A Review of Daspletis Loew, 1858 with the Addition of a Remarkable South African Species (Diptera: Asilidae: Stenopogoninae)

ABSTRACT Microstylum vespertilio Engel, 1932 is redescribed from the Northern Cape Province of South Africa and transferred to Daspletis Loew, 1858, a genus closely related to Microstylum Macquart, 1838, that now incorporates eight southern African species. Knowledge of the species is updated and a new identification key is provided.


INTRODUCTION
Daspletis Loew, 1858 is a genus of stenopogonine Asilidae confined to the southern parts of the Afrotropical Region, and although there were only seven described species at the outset of the present study, the taxonomic history of this small but interesting genus is fairly complicated and justifies the following brief summary. Loew (1858) -Described Daspletis on a female collected by Wahlberg from 'N'Gami' (in present day Botswana) to which he assigned the name vulpes. Ricardo (1925) -Redescribed vulpes on material from present day Zambia and added a new species, hirtus, from Zimbabwe. She also described Microstylum hermanni from South Africa as well as the new genus, Neodysmachus, founded on a single South African species, setithoracicus. Engel (1932) -Recorded additional material of Microstylum hermanni from the present day Eastern Cape (South Africa), describing vespertilio as a new variety of the species from the Northern Cape. Hull (1967) -Described Daspletis albosetatus on a single female from Namibia. Oldroyd (1974) -In reviewing the southern African Asilidae, produced a key to Daspletis that included four species-albosetatus (Angola, Namibia), hirtus (Botswana, South Africa, Zimbabwe), vulpes (Botswana, Zambia) and a new species, salicior (South Africa). Oldroyd (1980) -Catalogued the species mentioned above as Microstylum hermanni, M. vespertilio (as a full species and not as a variety of hermanni), Neodysmachus setithoracicus, Daspletis albosetatus, D. hirtus, D. salicior and D. vulpes. Londt (1983) -Revised Daspletis, presenting a key to six species: hermanni (transferred from Microstylum), hirtus, setithoracicus (established within the genus as a consequence of the synonymy of Neodysmachus), vulpes, and two new species, placodes and stenoura, both from South Africa. D. salicior was synonymised with hermanni, while albosetatus became the type species of a new genus, Remotomyia. Londt (1985) -Added lykos from Namibia and South Africa, bringing the number of species to seven (i.e. hermanni, hirtus, lykos, placodes, setithoracicus, stenoura, vulpes). Londt (1999) -Published a key to Afrotropical Stenopogoninae that included Daspletis. Dikow (2009a) -Included stenoura in a major cladistical study of the Asilidae that involved 158 species, representing 140 genera. The analysis supported the close relationship between Daspletis and Microstylum. The need for the present contribution emanates partly from the collection of a good series of a rather spectacular species from the Northern Cape Province of South Africa (Fig. 1). This species not only displays the most remarkable sexual dimorphism known for any within the genus, but it has the distinction of being by far the largest species in the group; males possessing a wingspan of some 65 mm. In addition, observations made on both the flight and resting habits of this species suggest an interesting reproductive behaviour worthy of further study. Although I initially believed this Northern Cape giant to be new to science, I subsequently discovered that it represented Engel's (1932) variety of Microstylum hermanni called vespertilio, subsequently catalogued as a full species (Oldroyd 1980). This paper is devoted to a redescription of vespertilio and its transfer to Daspletis and the reporting of observations made when collecting the new series. I also use this opportunity to report on significant additional material that has accumulated in the Natal Museum over the last twenty-five years of active collecting, and to provide a new key for the identification of the species.

Specimens
Material listed in this study is housed in the following institutions: BMNH -The Natural History Museum, London, UK; FMNH -Field Museum of Natural History, Chicago, USA; NMSA -Natal Museum, Pietermaritzburg, South Africa; SMNS -Staatliches Museum für Naturkunde Stuttgart, Germany. A few specimens that are to be used for dissection and molecular studies are in the private collection of Dr T. Dikow.

