overall shape of the outer gonostylus and the structure of the aedeagal complex.The female terminalia of R. (R.) hirticornis are well characterised by three reniform, medium-sized spermathecae.Differentiating between R. (R.) hirticornis and R. (R.) georgica sp.n. is covered in the discussion of the latter species.
Rhabdomastix (Rhabdomastix) georgica sp.n. (Figs 41,43,55) Diagnosis.General colouration dark greyish brown.Male antenna very long, longer than entire body.Wing narrow, infuscated.A2 ending before origin of Rs.Legs yellowish brown to brown, with coxae yellowish brown.Male terminalia with outer gonostylus straight, with distinct apical spine, and apical blade of interbase triangular, with sharp point at inner margin.Female terminalia with three reniform, large spermathecae.
Colour.General colouration dark greyish brown, dull, without conspicuous markings on thorax, more brownish on abdomen.Antenna brown throughout.Pleuron heavily suffused with grey pruinosity.Wing infuscated.Halter infuscated, especially on knob.Coxae yellowish brown, fore coxa more greyish pruinose.Trochanters and femora yellowish brown, the latter darkened distally.Rest of legs yellowish brown to brown.
Head.Male antenna very long, longer than entire body.Flagellomeres very long, first one rather conical, following ones cylindrical, longest near mid-length of antenna, terminal flagellomere minute.Verticils indistinct.Pubescence, long, erect, rather sparse, subequal to one-third length of longest flagellomeres.Female antenna considerably shorter than that of male, not reaching to base of wing.Flagellomeres oval to elongateoval, rather thin, not noticeably changing in size towards apex of antenna.Verticils on flagellomeres shorter than their respective segments.Pubescence indistinct.Palpus long, distinctly exceeding diameter of head; terminal palpomere nearly twice as long as penultimate.
Thorax.Wing long and narrow, about five times as long as broad, with long stalk.Sc1 ending at about half length of Rs.Sc2 faintly apparent shortly before tip of Sc1.R3 more than its own length beyond tip of R1.R4 with numerous macrotrichia both dorsally and ventrally.A2 long, sinuous, ending before origin of Rs.Halter comparatively short, not reaching to posterior margin of abdominal tergite 2.
Abdomen.Male terminalia (Fig. 41).Segment 9 longer than broad.Gonocoxite moderate in length and breadth.Outer gonostylus slightly more than half length of gonocoxite, gently bent at base, otherwise straight, nearly parallel-sided, with apical spine distinct.Inner gonostylus generally conical.Aedeagal complex as in Fig. 41.Interbase comparatively long, extending beyond half length of gonocoxite, with apical blade generally triangular, drawn out into sharp long point at inner distal margin, directed laterally.Aedeagus subequal in length to comparatively narrow vesica.Apodeme of vesica short, spine-like or somewhat bulb-shaped at apex from dorsal aspect, shorter than vesica.Female terminalia (Figs 43,55).Cercus slender, rather long, longer than tergite 10, gently upturned.Spermathecae three, large, reniform and considerably narrowed in portion closer to duct, practically without sclerotised parts of ducts.
Material examined.Holotype %: Georgia (Transcaucasia), Arsianskiy khrebet [mountain ridge], E slopes of Goderdzi Pass (1450-1500 m), 28.vi.1978(E.N.Savchenko leg.) (SMOC).Except for a printed inscription "Transcriptio" the data on the label are hand-written in Russian (in Cyrillic).The specimen (originally papered) is glued onto a triangular cardboard point, in nearly perfect condition, with only wings somewhat crumpled and stuck together and apex of abdomen missing.Terminalia dissected and placed in a sealed plastic tube with glycerine, pinned with the specimen.Paratypes: 3%, 1&, same data as for holotype (JSO).Etymology.The name of the new species, georgica, is derived from the name of the country of its occurrence, Georgia in Transcaucasia.The name is deemed to be and to be treated as a latinised adjective in nominative singular, in accordance with relevant provisions of the Article 11.9 of ICZN (1999).
