Revision of Tipula (Yamatotipula) stackelbergi Alexander (Diptera, Tipulidae), and a short discussion on subspecies among crane flies

Abstract All available type material of Tipula stackelbergi Alexander, Tipula usuriensis Alexander and Tipula subpruinosa Mannheims were examined. Tipula (Yamatotipula) stackelbergi stat. rev. is elevated from a subspecies of Tipula (Yamatotipula) pruinosa Wiedemann to a valid species. Two new synonyms are proposed: Tipula usuriensis syn. n. proved to be a junior synonym of. Tipula (Yamatotipula) pruinosa and Tipula subpruinosa syn. n. a junior synonym of Tipula (Yamatotipula) freyana Lackschewitz. Tipula (Yamatotipula) stackelbergi is redescribed, male and female terminalia of Tipula (Yamatotipula) pruinosa are illustrated and discussed. Female terminalia of Tipula (Yamatotipula) freyana are described and illustrated for the first time. A key to both sexes of Tipula (Yamatotipula) stackelbergi and Tipula (Yamatotipula) pruinosa, and a key to females of Tipula (Yamatotipula) chonsaniana, Tipula (Yamatotipula) freyana and Tipula (Yamatotipula) moesta are provided. Subspecies are not uncommon among crane flies, but their ranges and traits are poorly known. An interdisciplinary approach (genetics, ecology, taxonomy) is suggested if subspecific ranks are to be used in tipuloid systematics.


Introduction
The description of Tipula stackelbergi (Diptera, Tipulidae) was based on male holotype collected from Russian East Siberia (Alexander 1934). Later this species was also recorded from the Russian Far East (Savchenko 1961;Pilipenko 2009). Savchenko (1961) considered T. stackelbergi as a subspecies of T. pruinosa Wiedemann, based on the small structural differences between the two taxa. He also transferred T. stackelbergi to the subgenus Tipula (Yamatotipula) Matsumura. In the same publication, Savchenko suggested two synonyms for T. stackelbergi, namely T. usuriensis Alexander, 1925 andT. subpruinosa Mannheims, 1954. However, both synonyms were uncertain because Savchenko did not examine the type material of these two species. The description of T. usuriensis was based on male holotype, collected from Siberia (exact locality uncertain) and the description of T. subpruinosa was based on two females, collected from northern Finland (holotype) and Sweden (paratype) (Alexander 1925;Mannheims 1954). Since the taxonomic treatment by Savchenko (1961), both species have remained synonyms of T. stackelbergi (e.g. Oosterbroek and Theowald 1992) and apparently the type material has remained unstudied. In addition, Tipula stackelbergi has been used as an example of a tipulid (sub)species with a large but disjunct range in the Palaearctic region (Oosterbroek et al. 2001).
Subspecies are traditionally held as geographically separate and genetically distinct populations within the species' range, permitting gene flow in the area of contact (Wilson and Brown 1953;O'Brien and Mayr 1991;Patten and Unitt 2002). Despite possible interbreeding between subspecies, subspecies may retain differences in respective life cycles or other traits (Hewitt 2002;Kothera et al. 2009). Among birds, high subspecies richness was associated with large breeding ranges, island dwelling, inhabiting montane regions, habitat heterogeneity and low latitude; on the other hand, species phylogenetic age was a poor predictor of subspecies richness (Phillimore et al. 2007). Definition of subspecies, and propensity of naming subspecific taxa, vastly differs among taxonomic groups. High proportions of higher plants, mammals and birds have subspecies, less so compared to invertebrates (Haig et al. 2006). New molecular methods have revolutionized subspecific classifications: i) formerly held subspecies gain no support at all, ii) subspecies are proposed to be valid species or iii) their status as operational evolutionary units is supported (Ball and Avise 1992;Patten and Unitt 2002;Tsao and Yeh 2008;Miller et al. 2011). Despite problems in correct recognition and delineation of subspecies, subspecific taxa are seen as powerful tools in conservation and as meaningful biological entities (Haig et al. 2006;Phillimore and Owens 2006).