Label data
In recognition of the value of detailed lists of material examined (Dikow et al. 2009), standard formats have been employed when recording label information. Information is usually restricted to locality, date of collection, collector(s) (initials excluded), altitude (when available) and any other potentially useful information available. Should more detail be required, it may be available from the relevant institution. When material is considered to have special interest (e.g., type specimens), all label data are reproduced as appearing on labels, lines of information being separated by a slash (/). Square brackets are used when useful additional information, not found on labels, or comment, is provided. In this regard, coordinates are usually given when these, or a quarter-degree grid reference, do not appear on a label. Coordinates provided are usually for the populated place or geographic feature mentioned immediately before the added note, no attempt being made to estimate coordinates for places recorded as being a certain distance from a populated place. The use of question marks indicates unknown or questionable information. All specimens are arranged in geographical order according to latitude and within alphabetically ordered countries. TAXONOMY Genus Daspletis Loew, 1858Daspletis: Loew 1858Londt 1983: 285. Type species: D. vulpes Loew, 1858: 337, by monotypy. Neodysmachus Ricardo, 1925: 263. Type species: N. setithoracicus Ricardo, 1925 Diagnosis: The following characters were extracted from Londt's (1999) key to Afrotropical stenopogonine genera and serve as a brief generic diagnosis. Anatergites setose (a character effectively dividing the subfamily into two distinct groups); antennal postpedicel lacking a distinct style, and tipped with a small pit enclosing a sensory spine (this feature is shared only by Microstylum Macquart, 1838); facial swelling occupying lower three-quarters of face, entirely setose; presutural dorsocentral macrosetae well developed; vein M 1 not strongly arched anteriorly; postmetacoxal membrane long setose (these last few characters are used to separate the genus from Microstylum).
The wings are very broad and brown, with lighter spots in the cells of posterior margin, and closed cell R 5 . The femora of all legs and all bristles of body are black, also those of mystax. Occiput, pronotum and mesonotum with dirty yellowish hairs. The designs on mesonotum are identical with such in Hermanni, and the whitish middle stripe very conspicuous. The pubescence of abdomen black.
Length of body 30 mm.; length of wing 25 mm.; breadth of wing 8 mm.
posteromedially and with somewhat undulating hind margin. Epandrial lobes simple, in 'dorsal' view ( Fig. 6) deeply incised medially resulting in two lobes, weakly attached basally, in lateral view lobes project distally slightly beyond levels attained by other genital components. Proctiger simple, cerci deeply incised medially with weakly sclerotized basal parts. Gonocoxite well developed, distal end somewhat truncate (lateral view) and broadly bifurcate (ventral view); gonostyle well developed, curved toward epandrium (lateral view). Hypandrium well developed, in ventral view (Fig. 7) broad proximally, tapering fairly rapidly to narrowly-rounded distal lobe; laterally, hypandrial lobe almost parallel sided (although somewhat laterally compressed and consequently appearing paler in colour than surrounding structures), with smoothly rounded distal end. (Note: Londt (1983) incorrectly referred to the hypandrium as 'fused gonocoxites'.) Description of female: While females ( Fig. 1) share many features with males, there is significant sexual dimorphism, especially with respect to the coloration, size and shape of wings. For this reason the following notes relating to females are provided.
Head: Orange-brown to dark red-brown, macrosetae predominantly yellowish, smaller setae white (a few black setae may occur on scape, vertex, occelar tubercle and palpal tip). Antenna: Red-brown, white setose (some macrosetae situated ventrally on scape may be yellowish or dark brown). Mystax and occiput with yellowish macrosetae accompanied by smaller white setae.
Thorax: Less extensively dark red-brown when compared to males, posterior parts of mesonotum somewhat orange. Some pleura (e.g. katepisternum) somewhat orange. Setae mostly fine white, macrosetae yellowish except for dorsocentrals and some laterally situated macrosetae that may be black (there is great variation and no pattern can be established). Scutellar macrosetae also variable in colour, being either yellow or black.
Haltere yellow with darker stalk. Legs: Mostly dark red-brown (prothoracic femora and tibiae may be somewhat orange). Almost all minor setae white, macrosetae variable white or dark red-brown. Wings: Mean length (L) and breadth (B) 17.6×5.7 mm (n=6). Female wings shorter and narrower (L/B=3.1) than in males (L/B=2.7). Venation ( Fig. 4) similar to males, but little or no remnants of 'crossveins' on veins M 2 and M 3 . Membrane transparent, almost entirely lacking staining except narrowly along interior cell margins (this slight staining gives veins a characteristic 'shaded' appearance).