Discussion.R. (R.) georgica sp.n. is described from Georgia in Transcaucasia, hence from outside of Europe, representing thus an extra-limital species in terms of this revision.It was included here because of its affitinies to R. (R.) hirticornis.Within the Palaearctic Region, only three other Rhabdomastix species are known to be distinguished by correspondingly long male antennae, viz.R. (R.) hirticornis, Lackschewitz, 1940 (Europe), R. (R.) leucophaea Savchenko, 1976 (Transcaucasia: Azerbaijan) and R. (R.) omeina Alexander, 1932 (China: Sichuan), all formerly classified in the subgenus Palaeogonomyia (cf.Savchenko et al., 1992).In the latter two species the antennae are shorter than the body.Five Oriental species with greatly lengthened male antennae (himalayensis Alexander, 1960;manipurensis Alexander, 1964;nilgirica Alexander, 1949;schmidiana Alexander, 1958;trochanterata Edwards, 1928) all belong to the group centred around R. illudens Alexander, 1914, as discussed in the first part of this revision (Starý, 2003: 590), with the male antennae several times as long as the body.
R. (R.) georgica sp.n. apparently is closely related to R. (R.) hirticornis.It differs from Central European specimens of the latter by somewhat darker body colouration as well as having longer male antennae, which, in R. (R.) georgica sp.n., exceed the length of the entire body (subequal in length to the body in hirticornis).However, the female antennae of R. (R.) georgica sp.n. are shorter than those of R. (R.) hirticornis, not reaching to the bases of the wings (extending beyond the bases of the wings in hirticornis).The male terminalia of R. (R.) georgica sp.n. are characterised by a generally straight outer gonostylus, nearly parallel-sided, with the apical spine distinct (evenly arched in hirticornis, club-shaped at apex, without a distinct apical spine), and by the interbases with the apical blade triangular, drawn out into a sharp long point at inner margin (apical blade of interbases generally spoon-like in hirticornis).The female terminalia are distinctive in having the spermathecae comparatively large, reniform (smaller in hirticornis, often short-oval).
Diagnosis.General colouration grey throughout.Antenna short.Wing moderately broad, somewhat milky, with narrow darker seams along veins.A2 ending opposite to origin of Rs.Legs yellow to yellowish brown, femora darkened distally.Male terminalia with outer gonostylus unusually short and broad, inner gonostylus swollen, broadly rounded at apex and aedeagal complex generally slender.Female terminalia with three spherical, medium-sized spermathecae.
Colour.General colouration grey, dull, restrictedly tinged with brownish, without conspicuous markings on thorax.Antenna deep dark brown to black throughout.Pleuron heavily suffused with grey pruinosity.Wing slightly infuscated, somewhat milky, most veins vaguely and very narrowly seamed with darker.Halter bright pale yellow.Coxae, trochanters and proximal half of femora yellow, the latter darkened distally.Tibiae yellowish brown, tipped with darker.Tarsomere 1 yellowish brown, others darker.Abdomen a little darker than thorax, greyish brown.
Head.Antenna (Fig. 10) comparatively short, not reaching to base of wing.Flagellomeres short-oval.Longest verticils on flagellomeres slightly exceeding length of their respective segments.Pubescence rather long, suberect, subequal in length to breadth of respective segments, or slightly less so, distinct on almost all flagellomeres.Palpus short.
Thorax.Wing (Fig. 3) moderately broad, more than three times as long as broad, with short stalk.Sc1 ending at about half length of Rs.Sc2 faintly apparent shortly before tip of Sc1.R3 more than its own length beyond tip of R1.R4 with a few macrotrichia dorsally.A2 strongly sinuous, ending opposite to origin of Rs.Halter short, clearly not reaching to posterior margin of abdominal tergite 2.