In crane flies (Diptera, Tipuloidea) subspecific ranks are not uncommon. For example, out of 493 and 168 Palaearctic Tipulidae taxa described by C.P. Alexander (1889-1981) and E.N. Savchenko (1909-1994, respectively, 24 and 26 taxa are currently ranked as subspecies (data from Oosterbroek 2011). However, the recent tendency has been to elevate former subspecies to valid species (Starý 2006;Salmela and Autio 2009;Starý and Brodo 2009). In these cases, former subspecies are clearly separated upon differences in male and female hypopygial structures. In addition, due to the improved faunistic knowledge, range-sizes of former subspecies are in reality much larger than was previously known. On the other hand, some western Palaearctic (sub) species are most probably recent origin of Pleistocene glacial and interglacial periods; examples of such species are present in especially in the Iberian peninsula and Asia minor (Oosterbroek 1980). In general, tipuloid subspecies are elusive and very poorly known, and no rigorous assessment on the suitability of subspecific rank among crane flies has been carried out. Based on subjective opinion, perhaps a majority of the current Palaearctic tipuloid subspecies are in fact valid species. Furthermore, most allopatric or parapatric crane fly populations, that are genetically distinct from nominotypical (sub)species, are still to be found by biologists. Based on above mentioned references, subspecies should not be proposed on exiguous basis, relying on a small number of studied specimens and subtle differences in coloration or other structures. Instead, an interdisciplinary approach (genetics, ecology, taxonomy) is suggested if subspecific ranks are to be on a solid ground.
In this article I present the results of an examination of all available type material of T. stackelbergi, T. usuriensis and T. subpruinosa. I propose changes to the nomenclature of these species and I also review the morphology of T. pruinosa and T. stackelbergi, with an emphasis on male and female genitalia. In addition, female genitalia of T.
With except of the male hypopygium, the holotype specimen is in rather good condition (Figs 1b, c). All legs are detached from the specimen, but four legs are glued to the pin below the specimen. Tips of wings are broken. Right antenna is broken, only scape and pedicel are left; left antenna has seven flagellomeres. Tip of abdomen is broken; apparently hypopygium is mounted on a celluloid strip, which is attached on a pin. The surface of this strip is heavily cracked, and the structure of the hypopygium cannot be examined.
Abdominal tergites yellowish brown, slightly darkening toward tip of abdomen. 9 th tergite with two median projections, densely covered by dark bristles. Lateral corners of 9 th tergite glabrous, pointed (Fig. 2a). 9 th sternite with median incision, bearing two fleshy and hairy outgrowths in the margin of the incision. Outer gonostylus wormlike, apical half covered by light hairs (Fig. 2b). Inner gonostylus elongate (Figs 2b, c, 3e); beak rounded, with ten stout apical bristles and four subapical weaker bristles; central ridge with few weak bristles along its length; lower beak roundish, not angular. Posterior immovable apodeme of sperm pump almost straight (Fig. 2d). Distal end of compressor apodeme of sperm pump club-shaped, roundish (Fig. 2f ). Aedeagal guide as in Fig. 2e.
Material examined. Holotype of T. usuriensis: male, pinned specimen (USNM). "Kudia River/Amagu Siberia/Cockerell/July 1923" (white label, printed). "HOLO-TYPE /Tipula/ usuriensis/ C.P. Alexander" (red label, partly handwritten). Slide, permanently mounted wing. "Tipula usuriensis Alex./ ♀ Siberia, Amagu,/ Kudia River/ July 1923, (T.D.A. Cockerell) / The Alexander Collection of Crane-Flies/ HOLOTYPE 2967" (white label, partly handwritten). (Figs 3a, b). The holotype specimen of T. usuriensis is in quite bad condition (Fig. 3c). Half of the abdomen (distal part) and four legs are glued to a card. One wing (length 14.0 mm) is slide mounted and one wing is glued to a white card, one leg is also glued to the same card. Scape, pedicel and three flagellomeres of antennae are present. The holotype is also laterally flattened, perhaps due to compression of the freshly collected specimen. Hypopygium was detached by the author from the cardboard, macerated in KOH and finally preserved in glycerol in a microvial. Redescription of male and female terminalia. Male. 9 th tergite (Fig. 4a) essentially similar to T. (Y.) stackelbergi. 9 th tergite with two median projections, densely covered by dark bristles, lateral corners of the tergite glabrous, pointed (Fig. 4a). 9 th sternite with median incision, bearing two fleshy and hairy outgrowths. Outer gonostylus worm-like, apical half covered by dark hairs (Figs 4b, c). Inner gonostylus elongate. Beak rounded, rather wide, resembling helmet (Figs 4b, c, 3f-g). Apical portion of beak bearing around 20 stout bristles, central ridge with numerous weak bristles, along the whole length of the ridge. Lower beak angular. Posterior immovable apodeme of sperm pump curved in lateral and ventral view (Figs 4d, f ). Distal end of compressor apodeme of sperm pump truncated (Fig. 4f ). Aedeagal guide as in Fig. 4e.