Abdomen: Dark red-brown, but posterior margins of terga (T) somewhat orange. Apart from yellow macrosetae laterally on T1 all terga minutely white setose. Except for setae of T1 and anterolateral parts of T2 all abdominal setae directed anteriorly. Terminalia (Figs 8-10): T8 (epigynium) well developed, T9 not evident (reduced and membranous or fused with T10), T10 (acanthophorites) well developed, divided into two halves medially, each half bearing 5 well developed, dorsoventrally compressed macrosetae; cerci broad, well developed, fairly broadly fused basally. S8 (hypogynium) broadly rounded posteriorly, distally broad with medially incised anterior region. Three swollen, bladder-like spermathecal reservoirs visible on dissection, each with a long, narrow, weakly sclerotized duct ( Fig. 10)  . Note: Although all the Aggeneys material was collected on the same occasion, the FMNH specimens, which are slightly differently labelled, were not available to me when the redescription was drafted. There is, however, no doubt that these are correctly assigned to this taxon. In addition, the following specimens were also obtained and are available for DNA sequencing and dissection: 1_ 1^in 95 % ethanol (Dikow coll.), 1_ 2^Kahle's solution (Dikow coll.), 1_ in 95 % ethanol (NMSA).
Comparison: Species of Daspletis can be segregated into two species groups (see keycouplet 2). D. vespertilio clearly belongs to the hermanni group by virtue of its short dorsal occipital macrosetae, lack of macrosetae on abdominal terga (other than T1), and generally slender abdomen. In these respects vespertilio differs from species in the vulpes group (hirtus, placades, stenoura and vulpes). Although the males of vespertilio share the feature of darkly stained wings possessed by both hermanni and lykos their wings are much blacker and larger than in these species. Species in the hermanni group also share similar male genital characteristics (as do those in the vulpes group). The similarities are best appreciated through a comparison of published illustrations. The illustrations of the terminalia of vespertilio provided in this paper (Figs 5-7) should be compared with those of hermanni (figs 8-10 in Londt (1983)), setithoracicus (figs 19-21 in Londt (1983)) and lykos (figs 1-3 in Londt (1985)). Perhaps the most obvious differences displayed by vespertilio males are to be found in the somewhat reduced hypandrium which lacks the broad appearance of the laterally compressed distal lobe seen in the other species. Distribution (Fig. 14), phenology (Table 1) and biology: D. vespertilio has only been collected at three localities. The date of collection has not been recorded for the holotype, while all the other specimens were collected early in October. Both newly recorded sites were similar in that they feature poorly vegetated and fairly extensive red sand dune systems (Fig. 12). Although little biological information is available, some insights Figs 11, 12. Daspletis vespertilio (Engel, 1932): (11) male resting on red sand, (12) general view of habitat type, 5 km S Pella. may be inferred from field observations made by both Dr Dikow and me, as well as from their general morphology. Male individuals were first encountered after they had been disturbed and had taken to flight. Their highly characteristic, fairly slow, flapping flight was intriguing and reminiscent of some large antlions (e.g., species of Palpares). When pursued, often over considerable distances, they appeared to take evasive action by flying in an erratic manner. The fact that vespertilio males have large, broad, black, and therefore highly visible wings (as well as a long, slender, stabilizing abdomen), when compared with many other asilids, and that their flight is slow but sustained, strongly suggests that they probably perform an aerial display, presumably designed to attract the attention of females. Aerial displays are fairly well documented for Asilidae and include a number of species within the Stenopogoninae (Lavigne 2003). Females, with their smaller, narrower, and almost transparent wings, flew more rapidly, and in a relatively unsustained manner when compared with their male counterparts. Both sexes rested on the ground, wings folded one upon the other over their abdomens, and with their legs outstretched (Fig. 11).
Meteorological data indicate that much of the Northern Cape, including the localities recorded for vespertilio, receives most of its rain between February and April, and virtually no rain falls between July and October. D. vespertilio, therefore, flies at a time when the vegetation is parched and relatively little invertebrate life is evident. This probably accounts, in part, for the fact that this large species has been overlooked until now. Temperatures during October may be cool at night and relatively mild during the day. The dark coloration of males is almost certainly an adaptation to generally cool weather conditions and is reminiscent of insects associated with the winter rainfall areas of the Western Cape. This is in stark contrast with species in the vulpes species group (hirtus, placodes, stenoura and vulpes) that are mostly late summer active, and pale yellowish in colour. Londt's (1983) revision of Daspletis was based on 139 listed specimens, a further three being added with the description of lykos (Londt 1985). In order to more adequately define the distributions of the species, I list below all the material known to me, previously recorded and new (amounting to 317 specimens), and provide comments concerning distribution, phenology and biology of each species.