Abdomen.Male terminalia (Fig. 44).Segment 9 very short, broader than long, with conspicuous, more or less triangular lobe dorsally at posterior margin on each side of median interruption.Gonocoxite stout, short and broad.Outer gonostylus unusually short and broad, somewhat flattened at base, about one-third length of gonocoxite, subequal in length to inner gonostylus, nearly straight and parallel-sided, with distinct apical spine.Inner gonostylus very broad, swollen, broadly rounded at apex.Aedeagal complex as in Fig. 44.Interbase reaching to about half length of gonocoxite, generally slender, sinuous and only faintly dilated distally.Aedeagus very slender and rather long, about 1.5 times as long as moderately broad vesica.Apodeme of vesica rod-like or spinelike in dorsal aspect, about same length as vesica.Female terminalia (Figs 45,56).Cercus broad, subequal in length to tergite 10, generally straight, expanded and abruptly upturned before apex.Spermathecae three, rather large, spherical, with sclerotised parts of ducts thin and long, exceeding spermathecal diameter.
Material examined.The species was described from an unspecified number of males ["Banat, Orsova.Im Juni, %%, leg Th.Becker no.61829.(Typ. in der Samml.des Zool.Museums in Berlin.)"(Lackschewitz, 1935: 13)].I have examined two specimens that may be considered syntypes.Lectotype % (present designation): Romania, [Banat], Orsova, vi., (Th.Becker leg.) (ZMHB), labelled: "Orsova 61829.VI." (hand-written), "Gonomyia schistacea Schumm."(hand-written in pencil), "Sac.beckeri nov.sp.det.Lacksch."(printed, orange).Accordingly labelled as lectotype ("Lectotype Rhabdomastix (s.str.) beckeri (Lacksch.)% J. Starý 2003").The specimen is micro-pinned on a stage of plant parenchyma, with only left hind leg present; left antenna and apex of abdomen broken off.Terminalia dissected and placed in Canada balsam between celluloid slides, pinned with the specimen.Paralectotype: 1% (ZMHB) with same labels as lectotype (incl."61829"), except for the label with the schistacea identification.Terminalia dissected and placed in a sealed plastic tube with glycerine, pinned with the specimen.The paralectotype belongs to the species described here as R. (R.) eugeni sp.n. and is also listed as paratype under that species.The specimen was examined by me in 1978 (later not traced again in ZMHB), and it was not labelled by me as the paralectotype of beckeri, nor the paratype of eugeni.Hence, the type series of R. (R.) beckeri is a mixed one, and the lectotype is designated here to maintain the current usage of the name for the species with the male terminalia as illustrated by Lackschewitz (1935, Figs 6a,b).
Discussion.Some character traits of R. (R.) beckeri indicate a more distant relationship to the other species of the European Rhabdomastix s. str.In general appearance, R. (R.) beckeri is particularly distinctive by its somewhat milky wings with narrow darker seams along most of veins.The isolated position of R. (R.) beckeri is also indicated by some features in the structure of the generally rather robust male terminalia.Segment 9 is broad and short, broader than long, with a conspicuous triangular lobe dorsally at the posterior margin on each side of the median interruption (Fig. 44) (broader than long, but without any lobes, in the species centred around laeta; longer than broad, with only small lobes in the other species).In contrast to all the species, the outer gonostylus is unusually short and broad, the inner gonostylus conspicuously swollen and broadly rounded at apex and the interbases and the aedeagus is very slender.The female terminalia, although less distinctive, are well characterised by the shape of the cerci which generally are straight and broad, expanded and abruptly upturned before the apex.(Lackschewitz, 1935).Scale bars 0.25 mm.
It had been recorded from Slovakia by Starý (1987), but later this record was withdrawn (Starý, 1993(Starý, , 1996)).R. (R.) beckeri is very distinctive in the structure of the male terminalia, and the above specimens from Malé Trakany, Slovakia, had readily been identified with the figures by Lackschewitz (1935, Figs 6a,b).However, some doubts arose concerning the type [at that time, I had only examined the specimen listed here as paralectotype belonging to R. (R.) eugeni sp.n.].The records from the former Czechoslovakia by Savchenko (1989) and Savchenko et al. (1992) are based on the same unpublished material from Slovakia sent by me to E.N. Savchenko.Here I am publishing the first documented records from Slovakia.
DOUBTFUL SPECIES OF RHABDOMASTIX S. STR.
ACKNOWLEDGEMENTS.For invaluable information and/or for the loan and gift of specimens, I am much indebted to the following: R. Contreras-Lichtenberg, P.