Female. Female terminalia as in Fig. 6b. Basal part of hypogynial valves with dense black setae, proximal ends of valves rounded, widest sub-basally, not tapering toward proximal end (Fig. 6e). Stalk of genital fork gradually widening toward caudal and proximal ends, being narrowest around midpoint (Fig. 6f ). Dorsal view of vaginal apodeme as in Fig. 6f.
The holotype specimen is in good condition (Figs 5b, c, d). Left mid leg is missing, other legs are intact. Right wing has minor rupture proximal to the pterostigma, Costa is slightly damaged. Abdominal terminalia of the specimen were detached by me, macerated in KOH and later preserved in glycerol in a microvial. This microvial is attached to the same pin as the specimen. The name "stigma" has never been published, and it has most probably been a working title by Mannheims while compiling his first account of Finnish tipulids (Mannheims 1954).
Other material. Finland. Description of female terminalia. Female terminalia as in Fig. 7a. Basal part of hypogynial valves with modest setosity, proximal ends of valves pointed (Fig. 7b). Genital fork of vaginal apodeme dark brown, slightly sinuous in lateral view. Dorsal view of vaginal apodeme and genital fork as in Fig. 7c. 3 Basal part of hypogynial valves widest sub-basally, not tapering toward base (Fig. 6f ). Stalk of genital fork gradually widening toward caudal and proximal ends, being narrowest around midpoint (Fig. 6g)  remains questionable whether T. (Y.) pruinosa sinapruinosa is a valid subspecies. Based on the original description (Yang and Yang 1993) it is likely that Chinese specimens are conspecific with other eastern Palaearctic T. (Y.) pruinosa specimens. If these eastern Palaearctic specimens are to be ranked as subspecies below T. (Y.) pruinosa, T. usuriensis is the oldest available name for the taxon. However, as discussed above, subspecies should be delineated through several criteria, e.g. ecology and genetics. More data on Asian T. (Y.) pruinosa populations should be available for the assessment of speciation and reliable use of subspecific rank. Tipula (Y.) pruinosa and T. (Y.) stackelbergi are closely related but valid species. The species pair is well separated due to the differences in male genitalia (see the key to the species), but less so regarding female genitalia. More females of T. (Y.) stackelbergi should be studied in order to firmly validate the diagnostic differences presented here. Tipula (Y.) stackelbergi is a very rarely collected species, known only from East Siberia and the Russian Far East (Alexander 1934;Savchenko 1961;Pilipenko 2009).
Tipula subpruinosa, described from Finland and Sweden, was thought to be a synonym of T. (Y.) stackelbergi (Savchenko 1961;Oosterbroek and Theowald 1992). Due to this tentative synonymy, T. (Y.) stackelbergi was erroneously thought to be present in Fennoscandia. However, examination of the holotype of T. subpruinosa revealed that the species is a junior synonym of T. (Y.) freyana, not T. (Y.) stackelbergi. Hence, T. (Y.) stackelbergi should be removed from the list of European crane flies. It should be noted that the description of T. subpruinosa was very short and lacking any figures; it is not surprising it led to fallacious interpretation. In a similar vein, T. usuriensis was also tentatively synononymized by Savchenko (1961) with T. (Y.) stackelbergi. In his description of T. usuriensis Alexander (1925) provided figures depicting male 9 th tergite and lateral view of hypopygium, but these figures can now be considered too general to discriminate between T. Kuosmanen and Liisa Puhakka (Turku) are thanked for translations of Russian texts. Constructive comments by Fenja Brodo (Ottawa) and John Kramer (Leicester) improved the manuscript.