Annotated checklist of other Daspletis species
Daspletis hermanni (Ricardo, 1925) Fig. 13 Remarks: On rechecking the Natal Museum's holdings of Microstylum I found, under the name M. floccosum Hermann, a long series (17_ 23^) of hermanni from Willowmore, all collected by Dr Brauns. These are listed above and it will be noted that 1_ carries the same label data as the lectotype and paralectotype, housed in the BMNH. There is also 1^with the same label data as the 3_ and 5^previously listed by Londt (1983). Hermann did not describe a Microstylum with the name floccosum and so this is at best a manuscript name without standing. Of further interest is the fact that, together with these newly discovered hermanni specimens, were 5_ and 5^specimens that do not represent hermanni, but appear to truly represent a species of Microstylum that is very similar in general appearance to hermanni. The identity of these specimens will only become known after the publication of a much needed modern revision of Microstylum. Distribution, phenology and biology: D. hermanni is fairly widely distributed within the central region of South Africa (Fig. 13), occurring in both the Northern Cape, and Eastern Cape provinces. The species is active in the adult phase from August through to November (Table 1) Daspletis hirtus Ricardo, 1925 Fig. 14 Daspletis hirtus : Ricardo 1925: 263;Londt 1983: 289, figs 11-15. Londt (1983  Distribution, phenology and biology: D. hirtus has been collected in the northern parts of the Northern Cape province of South Africa and at Sawmills in Zimbabwe. It must be assumed that the species also occurs in the intervening area of Botswana where the species has not yet been recorded (Fig. 14). Adults have been collected in November, February and March (  (1985) did not give type status to the two female specimens as he was uncertain that they truly represented the species, there is little doubt that they should now be considered conspecific as the sexual dimorphism displayed by the closely related vespertilio is similar to that present in lykos.
Distribution, phenology and biology: Knowledge of the distribution of D. lykos is based only on three specimens from three localities, two in the Karasberg region of Namibia and one in the Northern Cape Province of South Africa (Fig. 13). All three specimens were taken on two successive days in August. The holotype was collected with prey: Hymenoptera: Anthophoridae.
Distribution, phenology and biology: D. setithoracicus is fairly widely distributed in the southern parts of South Africa (Fig. 14), occurring in the Northern Cape, Eastern Cape and Free State provinces. The species is active in the adult phase from October through to January, and there is a single record for May (which seems doubtful) (Table 1).
Daspletis stenoura Londt, 1983 Fig. 13 Daspletis stenoura : Londt 1983: 295, figs 22-24. Londt (1983  Note: Female specimens not associated with males are difficult to identify and so a few erroneous identifications may have been made. Distribution, phenology and biology: D. stenoura is fairly widely distributed in the central region of South Africa (Fig. 13), occurring at many places in the northern parts of the Northern Cape Province. Three ^specimens are recorded for Botswana, two from Serowe and one from Takatswane. This species flies at the same time as vulpes and can be found sympatrically. The species is active in the adult phase from November through to March (no records for February) (Table 1) Oldroyd (1974). Although I have not studied these specimens, I accept the identifications as correct.
Distribution, phenology and biology: D. vulpes appears to be fairly widely distributed in the central region of Southern Africa (Fig. 14), records occurring mainly from the northern parts of the Northern Cape Province of South Africa. One male is known from Botswana (Serowe) and a male was recorded by Ricardo (1925) from Livingstone (Zambia), that I have not studied. This species flies at the same time as D. stenoura and can be found sympatrically. It is active in the adult phase from November through to March (no records for February) (Table 1). Natal Museum prey records include (sex of predator in brackets): Orthoptera (2): Acrididae (^), ? Lentulidae (_); Hymenoptera (1): Sphecidae (^).

Taxonomy
Microstylum and Daspletis form a distinctive pair within the Stenopogoninae and are not easily confused with other Afrotropical genera in the subfamily (see key in Londt 1999). Dikow (2009a) established the sister-group relationship between these two genera within the Enigmomorphini based on a world-wide sample of Asilidae species and seven morphological characters (maxillary palpus one-segmented; sensory pit in distal palpomere absent; prementum laterally compressed proximally; labella reduced, fused entirely to prementum; lacinia same height throughout; anatergal setae composed of regular setae only; and superoposterior anepimeron with regular setae only). Unfortunately Dikow's (2009b) 'total evidence' study excluded the Daspletis species as no ethanol preserved specimens were available, so we do not know exactly where it would have been placed in that particular analysis.
Unfortunately Microstylum, with about 80 catalogued Afrotropical species (Oldroyd 1980) has not been adequately revised and so the validity of the characters used to differentiate these genera from one another have not been fully tested (see key in Londt 1999). The four main features used to separate Daspletis from Microstylum were as follows. All Daspletis species possess (1) a facial swelling (gibbosity) that occupies about three-quarters of the face and is entirely setose (i.e. occupied by the mystax); (2) well developed presutural dorsocentral macrosetae; (3) a vein M 1 that is not strongly arched anteriorly; and (4) a postmetacoxal membrane that is covered with long setae. While I have looked at many Microstylum specimens and can testify that most have less developed facial swellings, poorly developed presutural dorsocentral macrosetae, strongly arched M 1 veins, and largely asetose postmetacoxal membranes, I have to report having seen specimens that do not fully conform with all these characters.
As already mentioned above, when comparing vespertilio with other Daspletis species, there are two species groups within Daspletis, each with four species, and these are separated in the key to species that follows. The vulpes species group appears to be quite distinctive while the hermanni species group appears to represent a cluster of species that has more in common with species of Microstylum. What is needed is a detailed cladistic analysis that incorporates a much wider range of species from both genera and other Enigmomorphini in order to better establish the relationships that exist between these taxa.
Key to Daspletis species While this key allows for the identification of both males and females it must be noted that identification of males may require detailed examination of terminalia. Females that are not directly associated with males are sometimes difficult to identify with any certainty.   (Ricardo, 1925), lykos Londt, 1985 12 Abdominal T4 with setulae pointing anteriorly (a few on anterior margin may be differently orientated) ..  Londt, 1983 Distribution The generic distribution appears to be centred in the northern parts of the Northern Cape Province of South Africa where six of the eight species may be found (Figs 13,  14). The two species not found in this general area are however found in adjacent areas setithoracicus primarily in the Eastern Cape Province, and placodes in a fairly limited part of the Limpopo Province of South Africa. At least three species, hirtus, stenoura and vulpes, all members of the vulpes species group, probably have much more extensive distributions that will only be more fully appreciated when more sampling has been undertaken, especially in the generally poorly sampled central parts of Botswana. There does not appear to be any correlation with any of the better known centres of plant endemism (Van Wyk & Smith 2001).

Phenology
The seasonal occurrence of Daspletis species is shown in Table 1. Although more records would clarify the position, it is suspected that the species having dark-winged males (hermanni, lykos and vespertilio) are active in the adult stage during spring and early summer (around August -November). Other species are summer active (around October -May).

Biology
Daspletis is a genus associated mainly with the Nama-Karoo and Savannah biomes of southern Africa (see map of biomes in Van Wyk & Smith 2001). Personal experience suggests that these flies prefer dry, sandy habitats and many specimens have been collected at localities dominated by extensive sand-dune systems, particularly in the Northern Cape Province of South Africa. Personal experience also suggests that all Daspletis rest on open ground (habitat category 1c of Londt 1994), a behaviour that appears to be different from that of the closely related genus Microstylum, whose species tend to rest off the ground on rocks, or the tips of grass culms or bushes (habitat categories 3, 4b & 5b of Londt 1994). Being typical members of the Stenopogoninae, females possess acanthophorites and so oviposit in shallow scrapes in loose sand. Although most of the recorded prey items possess wings, indicating that Daspletis can capture flying prey, at least two items are wingless (an acridid nymph and a spider) which suggests that they can also capture ground-inhabiting prey. Although reproductive behaviour has not been observed, the males of some species may practice aerial displays designed to attract the attention of females (see comments under vespertilio).